From b67b4b117746aef686e527c3205792db0f2c9e16 Mon Sep 17 00:00:00 2001
From: Dominic Curran <dcurran@ti.com>
Date: Mon, 27 Oct 2008 22:30:53 -0400
Subject: Input: gpio-keys - add flag to allow auto repeat

This patch adds a flag to gpio-key driver to turn on the input subsystems
auto repeat feature if needed.

Signed-off-by: Dominic Curran <dcurran@ti.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 include/linux/gpio_keys.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/gpio_keys.h b/include/linux/gpio_keys.h
index ec6ecd74781d..1289fa7623ca 100644
--- a/include/linux/gpio_keys.h
+++ b/include/linux/gpio_keys.h
@@ -15,6 +15,7 @@ struct gpio_keys_button {
 struct gpio_keys_platform_data {
 	struct gpio_keys_button *buttons;
 	int nbuttons;
+	unsigned int rep:1;		/* enable input subsystem auto repeat */
 };
 
 #endif
-- 
cgit v1.2.3


From cae1c11414912bf77a62aebd65ced321f0b9da51 Mon Sep 17 00:00:00 2001
From: David Vrabel <david.vrabel@csr.com>
Date: Mon, 27 Oct 2008 15:22:46 +0000
Subject: uwb: reference count reservations

Reference counting the struct uwb_rsv's is safer and easier to get right than
the transferring ownership of the structures from the PAL to reservation
manager.

This fixes an oops in the debug PAL after a reservation timed out.

Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/usb/wusbcore/reservation.c |  7 +++---
 drivers/uwb/rsv.c                  | 48 +++++++++++++++++++++++++-------------
 drivers/uwb/uwb-debug.c            | 13 ++++++++---
 include/linux/uwb.h                |  1 +
 4 files changed, 46 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/wusbcore/reservation.c b/drivers/usb/wusbcore/reservation.c
index fc63e77ded2d..7b6525dac2f1 100644
--- a/drivers/usb/wusbcore/reservation.c
+++ b/drivers/usb/wusbcore/reservation.c
@@ -59,7 +59,6 @@ static void wusbhc_rsv_complete_cb(struct uwb_rsv *rsv)
 	case UWB_RSV_STATE_NONE:
 		dev_dbg(dev, "removed reservation\n");
 		wusbhc_bwa_set(wusbhc, 0, NULL);
-		wusbhc->rsv = NULL;
 		break;
 	default:
 		dev_dbg(dev, "unexpected reservation state: %d\n", rsv->state);
@@ -105,11 +104,11 @@ int wusbhc_rsv_establish(struct wusbhc *wusbhc)
 
 
 /**
- * wusbhc_rsv_terminate - terminate any cluster reservation
+ * wusbhc_rsv_terminate - terminate the cluster reservation
  * @wusbhc: the WUSB host whose reservation is to be terminated
  */
 void wusbhc_rsv_terminate(struct wusbhc *wusbhc)
 {
-	if (wusbhc->rsv)
-		uwb_rsv_terminate(wusbhc->rsv);
+	uwb_rsv_terminate(wusbhc->rsv);
+	uwb_rsv_destroy(wusbhc->rsv);
 }
diff --git a/drivers/uwb/rsv.c b/drivers/uwb/rsv.c
index e4facae46e0d..bcc41a4a6606 100644
--- a/drivers/uwb/rsv.c
+++ b/drivers/uwb/rsv.c
@@ -82,6 +82,23 @@ static void uwb_rsv_dump(struct uwb_rsv *rsv)
 	dev_dbg(dev, "rsv %s -> %s: %s\n", owner, target, uwb_rsv_state_str(rsv->state));
 }
 
+static void uwb_rsv_release(struct kref *kref)
+{
+	struct uwb_rsv *rsv = container_of(kref, struct uwb_rsv, kref);
+
+	kfree(rsv);
+}
+
+static void uwb_rsv_get(struct uwb_rsv *rsv)
+{
+	kref_get(&rsv->kref);
+}
+
+static void uwb_rsv_put(struct uwb_rsv *rsv)
+{
+	kref_put(&rsv->kref, uwb_rsv_release);
+}
+
 /*
  * Get a free stream index for a reservation.
  *
@@ -325,6 +342,7 @@ static struct uwb_rsv *uwb_rsv_alloc(struct uwb_rc *rc)
 
 	INIT_LIST_HEAD(&rsv->rc_node);
 	INIT_LIST_HEAD(&rsv->pal_node);
+	kref_init(&rsv->kref);
 	init_timer(&rsv->timer);
 	rsv->timer.function = uwb_rsv_timer;
 	rsv->timer.data     = (unsigned long)rsv;
@@ -334,14 +352,6 @@ static struct uwb_rsv *uwb_rsv_alloc(struct uwb_rc *rc)
 	return rsv;
 }
 
-static void uwb_rsv_free(struct uwb_rsv *rsv)
-{
-	uwb_dev_put(rsv->owner);
-	if (rsv->target.type == UWB_RSV_TARGET_DEV)
-		uwb_dev_put(rsv->target.dev);
-	kfree(rsv);
-}
-
 /**
  * uwb_rsv_create - allocate and initialize a UWB reservation structure
  * @rc: the radio controller
@@ -375,23 +385,23 @@ void uwb_rsv_remove(struct uwb_rsv *rsv)
 	if (rsv->state != UWB_RSV_STATE_NONE)
 		uwb_rsv_set_state(rsv, UWB_RSV_STATE_NONE);
 	del_timer_sync(&rsv->timer);
-	list_del(&rsv->rc_node);
-	uwb_rsv_free(rsv);
+	uwb_dev_put(rsv->owner);
+	if (rsv->target.type == UWB_RSV_TARGET_DEV)
+		uwb_dev_put(rsv->target.dev);
+
+	list_del_init(&rsv->rc_node);
+	uwb_rsv_put(rsv);
 }
 
 /**
  * uwb_rsv_destroy - free a UWB reservation structure
  * @rsv: the reservation to free
  *
- * The reservation will be terminated if it is pending or established.
+ * The reservation must already be terminated.
  */
 void uwb_rsv_destroy(struct uwb_rsv *rsv)
 {
-	struct uwb_rc *rc = rsv->rc;
-
-	mutex_lock(&rc->rsvs_mutex);
-	uwb_rsv_remove(rsv);
-	mutex_unlock(&rc->rsvs_mutex);
+	uwb_rsv_put(rsv);
 }
 EXPORT_SYMBOL_GPL(uwb_rsv_destroy);
 
@@ -423,6 +433,7 @@ int uwb_rsv_establish(struct uwb_rsv *rsv)
 		goto out;
 	}
 
+	uwb_rsv_get(rsv);
 	list_add_tail(&rsv->rc_node, &rc->reservations);
 	rsv->owner = &rc->uwb_dev;
 	uwb_dev_get(rsv->owner);
@@ -478,9 +489,14 @@ EXPORT_SYMBOL_GPL(uwb_rsv_terminate);
  *
  * Reservation requests from peers are denied unless a PAL accepts it
  * by calling this function.
+ *
+ * The PAL call uwb_rsv_destroy() for all accepted reservations before
+ * calling uwb_pal_unregister().
  */
 void uwb_rsv_accept(struct uwb_rsv *rsv, uwb_rsv_cb_f cb, void *pal_priv)
 {
+	uwb_rsv_get(rsv);
+
 	rsv->callback = cb;
 	rsv->pal_priv = pal_priv;
 	rsv->state    = UWB_RSV_STATE_T_ACCEPTED;
diff --git a/drivers/uwb/uwb-debug.c b/drivers/uwb/uwb-debug.c
index 6d232c35d07d..6db641e45313 100644
--- a/drivers/uwb/uwb-debug.c
+++ b/drivers/uwb/uwb-debug.c
@@ -104,6 +104,11 @@ static void uwb_dbg_rsv_cb(struct uwb_rsv *rsv)
 
 	dev_dbg(dev, "debug: rsv %s -> %s: %s\n",
 		owner, target, uwb_rsv_state_str(rsv->state));
+
+	if (rsv->state == UWB_RSV_STATE_NONE) {
+		list_del(&rsv->pal_node);
+		uwb_rsv_destroy(rsv);
+	}
 }
 
 static int cmd_rsv_establish(struct uwb_rc *rc,
@@ -153,11 +158,11 @@ static int cmd_rsv_terminate(struct uwb_rc *rc,
 			found = rsv;
 			break;
 		}
+		i++;
 	}
 	if (!found)
 		return -EINVAL;
 
-	list_del(&found->pal_node);
 	uwb_rsv_terminate(found);
 
 	return 0;
@@ -287,8 +292,10 @@ static void uwb_dbg_new_rsv(struct uwb_rsv *rsv)
 {
 	struct uwb_rc *rc = rsv->rc;
 
-	if (rc->dbg->accept)
+	if (rc->dbg->accept) {
+		list_add_tail(&rsv->pal_node, &rc->dbg->rsvs);
 		uwb_rsv_accept(rsv, uwb_dbg_rsv_cb, NULL);
+	}
 }
 
 /**
@@ -336,7 +343,7 @@ void uwb_dbg_del_rc(struct uwb_rc *rc)
 		return;
 
 	list_for_each_entry_safe(rsv, t, &rc->dbg->rsvs, pal_node) {
-		uwb_rsv_destroy(rsv);
+		uwb_rsv_terminate(rsv);
 	}
 
 	uwb_pal_unregister(rc, &rc->dbg->pal);
diff --git a/include/linux/uwb.h b/include/linux/uwb.h
index f9ccbd9a2ced..010ee708304d 100644
--- a/include/linux/uwb.h
+++ b/include/linux/uwb.h
@@ -201,6 +201,7 @@ struct uwb_rsv {
 	struct uwb_rc *rc;
 	struct list_head rc_node;
 	struct list_head pal_node;
+	struct kref kref;
 
 	struct uwb_dev *owner;
 	struct uwb_rsv_target target;
-- 
cgit v1.2.3


From 4d2bea4ca0adb4cebfbf89d34869c74081c42577 Mon Sep 17 00:00:00 2001
From: David Vrabel <david.vrabel@csr.com>
Date: Mon, 27 Oct 2008 15:42:31 +0000
Subject: wusb: do a proper channel stop

When stopping the WUSB channel the host should send Channel Stop IEs giving
the WUSB Channel Time of the last MMC.  Both WHCI and HWA hosts provide a
channel stop command for this.

Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/usb/host/hwa-hc.c       | 102 +++++++++++++++++++---------------------
 drivers/usb/host/whci/whcd.h    |   2 +-
 drivers/usb/host/whci/whci-hc.h |   2 +
 drivers/usb/host/whci/wusb.c    |  15 ++++--
 drivers/usb/wusbcore/mmc.c      |   8 +---
 drivers/usb/wusbcore/wusbhc.h   |  24 ++++++----
 include/linux/usb/wusb-wa.h     |   1 +
 7 files changed, 80 insertions(+), 74 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/host/hwa-hc.c b/drivers/usb/host/hwa-hc.c
index 64be4d88df11..0e18989e1658 100644
--- a/drivers/usb/host/hwa-hc.c
+++ b/drivers/usb/host/hwa-hc.c
@@ -171,11 +171,6 @@ static int hwahc_op_start(struct usb_hcd *usb_hcd)
 	if (result < 0)
 		goto error_set_cluster_id;
 
-	result = wa_nep_arm(&hwahc->wa, GFP_KERNEL);
-	if (result < 0) {
-		dev_err(dev, "cannot listen to notifications: %d\n", result);
-		goto error_stop;
-	}
 	usb_hcd->uses_new_polling = 1;
 	usb_hcd->poll_rh = 1;
 	usb_hcd->state = HC_STATE_RUNNING;
@@ -185,8 +180,6 @@ out:
 	d_fnend(4, dev, "(hwahc %p) = %d\n", hwahc, result);
 	return result;
 
-error_stop:
-	__wa_stop(&hwahc->wa);
 error_set_cluster_id:
 	wusb_cluster_id_put(wusbhc->cluster_id);
 error_cluster_id_get:
@@ -194,39 +187,6 @@ error_cluster_id_get:
 
 }
 
-/*
- * FIXME: break this function up
- */
-static int __hwahc_op_wusbhc_start(struct wusbhc *wusbhc)
-{
-	int result;
-	struct hwahc *hwahc = container_of(wusbhc, struct hwahc, wusbhc);
-	struct device *dev = &hwahc->wa.usb_iface->dev;
-
-	/* Set up a Host Info WUSB Information Element */
-	d_fnstart(4, dev, "(hwahc %p)\n", hwahc);
-	result = -ENOSPC;
-
-	result = __wa_set_feature(&hwahc->wa, WA_ENABLE);
-	if (result < 0) {
-		dev_err(dev, "error commanding HC to start: %d\n", result);
-		goto error_stop;
-	}
-	result = __wa_wait_status(&hwahc->wa, WA_ENABLE, WA_ENABLE);
-	if (result < 0) {
-		dev_err(dev, "error waiting for HC to start: %d\n", result);
-		goto error_stop;
-	}
-	result = 0;
-out:
-	d_fnend(4, dev, "(hwahc %p) = %d\n", hwahc, result);
-	return result;
-
-error_stop:
-	result = __wa_clear_feature(&hwahc->wa, WA_ENABLE);
-	goto out;
-}
-
 static int hwahc_op_suspend(struct usb_hcd *usb_hcd, pm_message_t msg)
 {
 	struct wusbhc *wusbhc = usb_hcd_to_wusbhc(usb_hcd);
@@ -246,18 +206,6 @@ static int hwahc_op_resume(struct usb_hcd *usb_hcd)
 	return -ENOSYS;
 }
 
-static void __hwahc_op_wusbhc_stop(struct wusbhc *wusbhc)
-{
-	int result;
-	struct hwahc *hwahc = container_of(wusbhc, struct hwahc, wusbhc);
-	struct device *dev = &hwahc->wa.usb_iface->dev;
-
-	d_fnstart(4, dev, "(hwahc %p)\n", hwahc);
-	/* Nothing for now */
-	d_fnend(4, dev, "(hwahc %p) = %d\n", hwahc, result);
-	return;
-}
-
 /*
  * No need to abort pipes, as when this is called, all the children
  * has been disconnected and that has done it [through
@@ -275,8 +223,6 @@ static void hwahc_op_stop(struct usb_hcd *usb_hcd)
 	d_fnstart(4, dev, "(hwahc %p)\n", hwahc);
 	mutex_lock(&wusbhc->mutex);
 	wusbhc_stop(wusbhc);
-	wa_nep_disarm(&hwahc->wa);
-	result = __wa_stop(&hwahc->wa);
 	wusb_cluster_id_put(wusbhc->cluster_id);
 	mutex_unlock(&wusbhc->mutex);
 	d_fnend(4, dev, "(hwahc %p) = %d\n", hwahc, result);
@@ -325,6 +271,54 @@ static void hwahc_op_endpoint_disable(struct usb_hcd *usb_hcd,
 	rpipe_ep_disable(&hwahc->wa, ep);
 }
 
+static int __hwahc_op_wusbhc_start(struct wusbhc *wusbhc)
+{
+	int result;
+	struct hwahc *hwahc = container_of(wusbhc, struct hwahc, wusbhc);
+	struct device *dev = &hwahc->wa.usb_iface->dev;
+
+	result = __wa_set_feature(&hwahc->wa, WA_ENABLE);
+	if (result < 0) {
+		dev_err(dev, "error commanding HC to start: %d\n", result);
+		goto error_stop;
+	}
+	result = __wa_wait_status(&hwahc->wa, WA_ENABLE, WA_ENABLE);
+	if (result < 0) {
+		dev_err(dev, "error waiting for HC to start: %d\n", result);
+		goto error_stop;
+	}
+	result = wa_nep_arm(&hwahc->wa, GFP_KERNEL);
+	if (result < 0) {
+		dev_err(dev, "cannot listen to notifications: %d\n", result);
+		goto error_stop;
+	}
+	return result;
+
+error_stop:
+	__wa_clear_feature(&hwahc->wa, WA_ENABLE);
+	return result;
+}
+
+static void __hwahc_op_wusbhc_stop(struct wusbhc *wusbhc, int delay)
+{
+	struct hwahc *hwahc = container_of(wusbhc, struct hwahc, wusbhc);
+	struct wahc *wa = &hwahc->wa;
+	u8 iface_no = wa->usb_iface->cur_altsetting->desc.bInterfaceNumber;
+	int ret;
+
+	ret = usb_control_msg(wa->usb_dev, usb_sndctrlpipe(wa->usb_dev, 0),
+			      WUSB_REQ_CHAN_STOP,
+			      USB_DIR_OUT | USB_TYPE_CLASS | USB_RECIP_INTERFACE,
+			      delay * 1000,
+			      iface_no,
+			      NULL, 0, 1000 /* FIXME: arbitrary */);
+	if (ret == 0)
+		msleep(delay);
+
+	wa_nep_disarm(&hwahc->wa);
+	__wa_stop(&hwahc->wa);
+}
+
 /*
  * Set the UWB MAS allocation for the WUSB cluster
  *
diff --git a/drivers/usb/host/whci/whcd.h b/drivers/usb/host/whci/whcd.h
index 1d2a53bd39fd..1bbb8cb6bf80 100644
--- a/drivers/usb/host/whci/whcd.h
+++ b/drivers/usb/host/whci/whcd.h
@@ -136,7 +136,7 @@ int whc_do_gencmd(struct whc *whc, u32 cmd, u32 params, void *addr, size_t len);
 
 /* wusb.c */
 int whc_wusbhc_start(struct wusbhc *wusbhc);
-void whc_wusbhc_stop(struct wusbhc *wusbhc);
+void whc_wusbhc_stop(struct wusbhc *wusbhc, int delay);
 int whc_mmcie_add(struct wusbhc *wusbhc, u8 interval, u8 repeat_cnt,
 		  u8 handle, struct wuie_hdr *wuie);
 int whc_mmcie_rm(struct wusbhc *wusbhc, u8 handle);
diff --git a/drivers/usb/host/whci/whci-hc.h b/drivers/usb/host/whci/whci-hc.h
index bff1eb7a35cf..51df7e313b38 100644
--- a/drivers/usb/host/whci/whci-hc.h
+++ b/drivers/usb/host/whci/whci-hc.h
@@ -410,6 +410,8 @@ struct dn_buf_entry {
 #  define WUSBDNTSCTRL_SLOTS(s)    ((s) << 0)
 
 #define WUSBTIME             0x68
+#  define WUSBTIME_CHANNEL_TIME_MASK 0x00ffffff
+
 #define WUSBBPST             0x6c
 #define WUSBDIBUPDATED       0x70
 
diff --git a/drivers/usb/host/whci/wusb.c b/drivers/usb/host/whci/wusb.c
index 66e4ddcd961d..2befd475def4 100644
--- a/drivers/usb/host/whci/wusb.c
+++ b/drivers/usb/host/whci/wusb.c
@@ -64,8 +64,9 @@ static int whc_update_di(struct whc *whc, int idx)
 }
 
 /*
- * WHCI starts and stops MMCs based on there being a valid GTK so
- * these need only start/stop the asynchronous and periodic schedules.
+ * WHCI starts MMCs based on there being a valid GTK so these need
+ * only start/stop the asynchronous and periodic schedules and send a
+ * channel stop command.
  */
 
 int whc_wusbhc_start(struct wusbhc *wusbhc)
@@ -78,12 +79,20 @@ int whc_wusbhc_start(struct wusbhc *wusbhc)
 	return 0;
 }
 
-void whc_wusbhc_stop(struct wusbhc *wusbhc)
+void whc_wusbhc_stop(struct wusbhc *wusbhc, int delay)
 {
 	struct whc *whc = wusbhc_to_whc(wusbhc);
+	u32 stop_time, now_time;
+	int ret;
 
 	pzl_stop(whc);
 	asl_stop(whc);
+
+	now_time = le_readl(whc->base + WUSBTIME) & WUSBTIME_CHANNEL_TIME_MASK;
+	stop_time = (now_time + ((delay * 8) << 7)) & 0x00ffffff;
+	ret = whc_do_gencmd(whc, WUSBGENCMDSTS_CHAN_STOP, stop_time, NULL, 0);
+	if (ret == 0)
+		msleep(delay);
 }
 
 int whc_mmcie_add(struct wusbhc *wusbhc, u8 interval, u8 repeat_cnt,
diff --git a/drivers/usb/wusbcore/mmc.c b/drivers/usb/wusbcore/mmc.c
index cfa77a01cebd..af2aee0fdffa 100644
--- a/drivers/usb/wusbcore/mmc.c
+++ b/drivers/usb/wusbcore/mmc.c
@@ -250,18 +250,14 @@ error_alloc:
  * wusbhc_stop - stop transmitting MMCs
  * @wusbhc: the HC to stop
  *
- * Send a Host Disconnect IE, wait, remove all the MMCs (stop sending MMCs).
- *
- * If we can't allocate a Host Stop IE, screw it, we don't notify the
- * devices we are disconnecting...
+ * Stops the WUSB channel and removes the cluster reservation.
  */
 void wusbhc_stop(struct wusbhc *wusbhc)
 {
 	if (wusbhc->active) {
 		wusbhc->active = 0;
-		wusbhc->stop(wusbhc);
+		wusbhc->stop(wusbhc, WUSB_CHANNEL_STOP_DELAY_MS);
 		wusbhc_sec_stop(wusbhc);
-		__wusbhc_host_disconnect_ie(wusbhc);
 		wusbhc_devconnect_stop(wusbhc);
 		wusbhc_rsv_terminate(wusbhc);
 	}
diff --git a/drivers/usb/wusbcore/wusbhc.h b/drivers/usb/wusbcore/wusbhc.h
index d0c132434f1b..b9bdf5a5f11b 100644
--- a/drivers/usb/wusbcore/wusbhc.h
+++ b/drivers/usb/wusbcore/wusbhc.h
@@ -64,6 +64,13 @@
 #include <linux/uwb.h>
 #include <linux/usb/wusb.h>
 
+/*
+ * Time from a WUSB channel stop request to the last transmitted MMC.
+ *
+ * This needs to be > 4.096 ms in case no MMCs can be transmitted in
+ * zone 0.
+ */
+#define WUSB_CHANNEL_STOP_DELAY_MS 8
 
 /**
  * Wireless USB device
@@ -198,21 +205,18 @@ struct wusb_port {
  * @mmcies_max	   Max number of Information Elements this HC can send
  *                 in its MMC. Read-only.
  *
+ * @start          Start the WUSB channel.
+ *
+ * @stop           Stop the WUSB channel after the specified number of
+ *                 milliseconds.  Channel Stop IEs should be transmitted
+ *                 as required by [WUSB] 4.16.2.1.
+ *
  * @mmcie_add	   HC specific operation (WHCI or HWA) for adding an
  *                 MMCIE.
  *
  * @mmcie_rm	   HC specific operation (WHCI or HWA) for removing an
  *                 MMCIE.
  *
- * @enc_types	   Array which describes the encryptions methods
- *                 supported by the host as described in WUSB1.0 --
- *                 one entry per supported method. As of WUSB1.0 there
- *                 is only four methods, we make space for eight just in
- *                 case they decide to add some more (and pray they do
- *                 it in sequential order). if 'enc_types[enc_method]
- *                 != 0', then it is supported by the host. enc_method
- *                 is USB_ENC_TYPE*.
- *
  * @set_ptk:       Set the PTK and enable encryption for a device. Or, if
  *                 the supplied key is NULL, disable encryption for that
  *                 device.
@@ -269,7 +273,7 @@ struct wusbhc {
 	u8 mmcies_max;
 	/* FIXME: make wusbhc_ops? */
 	int (*start)(struct wusbhc *wusbhc);
-	void (*stop)(struct wusbhc *wusbhc);
+	void (*stop)(struct wusbhc *wusbhc, int delay);
 	int (*mmcie_add)(struct wusbhc *wusbhc, u8 interval, u8 repeat_cnt,
 			 u8 handle, struct wuie_hdr *wuie);
 	int (*mmcie_rm)(struct wusbhc *wusbhc, u8 handle);
diff --git a/include/linux/usb/wusb-wa.h b/include/linux/usb/wusb-wa.h
index a102561e7026..fb7c359bdfba 100644
--- a/include/linux/usb/wusb-wa.h
+++ b/include/linux/usb/wusb-wa.h
@@ -51,6 +51,7 @@ enum {
 	WUSB_REQ_GET_TIME       = 25,
 	WUSB_REQ_SET_STREAM_IDX = 26,
 	WUSB_REQ_SET_WUSB_MAS   = 27,
+	WUSB_REQ_CHAN_STOP      = 28,
 };
 
 
-- 
cgit v1.2.3


From 1cde7f68ced8d10a20dd2370e9d1d22ab3c1ea5c Mon Sep 17 00:00:00 2001
From: David Vrabel <david.vrabel@csr.com>
Date: Mon, 27 Oct 2008 16:48:09 +0000
Subject: uwb: order IEs by element ID

ECMA-368 requires that IEs in a beacon must be sorted by element ID.  Most
hardware uses the ordering in the Set IE URC command so get the ordering
right on the host.

Also refactor the IE management code:
  - use uwb_ie_next() instead of uwb_ie_for_each().
  - remove unnecessary functions.
  - API is now only uwb_rc_ie_add() and uwb_rc_ie_rm().

Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/uwb/beacon.c           |  24 +--
 drivers/uwb/ie.c               | 463 +++++++++++++----------------------------
 drivers/uwb/lc-rc.c            |  25 ---
 drivers/uwb/uwb-internal.h     |  16 +-
 drivers/uwb/wlp/wlp-internal.h |   4 -
 include/linux/uwb.h            |  18 +-
 6 files changed, 172 insertions(+), 378 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/uwb/beacon.c b/drivers/uwb/beacon.c
index 46b18eec5026..ad823987cede 100644
--- a/drivers/uwb/beacon.c
+++ b/drivers/uwb/beacon.c
@@ -349,22 +349,22 @@ ssize_t uwb_bce_print_IEs(struct uwb_dev *uwb_dev, struct uwb_beca_e *bce,
 	ssize_t result = 0;
 	struct uwb_rc_evt_beacon *be;
 	struct uwb_beacon_frame *bf;
-	struct uwb_buf_ctx ctx = {
-		.buf = buf,
-		.bytes = 0,
-		.size = size
-	};
+	int ies_len;
+	struct uwb_ie_hdr *ies;
 
 	mutex_lock(&bce->mutex);
+
 	be = bce->be;
-	if (be == NULL)
-		goto out;
-	bf = (void *) be->BeaconInfo;
-	uwb_ie_for_each(uwb_dev, uwb_ie_dump_hex, &ctx,
-			bf->IEData, be->wBeaconInfoLength - sizeof(*bf));
-	result = ctx.bytes;
-out:
+	if (be) {
+		bf = (struct uwb_beacon_frame *)bce->be->BeaconInfo;
+		ies_len = be->wBeaconInfoLength - sizeof(struct uwb_beacon_frame);
+		ies = (struct uwb_ie_hdr *)bf->IEData;
+
+		result = uwb_ie_dump_hex(ies, ies_len, buf, size);
+	}
+
 	mutex_unlock(&bce->mutex);
+
 	return result;
 }
 
diff --git a/drivers/uwb/ie.c b/drivers/uwb/ie.c
index cf6f3d152b9d..ab976686175b 100644
--- a/drivers/uwb/ie.c
+++ b/drivers/uwb/ie.c
@@ -25,8 +25,6 @@
  */
 
 #include "uwb-internal.h"
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
 
 /**
  * uwb_ie_next - get the next IE in a buffer
@@ -60,6 +58,42 @@ struct uwb_ie_hdr *uwb_ie_next(void **ptr, size_t *len)
 }
 EXPORT_SYMBOL_GPL(uwb_ie_next);
 
+/**
+ * uwb_ie_dump_hex - print IEs to a character buffer
+ * @ies: the IEs to print.
+ * @len: length of all the IEs.
+ * @buf: the destination buffer.
+ * @size: size of @buf.
+ *
+ * Returns the number of characters written.
+ */
+int uwb_ie_dump_hex(const struct uwb_ie_hdr *ies, size_t len,
+		    char *buf, size_t size)
+{
+	void *ptr;
+	const struct uwb_ie_hdr *ie;
+	int r = 0;
+	u8 *d;
+
+	ptr = (void *)ies;
+	for (;;) {
+		ie = uwb_ie_next(&ptr, &len);
+		if (!ie)
+			break;
+
+		r += scnprintf(buf + r, size - r, "%02x %02x",
+			       (unsigned)ie->element_id,
+			       (unsigned)ie->length);
+		d = (uint8_t *)ie + sizeof(struct uwb_ie_hdr);
+		while (d != ptr && r < size)
+			r += scnprintf(buf + r, size - r, " %02x", (unsigned)*d++);
+		if (r < size)
+			buf[r++] = '\n';
+	};
+
+	return r;
+}
+
 /**
  * Get the IEs that a radio controller is sending in its beacon
  *
@@ -70,6 +104,7 @@ EXPORT_SYMBOL_GPL(uwb_ie_next);
  * anything. Once done with the iedata buffer, call
  * uwb_rc_ie_release(iedata). Don't call kfree on it.
  */
+static
 ssize_t uwb_rc_get_ie(struct uwb_rc *uwb_rc, struct uwb_rc_evt_get_ie **pget_ie)
 {
 	ssize_t result;
@@ -78,148 +113,35 @@ ssize_t uwb_rc_get_ie(struct uwb_rc *uwb_rc, struct uwb_rc_evt_get_ie **pget_ie)
 	struct uwb_rceb *reply = NULL;
 	struct uwb_rc_evt_get_ie *get_ie;
 
-	d_fnstart(3, dev, "(%p, %p)\n", uwb_rc, pget_ie);
-	result = -ENOMEM;
 	cmd = kzalloc(sizeof(*cmd), GFP_KERNEL);
 	if (cmd == NULL)
-		goto error_kzalloc;
+		return -ENOMEM;
+
 	cmd->bCommandType = UWB_RC_CET_GENERAL;
 	cmd->wCommand = cpu_to_le16(UWB_RC_CMD_GET_IE);
 	result = uwb_rc_vcmd(uwb_rc, "GET_IE", cmd, sizeof(*cmd),
 			     UWB_RC_CET_GENERAL, UWB_RC_CMD_GET_IE,
 			     &reply);
+	kfree(cmd);
 	if (result < 0)
-		goto error_cmd;
+		return result;
+
 	get_ie = container_of(reply, struct uwb_rc_evt_get_ie, rceb);
 	if (result < sizeof(*get_ie)) {
 		dev_err(dev, "not enough data returned for decoding GET IE "
 			"(%zu bytes received vs %zu needed)\n",
 			result, sizeof(*get_ie));
-		result = -EINVAL;
+		return -EINVAL;
 	} else if (result < sizeof(*get_ie) + le16_to_cpu(get_ie->wIELength)) {
 		dev_err(dev, "not enough data returned for decoding GET IE "
 			"payload (%zu bytes received vs %zu needed)\n", result,
 			sizeof(*get_ie) + le16_to_cpu(get_ie->wIELength));
-		result = -EINVAL;
-	} else
-		*pget_ie = get_ie;
-error_cmd:
-	kfree(cmd);
-error_kzalloc:
-	d_fnend(3, dev, "(%p, %p) = %d\n", uwb_rc, pget_ie, (int)result);
-	return result;
-}
-EXPORT_SYMBOL_GPL(uwb_rc_get_ie);
-
-
-/*
- * Given a pointer to an IE, print it in ASCII/hex followed by a new line
- *
- * @ie_hdr: pointer to the IE header. Length is in there, and it is
- *          guaranteed that the ie_hdr->length bytes following it are
- *          safely accesible.
- *
- * @_data: context data passed from uwb_ie_for_each(), an struct output_ctx
- */
-int uwb_ie_dump_hex(struct uwb_dev *uwb_dev, const struct uwb_ie_hdr *ie_hdr,
-		    size_t offset, void *_ctx)
-{
-	struct uwb_buf_ctx *ctx = _ctx;
-	const u8 *pl = (void *)(ie_hdr + 1);
-	u8 pl_itr;
-
-	ctx->bytes += scnprintf(ctx->buf + ctx->bytes, ctx->size - ctx->bytes,
-				"%02x %02x ", (unsigned) ie_hdr->element_id,
-				(unsigned) ie_hdr->length);
-	pl_itr = 0;
-	while (pl_itr < ie_hdr->length && ctx->bytes < ctx->size)
-		ctx->bytes += scnprintf(ctx->buf + ctx->bytes,
-					ctx->size - ctx->bytes,
-					"%02x ", (unsigned) pl[pl_itr++]);
-	if (ctx->bytes < ctx->size)
-		ctx->buf[ctx->bytes++] = '\n';
-	return 0;
-}
-EXPORT_SYMBOL_GPL(uwb_ie_dump_hex);
-
-
-/**
- * Verify that a pointer in a buffer points to valid IE
- *
- * @start: pointer to start of buffer in which IE appears
- * @itr:   pointer to IE inside buffer that will be verified
- * @top:   pointer to end of buffer
- *
- * @returns: 0 if IE is valid, <0 otherwise
- *
- * Verification involves checking that the buffer can contain a
- * header and the amount of data reported in the IE header can be found in
- * the buffer.
- */
-static
-int uwb_rc_ie_verify(struct uwb_dev *uwb_dev, const void *start,
-		     const void *itr, const void *top)
-{
-	struct device *dev = &uwb_dev->dev;
-	const struct uwb_ie_hdr *ie_hdr;
-
-	if (top - itr < sizeof(*ie_hdr)) {
-		dev_err(dev, "Bad IE: no data to decode header "
-			"(%zu bytes left vs %zu needed) at offset %zu\n",
-			top - itr, sizeof(*ie_hdr), itr - start);
-		return -EINVAL;
-	}
-	ie_hdr = itr;
-	itr += sizeof(*ie_hdr);
-	if (top - itr < ie_hdr->length) {
-		dev_err(dev, "Bad IE: not enough data for payload "
-			"(%zu bytes left vs %zu needed) at offset %zu\n",
-			top - itr, (size_t)ie_hdr->length,
-			(void *)ie_hdr - start);
 		return -EINVAL;
 	}
-	return 0;
-}
 
-
-/**
- * Walk a buffer filled with consecutive IE's a buffer
- *
- * @uwb_dev: UWB device this IEs belong to (for err messages mainly)
- *
- * @fn: function to call with each IE; if it returns 0, we keep
- *      traversing the buffer. If it returns !0, we'll stop and return
- *      that value.
- *
- * @data: pointer passed to @fn
- *
- * @buf: buffer where the consecutive IEs are located
- *
- * @size: size of @buf
- *
- * Each IE is checked for basic correctness (there is space left for
- * the header and the payload). If that test is failed, we stop
- * processing. For every good IE, @fn is called.
- */
-ssize_t uwb_ie_for_each(struct uwb_dev *uwb_dev, uwb_ie_f fn, void *data,
-			const void *buf, size_t size)
-{
-	ssize_t result = 0;
-	const struct uwb_ie_hdr *ie_hdr;
-	const void *itr = buf, *top = itr + size;
-
-	while (itr < top) {
-		if (uwb_rc_ie_verify(uwb_dev, buf, itr, top) != 0)
-			break;
-		ie_hdr = itr;
-		itr += sizeof(*ie_hdr) + ie_hdr->length;
-		result = fn(uwb_dev, ie_hdr, itr - buf, data);
-		if (result != 0)
-			break;
-	}
+	*pget_ie = get_ie;
 	return result;
 }
-EXPORT_SYMBOL_GPL(uwb_ie_for_each);
 
 
 /**
@@ -256,70 +178,6 @@ error_cmd:
 	return result;
 }
 
-/**
- * Determine by IE id if IE is host settable
- * WUSB 1.0 [8.6.2.8 Table 8.85]
- *
- * EXCEPTION:
- * All but UWB_IE_WLP appears in Table 8.85 from WUSB 1.0. Setting this IE
- * is required for the WLP substack to perform association with its WSS so
- * we hope that the WUSB spec will be changed to reflect this.
- */
-static
-int uwb_rc_ie_is_host_settable(enum uwb_ie element_id)
-{
-	if (element_id == UWB_PCA_AVAILABILITY ||
-	    element_id == UWB_BP_SWITCH_IE ||
-	    element_id == UWB_MAC_CAPABILITIES_IE ||
-	    element_id == UWB_PHY_CAPABILITIES_IE ||
-	    element_id == UWB_APP_SPEC_PROBE_IE ||
-	    element_id == UWB_IDENTIFICATION_IE ||
-	    element_id == UWB_MASTER_KEY_ID_IE ||
-	    element_id == UWB_IE_WLP ||
-	    element_id == UWB_APP_SPEC_IE)
-		return 1;
-	return 0;
-}
-
-
-/**
- * Extract Host Settable IEs from IE
- *
- * @ie_data: pointer to buffer containing all IEs
- * @size:    size of buffer
- *
- * @returns: length of buffer that only includes host settable IEs
- *
- * Given a buffer of IEs we move all Host Settable IEs to front of buffer
- * by overwriting the IEs that are not Host Settable.
- * Buffer length is adjusted accordingly.
- */
-static
-ssize_t uwb_rc_parse_host_settable_ie(struct uwb_dev *uwb_dev,
-				      void *ie_data, size_t size)
-{
-	size_t new_len = size;
-	struct uwb_ie_hdr *ie_hdr;
-	size_t ie_length;
-	void *itr = ie_data, *top = itr + size;
-
-	while (itr < top) {
-		if (uwb_rc_ie_verify(uwb_dev, ie_data, itr, top) != 0)
-			break;
-		ie_hdr = itr;
-		ie_length = sizeof(*ie_hdr) + ie_hdr->length;
-		if (uwb_rc_ie_is_host_settable(ie_hdr->element_id)) {
-			itr += ie_length;
-		} else {
-			memmove(itr, itr + ie_length, top - (itr + ie_length));
-			new_len -= ie_length;
-			top -= ie_length;
-		}
-	}
-	return new_len;
-}
-
-
 /* Cleanup the whole IE management subsystem */
 void uwb_rc_ie_init(struct uwb_rc *uwb_rc)
 {
@@ -328,49 +186,34 @@ void uwb_rc_ie_init(struct uwb_rc *uwb_rc)
 
 
 /**
- * Set up cache for host settable IEs currently being transmitted
+ * uwb_rc_ie_setup - setup a radio controller's IE manager
+ * @uwb_rc: the radio controller.
  *
- * First we just call GET-IE to get the current IEs being transmitted
- * (or we workaround and pretend we did) and (because the format is
- * the same) reuse that as the IE cache (with the command prefix, as
- * explained in 'struct uwb_rc').
+ * The current set of IEs are obtained from the hardware with a GET-IE
+ * command (since the radio controller is not yet beaconing this will
+ * be just the hardware's MAC and PHY Capability IEs).
  *
- * @returns: size of cache created
+ * Returns 0 on success; -ve on an error.
  */
-ssize_t uwb_rc_ie_setup(struct uwb_rc *uwb_rc)
+int uwb_rc_ie_setup(struct uwb_rc *uwb_rc)
 {
-	struct device *dev = &uwb_rc->uwb_dev.dev;
-	ssize_t result;
-	size_t capacity;
-	struct uwb_rc_evt_get_ie *ie_info;
+	struct uwb_rc_evt_get_ie *ie_info = NULL;
+	int capacity;
+
+	capacity = uwb_rc_get_ie(uwb_rc, &ie_info);
+	if (capacity < 0)
+		return capacity;
 
-	d_fnstart(3, dev, "(%p)\n", uwb_rc);
 	mutex_lock(&uwb_rc->ies_mutex);
-	result = uwb_rc_get_ie(uwb_rc, &ie_info);
-	if (result < 0)
-		goto error_get_ie;
-	capacity = result;
-	d_printf(5, dev, "Got IEs %zu bytes (%zu long at %p)\n", result,
-		 (size_t)le16_to_cpu(ie_info->wIELength), ie_info);
-
-	/* Remove IEs that host should not set. */
-	result = uwb_rc_parse_host_settable_ie(&uwb_rc->uwb_dev,
-			ie_info->IEData, le16_to_cpu(ie_info->wIELength));
-	if (result < 0)
-		goto error_parse;
-	d_printf(5, dev, "purged non-settable IEs to %zu bytes\n", result);
-	uwb_rc->ies = (void *) ie_info;
+
+	uwb_rc->ies = (struct uwb_rc_cmd_set_ie *)ie_info;
 	uwb_rc->ies->rccb.bCommandType = UWB_RC_CET_GENERAL;
 	uwb_rc->ies->rccb.wCommand = cpu_to_le16(UWB_RC_CMD_SET_IE);
 	uwb_rc->ies_capacity = capacity;
-	d_printf(5, dev, "IE cache at %p %zu bytes, %zu capacity\n",
-		 ie_info, result, capacity);
-	result = 0;
-error_parse:
-error_get_ie:
+
 	mutex_unlock(&uwb_rc->ies_mutex);
-	d_fnend(3, dev, "(%p) = %zu\n", uwb_rc, result);
-	return result;
+
+	return 0;
 }
 
 
@@ -383,26 +226,47 @@ void uwb_rc_ie_release(struct uwb_rc *uwb_rc)
 }
 
 
-static
-int __acc_size(struct uwb_dev *uwb_dev, const struct uwb_ie_hdr *ie_hdr,
-	       size_t offset, void *_ctx)
+static int uwb_rc_ie_add_one(struct uwb_rc *rc, const struct uwb_ie_hdr *new_ie)
 {
-	size_t *acc_size = _ctx;
-	*acc_size += sizeof(*ie_hdr) + ie_hdr->length;
-	d_printf(6, &uwb_dev->dev, "new acc size %zu\n", *acc_size);
+	struct uwb_rc_cmd_set_ie *new_ies;
+	void *ptr, *prev_ie;
+	struct uwb_ie_hdr *ie;
+	size_t length, new_ie_len, new_capacity, size, prev_size;
+
+	length = le16_to_cpu(rc->ies->wIELength);
+	new_ie_len = sizeof(struct uwb_ie_hdr) + new_ie->length;
+	new_capacity = sizeof(struct uwb_rc_cmd_set_ie) + length + new_ie_len;
+
+	if (new_capacity > rc->ies_capacity) {
+		new_ies = krealloc(rc->ies, new_capacity, GFP_KERNEL);
+		if (!new_ies)
+			return -ENOMEM;
+		rc->ies = new_ies;
+	}
+
+	ptr = rc->ies->IEData;
+	size = length;
+	for (;;) {
+		prev_ie = ptr;
+		prev_size = size;
+		ie = uwb_ie_next(&ptr, &size);
+		if (!ie || ie->element_id > new_ie->element_id)
+			break;
+	}
+
+	memmove(prev_ie + new_ie_len, prev_ie, prev_size);
+	memcpy(prev_ie, new_ie, new_ie_len);
+	rc->ies->wIELength = cpu_to_le16(length + new_ie_len);
+
 	return 0;
 }
 
-
 /**
- * Add a new IE to IEs currently being transmitted by device
- *
+ * uwb_rc_ie_add - add new IEs to the radio controller's beacon
+ * @uwb_rc: the radio controller.
  * @ies: the buffer containing the new IE or IEs to be added to
- *       the device's beacon. The buffer will be verified for
- *       consistence (meaning the headers should be right) and
- *       consistent with the buffer size.
- * @size: size of @ies (in bytes, total buffer size)
- * @returns: 0 if ok, <0 errno code on error
+ *       the device's beacon.
+ * @size: length of all the IEs.
  *
  * According to WHCI 0.95 [4.13.6] the driver will only receive the RCEB
  * after the device sent the first beacon that includes the IEs specified
@@ -411,66 +275,40 @@ int __acc_size(struct uwb_dev *uwb_dev, const struct uwb_ie_hdr *ie_hdr,
  * we start beaconing.
  *
  * Setting an IE on the device will overwrite all current IEs in device. So
- * we take the current IEs being transmitted by the device, append the
+ * we take the current IEs being transmitted by the device, insert the
  * new one, and call SET IE with all the IEs needed.
  *
- * The local IE cache will only be updated with the new IE if SET IE
- * completed successfully.
+ * Returns 0 on success; or -ENOMEM.
  */
 int uwb_rc_ie_add(struct uwb_rc *uwb_rc,
 		  const struct uwb_ie_hdr *ies, size_t size)
 {
 	int result = 0;
-	struct device *dev = &uwb_rc->uwb_dev.dev;
-	struct uwb_rc_cmd_set_ie *new_ies;
-	size_t ies_size, total_size, acc_size = 0;
-
-	if (uwb_rc->ies == NULL)
-		return -ESHUTDOWN;
-	uwb_ie_for_each(&uwb_rc->uwb_dev, __acc_size, &acc_size, ies, size);
-	if (acc_size != size) {
-		dev_err(dev, "BUG: bad IEs, misconstructed headers "
-			"[%zu bytes reported vs %zu calculated]\n",
-			size, acc_size);
-		WARN_ON(1);
-		return -EINVAL;
-	}
+	void *ptr;
+	const struct uwb_ie_hdr *ie;
+
 	mutex_lock(&uwb_rc->ies_mutex);
-	ies_size = le16_to_cpu(uwb_rc->ies->wIELength);
-	total_size = sizeof(*uwb_rc->ies) + ies_size;
-	if (total_size + size > uwb_rc->ies_capacity) {
-		d_printf(4, dev, "Reallocating IE cache from %p capacity %zu "
-			 "to capacity %zu\n", uwb_rc->ies, uwb_rc->ies_capacity,
-			 total_size + size);
-		new_ies = kzalloc(total_size + size, GFP_KERNEL);
-		if (new_ies == NULL) {
-			dev_err(dev, "No memory for adding new IE\n");
-			result = -ENOMEM;
-			goto error_alloc;
-		}
-		memcpy(new_ies, uwb_rc->ies, total_size);
-		uwb_rc->ies_capacity = total_size + size;
-		kfree(uwb_rc->ies);
-		uwb_rc->ies = new_ies;
-		d_printf(4, dev, "New IE cache at %p capacity %zu\n",
-			 uwb_rc->ies, uwb_rc->ies_capacity);
+
+	ptr = (void *)ies;
+	for (;;) {
+		ie = uwb_ie_next(&ptr, &size);
+		if (!ie)
+			break;
+
+		result = uwb_rc_ie_add_one(uwb_rc, ie);
+		if (result < 0)
+			break;
 	}
-	memcpy((void *)uwb_rc->ies + total_size, ies, size);
-	uwb_rc->ies->wIELength = cpu_to_le16(ies_size + size);
-	if (uwb_rc->beaconing != -1) {
-		result = uwb_rc_set_ie(uwb_rc, uwb_rc->ies);
-		if (result < 0) {
-			dev_err(dev, "Cannot set new IE on device: %d\n",
-				result);
-			uwb_rc->ies->wIELength = cpu_to_le16(ies_size);
+	if (result >= 0) {
+		if (size == 0) {
+			if (uwb_rc->beaconing != -1)
+				result = uwb_rc_set_ie(uwb_rc, uwb_rc->ies);
 		} else
-			result = 0;
+			result = -EINVAL;
 	}
-	d_printf(4, dev, "IEs now occupy %hu bytes of %zu capacity at %p\n",
-		 le16_to_cpu(uwb_rc->ies->wIELength), uwb_rc->ies_capacity,
-		 uwb_rc->ies);
-error_alloc:
+
 	mutex_unlock(&uwb_rc->ies_mutex);
+
 	return result;
 }
 EXPORT_SYMBOL_GPL(uwb_rc_ie_add);
@@ -489,53 +327,52 @@ EXPORT_SYMBOL_GPL(uwb_rc_ie_add);
  * beacon. We don't reallocate, we just mark the size smaller.
  */
 static
-int uwb_rc_ie_cache_rm(struct uwb_rc *uwb_rc, enum uwb_ie to_remove)
+void uwb_rc_ie_cache_rm(struct uwb_rc *uwb_rc, enum uwb_ie to_remove)
 {
-	struct uwb_ie_hdr *ie_hdr;
-	size_t new_len = le16_to_cpu(uwb_rc->ies->wIELength);
-	void *itr = uwb_rc->ies->IEData;
-	void *top = itr + new_len;
-
-	while (itr < top) {
-		ie_hdr = itr;
-		if (ie_hdr->element_id != to_remove) {
-			itr += sizeof(*ie_hdr) + ie_hdr->length;
-		} else {
-			int ie_length;
-			ie_length = sizeof(*ie_hdr) + ie_hdr->length;
-			if (top - itr != ie_length)
-				memmove(itr, itr + ie_length, top - itr + ie_length);
-			top -= ie_length;
-			new_len -= ie_length;
+	struct uwb_ie_hdr *ie;
+	size_t len = le16_to_cpu(uwb_rc->ies->wIELength);
+	void *ptr;
+	size_t size;
+
+	ptr = uwb_rc->ies->IEData;
+	size = len;
+	for (;;) {
+		ie = uwb_ie_next(&ptr, &size);
+		if (!ie)
+			break;
+		if (ie->element_id == to_remove) {
+			len -= sizeof(struct uwb_ie_hdr) + ie->length;
+			memmove(ie, ptr, size);
+			ptr = ie;
 		}
 	}
-	uwb_rc->ies->wIELength = cpu_to_le16(new_len);
-	return 0;
+	uwb_rc->ies->wIELength = cpu_to_le16(len);
 }
 
 
 /**
- * Remove an IE currently being transmitted by device
+ * uwb_rc_ie_rm - remove an IE from the radio controller's beacon
+ * @uwb_rc: the radio controller.
+ * @element_id: the element ID of the IE to remove.
  *
- * @element_id: id of IE to be removed from device's beacon
+ * Only IEs previously added with uwb_rc_ie_add() may be removed.
+ *
+ * Returns 0 on success; or -ve the SET-IE command to the radio
+ * controller failed.
  */
 int uwb_rc_ie_rm(struct uwb_rc *uwb_rc, enum uwb_ie element_id)
 {
-	struct device *dev = &uwb_rc->uwb_dev.dev;
-	int result;
+	int result = 0;
 
-	if (uwb_rc->ies == NULL)
-		return -ESHUTDOWN;
 	mutex_lock(&uwb_rc->ies_mutex);
-	result = uwb_rc_ie_cache_rm(uwb_rc, element_id);
-	if (result < 0)
-		dev_err(dev, "Cannot remove IE from cache.\n");
-	if (uwb_rc->beaconing != -1) {
+
+	uwb_rc_ie_cache_rm(uwb_rc, element_id);
+
+	if (uwb_rc->beaconing != -1)
 		result = uwb_rc_set_ie(uwb_rc, uwb_rc->ies);
-		if (result < 0)
-			dev_err(dev, "Cannot set new IE on device.\n");
-	}
+
 	mutex_unlock(&uwb_rc->ies_mutex);
+
 	return result;
 }
 EXPORT_SYMBOL_GPL(uwb_rc_ie_rm);
diff --git a/drivers/uwb/lc-rc.c b/drivers/uwb/lc-rc.c
index ee5772f00d42..1129e8767b58 100644
--- a/drivers/uwb/lc-rc.c
+++ b/drivers/uwb/lc-rc.c
@@ -468,28 +468,3 @@ void uwb_rc_put(struct uwb_rc *rc)
 	__uwb_rc_put(rc);
 }
 EXPORT_SYMBOL_GPL(uwb_rc_put);
-
-/*
- *
- *
- */
-ssize_t uwb_rc_print_IEs(struct uwb_rc *uwb_rc, char *buf, size_t size)
-{
-	ssize_t result;
-	struct uwb_rc_evt_get_ie *ie_info;
-	struct uwb_buf_ctx ctx;
-
-	result = uwb_rc_get_ie(uwb_rc, &ie_info);
-	if (result < 0)
-		goto error_get_ie;
-	ctx.buf = buf;
-	ctx.size = size;
-	ctx.bytes = 0;
-	uwb_ie_for_each(&uwb_rc->uwb_dev, uwb_ie_dump_hex, &ctx,
-			ie_info->IEData, result - sizeof(*ie_info));
-	result = ctx.bytes;
-	kfree(ie_info);
-error_get_ie:
-	return result;
-}
-
diff --git a/drivers/uwb/uwb-internal.h b/drivers/uwb/uwb-internal.h
index 2ad307d12961..983ebc4dd8d5 100644
--- a/drivers/uwb/uwb-internal.h
+++ b/drivers/uwb/uwb-internal.h
@@ -66,14 +66,14 @@ extern int uwb_rc_scan(struct uwb_rc *rc,
 		       unsigned channel, enum uwb_scan_type type,
 		       unsigned bpst_offset);
 extern int uwb_rc_send_all_drp_ie(struct uwb_rc *rc);
-extern ssize_t uwb_rc_print_IEs(struct uwb_rc *rc, char *, size_t);
-extern void uwb_rc_ie_init(struct uwb_rc *);
-extern void uwb_rc_ie_init(struct uwb_rc *);
-extern ssize_t uwb_rc_ie_setup(struct uwb_rc *);
-extern void uwb_rc_ie_release(struct uwb_rc *);
-extern int uwb_rc_ie_add(struct uwb_rc *,
-			 const struct uwb_ie_hdr *, size_t);
-extern int uwb_rc_ie_rm(struct uwb_rc *, enum uwb_ie);
+
+void uwb_rc_ie_init(struct uwb_rc *);
+int uwb_rc_ie_setup(struct uwb_rc *);
+void uwb_rc_ie_release(struct uwb_rc *);
+int uwb_ie_dump_hex(const struct uwb_ie_hdr *ies, size_t len,
+		    char *buf, size_t size);
+int uwb_rc_set_ie(struct uwb_rc *, struct uwb_rc_cmd_set_ie *);
+
 
 extern const char *uwb_rc_strerror(unsigned code);
 
diff --git a/drivers/uwb/wlp/wlp-internal.h b/drivers/uwb/wlp/wlp-internal.h
index 1c94fabfb1a7..3e8d5de7c5b9 100644
--- a/drivers/uwb/wlp/wlp-internal.h
+++ b/drivers/uwb/wlp/wlp-internal.h
@@ -42,10 +42,6 @@ enum wlp_wss_connect {
 extern struct kobj_type wss_ktype;
 extern struct attribute_group wss_attr_group;
 
-extern int uwb_rc_ie_add(struct uwb_rc *, const struct uwb_ie_hdr *, size_t);
-extern int uwb_rc_ie_rm(struct uwb_rc *, enum uwb_ie);
-
-
 /* This should be changed to a dynamic array where entries are sorted
  * by eth_addr and search is done in a binary form
  *
diff --git a/include/linux/uwb.h b/include/linux/uwb.h
index 010ee708304d..6d93f54b8879 100644
--- a/include/linux/uwb.h
+++ b/include/linux/uwb.h
@@ -444,7 +444,6 @@ ssize_t uwb_rc_vcmd(struct uwb_rc *rc, const char *cmd_name,
 		    struct uwb_rccb *cmd, size_t cmd_size,
 		    u8 expected_type, u16 expected_event,
 		    struct uwb_rceb **preply);
-ssize_t uwb_rc_get_ie(struct uwb_rc *, struct uwb_rc_evt_get_ie **);
 int uwb_bg_joined(struct uwb_rc *rc);
 
 size_t __uwb_addr_print(char *, size_t, const unsigned char *, int);
@@ -653,22 +652,9 @@ static inline int edc_inc(struct edc *err_hist, u16 max_err, u16 timeframe)
 
 /* Information Element handling */
 
-/* For representing the state of writing to a buffer when iterating */
-struct uwb_buf_ctx {
-	char *buf;
-	size_t bytes, size;
-};
-
-typedef int (*uwb_ie_f)(struct uwb_dev *, const struct uwb_ie_hdr *,
-			size_t, void *);
 struct uwb_ie_hdr *uwb_ie_next(void **ptr, size_t *len);
-ssize_t uwb_ie_for_each(struct uwb_dev *uwb_dev, uwb_ie_f fn, void *data,
-			const void *buf, size_t size);
-int uwb_ie_dump_hex(struct uwb_dev *, const struct uwb_ie_hdr *,
-		    size_t, void *);
-int uwb_rc_set_ie(struct uwb_rc *, struct uwb_rc_cmd_set_ie *);
-struct uwb_ie_hdr *uwb_ie_next(void **ptr, size_t *len);
-
+int uwb_rc_ie_add(struct uwb_rc *uwb_rc, const struct uwb_ie_hdr *ies, size_t size);
+int uwb_rc_ie_rm(struct uwb_rc *uwb_rc, enum uwb_ie element_id);
 
 /*
  * Transmission statistics
-- 
cgit v1.2.3


From 5e1f8c9e20a92743eefc9a82c2db835213905e26 Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Tue, 28 Oct 2008 13:21:55 -0400
Subject: ext3: Add support for non-native signed/unsigned htree hash
 algorithms

The original ext3 hash algorithms assumed that variables of type char
were signed, as God and K&R intended.  Unfortunately, this assumption
is not true on some architectures.  Userspace support for marking
filesystems with non-native signed/unsigned chars was added two years
ago, but the kernel-side support was never added (until now).

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: akpm@linux-foundation.org
Cc: linux-kernel@vger.kernel.org
---
 fs/ext3/hash.c             | 77 ++++++++++++++++++++++++++++++++++++++++------
 fs/ext3/namei.c            |  7 +++++
 fs/ext3/super.c            | 12 ++++++++
 include/linux/ext3_fs.h    | 28 ++++++++++++++++-
 include/linux/ext3_fs_sb.h |  1 +
 5 files changed, 114 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/fs/ext3/hash.c b/fs/ext3/hash.c
index c30e149fbd2e..7d215b4d4f2e 100644
--- a/fs/ext3/hash.c
+++ b/fs/ext3/hash.c
@@ -35,23 +35,71 @@ static void TEA_transform(__u32 buf[4], __u32 const in[])
 
 
 /* The old legacy hash */
-static __u32 dx_hack_hash (const char *name, int len)
+static __u32 dx_hack_hash_unsigned(const char *name, int len)
 {
-	__u32 hash0 = 0x12a3fe2d, hash1 = 0x37abe8f9;
+	__u32 hash, hash0 = 0x12a3fe2d, hash1 = 0x37abe8f9;
+	const unsigned char *ucp = (const unsigned char *) name;
+
+	while (len--) {
+		hash = hash1 + (hash0 ^ (((int) *ucp++) * 7152373));
+
+		if (hash & 0x80000000)
+			hash -= 0x7fffffff;
+		hash1 = hash0;
+		hash0 = hash;
+	}
+	return hash0 << 1;
+}
+
+static __u32 dx_hack_hash_signed(const char *name, int len)
+{
+	__u32 hash, hash0 = 0x12a3fe2d, hash1 = 0x37abe8f9;
+	const signed char *scp = (const signed char *) name;
+
 	while (len--) {
-		__u32 hash = hash1 + (hash0 ^ (*name++ * 7152373));
+		hash = hash1 + (hash0 ^ (((int) *scp++) * 7152373));
 
-		if (hash & 0x80000000) hash -= 0x7fffffff;
+		if (hash & 0x80000000)
+			hash -= 0x7fffffff;
 		hash1 = hash0;
 		hash0 = hash;
 	}
-	return (hash0 << 1);
+	return hash0 << 1;
 }
 
-static void str2hashbuf(const char *msg, int len, __u32 *buf, int num)
+static void str2hashbuf_signed(const char *msg, int len, __u32 *buf, int num)
 {
 	__u32	pad, val;
 	int	i;
+	const signed char *scp = (const signed char *) msg;
+
+	pad = (__u32)len | ((__u32)len << 8);
+	pad |= pad << 16;
+
+	val = pad;
+	if (len > num*4)
+		len = num * 4;
+	for (i = 0; i < len; i++) {
+		if ((i % 4) == 0)
+			val = pad;
+		val = ((int) scp[i]) + (val << 8);
+		if ((i % 4) == 3) {
+			*buf++ = val;
+			val = pad;
+			num--;
+		}
+	}
+	if (--num >= 0)
+		*buf++ = val;
+	while (--num >= 0)
+		*buf++ = pad;
+}
+
+static void str2hashbuf_unsigned(const char *msg, int len, __u32 *buf, int num)
+{
+	__u32	pad, val;
+	int	i;
+	const unsigned char *ucp = (const unsigned char *) msg;
 
 	pad = (__u32)len | ((__u32)len << 8);
 	pad |= pad << 16;
@@ -62,7 +110,7 @@ static void str2hashbuf(const char *msg, int len, __u32 *buf, int num)
 	for (i=0; i < len; i++) {
 		if ((i % 4) == 0)
 			val = pad;
-		val = msg[i] + (val << 8);
+		val = ((int) ucp[i]) + (val << 8);
 		if ((i % 4) == 3) {
 			*buf++ = val;
 			val = pad;
@@ -95,6 +143,8 @@ int ext3fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo)
 	const char	*p;
 	int		i;
 	__u32		in[8], buf[4];
+	void		(*str2hashbuf)(const char *, int, __u32 *, int) =
+				str2hashbuf_signed;
 
 	/* Initialize the default seed for the hash checksum functions */
 	buf[0] = 0x67452301;
@@ -113,13 +163,18 @@ int ext3fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo)
 	}
 
 	switch (hinfo->hash_version) {
+	case DX_HASH_LEGACY_UNSIGNED:
+		hash = dx_hack_hash_unsigned(name, len);
+		break;
 	case DX_HASH_LEGACY:
-		hash = dx_hack_hash(name, len);
+		hash = dx_hack_hash_signed(name, len);
 		break;
+	case DX_HASH_HALF_MD4_UNSIGNED:
+		str2hashbuf = str2hashbuf_unsigned;
 	case DX_HASH_HALF_MD4:
 		p = name;
 		while (len > 0) {
-			str2hashbuf(p, len, in, 8);
+			(*str2hashbuf)(p, len, in, 8);
 			half_md4_transform(buf, in);
 			len -= 32;
 			p += 32;
@@ -127,10 +182,12 @@ int ext3fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo)
 		minor_hash = buf[2];
 		hash = buf[1];
 		break;
+	case DX_HASH_TEA_UNSIGNED:
+		str2hashbuf = str2hashbuf_unsigned;
 	case DX_HASH_TEA:
 		p = name;
 		while (len > 0) {
-			str2hashbuf(p, len, in, 4);
+			(*str2hashbuf)(p, len, in, 4);
 			TEA_transform(buf, in);
 			len -= 16;
 			p += 16;
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 1dd2abe6313e..287b304d42a7 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -368,6 +368,8 @@ dx_probe(struct qstr *entry, struct inode *dir,
 		goto fail;
 	}
 	hinfo->hash_version = root->info.hash_version;
+	if (hinfo->hash_version <= DX_HASH_TEA)
+		hinfo->hash_version += EXT3_SB(dir->i_sb)->s_hash_unsigned;
 	hinfo->seed = EXT3_SB(dir->i_sb)->s_hash_seed;
 	if (entry)
 		ext3fs_dirhash(entry->name, entry->len, hinfo);
@@ -636,6 +638,9 @@ int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash,
 	dir = dir_file->f_path.dentry->d_inode;
 	if (!(EXT3_I(dir)->i_flags & EXT3_INDEX_FL)) {
 		hinfo.hash_version = EXT3_SB(dir->i_sb)->s_def_hash_version;
+		if (hinfo.hash_version <= DX_HASH_TEA)
+			hinfo.hash_version +=
+				EXT3_SB(dir->i_sb)->s_hash_unsigned;
 		hinfo.seed = EXT3_SB(dir->i_sb)->s_hash_seed;
 		count = htree_dirblock_to_tree(dir_file, dir, 0, &hinfo,
 					       start_hash, start_minor_hash);
@@ -1398,6 +1403,8 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry,
 
 	/* Initialize as for dx_probe */
 	hinfo.hash_version = root->info.hash_version;
+	if (hinfo.hash_version <= DX_HASH_TEA)
+		hinfo.hash_version += EXT3_SB(dir->i_sb)->s_hash_unsigned;
 	hinfo.seed = EXT3_SB(dir->i_sb)->s_hash_seed;
 	ext3fs_dirhash(name, namelen, &hinfo);
 	frame = frames;
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index f6c94f232ec1..541d5e4f7f6e 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -1744,6 +1744,18 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	for (i=0; i < 4; i++)
 		sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]);
 	sbi->s_def_hash_version = es->s_def_hash_version;
+	i = le32_to_cpu(es->s_flags);
+	if (i & EXT2_FLAGS_UNSIGNED_HASH)
+		sbi->s_hash_unsigned = 3;
+	else if ((i & EXT2_FLAGS_SIGNED_HASH) == 0) {
+#ifdef __CHAR_UNSIGNED__
+		es->s_flags |= cpu_to_le32(EXT2_FLAGS_UNSIGNED_HASH);
+		sbi->s_hash_unsigned = 3;
+#else
+		es->s_flags |= cpu_to_le32(EXT2_FLAGS_SIGNED_HASH);
+#endif
+		sb->s_dirt = 1;
+	}
 
 	if (sbi->s_blocks_per_group > blocksize * 8) {
 		printk (KERN_ERR
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index d14f02918483..9004794a35fe 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -353,6 +353,13 @@ struct ext3_inode {
 #define	EXT3_ERROR_FS			0x0002	/* Errors detected */
 #define	EXT3_ORPHAN_FS			0x0004	/* Orphans being recovered */
 
+/*
+ * Misc. filesystem flags
+ */
+#define EXT2_FLAGS_SIGNED_HASH		0x0001  /* Signed dirhash in use */
+#define EXT2_FLAGS_UNSIGNED_HASH	0x0002  /* Unsigned dirhash in use */
+#define EXT2_FLAGS_TEST_FILESYS		0x0004	/* to test development code */
+
 /*
  * Mount flags
  */
@@ -489,7 +496,23 @@ struct ext3_super_block {
 	__u16	s_reserved_word_pad;
 	__le32	s_default_mount_opts;
 	__le32	s_first_meta_bg;	/* First metablock block group */
-	__u32	s_reserved[190];	/* Padding to the end of the block */
+	__le32	s_mkfs_time;		/* When the filesystem was created */
+	__le32	s_jnl_blocks[17];	/* Backup of the journal inode */
+	/* 64bit support valid if EXT4_FEATURE_COMPAT_64BIT */
+/*150*/	__le32	s_blocks_count_hi;	/* Blocks count */
+	__le32	s_r_blocks_count_hi;	/* Reserved blocks count */
+	__le32	s_free_blocks_count_hi;	/* Free blocks count */
+	__le16	s_min_extra_isize;	/* All inodes have at least # bytes */
+	__le16	s_want_extra_isize; 	/* New inodes should reserve # bytes */
+	__le32	s_flags;		/* Miscellaneous flags */
+	__le16  s_raid_stride;		/* RAID stride */
+	__le16  s_mmp_interval;         /* # seconds to wait in MMP checking */
+	__le64  s_mmp_block;            /* Block for multi-mount protection */
+	__le32  s_raid_stripe_width;    /* blocks on all data disks (N*stride)*/
+	__u8	s_log_groups_per_flex;  /* FLEX_BG group size */
+	__u8	s_reserved_char_pad2;
+	__le16  s_reserved_pad;
+	__u32   s_reserved[162];        /* Padding to the end of the block */
 };
 
 #ifdef __KERNEL__
@@ -694,6 +717,9 @@ static inline __le16 ext3_rec_len_to_disk(unsigned len)
 #define DX_HASH_LEGACY		0
 #define DX_HASH_HALF_MD4	1
 #define DX_HASH_TEA		2
+#define DX_HASH_LEGACY_UNSIGNED	3
+#define DX_HASH_HALF_MD4_UNSIGNED	4
+#define DX_HASH_TEA_UNSIGNED		5
 
 #ifdef __KERNEL__
 
diff --git a/include/linux/ext3_fs_sb.h b/include/linux/ext3_fs_sb.h
index e024e38248ff..a4e9216b3a6d 100644
--- a/include/linux/ext3_fs_sb.h
+++ b/include/linux/ext3_fs_sb.h
@@ -57,6 +57,7 @@ struct ext3_sb_info {
 	u32 s_next_generation;
 	u32 s_hash_seed[4];
 	int s_def_hash_version;
+	int s_hash_unsigned;	/* 3 if hash should be signed, 0 if not */
 	struct percpu_counter s_freeblocks_counter;
 	struct percpu_counter s_freeinodes_counter;
 	struct percpu_counter s_dirs_counter;
-- 
cgit v1.2.3


From a20c7ab570ffdce1d6f67c7acf8c1c502a3b3839 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dbaryshkov@gmail.com>
Date: Thu, 16 Oct 2008 18:43:48 +0400
Subject: [MTD] sharpsl-nand: use platform_data for model-specific values

Add platform_data which holds all model-specific values, like badblocks
pattern, oobinfo, partitions.

Signed-off-by: Dmitry Baryshkov <dbaryshkov@gmail.com>
---
 drivers/mtd/nand/sharpsl.c  | 158 +++++++++++++++++++++++---------------------
 include/linux/mtd/sharpsl.h |  20 ++++++
 2 files changed, 104 insertions(+), 74 deletions(-)
 create mode 100644 include/linux/mtd/sharpsl.h

(limited to 'include/linux')

diff --git a/drivers/mtd/nand/sharpsl.c b/drivers/mtd/nand/sharpsl.c
index 17625c0a8f61..698378ca8e0a 100644
--- a/drivers/mtd/nand/sharpsl.c
+++ b/drivers/mtd/nand/sharpsl.c
@@ -20,6 +20,7 @@
 #include <linux/mtd/nand.h>
 #include <linux/mtd/nand_ecc.h>
 #include <linux/mtd/partitions.h>
+#include <linux/mtd/sharpsl.h>
 #include <linux/interrupt.h>
 #include <linux/platform_device.h>
 
@@ -53,31 +54,6 @@ struct sharpsl_nand {
 #define FLCLE		(1 << 1)
 #define FLCE0		(1 << 0)
 
-#ifdef CONFIG_MTD_PARTITIONS
-/*
- * Define partitions for flash device
- */
-#define DEFAULT_NUM_PARTITIONS 3
-
-static struct mtd_partition sharpsl_nand_default_partition_info[] = {
-	{
-	 .name = "System Area",
-	 .offset = 0,
-	 .size = 7 * 1024 * 1024,
-	 },
-	{
-	 .name = "Root Filesystem",
-	 .offset = 7 * 1024 * 1024,
-	 .size = 30 * 1024 * 1024,
-	 },
-	{
-	 .name = "Home Filesystem",
-	 .offset = MTDPART_OFS_APPEND,
-	 .size = MTDPART_SIZ_FULL,
-	 },
-};
-#endif
-
 /*
  *	hardware specific access to control-lines
  *	ctrl:
@@ -106,31 +82,6 @@ static void sharpsl_nand_hwcontrol(struct mtd_info *mtd, int cmd,
 		writeb(cmd, chip->IO_ADDR_W);
 }
 
-static uint8_t scan_ff_pattern[] = { 0xff, 0xff };
-
-static struct nand_bbt_descr sharpsl_bbt = {
-	.options = 0,
-	.offs = 4,
-	.len = 2,
-	.pattern = scan_ff_pattern
-};
-
-static struct nand_bbt_descr sharpsl_akita_bbt = {
-	.options = 0,
-	.offs = 4,
-	.len = 1,
-	.pattern = scan_ff_pattern
-};
-
-static struct nand_ecclayout akita_oobinfo = {
-	.eccbytes = 24,
-	.eccpos = {
-		   0x5, 0x1, 0x2, 0x3, 0x6, 0x7, 0x15, 0x11,
-		   0x12, 0x13, 0x16, 0x17, 0x25, 0x21, 0x22, 0x23,
-		   0x26, 0x27, 0x35, 0x31, 0x32, 0x33, 0x36, 0x37},
-	.oobfree = {{0x08, 0x09}}
-};
-
 static int sharpsl_nand_dev_ready(struct mtd_info *mtd)
 {
 	struct sharpsl_nand *sharpsl = mtd_to_sharpsl(mtd);
@@ -169,6 +120,12 @@ static int __devinit sharpsl_nand_probe(struct platform_device *pdev)
 	struct resource *r;
 	int err = 0;
 	struct sharpsl_nand *sharpsl;
+	struct sharpsl_nand_platform_data *data = pdev->dev.platform_data;
+
+	if (!data) {
+		dev_err(&pdev->dev, "no platform data!\n");
+		return -EINVAL;
+	}
 
 	/* Allocate memory for MTD device structure and private data */
 	sharpsl = kzalloc(sizeof(struct sharpsl_nand), GFP_KERNEL);
@@ -218,11 +175,8 @@ static int __devinit sharpsl_nand_probe(struct platform_device *pdev)
 	this->ecc.mode = NAND_ECC_HW;
 	this->ecc.size = 256;
 	this->ecc.bytes = 3;
-	this->badblock_pattern = &sharpsl_bbt;
-	if (machine_is_akita() || machine_is_borzoi()) {
-		this->badblock_pattern = &sharpsl_akita_bbt;
-		this->ecc.layout = &akita_oobinfo;
-	}
+	this->badblock_pattern = data->badblock_pattern;
+	this->ecc.layout = data->ecc_layout;
 	this->ecc.hwctl = sharpsl_nand_enable_hwecc;
 	this->ecc.calculate = sharpsl_nand_calculate_ecc;
 	this->ecc.correct = nand_correct_data;
@@ -236,29 +190,16 @@ static int __devinit sharpsl_nand_probe(struct platform_device *pdev)
 	sharpsl->mtd.name = "sharpsl-nand";
 #ifdef CONFIG_MTD_PARTITIONS
 	nr_partitions = parse_mtd_partitions(&sharpsl->mtd, part_probes, &sharpsl_partition_info, 0);
-
 	if (nr_partitions <= 0) {
-		nr_partitions = ARRAY_SIZE(sharpsl_nand_default_partition_info);
-		sharpsl_partition_info = sharpsl_nand_default_partition_info;
-		if (machine_is_poodle()) {
-			sharpsl_partition_info[1].size = 22 * 1024 * 1024;
-		} else if (machine_is_corgi() || machine_is_shepherd()) {
-			sharpsl_partition_info[1].size = 25 * 1024 * 1024;
-		} else if (machine_is_husky()) {
-			sharpsl_partition_info[1].size = 53 * 1024 * 1024;
-		} else if (machine_is_spitz()) {
-			sharpsl_partition_info[1].size = 5 * 1024 * 1024;
-		} else if (machine_is_akita()) {
-			sharpsl_partition_info[1].size = 58 * 1024 * 1024;
-		} else if (machine_is_borzoi()) {
-			sharpsl_partition_info[1].size = 32 * 1024 * 1024;
-		}
+		nr_partitions = data->nr_partitions;
+		sharpsl_partition_info = data->partitions;
 	}
 
-	err = add_mtd_partitions(&sharpsl->mtd, sharpsl_partition_info, nr_partitions);
-#else
-	err = add_mtd_device(&sharpsl->mtd);
+	if (nr_partitions > 0)
+		err = add_mtd_partitions(&sharpsl->mtd, sharpsl_partition_info, nr_partitions);
+	else
 #endif
+	err = add_mtd_device(&sharpsl->mtd);
 	if (err)
 		goto err_add;
 
@@ -306,6 +247,58 @@ static struct platform_driver sharpsl_nand_driver = {
 	.remove		= __devexit_p(sharpsl_nand_remove),
 };
 
+/*
+ * Define partitions for flash device
+ */
+static struct mtd_partition sharpsl_nand_partitions[] = {
+	{
+	 .name = "System Area",
+	 .offset = 0,
+	 .size = 7 * 1024 * 1024,
+	 },
+	{
+	 .name = "Root Filesystem",
+	 .offset = 7 * 1024 * 1024,
+	 .size = 30 * 1024 * 1024,
+	 },
+	{
+	 .name = "Home Filesystem",
+	 .offset = MTDPART_OFS_APPEND,
+	 .size = MTDPART_SIZ_FULL,
+	 },
+};
+
+static uint8_t scan_ff_pattern[] = { 0xff, 0xff };
+
+static struct nand_bbt_descr sharpsl_bbt = {
+	.options = 0,
+	.offs = 4,
+	.len = 2,
+	.pattern = scan_ff_pattern
+};
+
+static struct nand_bbt_descr sharpsl_akita_bbt = {
+	.options = 0,
+	.offs = 4,
+	.len = 1,
+	.pattern = scan_ff_pattern
+};
+
+static struct nand_ecclayout akita_oobinfo = {
+	.eccbytes = 24,
+	.eccpos = {
+		   0x5, 0x1, 0x2, 0x3, 0x6, 0x7, 0x15, 0x11,
+		   0x12, 0x13, 0x16, 0x17, 0x25, 0x21, 0x22, 0x23,
+		   0x26, 0x27, 0x35, 0x31, 0x32, 0x33, 0x36, 0x37},
+	.oobfree = {{0x08, 0x09}}
+};
+
+static struct sharpsl_nand_platform_data sharpsl_nand_platform_data = {
+	.badblock_pattern	= &sharpsl_bbt,
+	.partitions		= sharpsl_nand_partitions,
+	.nr_partitions		= ARRAY_SIZE(sharpsl_nand_partitions),
+};
+
 static struct resource sharpsl_nand_resources[] = {
 	{
 		.start	= 0x0C000000,
@@ -319,10 +312,27 @@ static struct platform_device sharpsl_nand_device = {
 	.id		= -1,
 	.resource	= sharpsl_nand_resources,
 	.num_resources	= ARRAY_SIZE(sharpsl_nand_resources),
+	.dev.platform_data	= &sharpsl_nand_platform_data,
 };
 
 static int __init sharpsl_nand_init(void)
 {
+	if (machine_is_poodle()) {
+		sharpsl_nand_partitions[1].size = 22 * 1024 * 1024;
+	} else if (machine_is_corgi() || machine_is_shepherd()) {
+		sharpsl_nand_partitions[1].size = 25 * 1024 * 1024;
+	} else if (machine_is_husky()) {
+		sharpsl_nand_partitions[1].size = 53 * 1024 * 1024;
+	} else if (machine_is_spitz()) {
+		sharpsl_nand_partitions[1].size = 5 * 1024 * 1024;
+	} else if (machine_is_akita()) {
+		sharpsl_nand_partitions[1].size = 58 * 1024 * 1024;
+		sharpsl_nand_platform_data.badblock_pattern = &sharpsl_akita_bbt;
+		sharpsl_nand_platform_data.ecc_layout = &akita_oobinfo;
+	} else if (machine_is_borzoi()) {
+		sharpsl_nand_partitions[1].size = 32 * 1024 * 1024;
+	}
+
 	platform_device_register(&sharpsl_nand_device);
 	return platform_driver_register(&sharpsl_nand_driver);
 }
diff --git a/include/linux/mtd/sharpsl.h b/include/linux/mtd/sharpsl.h
new file mode 100644
index 000000000000..25f4d2a845c1
--- /dev/null
+++ b/include/linux/mtd/sharpsl.h
@@ -0,0 +1,20 @@
+/*
+ * SharpSL NAND support
+ *
+ * Copyright (C) 2008 Dmitry Baryshkov
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/mtd/nand.h>
+#include <linux/mtd/nand_ecc.h>
+#include <linux/mtd/partitions.h>
+
+struct sharpsl_nand_platform_data {
+	struct nand_bbt_descr	*badblock_pattern;
+	struct nand_ecclayout	*ecc_layout;
+	struct mtd_partition	*partitions;
+	unsigned int		nr_partitions;
+};
-- 
cgit v1.2.3


From ae9eba0e2744f1aa15cdc97cd39277a84723ae23 Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Thu, 30 Oct 2008 20:06:16 +0100
Subject: uwb: struct device - replace bus_id with dev_name(), dev_set_name()

Cc: David Vrabel <david.vrabel@csr.com>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-Off-By: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/uwb/umc-dev.c     | 3 +--
 drivers/uwb/whci.c        | 2 +-
 include/linux/uwb/debug.h | 2 +-
 3 files changed, 3 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/uwb/umc-dev.c b/drivers/uwb/umc-dev.c
index aa44e1c1a102..53207e14cd8f 100644
--- a/drivers/uwb/umc-dev.c
+++ b/drivers/uwb/umc-dev.c
@@ -31,8 +31,7 @@ struct umc_dev *umc_device_create(struct device *parent, int n)
 
 	umc = kzalloc(sizeof(struct umc_dev), GFP_KERNEL);
 	if (umc) {
-		snprintf(umc->dev.bus_id, sizeof(umc->dev.bus_id), "%s-%d",
-			 parent->bus_id, n);
+		dev_set_name(&umc->dev, "%s-%d", dev_name(parent), n);
 		umc->dev.parent  = parent;
 		umc->dev.bus     = &umc_bus_type;
 		umc->dev.release = umc_device_release;
diff --git a/drivers/uwb/whci.c b/drivers/uwb/whci.c
index 3df2388f908f..e626467f95e3 100644
--- a/drivers/uwb/whci.c
+++ b/drivers/uwb/whci.c
@@ -111,7 +111,7 @@ static int whci_add_cap(struct whci_card *card, int n)
 		+ UWBCAPDATA_TO_OFFSET(capdata);
 	umc->resource.end    = umc->resource.start
 		+ (n == 0 ? 0x20 : UWBCAPDATA_TO_SIZE(capdata)) - 1;
-	umc->resource.name   = umc->dev.bus_id;
+	umc->resource.name   = dev_name(&umc->dev);
 	umc->resource.flags  = card->pci->resource[bar].flags;
 	umc->resource.parent = &card->pci->resource[bar];
 	umc->irq             = card->pci->irq;
diff --git a/include/linux/uwb/debug.h b/include/linux/uwb/debug.h
index a86a73fe303f..67a240527145 100644
--- a/include/linux/uwb/debug.h
+++ b/include/linux/uwb/debug.h
@@ -60,7 +60,7 @@ do {									\
 				snprintf(__head, sizeof(__head),	\
 					 "%s %s: ",			\
 					 dev_driver_string(__dev),	\
-					 __dev->bus_id);		\
+					 dev_name(__dev));		\
 		}							\
 		printk(KERN_ERR "%s%s" _tag ": " f, __head,		\
 			__func__, ## a);				\
-- 
cgit v1.2.3


From c5995bd2819dc577d0b32b26be0836d16c977e24 Mon Sep 17 00:00:00 2001
From: Stefano Panella <stefano.panella@csr.com>
Date: Tue, 4 Nov 2008 14:06:31 +0000
Subject: uwb: infrastructure for handling Relinquish Request IEs

The structures and event handler needed to handle Relinish Request IEs
received from neighbors.  Nothing is done with these IEs yet.

Signed-off-by: Stefano Panella <stefano.panella@csr.com>
Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/uwb/Makefile       |  1 +
 drivers/uwb/ie-rcv.c       | 55 ++++++++++++++++++++++++++++++++++++++++++++++
 drivers/uwb/uwb-internal.h |  1 +
 drivers/uwb/uwbd.c         |  4 ++++
 include/linux/uwb/spec.h   | 28 +++++++++++++++++++++++
 5 files changed, 89 insertions(+)
 create mode 100644 drivers/uwb/ie-rcv.c

(limited to 'include/linux')

diff --git a/drivers/uwb/Makefile b/drivers/uwb/Makefile
index 257e6908304c..2b99c3e61671 100644
--- a/drivers/uwb/Makefile
+++ b/drivers/uwb/Makefile
@@ -13,6 +13,7 @@ uwb-objs :=		\
 	drp-ie.o	\
 	est.o		\
 	ie.o		\
+	ie-rcv.o	\
 	lc-dev.o	\
 	lc-rc.o		\
 	neh.o		\
diff --git a/drivers/uwb/ie-rcv.c b/drivers/uwb/ie-rcv.c
new file mode 100644
index 000000000000..917e6d78a798
--- /dev/null
+++ b/drivers/uwb/ie-rcv.c
@@ -0,0 +1,55 @@
+/*
+ * Ultra Wide Band
+ * IE Received notification handling.
+ *
+ * Copyright (C) 2008 Cambridge Silicon Radio Ltd.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/errno.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/bitmap.h>
+#include "uwb-internal.h"
+
+/*
+ * Process an incoming IE Received notification.
+ */
+int uwbd_evt_handle_rc_ie_rcv(struct uwb_event *evt)
+{
+	int result = -EINVAL;
+	struct device *dev = &evt->rc->uwb_dev.dev;
+	struct uwb_rc_evt_ie_rcv *iercv;
+	size_t iesize;
+
+	/* Is there enough data to decode it? */
+	if (evt->notif.size < sizeof(*iercv)) {
+		dev_err(dev, "IE Received notification: Not enough data to "
+			"decode (%zu vs %zu bytes needed)\n",
+			evt->notif.size, sizeof(*iercv));
+		goto error;
+	}
+	iercv = container_of(evt->notif.rceb, struct uwb_rc_evt_ie_rcv, rceb);
+	iesize = le16_to_cpu(iercv->wIELength);
+
+	dev_dbg(dev, "IE received, element ID=%d\n", iercv->IEData[0]);
+
+	if (iercv->IEData[0] == UWB_RELINQUISH_REQUEST_IE) {
+		dev_warn(dev, "unhandled Relinquish Request IE\n");
+	}
+
+	return 0;
+error:
+	return result;
+}
diff --git a/drivers/uwb/uwb-internal.h b/drivers/uwb/uwb-internal.h
index 983ebc4dd8d5..031e8a885681 100644
--- a/drivers/uwb/uwb-internal.h
+++ b/drivers/uwb/uwb-internal.h
@@ -167,6 +167,7 @@ extern void uwbd_event_queue(struct uwb_event *);
 void uwbd_flush(struct uwb_rc *rc);
 
 /* UWB event handlers */
+extern int uwbd_evt_handle_rc_ie_rcv(struct uwb_event *);
 extern int uwbd_evt_handle_rc_beacon(struct uwb_event *);
 extern int uwbd_evt_handle_rc_beacon_size(struct uwb_event *);
 extern int uwbd_evt_handle_rc_bpoie_change(struct uwb_event *);
diff --git a/drivers/uwb/uwbd.c b/drivers/uwb/uwbd.c
index 78908416e42c..f75113571f4a 100644
--- a/drivers/uwb/uwbd.c
+++ b/drivers/uwb/uwbd.c
@@ -104,6 +104,10 @@ struct uwbd_event {
 /** Table of handlers for and properties of the UWBD Radio Control Events */
 static
 struct uwbd_event uwbd_events[] = {
+	[UWB_RC_EVT_IE_RCV] = {
+		.handler = uwbd_evt_handle_rc_ie_rcv,
+		.name = "IE_RECEIVED"
+	},
 	[UWB_RC_EVT_BEACON] = {
 		.handler = uwbd_evt_handle_rc_beacon,
 		.name = "BEACON_RECEIVED"
diff --git a/include/linux/uwb/spec.h b/include/linux/uwb/spec.h
index 198c15f8e251..a30436ea53aa 100644
--- a/include/linux/uwb/spec.h
+++ b/include/linux/uwb/spec.h
@@ -200,6 +200,12 @@ enum uwb_drp_reason {
 	UWB_DRP_REASON_MODIFIED,
 };
 
+/** Relinquish Request Reason Codes ([ECMA-368] table 113) */
+enum uwb_relinquish_req_reason {
+	UWB_RELINQUISH_REQ_REASON_NON_SPECIFIC = 0,
+	UWB_RELINQUISH_REQ_REASON_OVER_ALLOCATION,
+};
+
 /**
  *  DRP Notification Reason Codes (WHCI 0.95 [3.1.4.9])
  */
@@ -252,6 +258,7 @@ enum uwb_ie {
 	UWB_APP_SPEC_PROBE_IE = 15,
 	UWB_IDENTIFICATION_IE = 19,
 	UWB_MASTER_KEY_ID_IE = 20,
+	UWB_RELINQUISH_REQUEST_IE = 21,
 	UWB_IE_WLP = 250, /* WiMedia Logical Link Control Protocol WLP 0.99 */
 	UWB_APP_SPEC_IE = 255,
 };
@@ -365,6 +372,27 @@ struct uwb_ie_drp_avail {
 	DECLARE_BITMAP(bmp, UWB_NUM_MAS);
 } __attribute__((packed));
 
+/* Relinqish Request IE ([ECMA-368] section 16.8.19). */
+struct uwb_relinquish_request_ie {
+        struct uwb_ie_hdr       hdr;
+        __le16                  relinquish_req_control;
+        struct uwb_dev_addr     dev_addr;
+        struct uwb_drp_alloc    allocs[];
+} __attribute__((packed));
+
+static inline int uwb_ie_relinquish_req_reason_code(struct uwb_relinquish_request_ie *ie)
+{
+	return (le16_to_cpu(ie->relinquish_req_control) >> 0) & 0xf;
+}
+
+static inline void uwb_ie_relinquish_req_set_reason_code(struct uwb_relinquish_request_ie *ie,
+							 int reason_code)
+{
+	u16 ctrl = le16_to_cpu(ie->relinquish_req_control);
+	ctrl = (ctrl & ~(0xf << 0)) | (reason_code << 0);
+	ie->relinquish_req_control = cpu_to_le16(ctrl);
+}
+
 /**
  * The Vendor ID is set to an OUI that indicates the vendor of the device.
  * ECMA-368 [16.8.10]
-- 
cgit v1.2.3


From 6d5a681dfb583b2f1eefe7cd5505419ca2d4d6c8 Mon Sep 17 00:00:00 2001
From: Stefano Panella <stefano.panella@csr.com>
Date: Tue, 4 Nov 2008 14:24:57 +0000
Subject: uwb: add commands to add/remove IEs to the debug interface

Add the commands UWB_DBG_CMD_IE_ADD and UWB_DBG_CMD_IE_RM to the debug
interface and make them call uwb_rc_ie_add() and uwb_rc_ie_rm().

Signed-off-by: Stefano Panella <stefano.panella@csr.com>
Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/uwb/uwb-debug.c       | 20 +++++++++++++++++++-
 include/linux/uwb/debug-cmd.h |  9 +++++++++
 2 files changed, 28 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/uwb/uwb-debug.c b/drivers/uwb/uwb-debug.c
index 6db641e45313..88e6ac713817 100644
--- a/drivers/uwb/uwb-debug.c
+++ b/drivers/uwb/uwb-debug.c
@@ -168,6 +168,18 @@ static int cmd_rsv_terminate(struct uwb_rc *rc,
 	return 0;
 }
 
+static int cmd_ie_add(struct uwb_rc *rc, struct uwb_dbg_cmd_ie *ie_to_add)
+{
+	return uwb_rc_ie_add(rc,
+			     (const struct uwb_ie_hdr *) ie_to_add->data,
+			     ie_to_add->len);
+}
+
+static int cmd_ie_rm(struct uwb_rc *rc, struct uwb_dbg_cmd_ie *ie_to_rm)
+{
+	return uwb_rc_ie_rm(rc, ie_to_rm->data[0]);
+}
+
 static int command_open(struct inode *inode, struct file *file)
 {
 	file->private_data = inode->i_private;
@@ -195,6 +207,12 @@ static ssize_t command_write(struct file *file, const char __user *buf,
 	case UWB_DBG_CMD_RSV_TERMINATE:
 		ret = cmd_rsv_terminate(rc, &cmd.rsv_terminate);
 		break;
+	case UWB_DBG_CMD_IE_ADD:
+		ret = cmd_ie_add(rc, &cmd.ie_add);
+		break;
+	case UWB_DBG_CMD_IE_RM:
+		ret = cmd_ie_rm(rc, &cmd.ie_rm);
+		break;
 	default:
 		return -EINVAL;
 	}
@@ -332,7 +350,7 @@ void uwb_dbg_add_rc(struct uwb_rc *rc)
 }
 
 /**
- * uwb_dbg_add_rc - remove a radio controller's debug interface
+ * uwb_dbg_del_rc - remove a radio controller's debug interface
  * @rc: the radio controller
  */
 void uwb_dbg_del_rc(struct uwb_rc *rc)
diff --git a/include/linux/uwb/debug-cmd.h b/include/linux/uwb/debug-cmd.h
index 1141f41bab5c..6a16566f0221 100644
--- a/include/linux/uwb/debug-cmd.h
+++ b/include/linux/uwb/debug-cmd.h
@@ -32,6 +32,8 @@
 enum uwb_dbg_cmd_type {
 	UWB_DBG_CMD_RSV_ESTABLISH = 1,
 	UWB_DBG_CMD_RSV_TERMINATE = 2,
+	UWB_DBG_CMD_IE_ADD = 3,
+	UWB_DBG_CMD_IE_RM = 4,
 };
 
 struct uwb_dbg_cmd_rsv_establish {
@@ -46,11 +48,18 @@ struct uwb_dbg_cmd_rsv_terminate {
 	int index;
 };
 
+struct uwb_dbg_cmd_ie {
+	__u8 data[128];
+	int len;
+};
+
 struct uwb_dbg_cmd {
 	__u32 type;
 	union {
 		struct uwb_dbg_cmd_rsv_establish rsv_establish;
 		struct uwb_dbg_cmd_rsv_terminate rsv_terminate;
+		struct uwb_dbg_cmd_ie ie_add;
+		struct uwb_dbg_cmd_ie ie_rm;
 	};
 };
 
-- 
cgit v1.2.3


From fec1a5932f16c0eb1b3f5ca2e18d81d860924088 Mon Sep 17 00:00:00 2001
From: Stefano Panella <stefano.panella@csr.com>
Date: Tue, 4 Nov 2008 15:39:08 +0000
Subject: uwb: per-radio controller event thread and beacon cache

Use an event thread per-radio controller so processing events from one
radio controller doesn't delay another.

A radio controller shouldn't have information on devices seen by a
different radio controller (they may be on different channels) so make the
beacon cache per-radio controller.

Signed-off-by: Stefano Panella <stefano.panella@csr.com>
Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/uwb/beacon.c       | 61 ++++++++++++++---------------
 drivers/uwb/driver.c       |  2 -
 drivers/uwb/lc-rc.c        | 21 ++++++----
 drivers/uwb/uwb-internal.h | 20 ++--------
 drivers/uwb/uwbd.c         | 98 +++++++++++++++-------------------------------
 include/linux/uwb.h        | 20 ++++++++++
 6 files changed, 100 insertions(+), 122 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/uwb/beacon.c b/drivers/uwb/beacon.c
index ad823987cede..d9f2a8acc593 100644
--- a/drivers/uwb/beacon.c
+++ b/drivers/uwb/beacon.c
@@ -168,12 +168,6 @@ out_up:
  * FIXME: use something faster for search than a list
  */
 
-struct uwb_beca uwb_beca = {
-	.list = LIST_HEAD_INIT(uwb_beca.list),
-	.mutex = __MUTEX_INITIALIZER(uwb_beca.mutex)
-};
-
-
 void uwb_bce_kfree(struct kref *_bce)
 {
 	struct uwb_beca_e *bce = container_of(_bce, struct uwb_beca_e, refcnt);
@@ -185,10 +179,11 @@ void uwb_bce_kfree(struct kref *_bce)
 
 /* Find a beacon by dev addr in the cache */
 static
-struct uwb_beca_e *__uwb_beca_find_bydev(const struct uwb_dev_addr *dev_addr)
+struct uwb_beca_e *__uwb_beca_find_bydev(struct uwb_rc *rc,
+					 const struct uwb_dev_addr *dev_addr)
 {
 	struct uwb_beca_e *bce, *next;
-	list_for_each_entry_safe(bce, next, &uwb_beca.list, node) {
+	list_for_each_entry_safe(bce, next, &rc->uwb_beca.list, node) {
 		d_printf(6, NULL, "looking for addr %02x:%02x in %02x:%02x\n",
 			 dev_addr->data[0], dev_addr->data[1],
 			 bce->dev_addr.data[0], bce->dev_addr.data[1]);
@@ -202,10 +197,11 @@ out:
 
 /* Find a beacon by dev addr in the cache */
 static
-struct uwb_beca_e *__uwb_beca_find_bymac(const struct uwb_mac_addr *mac_addr)
+struct uwb_beca_e *__uwb_beca_find_bymac(struct uwb_rc *rc, 
+					 const struct uwb_mac_addr *mac_addr)
 {
 	struct uwb_beca_e *bce, *next;
-	list_for_each_entry_safe(bce, next, &uwb_beca.list, node) {
+	list_for_each_entry_safe(bce, next, &rc->uwb_beca.list, node) {
 		if (!memcmp(bce->mac_addr, mac_addr->data,
 			    sizeof(struct uwb_mac_addr)))
 			goto out;
@@ -229,11 +225,11 @@ struct uwb_dev *uwb_dev_get_by_devaddr(struct uwb_rc *rc,
 	struct uwb_dev *found = NULL;
 	struct uwb_beca_e *bce;
 
-	mutex_lock(&uwb_beca.mutex);
-	bce = __uwb_beca_find_bydev(devaddr);
+	mutex_lock(&rc->uwb_beca.mutex);
+	bce = __uwb_beca_find_bydev(rc, devaddr);
 	if (bce)
 		found = uwb_dev_try_get(rc, bce->uwb_dev);
-	mutex_unlock(&uwb_beca.mutex);
+	mutex_unlock(&rc->uwb_beca.mutex);
 
 	return found;
 }
@@ -249,11 +245,11 @@ struct uwb_dev *uwb_dev_get_by_macaddr(struct uwb_rc *rc,
 	struct uwb_dev *found = NULL;
 	struct uwb_beca_e *bce;
 
-	mutex_lock(&uwb_beca.mutex);
-	bce = __uwb_beca_find_bymac(macaddr);
+	mutex_lock(&rc->uwb_beca.mutex);
+	bce = __uwb_beca_find_bymac(rc, macaddr);
 	if (bce)
 		found = uwb_dev_try_get(rc, bce->uwb_dev);
-	mutex_unlock(&uwb_beca.mutex);
+	mutex_unlock(&rc->uwb_beca.mutex);
 
 	return found;
 }
@@ -274,7 +270,9 @@ static void uwb_beca_e_init(struct uwb_beca_e *bce)
  * @bf:         Beacon frame (part of b, really)
  * @ts_jiffies: Timestamp (in jiffies) when the beacon was received
  */
-struct uwb_beca_e *__uwb_beca_add(struct uwb_rc_evt_beacon *be,
+static
+struct uwb_beca_e *__uwb_beca_add(struct uwb_rc *rc,
+				  struct uwb_rc_evt_beacon *be,
 				  struct uwb_beacon_frame *bf,
 				  unsigned long ts_jiffies)
 {
@@ -286,7 +284,7 @@ struct uwb_beca_e *__uwb_beca_add(struct uwb_rc_evt_beacon *be,
 	uwb_beca_e_init(bce);
 	bce->ts_jiffies = ts_jiffies;
 	bce->uwb_dev = NULL;
-	list_add(&bce->node, &uwb_beca.list);
+	list_add(&bce->node, &rc->uwb_beca.list);
 	return bce;
 }
 
@@ -295,13 +293,13 @@ struct uwb_beca_e *__uwb_beca_add(struct uwb_rc_evt_beacon *be,
  *
  * Remove associated devicest too.
  */
-void uwb_beca_purge(void)
+void uwb_beca_purge(struct uwb_rc *rc)
 {
 	struct uwb_beca_e *bce, *next;
 	unsigned long expires;
 
-	mutex_lock(&uwb_beca.mutex);
-	list_for_each_entry_safe(bce, next, &uwb_beca.list, node) {
+	mutex_lock(&rc->uwb_beca.mutex);
+	list_for_each_entry_safe(bce, next, &rc->uwb_beca.list, node) {
 		expires = bce->ts_jiffies + msecs_to_jiffies(beacon_timeout_ms);
 		if (time_after(jiffies, expires)) {
 			uwbd_dev_offair(bce);
@@ -309,19 +307,20 @@ void uwb_beca_purge(void)
 			uwb_bce_put(bce);
 		}
 	}
-	mutex_unlock(&uwb_beca.mutex);
+	mutex_unlock(&rc->uwb_beca.mutex);
 }
 
 /* Clean up the whole beacon cache. Called on shutdown */
-void uwb_beca_release(void)
+void uwb_beca_release(struct uwb_rc *rc)
 {
 	struct uwb_beca_e *bce, *next;
-	mutex_lock(&uwb_beca.mutex);
-	list_for_each_entry_safe(bce, next, &uwb_beca.list, node) {
+
+	mutex_lock(&rc->uwb_beca.mutex);
+	list_for_each_entry_safe(bce, next, &rc->uwb_beca.list, node) {
 		list_del(&bce->node);
 		uwb_bce_put(bce);
 	}
-	mutex_unlock(&uwb_beca.mutex);
+	mutex_unlock(&rc->uwb_beca.mutex);
 }
 
 static void uwb_beacon_print(struct uwb_rc *rc, struct uwb_rc_evt_beacon *be,
@@ -437,18 +436,18 @@ int uwbd_evt_handle_rc_beacon(struct uwb_event *evt)
 	if (uwb_mac_addr_bcast(&bf->Device_Identifier))
 		return 0;
 
-	mutex_lock(&uwb_beca.mutex);
-	bce = __uwb_beca_find_bymac(&bf->Device_Identifier);
+	mutex_lock(&rc->uwb_beca.mutex);
+	bce = __uwb_beca_find_bymac(rc, &bf->Device_Identifier);
 	if (bce == NULL) {
 		/* Not in there, a new device is pinging */
 		uwb_beacon_print(evt->rc, be, bf);
-		bce = __uwb_beca_add(be, bf, evt->ts_jiffies);
+		bce = __uwb_beca_add(rc, be, bf, evt->ts_jiffies);
 		if (bce == NULL) {
-			mutex_unlock(&uwb_beca.mutex);
+			mutex_unlock(&rc->uwb_beca.mutex);
 			return -ENOMEM;
 		}
 	}
-	mutex_unlock(&uwb_beca.mutex);
+	mutex_unlock(&rc->uwb_beca.mutex);
 
 	mutex_lock(&bce->mutex);
 	/* purge old beacon data */
diff --git a/drivers/uwb/driver.c b/drivers/uwb/driver.c
index 521cdeb84971..f57c26580de2 100644
--- a/drivers/uwb/driver.c
+++ b/drivers/uwb/driver.c
@@ -118,7 +118,6 @@ static int __init uwb_subsys_init(void)
 	result = class_register(&uwb_rc_class);
 	if (result < 0)
 		goto error_uwb_rc_class_register;
-	uwbd_start();
 	uwb_dbg_init();
 	return 0;
 
@@ -132,7 +131,6 @@ module_init(uwb_subsys_init);
 static void __exit uwb_subsys_exit(void)
 {
 	uwb_dbg_exit();
-	uwbd_stop();
 	class_unregister(&uwb_rc_class);
 	uwb_est_destroy();
 	return;
diff --git a/drivers/uwb/lc-rc.c b/drivers/uwb/lc-rc.c
index 1129e8767b58..38e3d57ec8f7 100644
--- a/drivers/uwb/lc-rc.c
+++ b/drivers/uwb/lc-rc.c
@@ -36,8 +36,6 @@
 #include <linux/etherdevice.h>
 #include <linux/usb.h>
 
-#define D_LOCAL 1
-#include <linux/uwb/debug.h>
 #include "uwb-internal.h"
 
 static int uwb_rc_index_match(struct device *dev, void *data)
@@ -83,7 +81,6 @@ static void uwb_rc_sys_release(struct device *dev)
 
 	uwb_rc_neh_destroy(rc);
 	uwb_rc_ie_release(rc);
-	d_printf(1, dev, "freed uwb_rc %p\n", rc);
 	kfree(rc);
 }
 
@@ -100,6 +97,8 @@ void uwb_rc_init(struct uwb_rc *rc)
 	rc->scan_type = UWB_SCAN_DISABLED;
 	INIT_LIST_HEAD(&rc->notifs_chain.list);
 	mutex_init(&rc->notifs_chain.mutex);
+	INIT_LIST_HEAD(&rc->uwb_beca.list);
+	mutex_init(&rc->uwb_beca.mutex);
 	uwb_drp_avail_init(rc);
 	uwb_rc_ie_init(rc);
 	uwb_rsv_init(rc);
@@ -250,6 +249,12 @@ int uwb_rc_add(struct uwb_rc *rc, struct device *parent_dev, void *priv)
 
 	rc->priv = priv;
 
+	init_waitqueue_head(&rc->uwbd.wq);
+	INIT_LIST_HEAD(&rc->uwbd.event_list);
+	spin_lock_init(&rc->uwbd.event_list_lock);
+
+	uwbd_start(rc);
+
 	result = rc->start(rc);
 	if (result < 0)
 		goto error_rc_start;
@@ -284,7 +289,7 @@ error_sys_add:
 error_dev_add:
 error_rc_setup:
 	rc->stop(rc);
-	uwbd_flush(rc);
+	uwbd_stop(rc);
 error_rc_start:
 	return result;
 }
@@ -315,16 +320,18 @@ void uwb_rc_rm(struct uwb_rc *rc)
 	uwb_rc_reset(rc);
 
 	rc->stop(rc);
-	uwbd_flush(rc);
+
+	uwbd_stop(rc);
 
 	uwb_dev_lock(&rc->uwb_dev);
 	rc->priv = NULL;
 	rc->cmd = NULL;
 	uwb_dev_unlock(&rc->uwb_dev);
-	mutex_lock(&uwb_beca.mutex);
+	mutex_lock(&rc->uwb_beca.mutex);
 	uwb_dev_for_each(rc, uwb_dev_offair_helper, NULL);
 	__uwb_rc_sys_rm(rc);
-	mutex_unlock(&uwb_beca.mutex);
+	mutex_unlock(&rc->uwb_beca.mutex);
+ 	uwb_beca_release(rc);
 	uwb_dev_rm(&rc->uwb_dev);
 }
 EXPORT_SYMBOL_GPL(uwb_rc_rm);
diff --git a/drivers/uwb/uwb-internal.h b/drivers/uwb/uwb-internal.h
index 031e8a885681..4c2449679978 100644
--- a/drivers/uwb/uwb-internal.h
+++ b/drivers/uwb/uwb-internal.h
@@ -160,8 +160,8 @@ struct uwb_event {
 	};
 };
 
-extern void uwbd_start(void);
-extern void uwbd_stop(void);
+extern void uwbd_start(struct uwb_rc *rc);
+extern void uwbd_stop(struct uwb_rc *rc);
 extern struct uwb_event *uwb_event_alloc(size_t, gfp_t gfp_mask);
 extern void uwbd_event_queue(struct uwb_event *);
 void uwbd_flush(struct uwb_rc *rc);
@@ -194,15 +194,6 @@ int uwbd_evt_handle_rc_dev_addr_conflict(struct uwb_event *evt);
 
 extern unsigned long beacon_timeout_ms;
 
-/** Beacon cache list */
-struct uwb_beca {
-	struct list_head list;
-	size_t entries;
-	struct mutex mutex;
-};
-
-extern struct uwb_beca uwb_beca;
-
 /**
  * Beacon cache entry
  *
@@ -229,9 +220,6 @@ struct uwb_beca_e {
 struct uwb_beacon_frame;
 extern ssize_t uwb_bce_print_IEs(struct uwb_dev *, struct uwb_beca_e *,
 				 char *, size_t);
-extern struct uwb_beca_e *__uwb_beca_add(struct uwb_rc_evt_beacon *,
-					 struct uwb_beacon_frame *,
-					 unsigned long);
 
 extern void uwb_bce_kfree(struct kref *_bce);
 static inline void uwb_bce_get(struct uwb_beca_e *bce)
@@ -242,8 +230,8 @@ static inline void uwb_bce_put(struct uwb_beca_e *bce)
 {
 	kref_put(&bce->refcnt, uwb_bce_kfree);
 }
-extern void uwb_beca_purge(void);
-extern void uwb_beca_release(void);
+extern void uwb_beca_purge(struct uwb_rc *rc);
+extern void uwb_beca_release(struct uwb_rc *rc);
 
 struct uwb_dev *uwb_dev_get_by_devaddr(struct uwb_rc *rc,
 				       const struct uwb_dev_addr *devaddr);
diff --git a/drivers/uwb/uwbd.c b/drivers/uwb/uwbd.c
index f75113571f4a..ec42ce92dbce 100644
--- a/drivers/uwb/uwbd.c
+++ b/drivers/uwb/uwbd.c
@@ -170,8 +170,6 @@ static const struct uwbd_event uwbd_message_handlers[] = {
 	},
 };
 
-static DEFINE_MUTEX(uwbd_event_mutex);
-
 /**
  * Handle an URC event passed to the UWB Daemon
  *
@@ -235,19 +233,10 @@ static void uwbd_event_handle_message(struct uwb_event *evt)
 		return;
 	}
 
-	/* If this is a reset event we need to drop the
-	 * uwbd_event_mutex or it deadlocks when the reset handler
-	 * attempts to flush the uwbd events. */
-	if (evt->message == UWB_EVT_MSG_RESET)
-		mutex_unlock(&uwbd_event_mutex);
-
 	result = uwbd_message_handlers[evt->message].handler(evt);
 	if (result < 0)
 		dev_err(&rc->uwb_dev.dev, "UWBD: '%s' message failed: %d\n",
 			uwbd_message_handlers[evt->message].name, result);
-
-	if (evt->message == UWB_EVT_MSG_RESET)
-		mutex_lock(&uwbd_event_mutex);
 }
 
 static void uwbd_event_handle(struct uwb_event *evt)
@@ -275,20 +264,6 @@ static void uwbd_event_handle(struct uwb_event *evt)
 
 	__uwb_rc_put(rc);	/* for the __uwb_rc_get() in uwb_rc_notif_cb() */
 }
-/* The UWB Daemon */
-
-
-/** Daemon's PID: used to decide if we can queue or not */
-static int uwbd_pid;
-/** Daemon's task struct for managing the kthread */
-static struct task_struct *uwbd_task;
-/** Daemon's waitqueue for waiting for new events */
-static DECLARE_WAIT_QUEUE_HEAD(uwbd_wq);
-/** Daemon's list of events; we queue/dequeue here */
-static struct list_head uwbd_event_list = LIST_HEAD_INIT(uwbd_event_list);
-/** Daemon's list lock to protect concurent access */
-static DEFINE_SPINLOCK(uwbd_event_list_lock);
-
 
 /**
  * UWB Daemon
@@ -302,65 +277,58 @@ static DEFINE_SPINLOCK(uwbd_event_list_lock);
  * FIXME: should change so we don't have a 1HZ timer all the time, but
  *        only if there are devices.
  */
-static int uwbd(void *unused)
+static int uwbd(void *param)
 {
+	struct uwb_rc *rc = param;
 	unsigned long flags;
-	struct list_head list = LIST_HEAD_INIT(list);
-	struct uwb_event *evt, *nxt;
+	struct uwb_event *evt;
 	int should_stop = 0;
+
 	while (1) {
 		wait_event_interruptible_timeout(
-			uwbd_wq,
-			!list_empty(&uwbd_event_list)
+			rc->uwbd.wq,
+			!list_empty(&rc->uwbd.event_list)
 			  || (should_stop = kthread_should_stop()),
 			HZ);
 		if (should_stop)
 			break;
 		try_to_freeze();
 
-		mutex_lock(&uwbd_event_mutex);
-		spin_lock_irqsave(&uwbd_event_list_lock, flags);
-		list_splice_init(&uwbd_event_list, &list);
-		spin_unlock_irqrestore(&uwbd_event_list_lock, flags);
-		list_for_each_entry_safe(evt, nxt, &list, list_node) {
+		spin_lock_irqsave(&rc->uwbd.event_list_lock, flags);
+		if (!list_empty(&rc->uwbd.event_list)) {
+			evt = list_first_entry(&rc->uwbd.event_list, struct uwb_event, list_node);
 			list_del(&evt->list_node);
+		} else
+			evt = NULL;
+		spin_unlock_irqrestore(&rc->uwbd.event_list_lock, flags);
+
+		if (evt) {
 			uwbd_event_handle(evt);
 			kfree(evt);
 		}
-		mutex_unlock(&uwbd_event_mutex);
 
-		uwb_beca_purge();	/* Purge devices that left */
+		uwb_beca_purge(rc);	/* Purge devices that left */
 	}
 	return 0;
 }
 
 
 /** Start the UWB daemon */
-void uwbd_start(void)
+void uwbd_start(struct uwb_rc *rc)
 {
-	uwbd_task = kthread_run(uwbd, NULL, "uwbd");
-	if (uwbd_task == NULL)
+	rc->uwbd.task = kthread_run(uwbd, rc, "uwbd");
+	if (rc->uwbd.task == NULL)
 		printk(KERN_ERR "UWB: Cannot start management daemon; "
 		       "UWB won't work\n");
 	else
-		uwbd_pid = uwbd_task->pid;
+		rc->uwbd.pid = rc->uwbd.task->pid;
 }
 
 /* Stop the UWB daemon and free any unprocessed events */
-void uwbd_stop(void)
+void uwbd_stop(struct uwb_rc *rc)
 {
-	unsigned long flags;
-	struct uwb_event *evt, *nxt;
-	kthread_stop(uwbd_task);
-	spin_lock_irqsave(&uwbd_event_list_lock, flags);
-	uwbd_pid = 0;
-	list_for_each_entry_safe(evt, nxt, &uwbd_event_list, list_node) {
-		if (evt->type == UWB_EVT_TYPE_NOTIF)
-			kfree(evt->notif.rceb);
-		kfree(evt);
-	}
-	spin_unlock_irqrestore(&uwbd_event_list_lock, flags);
-	uwb_beca_release();
+	kthread_stop(rc->uwbd.task);
+	uwbd_flush(rc);
 }
 
 /*
@@ -377,18 +345,20 @@ void uwbd_stop(void)
  */
 void uwbd_event_queue(struct uwb_event *evt)
 {
+	struct uwb_rc *rc = evt->rc;
 	unsigned long flags;
-	spin_lock_irqsave(&uwbd_event_list_lock, flags);
-	if (uwbd_pid != 0) {
-		list_add(&evt->list_node, &uwbd_event_list);
-		wake_up_all(&uwbd_wq);
+
+	spin_lock_irqsave(&rc->uwbd.event_list_lock, flags);
+	if (rc->uwbd.pid != 0) {
+		list_add(&evt->list_node, &rc->uwbd.event_list);
+		wake_up_all(&rc->uwbd.wq);
 	} else {
 		__uwb_rc_put(evt->rc);
 		if (evt->type == UWB_EVT_TYPE_NOTIF)
 			kfree(evt->notif.rceb);
 		kfree(evt);
 	}
-	spin_unlock_irqrestore(&uwbd_event_list_lock, flags);
+	spin_unlock_irqrestore(&rc->uwbd.event_list_lock, flags);
 	return;
 }
 
@@ -396,10 +366,8 @@ void uwbd_flush(struct uwb_rc *rc)
 {
 	struct uwb_event *evt, *nxt;
 
-	mutex_lock(&uwbd_event_mutex);
-
-	spin_lock_irq(&uwbd_event_list_lock);
-	list_for_each_entry_safe(evt, nxt, &uwbd_event_list, list_node) {
+	spin_lock_irq(&rc->uwbd.event_list_lock);
+	list_for_each_entry_safe(evt, nxt, &rc->uwbd.event_list, list_node) {
 		if (evt->rc == rc) {
 			__uwb_rc_put(rc);
 			list_del(&evt->list_node);
@@ -408,7 +376,5 @@ void uwbd_flush(struct uwb_rc *rc)
 			kfree(evt);
 		}
 	}
-	spin_unlock_irq(&uwbd_event_list_lock);
-
-	mutex_unlock(&uwbd_event_mutex);
+	spin_unlock_irq(&rc->uwbd.event_list_lock);
 }
diff --git a/include/linux/uwb.h b/include/linux/uwb.h
index 6d93f54b8879..881f0c5b6d28 100644
--- a/include/linux/uwb.h
+++ b/include/linux/uwb.h
@@ -30,6 +30,7 @@
 #include <linux/device.h>
 #include <linux/mutex.h>
 #include <linux/timer.h>
+#include <linux/wait.h>
 #include <linux/workqueue.h>
 #include <linux/uwb/spec.h>
 
@@ -86,6 +87,22 @@ struct uwb_notifs_chain {
 	struct mutex mutex;
 };
 
+/* Beacon cache list */
+struct uwb_beca {
+	struct list_head list;
+	size_t entries;
+	struct mutex mutex;
+};
+
+/* Event handling thread. */
+struct uwbd {
+	int pid;
+	struct task_struct *task;
+	wait_queue_head_t wq;
+	struct list_head event_list;
+	spinlock_t event_list_lock;
+};
+
 /**
  * struct uwb_mas_bm - a bitmap of all MAS in a superframe
  * @bm: a bitmap of length #UWB_NUM_MAS
@@ -342,6 +359,9 @@ struct uwb_rc {
 	enum uwb_scan_type scan_type:3;
 	unsigned ready:1;
 	struct uwb_notifs_chain notifs_chain;
+	struct uwb_beca uwb_beca;
+
+	struct uwbd uwbd;
 
 	struct uwb_drp_avail drp_avail;
 	struct list_head reservations;
-- 
cgit v1.2.3


From 171bbfbeab7730031eec8025341401fabe540bd5 Mon Sep 17 00:00:00 2001
From: Mark Fasheh <mfasheh@suse.com>
Date: Tue, 25 Nov 2008 17:42:31 -0500
Subject: jbd2: Add BH_JBDPrivateStart

Add this so that file systems using JBD2 can safely allocate unused b_state
bits.

In this case, we add it so that Ocfs2 can define a single bit for tracking
the validation state of a buffer.

Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 include/linux/jbd2.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index c7d106ef22e2..f36645745489 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -329,6 +329,7 @@ enum jbd_state_bits {
 	BH_State,		/* Pins most journal_head state */
 	BH_JournalHead,		/* Pins bh->b_private and jh->b_bh */
 	BH_Unshadow,		/* Dummy bit, for BJ_Shadow wakeup filtering */
+	BH_JBDPrivateStart,	/* First bit available for private use by FS */
 };
 
 BUFFER_FNS(JBD, jbd)
-- 
cgit v1.2.3


From e07f7183a486cf9783d1f8c9d2997b5b39eeb2d4 Mon Sep 17 00:00:00 2001
From: Josef Bacik <jbacik@redhat.com>
Date: Wed, 26 Nov 2008 01:14:26 -0500
Subject: jbd2: improve jbd2 fsync batching

This patch removes the static sleep time in favor of a more self
optimizing approach where we measure the average amount of time it
takes to commit a transaction to disk and the ammount of time a
transaction has been running.  If somebody does a sync write or an
fsync() traditionally we would sleep for 1 jiffies, which depending on
the value of HZ could be a significant amount of time compared to how
long it takes to commit a transaction to the underlying storage.  With
this patch instead of sleeping for a jiffie, we check to see if the
amount of time this transaction has been running is less than the
average commit time, and if it is we sleep for the delta using
schedule_hrtimeout to give us a higher precision sleep time.  This
greatly benefits high end storage where you could end up sleeping for
longer than it takes to commit the transaction and therefore sitting
idle instead of allowing the transaction to be committed by keeping
the sleep time to a minimum so you are sure to always be doing
something.

Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 fs/jbd2/commit.c      | 14 +++++++++++++
 fs/jbd2/transaction.c | 58 ++++++++++++++++++++++++++++++++++++++-------------
 include/linux/jbd2.h  | 15 +++++++++++++
 3 files changed, 73 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index 6393fd0d804e..f22d1828ea85 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -355,6 +355,8 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	int flags;
 	int err;
 	unsigned long long blocknr;
+	ktime_t start_time;
+	u64 commit_time;
 	char *tagp = NULL;
 	journal_header_t *header;
 	journal_block_tag_t *tag = NULL;
@@ -481,6 +483,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	commit_transaction->t_state = T_FLUSH;
 	journal->j_committing_transaction = commit_transaction;
 	journal->j_running_transaction = NULL;
+	start_time = ktime_get();
 	commit_transaction->t_log_start = journal->j_head;
 	wake_up(&journal->j_wait_transaction_locked);
 	spin_unlock(&journal->j_state_lock);
@@ -995,6 +998,17 @@ restart_loop:
 	J_ASSERT(commit_transaction == journal->j_committing_transaction);
 	journal->j_commit_sequence = commit_transaction->t_tid;
 	journal->j_committing_transaction = NULL;
+	commit_time = ktime_to_ns(ktime_sub(ktime_get(), start_time));
+
+	/*
+	 * weight the commit time higher than the average time so we don't
+	 * react too strongly to vast changes in the commit time
+	 */
+	if (likely(journal->j_average_commit_time))
+		journal->j_average_commit_time = (commit_time +
+				journal->j_average_commit_time*3) / 4;
+	else
+		journal->j_average_commit_time = commit_time;
 	spin_unlock(&journal->j_state_lock);
 
 	if (journal->j_commit_callback)
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index 39b7805a599a..13dcbc990f41 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -25,6 +25,7 @@
 #include <linux/timer.h>
 #include <linux/mm.h>
 #include <linux/highmem.h>
+#include <linux/hrtimer.h>
 
 static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh);
 
@@ -48,6 +49,7 @@ jbd2_get_transaction(journal_t *journal, transaction_t *transaction)
 {
 	transaction->t_journal = journal;
 	transaction->t_state = T_RUNNING;
+	transaction->t_start_time = ktime_get();
 	transaction->t_tid = journal->j_transaction_sequence++;
 	transaction->t_expires = jiffies + journal->j_commit_interval;
 	spin_lock_init(&transaction->t_handle_lock);
@@ -1193,7 +1195,7 @@ int jbd2_journal_stop(handle_t *handle)
 {
 	transaction_t *transaction = handle->h_transaction;
 	journal_t *journal = transaction->t_journal;
-	int old_handle_count, err;
+	int err;
 	pid_t pid;
 
 	J_ASSERT(journal_current_handle() == handle);
@@ -1216,24 +1218,52 @@ int jbd2_journal_stop(handle_t *handle)
 	/*
 	 * Implement synchronous transaction batching.  If the handle
 	 * was synchronous, don't force a commit immediately.  Let's
-	 * yield and let another thread piggyback onto this transaction.
-	 * Keep doing that while new threads continue to arrive.
-	 * It doesn't cost much - we're about to run a commit and sleep
-	 * on IO anyway.  Speeds up many-threaded, many-dir operations
-	 * by 30x or more...
+	 * yield and let another thread piggyback onto this
+	 * transaction.  Keep doing that while new threads continue to
+	 * arrive.  It doesn't cost much - we're about to run a commit
+	 * and sleep on IO anyway.  Speeds up many-threaded, many-dir
+	 * operations by 30x or more...
+	 *
+	 * We try and optimize the sleep time against what the
+	 * underlying disk can do, instead of having a static sleep
+	 * time.  This is useful for the case where our storage is so
+	 * fast that it is more optimal to go ahead and force a flush
+	 * and wait for the transaction to be committed than it is to
+	 * wait for an arbitrary amount of time for new writers to
+	 * join the transaction.  We achieve this by measuring how
+	 * long it takes to commit a transaction, and compare it with
+	 * how long this transaction has been running, and if run time
+	 * < commit time then we sleep for the delta and commit.  This
+	 * greatly helps super fast disks that would see slowdowns as
+	 * more threads started doing fsyncs.
 	 *
-	 * But don't do this if this process was the most recent one to
-	 * perform a synchronous write.  We do this to detect the case where a
-	 * single process is doing a stream of sync writes.  No point in waiting
-	 * for joiners in that case.
+	 * But don't do this if this process was the most recent one
+	 * to perform a synchronous write.  We do this to detect the
+	 * case where a single process is doing a stream of sync
+	 * writes.  No point in waiting for joiners in that case.
 	 */
 	pid = current->pid;
 	if (handle->h_sync && journal->j_last_sync_writer != pid) {
+		u64 commit_time, trans_time;
+
 		journal->j_last_sync_writer = pid;
-		do {
-			old_handle_count = transaction->t_handle_count;
-			schedule_timeout_uninterruptible(1);
-		} while (old_handle_count != transaction->t_handle_count);
+
+		spin_lock(&journal->j_state_lock);
+		commit_time = journal->j_average_commit_time;
+		spin_unlock(&journal->j_state_lock);
+
+		trans_time = ktime_to_ns(ktime_sub(ktime_get(),
+						   transaction->t_start_time));
+
+		commit_time = min_t(u64, commit_time,
+				    1000*jiffies_to_usecs(1));
+
+		if (trans_time < commit_time) {
+			ktime_t expires = ktime_add_ns(ktime_get(),
+						       commit_time);
+			set_current_state(TASK_UNINTERRUPTIBLE);
+			schedule_hrtimeout(&expires, HRTIMER_MODE_ABS);
+		}
 	}
 
 	current->journal_info = NULL;
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index f36645745489..ab8cef130c28 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -637,6 +637,11 @@ struct transaction_s
 	 */
 	unsigned long		t_expires;
 
+	/*
+	 * When this transaction started, in nanoseconds [no locking]
+	 */
+	ktime_t			t_start_time;
+
 	/*
 	 * How many handles used this transaction? [t_handle_lock]
 	 */
@@ -939,8 +944,18 @@ struct journal_s
 	struct buffer_head	**j_wbuf;
 	int			j_wbufsize;
 
+	/*
+	 * this is the pid of hte last person to run a synchronous operation
+	 * through the journal
+	 */
 	pid_t			j_last_sync_writer;
 
+	/*
+	 * the average amount of time in nanoseconds it takes to commit a
+	 * transaction to disk. [j_state_lock]
+	 */
+	u64			j_average_commit_time;
+
 	/* This function is called when a transaction is closed */
 	void			(*j_commit_callback)(journal_t *,
 						     transaction_t *);
-- 
cgit v1.2.3


From 30773840c19cea60dcef39545960d541b1ac1cf8 Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Sat, 3 Jan 2009 20:27:38 -0500
Subject: ext4: add fsync batch tuning knobs

Add new mount options, min_batch_time and max_batch_time, which
controls how long the jbd2 layer should wait for additional filesystem
operations to get batched with a synchronous write transaction.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 Documentation/filesystems/ext4.txt | 29 +++++++++++++++++++++++
 fs/ext4/ext4.h                     |  7 ++++++
 fs/ext4/ext4_sb.h                  |  2 ++
 fs/ext4/super.c                    | 47 ++++++++++++++++++++++++++++++++------
 fs/jbd2/journal.c                  |  2 ++
 fs/jbd2/transaction.c              |  4 +++-
 include/linux/jbd2.h               |  8 +++++++
 7 files changed, 91 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt
index f75ab101c00a..e3fcbea3ec8c 100644
--- a/Documentation/filesystems/ext4.txt
+++ b/Documentation/filesystems/ext4.txt
@@ -283,6 +283,35 @@ delalloc	(*)	Deferring block allocation until write-out time.
 nodelalloc		Disable delayed allocation. Blocks are allocation
 			when data is copied from user to page cache.
 
+max_batch_time=usec	Maximum amount of time ext4 should wait for
+			additional filesystem operations to be batch
+			together with a synchronous write operation.
+			Since a synchronous write operation is going to
+			force a commit and then a wait for the I/O
+			complete, it doesn't cost much, and can be a
+			huge throughput win, we wait for a small amount
+			of time to see if any other transactions can
+			piggyback on the synchronous write.   The
+			algorithm used is designed to automatically tune
+			for the speed of the disk, by measuring the
+			amount of time (on average) that it takes to
+			finish committing a transaction.  Call this time
+			the "commit time".  If the time that the
+			transactoin has been running is less than the
+			commit time, ext4 will try sleeping for the
+			commit time to see if other operations will join
+			the transaction.   The commit time is capped by
+			the max_batch_time, which defaults to 15000us
+			(15ms).   This optimization can be turned off
+			entirely by setting max_batch_time to 0.
+
+min_batch_time=usec	This parameter sets the commit time (as
+			described above) to be at least min_batch_time.
+			It defaults to zero microseconds.  Increasing
+			this parameter may improve the throughput of
+			multi-threaded, synchronous workloads on very
+			fast disks, at the cost of increasing latency.
+
 Data Mode
 =========
 There are 3 different data modes:
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index ac8551e0b70a..9ba9fd6d14da 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -328,6 +328,7 @@ struct ext4_mount_options {
 	uid_t s_resuid;
 	gid_t s_resgid;
 	unsigned long s_commit_interval;
+	u32 s_min_batch_time, s_max_batch_time;
 #ifdef CONFIG_QUOTA
 	int s_jquota_fmt;
 	char *s_qf_names[MAXQUOTAS];
@@ -805,6 +806,12 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino)
 #define EXT4_DEFM_JMODE_ORDERED	0x0040
 #define EXT4_DEFM_JMODE_WBACK	0x0060
 
+/*
+ * Default journal batch times
+ */
+#define EXT4_DEF_MIN_BATCH_TIME	0
+#define EXT4_DEF_MAX_BATCH_TIME	15000 /* 15ms */
+
 /*
  * Structure of a directory entry
  */
diff --git a/fs/ext4/ext4_sb.h b/fs/ext4/ext4_sb.h
index 3db800f399a6..039b6ea1a042 100644
--- a/fs/ext4/ext4_sb.h
+++ b/fs/ext4/ext4_sb.h
@@ -74,6 +74,8 @@ struct ext4_sb_info {
 	struct journal_s *s_journal;
 	struct list_head s_orphan;
 	unsigned long s_commit_interval;
+	u32 s_max_batch_time;
+	u32 s_min_batch_time;
 	struct block_device *journal_bdev;
 #ifdef CONFIG_JBD2_DEBUG
 	struct timer_list turn_ro_timer;	/* For turning read-only (crash simulation) */
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index dc27d4c613c0..da377f9521bb 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -705,10 +705,19 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs)
 #endif
 	if (!test_opt(sb, RESERVATION))
 		seq_puts(seq, ",noreservation");
-	if (sbi->s_commit_interval) {
+	if (sbi->s_commit_interval != JBD2_DEFAULT_MAX_COMMIT_AGE*HZ) {
 		seq_printf(seq, ",commit=%u",
 			   (unsigned) (sbi->s_commit_interval / HZ));
 	}
+	if (sbi->s_min_batch_time != EXT4_DEF_MIN_BATCH_TIME) {
+		seq_printf(seq, ",min_batch_time=%u",
+			   (unsigned) sbi->s_min_batch_time);
+	}
+	if (sbi->s_max_batch_time != EXT4_DEF_MAX_BATCH_TIME) {
+		seq_printf(seq, ",max_batch_time=%u",
+			   (unsigned) sbi->s_min_batch_time);
+	}
+
 	/*
 	 * We're changing the default of barrier mount option, so
 	 * let's always display its mount state so it's clear what its
@@ -874,7 +883,8 @@ enum {
 	Opt_nouid32, Opt_debug, Opt_oldalloc, Opt_orlov,
 	Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl,
 	Opt_reservation, Opt_noreservation, Opt_noload, Opt_nobh, Opt_bh,
-	Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_journal_dev,
+	Opt_commit, Opt_min_batch_time, Opt_max_batch_time,
+	Opt_journal_update, Opt_journal_inum, Opt_journal_dev,
 	Opt_journal_checksum, Opt_journal_async_commit,
 	Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
 	Opt_data_err_abort, Opt_data_err_ignore,
@@ -913,6 +923,8 @@ static const match_table_t tokens = {
 	{Opt_nobh, "nobh"},
 	{Opt_bh, "bh"},
 	{Opt_commit, "commit=%u"},
+	{Opt_min_batch_time, "min_batch_time=%u"},
+	{Opt_max_batch_time, "max_batch_time=%u"},
 	{Opt_journal_update, "journal=update"},
 	{Opt_journal_inum, "journal=%u"},
 	{Opt_journal_dev, "journal_dev=%u"},
@@ -1131,6 +1143,22 @@ static int parse_options(char *options, struct super_block *sb,
 				option = JBD2_DEFAULT_MAX_COMMIT_AGE;
 			sbi->s_commit_interval = HZ * option;
 			break;
+		case Opt_max_batch_time:
+			if (match_int(&args[0], &option))
+				return 0;
+			if (option < 0)
+				return 0;
+			if (option == 0)
+				option = EXT4_DEF_MAX_BATCH_TIME;
+			sbi->s_max_batch_time = option;
+			break;
+		case Opt_min_batch_time:
+			if (match_int(&args[0], &option))
+				return 0;
+			if (option < 0)
+				return 0;
+			sbi->s_min_batch_time = option;
+			break;
 		case Opt_data_journal:
 			data_opt = EXT4_MOUNT_JOURNAL_DATA;
 			goto datacheck;
@@ -1979,6 +2007,9 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 
 	sbi->s_resuid = le16_to_cpu(es->s_def_resuid);
 	sbi->s_resgid = le16_to_cpu(es->s_def_resgid);
+	sbi->s_commit_interval = JBD2_DEFAULT_MAX_COMMIT_AGE * HZ;
+	sbi->s_min_batch_time = EXT4_DEF_MIN_BATCH_TIME;
+	sbi->s_max_batch_time = EXT4_DEF_MAX_BATCH_TIME;
 
 	set_opt(sbi->s_mount_opt, RESERVATION);
 	set_opt(sbi->s_mount_opt, BARRIER);
@@ -2524,11 +2555,9 @@ static void ext4_init_journal_params(struct super_block *sb, journal_t *journal)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 
-	if (sbi->s_commit_interval)
-		journal->j_commit_interval = sbi->s_commit_interval;
-	/* We could also set up an ext4-specific default for the commit
-	 * interval here, but for now we'll just fall back to the jbd
-	 * default. */
+	journal->j_commit_interval = sbi->s_commit_interval;
+	journal->j_min_batch_time = sbi->s_min_batch_time;
+	journal->j_max_batch_time = sbi->s_max_batch_time;
 
 	spin_lock(&journal->j_state_lock);
 	if (test_opt(sb, BARRIER))
@@ -3042,6 +3071,8 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data)
 	old_opts.s_resuid = sbi->s_resuid;
 	old_opts.s_resgid = sbi->s_resgid;
 	old_opts.s_commit_interval = sbi->s_commit_interval;
+	old_opts.s_min_batch_time = sbi->s_min_batch_time;
+	old_opts.s_max_batch_time = sbi->s_max_batch_time;
 #ifdef CONFIG_QUOTA
 	old_opts.s_jquota_fmt = sbi->s_jquota_fmt;
 	for (i = 0; i < MAXQUOTAS; i++)
@@ -3178,6 +3209,8 @@ restore_opts:
 	sbi->s_resuid = old_opts.s_resuid;
 	sbi->s_resgid = old_opts.s_resgid;
 	sbi->s_commit_interval = old_opts.s_commit_interval;
+	sbi->s_min_batch_time = old_opts.s_min_batch_time;
+	sbi->s_max_batch_time = old_opts.s_max_batch_time;
 #ifdef CONFIG_QUOTA
 	sbi->s_jquota_fmt = old_opts.s_jquota_fmt;
 	for (i = 0; i < MAXQUOTAS; i++) {
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 74d87290381c..fd1d7557a098 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -964,6 +964,8 @@ static journal_t * journal_init_common (void)
 	spin_lock_init(&journal->j_state_lock);
 
 	journal->j_commit_interval = (HZ * JBD2_DEFAULT_MAX_COMMIT_AGE);
+	journal->j_min_batch_time = 0;
+	journal->j_max_batch_time = 15000; /* 15ms */
 
 	/* The journal is marked for error until we succeed with recovery! */
 	journal->j_flags = JBD2_ABORT;
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index 13dcbc990f41..48c21bac5a56 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -1255,8 +1255,10 @@ int jbd2_journal_stop(handle_t *handle)
 		trans_time = ktime_to_ns(ktime_sub(ktime_get(),
 						   transaction->t_start_time));
 
+		commit_time = max_t(u64, commit_time,
+				    1000*journal->j_min_batch_time);
 		commit_time = min_t(u64, commit_time,
-				    1000*jiffies_to_usecs(1));
+				    1000*journal->j_max_batch_time);
 
 		if (trans_time < commit_time) {
 			ktime_t expires = ktime_add_ns(ktime_get(),
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index ab8cef130c28..a3cd647ea1bc 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -956,6 +956,14 @@ struct journal_s
 	 */
 	u64			j_average_commit_time;
 
+	/*
+	 * minimum and maximum times that we should wait for
+	 * additional filesystem operations to get batched into a
+	 * synchronous handle in microseconds
+	 */
+	u32			j_min_batch_time;
+	u32			j_max_batch_time;
+
 	/* This function is called when a transaction is closed */
 	void			(*j_commit_callback)(journal_t *,
 						     transaction_t *);
-- 
cgit v1.2.3


From 1a0d3786dd57dbd74f340322054c3d618b999dcf Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Wed, 5 Nov 2008 00:09:22 -0500
Subject: jbd2: Remove a large array of bh's from the stack of the checkpoint
 routine

jbd2_log_do_checkpoint()n is one of the kernel's largest stack users.
Move the array of buffer head's from the stack of jbd2_log_do_checkpoint()
to the in-core journal structure.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 fs/jbd2/checkpoint.c | 22 +++++++++-------------
 fs/jbd2/journal.c    |  2 ++
 include/linux/jbd2.h | 10 ++++++++++
 3 files changed, 21 insertions(+), 13 deletions(-)

(limited to 'include/linux')

diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index 9497718fe920..adc08ec875ed 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -249,16 +249,14 @@ restart:
 	return ret;
 }
 
-#define NR_BATCH	64
-
 static void
-__flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count)
+__flush_batch(journal_t *journal, int *batch_count)
 {
 	int i;
 
-	ll_rw_block(SWRITE, *batch_count, bhs);
+	ll_rw_block(SWRITE, *batch_count, journal->j_chkpt_bhs);
 	for (i = 0; i < *batch_count; i++) {
-		struct buffer_head *bh = bhs[i];
+		struct buffer_head *bh = journal->j_chkpt_bhs[i];
 		clear_buffer_jwrite(bh);
 		BUFFER_TRACE(bh, "brelse");
 		__brelse(bh);
@@ -277,8 +275,7 @@ __flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count)
  * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it
  */
 static int __process_buffer(journal_t *journal, struct journal_head *jh,
-			struct buffer_head **bhs, int *batch_count,
-			transaction_t *transaction)
+			    int *batch_count, transaction_t *transaction)
 {
 	struct buffer_head *bh = jh2bh(jh);
 	int ret = 0;
@@ -325,14 +322,14 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
 		get_bh(bh);
 		J_ASSERT_BH(bh, !buffer_jwrite(bh));
 		set_buffer_jwrite(bh);
-		bhs[*batch_count] = bh;
+		journal->j_chkpt_bhs[*batch_count] = bh;
 		__buffer_relink_io(jh);
 		jbd_unlock_bh_state(bh);
 		transaction->t_chp_stats.cs_written++;
 		(*batch_count)++;
-		if (*batch_count == NR_BATCH) {
+		if (*batch_count == JBD2_NR_BATCH) {
 			spin_unlock(&journal->j_list_lock);
-			__flush_batch(journal, bhs, batch_count);
+			__flush_batch(journal, batch_count);
 			ret = 1;
 		}
 	}
@@ -388,7 +385,6 @@ restart:
 	if (journal->j_checkpoint_transactions == transaction &&
 			transaction->t_tid == this_tid) {
 		int batch_count = 0;
-		struct buffer_head *bhs[NR_BATCH];
 		struct journal_head *jh;
 		int retry = 0, err;
 
@@ -402,7 +398,7 @@ restart:
 				retry = 1;
 				break;
 			}
-			retry = __process_buffer(journal, jh, bhs, &batch_count,
+			retry = __process_buffer(journal, jh, &batch_count,
 						 transaction);
 			if (retry < 0 && !result)
 				result = retry;
@@ -419,7 +415,7 @@ restart:
 				spin_unlock(&journal->j_list_lock);
 				retry = 1;
 			}
-			__flush_batch(journal, bhs, &batch_count);
+			__flush_batch(journal, &batch_count);
 		}
 
 		if (retry) {
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index fd1d7557a098..34ef98057202 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -1477,7 +1477,9 @@ int jbd2_journal_destroy(journal_t *journal)
 	spin_lock(&journal->j_list_lock);
 	while (journal->j_checkpoint_transactions != NULL) {
 		spin_unlock(&journal->j_list_lock);
+		mutex_lock(&journal->j_checkpoint_mutex);
 		jbd2_log_do_checkpoint(journal);
+		mutex_unlock(&journal->j_checkpoint_mutex);
 		spin_lock(&journal->j_list_lock);
 	}
 
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index a3cd647ea1bc..004c9a8d63ed 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -687,6 +687,8 @@ jbd2_time_diff(unsigned long start, unsigned long end)
 	return end + (MAX_JIFFY_OFFSET - start);
 }
 
+#define JBD2_NR_BATCH	64
+
 /**
  * struct journal_s - The journal_s type is the concrete type associated with
  *     journal_t.
@@ -830,6 +832,14 @@ struct journal_s
 	/* Semaphore for locking against concurrent checkpoints */
 	struct mutex		j_checkpoint_mutex;
 
+	/*
+	 * List of buffer heads used by the checkpoint routine.  This
+	 * was moved from jbd2_log_do_checkpoint() to reduce stack
+	 * usage.  Access to this array is controlled by the
+	 * j_checkpoint_mutex.  [j_checkpoint_mutex]
+	 */
+	struct buffer_head	*j_chkpt_bhs[JBD2_NR_BATCH];
+	
 	/*
 	 * Journal head: identifies the first unused block in the journal.
 	 * [j_state_lock]
-- 
cgit v1.2.3


From fb68407b0d9efba962c03f55009c797e22f024bc Mon Sep 17 00:00:00 2001
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Date: Thu, 6 Nov 2008 17:50:21 -0500
Subject: jbd2: Call journal commit callback without holding j_list_lock

Avoid freeing the transaction in __jbd2_journal_drop_transaction() so
the journal commit callback can run without holding j_list_lock, to
avoid lock contention on this spinlock.

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 fs/jbd2/checkpoint.c |  2 +-
 fs/jbd2/commit.c     | 13 ++++++++-----
 include/linux/jbd2.h |  4 ++--
 3 files changed, 11 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index adc08ec875ed..17159cacbd9e 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -682,6 +682,7 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh)
 	   safely remove this transaction from the log */
 
 	__jbd2_journal_drop_transaction(journal, transaction);
+	kfree(transaction);
 
 	/* Just in case anybody was waiting for more transactions to be
            checkpointed... */
@@ -756,5 +757,4 @@ void __jbd2_journal_drop_transaction(journal_t *journal, transaction_t *transact
 	J_ASSERT(journal->j_running_transaction != transaction);
 
 	jbd_debug(1, "Dropping transaction %d, all done\n", transaction->t_tid);
-	kfree(transaction);
 }
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index f22d1828ea85..0ad84162c425 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -363,7 +363,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	int space_left = 0;
 	int first_tag = 0;
 	int tag_flag;
-	int i;
+	int i, to_free = 0;
 	int tag_bytes = journal_tag_bytes(journal);
 	struct buffer_head *cbh = NULL; /* For transactional checksums */
 	__u32 crc32_sum = ~0;
@@ -1011,12 +1011,10 @@ restart_loop:
 		journal->j_average_commit_time = commit_time;
 	spin_unlock(&journal->j_state_lock);
 
-	if (journal->j_commit_callback)
-		journal->j_commit_callback(journal, commit_transaction);
-
 	if (commit_transaction->t_checkpoint_list == NULL &&
 	    commit_transaction->t_checkpoint_io_list == NULL) {
 		__jbd2_journal_drop_transaction(journal, commit_transaction);
+		to_free = 1;
 	} else {
 		if (journal->j_checkpoint_transactions == NULL) {
 			journal->j_checkpoint_transactions = commit_transaction;
@@ -1035,11 +1033,16 @@ restart_loop:
 	}
 	spin_unlock(&journal->j_list_lock);
 
+	if (journal->j_commit_callback)
+		journal->j_commit_callback(journal, commit_transaction);
+
 	trace_mark(jbd2_end_commit, "dev %s transaction %d head %d",
-		   journal->j_devname, journal->j_commit_sequence,
+		   journal->j_devname, commit_transaction->t_tid,
 		   journal->j_tail_sequence);
 	jbd_debug(1, "JBD: commit %d complete, head %d\n",
 		  journal->j_commit_sequence, journal->j_tail_sequence);
+	if (to_free)
+		kfree(commit_transaction);
 
 	wake_up(&journal->j_wait_done_commit);
 }
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index 004c9a8d63ed..9d82084a1605 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -1179,8 +1179,8 @@ int jbd2_log_wait_commit(journal_t *journal, tid_t tid);
 int jbd2_log_do_checkpoint(journal_t *journal);
 
 void __jbd2_log_wait_for_space(journal_t *journal);
-extern void	__jbd2_journal_drop_transaction(journal_t *, transaction_t *);
-extern int	jbd2_cleanup_journal_tail(journal_t *);
+extern void __jbd2_journal_drop_transaction(journal_t *, transaction_t *);
+extern int jbd2_cleanup_journal_tail(journal_t *);
 
 /* Debugging code only: */
 
-- 
cgit v1.2.3


From 307ba6dd73254fe7d2ce27db64ffd90e1bb3c6c0 Mon Sep 17 00:00:00 2001
From: David Vrabel <david.vrabel@csr.com>
Date: Fri, 7 Nov 2008 17:37:33 +0000
Subject: uwb: don't unbind the radio controller driver when resetting

Use pre_reset and post_reset methods to avoid unbinding the radio
controller driver after a uwb_rc_reset_all() call.  This avoids a
deadlock in uwb_rc_rm() when waiting for the uwb event thread to stop.

Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/uwb/hwa-rc.c       | 22 +++++++++++++++-
 drivers/uwb/reset.c        | 49 ++++++++++++++++++++++++++++++------
 drivers/uwb/rsv.c          | 29 +++++++++++++++-------
 drivers/uwb/umc-bus.c      | 62 ++++++++++++++++++++++++++++++++--------------
 drivers/uwb/uwb-internal.h |  1 +
 drivers/uwb/whc-rc.c       | 30 ++++++++++++++++++----
 include/linux/uwb.h        |  2 ++
 include/linux/uwb/umc.h    |  2 ++
 8 files changed, 155 insertions(+), 42 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/uwb/hwa-rc.c b/drivers/uwb/hwa-rc.c
index 18009c99577d..158e98d08af9 100644
--- a/drivers/uwb/hwa-rc.c
+++ b/drivers/uwb/hwa-rc.c
@@ -881,6 +881,24 @@ static void hwarc_disconnect(struct usb_interface *iface)
 	uwb_rc_put(uwb_rc);	/* when creating the device, refcount = 1 */
 }
 
+static int hwarc_pre_reset(struct usb_interface *iface)
+{
+	struct hwarc *hwarc = usb_get_intfdata(iface);
+	struct uwb_rc *uwb_rc = hwarc->uwb_rc;
+
+	uwb_rc_pre_reset(uwb_rc);
+	return 0;
+}
+
+static int hwarc_post_reset(struct usb_interface *iface)
+{
+	struct hwarc *hwarc = usb_get_intfdata(iface);
+	struct uwb_rc *uwb_rc = hwarc->uwb_rc;
+
+	uwb_rc_post_reset(uwb_rc);
+	return 0;
+}
+
 /** USB device ID's that we handle */
 static struct usb_device_id hwarc_id_table[] = {
 	/* D-Link DUB-1210 */
@@ -897,9 +915,11 @@ MODULE_DEVICE_TABLE(usb, hwarc_id_table);
 
 static struct usb_driver hwarc_driver = {
 	.name =		"hwa-rc",
+	.id_table =	hwarc_id_table,
 	.probe =	hwarc_probe,
 	.disconnect =	hwarc_disconnect,
-	.id_table =	hwarc_id_table,
+	.pre_reset =    hwarc_pre_reset,
+	.post_reset =   hwarc_post_reset,
 };
 
 static int __init hwarc_driver_init(void)
diff --git a/drivers/uwb/reset.c b/drivers/uwb/reset.c
index 8de856fa7958..e39b32099af3 100644
--- a/drivers/uwb/reset.c
+++ b/drivers/uwb/reset.c
@@ -323,17 +323,16 @@ int uwbd_msg_handle_reset(struct uwb_event *evt)
 	struct uwb_rc *rc = evt->rc;
 	int ret;
 
-	/* Need to prevent the RC hardware module going away while in
-	   the rc->reset() call. */
-	if (!try_module_get(rc->owner))
-		return 0;
-
 	dev_info(&rc->uwb_dev.dev, "resetting radio controller\n");
 	ret = rc->reset(rc);
-	if (ret)
+	if (ret) {
 		dev_err(&rc->uwb_dev.dev, "failed to reset hardware: %d\n", ret);
-
-	module_put(rc->owner);
+		goto error;
+	}
+	return 0;
+error:
+	/* Nothing can be done except try the reset again. */
+	uwb_rc_reset_all(rc);
 	return ret;
 }
 
@@ -360,3 +359,37 @@ void uwb_rc_reset_all(struct uwb_rc *rc)
 	uwbd_event_queue(evt);
 }
 EXPORT_SYMBOL_GPL(uwb_rc_reset_all);
+
+void uwb_rc_pre_reset(struct uwb_rc *rc)
+{
+	rc->stop(rc);
+	uwbd_flush(rc);
+
+	mutex_lock(&rc->uwb_dev.mutex);
+	rc->beaconing = -1;
+	rc->scanning = -1;
+	mutex_unlock(&rc->uwb_dev.mutex);
+
+	uwb_rsv_remove_all(rc);
+}
+EXPORT_SYMBOL_GPL(uwb_rc_pre_reset);
+
+void uwb_rc_post_reset(struct uwb_rc *rc)
+{
+	int ret;
+
+	ret = rc->start(rc);
+	if (ret)
+		goto error;
+	ret = uwb_rc_mac_addr_set(rc, &rc->uwb_dev.mac_addr);
+	if (ret)
+		goto error;
+	ret = uwb_rc_dev_addr_set(rc, &rc->uwb_dev.dev_addr);
+	if (ret)
+		goto error;
+	return;
+error:
+	/* Nothing can be done except try the reset again. */
+	uwb_rc_reset_all(rc);
+}
+EXPORT_SYMBOL_GPL(uwb_rc_post_reset);
diff --git a/drivers/uwb/rsv.c b/drivers/uwb/rsv.c
index ce0094657d3d..3d76efe26acc 100644
--- a/drivers/uwb/rsv.c
+++ b/drivers/uwb/rsv.c
@@ -659,6 +659,25 @@ static void uwb_rsv_timer(unsigned long arg)
 	uwb_rsv_sched_update(rsv->rc);
 }
 
+/**
+ * uwb_rsv_remove_all - remove all reservations
+ * @rc: the radio controller
+ *
+ * A DRP IE update is not done.
+ */
+void uwb_rsv_remove_all(struct uwb_rc *rc)
+{
+	struct uwb_rsv *rsv, *t;
+
+	mutex_lock(&rc->rsvs_mutex);
+	list_for_each_entry_safe(rsv, t, &rc->reservations, rc_node) {
+		uwb_rsv_remove(rsv);
+	}
+	mutex_unlock(&rc->rsvs_mutex);
+
+	cancel_work_sync(&rc->rsv_update_work);
+}
+
 void uwb_rsv_init(struct uwb_rc *rc)
 {
 	INIT_LIST_HEAD(&rc->reservations);
@@ -682,14 +701,6 @@ int uwb_rsv_setup(struct uwb_rc *rc)
 
 void uwb_rsv_cleanup(struct uwb_rc *rc)
 {
-	struct uwb_rsv *rsv, *t;
-
-	mutex_lock(&rc->rsvs_mutex);
-	list_for_each_entry_safe(rsv, t, &rc->reservations, rc_node) {
-		uwb_rsv_remove(rsv);
-	}
-	mutex_unlock(&rc->rsvs_mutex);
-
-	cancel_work_sync(&rc->rsv_update_work);
+	uwb_rsv_remove_all(rc);
 	destroy_workqueue(rc->rsv_workq);
 }
diff --git a/drivers/uwb/umc-bus.c b/drivers/uwb/umc-bus.c
index 2d8d62d9f53e..5ad36164c13b 100644
--- a/drivers/uwb/umc-bus.c
+++ b/drivers/uwb/umc-bus.c
@@ -11,23 +11,48 @@
 #include <linux/uwb/umc.h>
 #include <linux/pci.h>
 
-static int umc_bus_unbind_helper(struct device *dev, void *data)
+static int umc_bus_pre_reset_helper(struct device *dev, void *data)
 {
-	struct device *parent = data;
+	int ret = 0;
 
-	if (dev->parent == parent && dev->driver)
-		device_release_driver(dev);
-	return 0;
+	if (dev->driver) {
+		struct umc_dev *umc = to_umc_dev(dev);
+		struct umc_driver *umc_drv = to_umc_driver(dev->driver);
+
+		if (umc_drv->pre_reset)
+			ret = umc_drv->pre_reset(umc);
+		else
+			device_release_driver(dev);
+	}
+	return ret;
+}
+
+static int umc_bus_post_reset_helper(struct device *dev, void *data)
+{
+	int ret = 0;
+
+	if (dev->driver) {
+		struct umc_dev *umc = to_umc_dev(dev);
+		struct umc_driver *umc_drv = to_umc_driver(dev->driver);
+
+		if (umc_drv->post_reset)
+			ret = umc_drv->post_reset(umc);
+	} else
+		ret = device_attach(dev);
+
+	return ret;
 }
 
 /**
  * umc_controller_reset - reset the whole UMC controller
  * @umc: the UMC device for the radio controller.
  *
- * Drivers will be unbound from all UMC devices belonging to the
- * controller and then the radio controller will be rebound.  The
- * radio controller is expected to do a full hardware reset when it is
- * probed.
+ * Drivers or all capabilities of the controller will have their
+ * pre_reset methods called or be unbound from their device.  Then all
+ * post_reset methods will be called or the drivers will be rebound.
+ *
+ * Radio controllers must provide pre_reset and post_reset methods and
+ * reset the hardware in their start method.
  *
  * If this is called while a probe() or remove() is in progress it
  * will return -EAGAIN and not perform the reset.
@@ -35,14 +60,13 @@ static int umc_bus_unbind_helper(struct device *dev, void *data)
 int umc_controller_reset(struct umc_dev *umc)
 {
 	struct device *parent = umc->dev.parent;
-	int ret;
+	int ret = 0;
 
-	if (down_trylock(&parent->sem))
+	if(down_trylock(&parent->sem))
 		return -EAGAIN;
-	bus_for_each_dev(&umc_bus_type, NULL, parent, umc_bus_unbind_helper);
-	ret = device_attach(&umc->dev);
-	if (ret == 1)
-		ret = 0;
+	ret = device_for_each_child(parent, parent, umc_bus_pre_reset_helper);
+	if (ret >= 0)
+		device_for_each_child(parent, parent, umc_bus_post_reset_helper);
 	up(&parent->sem);
 
 	return ret;
@@ -75,10 +99,10 @@ static int umc_bus_rescan_helper(struct device *dev, void *data)
 	if (!dev->driver)
 		ret = device_attach(dev);
 
-	return ret < 0 ? ret : 0;
+	return ret;
 }
 
-static void umc_bus_rescan(void)
+static void umc_bus_rescan(struct device *parent)
 {
 	int err;
 
@@ -86,7 +110,7 @@ static void umc_bus_rescan(void)
 	 * We can't use bus_rescan_devices() here as it deadlocks when
 	 * it tries to retake the dev->parent semaphore.
 	 */
-	err = bus_for_each_dev(&umc_bus_type, NULL, NULL, umc_bus_rescan_helper);
+	err = device_for_each_child(parent, NULL, umc_bus_rescan_helper);
 	if (err < 0)
 		printk(KERN_WARNING "%s: rescan of bus failed: %d\n",
 		       KBUILD_MODNAME, err);
@@ -120,7 +144,7 @@ static int umc_device_probe(struct device *dev)
 	if (err)
 		put_device(dev);
 	else
-		umc_bus_rescan();
+		umc_bus_rescan(dev->parent);
 
 	return err;
 }
diff --git a/drivers/uwb/uwb-internal.h b/drivers/uwb/uwb-internal.h
index 4c2449679978..af95541dabcd 100644
--- a/drivers/uwb/uwb-internal.h
+++ b/drivers/uwb/uwb-internal.h
@@ -248,6 +248,7 @@ extern struct device_attribute dev_attr_scan;
 void uwb_rsv_init(struct uwb_rc *rc);
 int uwb_rsv_setup(struct uwb_rc *rc);
 void uwb_rsv_cleanup(struct uwb_rc *rc);
+void uwb_rsv_remove_all(struct uwb_rc *rc);
 
 void uwb_rsv_set_state(struct uwb_rsv *rsv, enum uwb_rsv_state new_state);
 void uwb_rsv_remove(struct uwb_rsv *rsv);
diff --git a/drivers/uwb/whc-rc.c b/drivers/uwb/whc-rc.c
index 6c454eadd307..e0d66938ccd8 100644
--- a/drivers/uwb/whc-rc.c
+++ b/drivers/uwb/whc-rc.c
@@ -394,7 +394,7 @@ void whcrc_stop_rc(struct uwb_rc *rc)
 
 	le_writel(0, whcrc->rc_base + URCCMD);
 	whci_wait_for(&umc_dev->dev, whcrc->rc_base + URCSTS,
-		      URCSTS_HALTED, 0, 40, "URCSTS.HALTED");
+		      URCSTS_HALTED, URCSTS_HALTED, 100, "URCSTS.HALTED");
 }
 
 static void whcrc_init(struct whcrc *whcrc)
@@ -488,6 +488,24 @@ static void whcrc_remove(struct umc_dev *umc_dev)
 	d_printf(1, &umc_dev->dev, "freed whcrc %p\n", whcrc);
 }
 
+static int whcrc_pre_reset(struct umc_dev *umc)
+{
+	struct whcrc *whcrc = umc_get_drvdata(umc);
+	struct uwb_rc *uwb_rc = whcrc->uwb_rc;
+
+	uwb_rc_pre_reset(uwb_rc);
+	return 0;
+}
+
+static int whcrc_post_reset(struct umc_dev *umc)
+{
+	struct whcrc *whcrc = umc_get_drvdata(umc);
+	struct uwb_rc *uwb_rc = whcrc->uwb_rc;
+
+	uwb_rc_post_reset(uwb_rc);
+	return 0;
+}
+
 /* PCI device ID's that we handle [so it gets loaded] */
 static struct pci_device_id whcrc_id_table[] = {
 	{ PCI_DEVICE_CLASS(PCI_CLASS_WIRELESS_WHCI, ~0) },
@@ -496,10 +514,12 @@ static struct pci_device_id whcrc_id_table[] = {
 MODULE_DEVICE_TABLE(pci, whcrc_id_table);
 
 static struct umc_driver whcrc_driver = {
-	.name   = "whc-rc",
-	.cap_id = UMC_CAP_ID_WHCI_RC,
-	.probe  = whcrc_probe,
-	.remove = whcrc_remove,
+	.name       = "whc-rc",
+	.cap_id     = UMC_CAP_ID_WHCI_RC,
+	.probe      = whcrc_probe,
+	.remove     = whcrc_remove,
+	.pre_reset  = whcrc_pre_reset,
+	.post_reset = whcrc_post_reset,
 };
 
 static int __init whcrc_driver_init(void)
diff --git a/include/linux/uwb.h b/include/linux/uwb.h
index 881f0c5b6d28..c4854848999d 100644
--- a/include/linux/uwb.h
+++ b/include/linux/uwb.h
@@ -540,6 +540,8 @@ void uwb_rc_rm(struct uwb_rc *);
 void uwb_rc_neh_grok(struct uwb_rc *, void *, size_t);
 void uwb_rc_neh_error(struct uwb_rc *, int);
 void uwb_rc_reset_all(struct uwb_rc *rc);
+void uwb_rc_pre_reset(struct uwb_rc *rc);
+void uwb_rc_post_reset(struct uwb_rc *rc);
 
 /**
  * uwb_rsv_is_owner - is the owner of this reservation the RC?
diff --git a/include/linux/uwb/umc.h b/include/linux/uwb/umc.h
index 36a39e34f8d7..4b4fc0f43855 100644
--- a/include/linux/uwb/umc.h
+++ b/include/linux/uwb/umc.h
@@ -89,6 +89,8 @@ struct umc_driver {
 	void (*remove)(struct umc_dev *);
 	int  (*suspend)(struct umc_dev *, pm_message_t state);
 	int  (*resume)(struct umc_dev *);
+	int  (*pre_reset)(struct umc_dev *);
+	int  (*post_reset)(struct umc_dev *);
 
 	struct device_driver driver;
 };
-- 
cgit v1.2.3


From e17be2b2a95b43fe0d5878adf330701bb7a77115 Mon Sep 17 00:00:00 2001
From: David Vrabel <david.vrabel@csr.com>
Date: Mon, 17 Nov 2008 15:24:14 +0000
Subject: uwb: add pal parameter to new reservation callback

The pal parameter allows PALs to retrieve their PAL-specific data
structure.

Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/uwb/rsv.c       |  2 +-
 drivers/uwb/uwb-debug.c | 10 +++++-----
 include/linux/uwb.h     |  2 +-
 3 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/uwb/rsv.c b/drivers/uwb/rsv.c
index 3d76efe26acc..935d5b536db7 100644
--- a/drivers/uwb/rsv.c
+++ b/drivers/uwb/rsv.c
@@ -558,7 +558,7 @@ static struct uwb_rsv *uwb_rsv_new_target(struct uwb_rc *rc,
 	spin_lock(&rc->pal_lock);
 	list_for_each_entry(pal, &rc->pals, node) {
 		if (pal->new_rsv)
-			pal->new_rsv(rsv);
+			pal->new_rsv(pal, rsv);
 		if (rsv->state == UWB_RSV_STATE_T_ACCEPTED)
 			break;
 	}
diff --git a/drivers/uwb/uwb-debug.c b/drivers/uwb/uwb-debug.c
index 88e6ac713817..217ebaac128d 100644
--- a/drivers/uwb/uwb-debug.c
+++ b/drivers/uwb/uwb-debug.c
@@ -306,13 +306,13 @@ static struct file_operations drp_avail_fops = {
 	.owner   = THIS_MODULE,
 };
 
-static void uwb_dbg_new_rsv(struct uwb_rsv *rsv)
+static void uwb_dbg_new_rsv(struct uwb_pal *pal, struct uwb_rsv *rsv)
 {
-	struct uwb_rc *rc = rsv->rc;
+	struct uwb_dbg *dbg = container_of(pal, struct uwb_dbg, pal);
 
-	if (rc->dbg->accept) {
-		list_add_tail(&rsv->pal_node, &rc->dbg->rsvs);
-		uwb_rsv_accept(rsv, uwb_dbg_rsv_cb, NULL);
+	if (dbg->accept) {
+		list_add_tail(&rsv->pal_node, &dbg->rsvs);
+		uwb_rsv_accept(rsv, uwb_dbg_rsv_cb, dbg);
 	}
 }
 
diff --git a/include/linux/uwb.h b/include/linux/uwb.h
index c4854848999d..effd97998fd1 100644
--- a/include/linux/uwb.h
+++ b/include/linux/uwb.h
@@ -405,7 +405,7 @@ struct uwb_pal {
 	struct list_head node;
 	const char *name;
 	struct device *device;
-	void (*new_rsv)(struct uwb_rsv *rsv);
+	void (*new_rsv)(struct uwb_pal *pal, struct uwb_rsv *rsv);
 };
 
 void uwb_pal_init(struct uwb_pal *pal);
-- 
cgit v1.2.3


From 6fae35f9cea92793a98b2d9ab21235e5ae035581 Mon Sep 17 00:00:00 2001
From: David Vrabel <david.vrabel@csr.com>
Date: Mon, 17 Nov 2008 15:53:42 +0000
Subject: uwb: add basic radio manager

The UWB radio manager coordinates the use of the radio between the
PALs that may be using it.  PALs request use of the radio with
uwb_radio_start() and the radio manager will start beaconing if its
not already doing so.  When the last PAL has called uwb_radio_stop()
beaconing will be stopped.

In the future, the radio manager will have a more sophisticated channel
selection algorithm, probably following the Channel Selection Policy
from the WiMedia Alliance when it is finalized.  For now, channel 9
(BG1, TFC1) is selected.

The user may override the channel selected by the radio manager and may
force the radio to stop beaconing.

The WUSB Host Controller PAL makes use of this and there are two new
debug PAL commands that can be used for testing.

Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 Documentation/ABI/testing/sysfs-class-uwb_rc |  14 +-
 Documentation/usb/wusb-cbaf                  |   9 --
 drivers/usb/host/hwa-hc.c                    |   1 -
 drivers/usb/host/whci/hcd.c                  |   2 -
 drivers/usb/wusbcore/devconnect.c            |   5 +-
 drivers/usb/wusbcore/mmc.c                   |  75 ++--------
 drivers/usb/wusbcore/pal.c                   |  16 ++-
 drivers/usb/wusbcore/wusbhc.h                |   8 +-
 drivers/uwb/Makefile                         |   1 +
 drivers/uwb/beacon.c                         |  26 ++--
 drivers/uwb/drp.c                            |  24 +---
 drivers/uwb/lc-rc.c                          |  11 +-
 drivers/uwb/pal.c                            |  20 +--
 drivers/uwb/radio.c                          | 202 +++++++++++++++++++++++++++
 drivers/uwb/reset.c                          |   6 +-
 drivers/uwb/rsv.c                            |   4 +-
 drivers/uwb/uwb-debug.c                      |  26 +++-
 drivers/uwb/uwb-internal.h                   |   5 +
 drivers/uwb/wlp/wlp-lc.c                     |   5 +-
 include/linux/uwb.h                          |  23 ++-
 include/linux/uwb/debug-cmd.h                |   2 +
 21 files changed, 323 insertions(+), 162 deletions(-)
 create mode 100644 drivers/uwb/radio.c

(limited to 'include/linux')

diff --git a/Documentation/ABI/testing/sysfs-class-uwb_rc b/Documentation/ABI/testing/sysfs-class-uwb_rc
index a0d18dbeb7a9..6a5fd072849d 100644
--- a/Documentation/ABI/testing/sysfs-class-uwb_rc
+++ b/Documentation/ABI/testing/sysfs-class-uwb_rc
@@ -32,14 +32,16 @@ Contact:        linux-usb@vger.kernel.org
 Description:
                 Write:
 
-                <channel> [<bpst offset>]
+                <channel>
 
-                to start beaconing on a specific channel, or stop
-                beaconing if <channel> is -1.  Valid channels depends
-                on the radio controller's supported band groups.
+                to force a specific channel to be used when beaconing,
+                or, if <channel> is -1, to prohibit beaconing.  If
+                <channel> is 0, then the default channel selection
+                algorithm will be used.  Valid channels depends on the
+                radio controller's supported band groups.
 
-                <bpst offset> may be used to try and join a specific
-                beacon group if more than one was found during a scan.
+                Reading returns the currently active channel, or -1 if
+                the radio controller is not beaconing.
 
 What:           /sys/class/uwb_rc/uwbN/scan
 Date:           July 2008
diff --git a/Documentation/usb/wusb-cbaf b/Documentation/usb/wusb-cbaf
index 2e78b70f3adc..426ddaaef96f 100644
--- a/Documentation/usb/wusb-cbaf
+++ b/Documentation/usb/wusb-cbaf
@@ -80,12 +80,6 @@ case $1 in
     start)
         for dev in ${2:-$hdevs}
           do
-          uwb_rc=$(readlink -f $dev/uwb_rc)
-          if cat $uwb_rc/beacon | grep -q -- "-1"
-              then
-              echo 13 0 > $uwb_rc/beacon
-              echo I: started beaconing on ch 13 on $(basename $uwb_rc) >&2
-          fi
           echo $host_CHID > $dev/wusb_chid
           echo I: started host $(basename $dev) >&2
         done
@@ -95,9 +89,6 @@ case $1 in
           do
           echo 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > $dev/wusb_chid
           echo I: stopped host $(basename $dev) >&2
-          uwb_rc=$(readlink -f $dev/uwb_rc)
-          echo -1 | cat > $uwb_rc/beacon
-          echo I: stopped beaconing on $(basename $uwb_rc) >&2
         done
         ;;
     set-chid)
diff --git a/drivers/usb/host/hwa-hc.c b/drivers/usb/host/hwa-hc.c
index 2827353e97e1..2a4d36fa70b0 100644
--- a/drivers/usb/host/hwa-hc.c
+++ b/drivers/usb/host/hwa-hc.c
@@ -221,7 +221,6 @@ static void hwahc_op_stop(struct usb_hcd *usb_hcd)
 
 	d_fnstart(4, dev, "(hwahc %p)\n", hwahc);
 	mutex_lock(&wusbhc->mutex);
-	wusbhc_stop(wusbhc);
 	wusb_cluster_id_put(wusbhc->cluster_id);
 	mutex_unlock(&wusbhc->mutex);
 	d_fnend(4, dev, "(hwahc %p) = %d\n", hwahc, result);
diff --git a/drivers/usb/host/whci/hcd.c b/drivers/usb/host/whci/hcd.c
index de1e07271b81..f599f89d3be1 100644
--- a/drivers/usb/host/whci/hcd.c
+++ b/drivers/usb/host/whci/hcd.c
@@ -91,8 +91,6 @@ static void whc_stop(struct usb_hcd *usb_hcd)
 
 	mutex_lock(&wusbhc->mutex);
 
-	wusbhc_stop(wusbhc);
-
 	/* stop HC */
 	le_writel(0, whc->base + WUSBINTR);
 	whc_write_wusbcmd(whc, WUSBCMD_RUN, 0);
diff --git a/drivers/usb/wusbcore/devconnect.c b/drivers/usb/wusbcore/devconnect.c
index c01c7a80744c..08a1ec903867 100644
--- a/drivers/usb/wusbcore/devconnect.c
+++ b/drivers/usb/wusbcore/devconnect.c
@@ -1124,8 +1124,7 @@ void wusbhc_devconnect_destroy(struct wusbhc *wusbhc)
  * FIXME: This also enables the keep alives but this is not necessary
  * until there are connected and authenticated devices.
  */
-int wusbhc_devconnect_start(struct wusbhc *wusbhc,
-			    const struct wusb_ckhdid *chid)
+int wusbhc_devconnect_start(struct wusbhc *wusbhc)
 {
 	struct device *dev = wusbhc->dev;
 	struct wuie_host_info *hi;
@@ -1138,7 +1137,7 @@ int wusbhc_devconnect_start(struct wusbhc *wusbhc,
 	hi->hdr.bLength       = sizeof(*hi);
 	hi->hdr.bIEIdentifier = WUIE_ID_HOST_INFO;
 	hi->attributes        = cpu_to_le16((wusbhc->rsv->stream << 3) | WUIE_HI_CAP_ALL);
-	hi->CHID              = *chid;
+	hi->CHID              = wusbhc->chid;
 	result = wusbhc_mmcie_set(wusbhc, 0, 0, &hi->hdr);
 	if (result < 0) {
 		dev_err(dev, "Cannot add Host Info MMCIE: %d\n", result);
diff --git a/drivers/usb/wusbcore/mmc.c b/drivers/usb/wusbcore/mmc.c
index af2aee0fdffa..5463ecebafdf 100644
--- a/drivers/usb/wusbcore/mmc.c
+++ b/drivers/usb/wusbcore/mmc.c
@@ -162,12 +162,11 @@ EXPORT_SYMBOL_GPL(wusbhc_mmcie_rm);
 /*
  * wusbhc_start - start transmitting MMCs and accepting connections
  * @wusbhc: the HC to start
- * @chid: the CHID to use for this host
  *
  * Establishes a cluster reservation, enables device connections, and
  * starts MMCs with appropriate DNTS parameters.
  */
-int wusbhc_start(struct wusbhc *wusbhc, const struct wusb_ckhdid *chid)
+int wusbhc_start(struct wusbhc *wusbhc)
 {
 	int result;
 	struct device *dev = wusbhc->dev;
@@ -181,7 +180,7 @@ int wusbhc_start(struct wusbhc *wusbhc, const struct wusb_ckhdid *chid)
 		goto error_rsv_establish;
 	}
 
-	result = wusbhc_devconnect_start(wusbhc, chid);
+	result = wusbhc_devconnect_start(wusbhc);
 	if (result < 0) {
 		dev_err(dev, "error enabling device connections: %d\n", result);
 		goto error_devconnect_start;
@@ -218,34 +217,6 @@ error_rsv_establish:
 	return result;
 }
 
-/*
- * Disconnect all from the WUSB Channel
- *
- * Send a Host Disconnect IE in the MMC, wait, don't send it any more
- */
-static int __wusbhc_host_disconnect_ie(struct wusbhc *wusbhc)
-{
-	int result = -ENOMEM;
-	struct wuie_host_disconnect *host_disconnect_ie;
-	might_sleep();
-	host_disconnect_ie = kmalloc(sizeof(*host_disconnect_ie), GFP_KERNEL);
-	if (host_disconnect_ie == NULL)
-		goto error_alloc;
-	host_disconnect_ie->hdr.bLength       = sizeof(*host_disconnect_ie);
-	host_disconnect_ie->hdr.bIEIdentifier = WUIE_ID_HOST_DISCONNECT;
-	result = wusbhc_mmcie_set(wusbhc, 0, 0, &host_disconnect_ie->hdr);
-	if (result < 0)
-		goto error_mmcie_set;
-
-	/* WUSB1.0[8.5.3.1 & 7.5.2] */
-	msleep(100);
-	wusbhc_mmcie_rm(wusbhc, &host_disconnect_ie->hdr);
-error_mmcie_set:
-	kfree(host_disconnect_ie);
-error_alloc:
-	return result;
-}
-
 /*
  * wusbhc_stop - stop transmitting MMCs
  * @wusbhc: the HC to stop
@@ -264,29 +235,6 @@ void wusbhc_stop(struct wusbhc *wusbhc)
 }
 EXPORT_SYMBOL_GPL(wusbhc_stop);
 
-/*
- * Change the CHID in a WUSB Channel
- *
- * If it is just a new CHID, send a Host Disconnect IE and then change
- * the CHID IE.
- */
-static int __wusbhc_chid_change(struct wusbhc *wusbhc,
-				const struct wusb_ckhdid *chid)
-{
-	int result = -ENOSYS;
-	struct device *dev = wusbhc->dev;
-	dev_err(dev, "%s() not implemented yet\n", __func__);
-	return result;
-
-	BUG_ON(wusbhc->wuie_host_info == NULL);
-	__wusbhc_host_disconnect_ie(wusbhc);
-	wusbhc->wuie_host_info->CHID = *chid;
-	result = wusbhc_mmcie_set(wusbhc, 0, 0, &wusbhc->wuie_host_info->hdr);
-	if (result < 0)
-		dev_err(dev, "Can't update Host Info WUSB IE: %d\n", result);
-	return result;
-}
-
 /*
  * Set/reset/update a new CHID
  *
@@ -302,16 +250,19 @@ int wusbhc_chid_set(struct wusbhc *wusbhc, const struct wusb_ckhdid *chid)
 		chid = NULL;
 
 	mutex_lock(&wusbhc->mutex);
-	if (wusbhc->active) {
-		if (chid)
-			result = __wusbhc_chid_change(wusbhc, chid);
-		else
-			wusbhc_stop(wusbhc);
-	} else {
-		if (chid)
-			wusbhc_start(wusbhc, chid);
+	if (chid) {
+		if (wusbhc->active) {
+			mutex_unlock(&wusbhc->mutex);
+			return -EBUSY;
+		}
+		wusbhc->chid = *chid;
 	}
 	mutex_unlock(&wusbhc->mutex);
+
+	if (chid)
+		result = uwb_radio_start(&wusbhc->pal);
+	else
+		uwb_radio_stop(&wusbhc->pal);
 	return result;
 }
 EXPORT_SYMBOL_GPL(wusbhc_chid_set);
diff --git a/drivers/usb/wusbcore/pal.c b/drivers/usb/wusbcore/pal.c
index 7cc51e9905cf..d0b172c5ecc7 100644
--- a/drivers/usb/wusbcore/pal.c
+++ b/drivers/usb/wusbcore/pal.c
@@ -18,6 +18,16 @@
  */
 #include "wusbhc.h"
 
+static void wusbhc_channel_changed(struct uwb_pal *pal, int channel)
+{
+	struct wusbhc *wusbhc = container_of(pal, struct wusbhc, pal);
+
+	if (channel < 0)
+		wusbhc_stop(wusbhc);
+	else
+		wusbhc_start(wusbhc);
+}
+
 /**
  * wusbhc_pal_register - register the WUSB HC as a UWB PAL
  * @wusbhc: the WUSB HC
@@ -28,8 +38,10 @@ int wusbhc_pal_register(struct wusbhc *wusbhc)
 
 	wusbhc->pal.name   = "wusbhc";
 	wusbhc->pal.device = wusbhc->usb_hcd.self.controller;
+	wusbhc->pal.rc     = wusbhc->uwb_rc;
+	wusbhc->pal.channel_changed = wusbhc_channel_changed;
 
-	return uwb_pal_register(wusbhc->uwb_rc, &wusbhc->pal);
+	return uwb_pal_register(&wusbhc->pal);
 }
 
 /**
@@ -38,5 +50,5 @@ int wusbhc_pal_register(struct wusbhc *wusbhc)
  */
 void wusbhc_pal_unregister(struct wusbhc *wusbhc)
 {
-	uwb_pal_unregister(wusbhc->uwb_rc, &wusbhc->pal);
+	uwb_pal_unregister(&wusbhc->pal);
 }
diff --git a/drivers/usb/wusbcore/wusbhc.h b/drivers/usb/wusbcore/wusbhc.h
index 8fef934ad2f3..797c2453a35b 100644
--- a/drivers/usb/wusbcore/wusbhc.h
+++ b/drivers/usb/wusbcore/wusbhc.h
@@ -252,7 +252,8 @@ struct wusbhc {
 	struct uwb_pal pal;
 
 	unsigned trust_timeout;			/* in jiffies */
-	struct wuie_host_info *wuie_host_info;	/* Includes CHID */
+	struct wusb_ckhdid chid;
+	struct wuie_host_info *wuie_host_info;
 
 	struct mutex mutex;			/* locks everything else */
 	u16 cluster_id;				/* Wireless USB Cluster ID */
@@ -376,15 +377,14 @@ static inline void wusbhc_put(struct wusbhc *wusbhc)
 	usb_put_hcd(&wusbhc->usb_hcd);
 }
 
-int wusbhc_start(struct wusbhc *wusbhc, const struct wusb_ckhdid *chid);
+int wusbhc_start(struct wusbhc *wusbhc);
 void wusbhc_stop(struct wusbhc *wusbhc);
 extern int wusbhc_chid_set(struct wusbhc *, const struct wusb_ckhdid *);
 
 /* Device connect handling */
 extern int wusbhc_devconnect_create(struct wusbhc *);
 extern void wusbhc_devconnect_destroy(struct wusbhc *);
-extern int wusbhc_devconnect_start(struct wusbhc *wusbhc,
-				   const struct wusb_ckhdid *chid);
+extern int wusbhc_devconnect_start(struct wusbhc *wusbhc);
 extern void wusbhc_devconnect_stop(struct wusbhc *wusbhc);
 extern void wusbhc_handle_dn(struct wusbhc *, u8 srcaddr,
 			     struct wusb_dn_hdr *dn_hdr, size_t size);
diff --git a/drivers/uwb/Makefile b/drivers/uwb/Makefile
index 2b99c3e61671..ce21a95da04a 100644
--- a/drivers/uwb/Makefile
+++ b/drivers/uwb/Makefile
@@ -18,6 +18,7 @@ uwb-objs :=		\
 	lc-rc.o		\
 	neh.o		\
 	pal.o		\
+	radio.o		\
 	reset.o		\
 	rsv.o		\
 	scan.o		\
diff --git a/drivers/uwb/beacon.c b/drivers/uwb/beacon.c
index d9f2a8acc593..247956098afa 100644
--- a/drivers/uwb/beacon.c
+++ b/drivers/uwb/beacon.c
@@ -119,7 +119,6 @@ int uwb_rc_beacon(struct uwb_rc *rc, int channel, unsigned bpst_offset)
 	int result;
 	struct device *dev = &rc->uwb_dev.dev;
 
-	mutex_lock(&rc->uwb_dev.mutex);
 	if (channel < 0)
 		channel = -1;
 	if (channel == -1)
@@ -128,7 +127,7 @@ int uwb_rc_beacon(struct uwb_rc *rc, int channel, unsigned bpst_offset)
 		/* channel >= 0...dah */
 		result = uwb_rc_start_beacon(rc, bpst_offset, channel);
 		if (result < 0)
-			goto out_up;
+			return result;
 		if (le16_to_cpu(rc->ies->wIELength) > 0) {
 			result = uwb_rc_set_ie(rc, rc->ies);
 			if (result < 0) {
@@ -137,19 +136,14 @@ int uwb_rc_beacon(struct uwb_rc *rc, int channel, unsigned bpst_offset)
 				result = uwb_rc_stop_beacon(rc);
 				channel = -1;
 				bpst_offset = 0;
-			} else
-				result = 0;
+			}
 		}
 	}
 
-	if (result < 0)
-		goto out_up;
-	rc->beaconing = channel;
-
-	uwb_notify(rc, NULL, uwb_bg_joined(rc) ? UWB_NOTIF_BG_JOIN : UWB_NOTIF_BG_LEAVE);
-
-out_up:
-	mutex_unlock(&rc->uwb_dev.mutex);
+	if (result >= 0) {
+		rc->beaconing = channel;
+		uwb_notify(rc, NULL, uwb_bg_joined(rc) ? UWB_NOTIF_BG_JOIN : UWB_NOTIF_BG_LEAVE);
+	}
 	return result;
 }
 
@@ -618,9 +612,6 @@ static ssize_t uwb_rc_beacon_show(struct device *dev,
 
 /*
  * Start beaconing on the specified channel, or stop beaconing.
- *
- * The BPST offset of when to start searching for a beacon group to
- * join may be specified.
  */
 static ssize_t uwb_rc_beacon_store(struct device *dev,
 				   struct device_attribute *attr,
@@ -629,12 +620,11 @@ static ssize_t uwb_rc_beacon_store(struct device *dev,
 	struct uwb_dev *uwb_dev = to_uwb_dev(dev);
 	struct uwb_rc *rc = uwb_dev->rc;
 	int channel;
-	unsigned bpst_offset = 0;
 	ssize_t result = -EINVAL;
 
-	result = sscanf(buf, "%d %u\n", &channel, &bpst_offset);
+	result = sscanf(buf, "%d", &channel);
 	if (result >= 1)
-		result = uwb_rc_beacon(rc, channel, bpst_offset);
+		result = uwb_radio_force_channel(rc, channel);
 
 	return result < 0 ? result : size;
 }
diff --git a/drivers/uwb/drp.c b/drivers/uwb/drp.c
index c0b1e5e2bd08..fe328146adb7 100644
--- a/drivers/uwb/drp.c
+++ b/drivers/uwb/drp.c
@@ -37,14 +37,13 @@
  *
  * A DRP Availability IE is appended.
  *
- * rc->uwb_dev.mutex is held
+ * rc->rsvs_mutex is held
  *
  * FIXME We currently ignore the returned value indicating the remaining space
  * in beacon. This could be used to deny reservation requests earlier if
  * determined that they would cause the beacon space to be exceeded.
  */
-static
-int uwb_rc_gen_send_drp_ie(struct uwb_rc *rc)
+int uwb_rc_send_all_drp_ie(struct uwb_rc *rc)
 {
 	int result;
 	struct device *dev = &rc->uwb_dev.dev;
@@ -102,25 +101,6 @@ error_cmd:
 	kfree(cmd);
 error:
 	return result;
-
-}
-/**
- * Send all DRP IEs associated with this host
- *
- * @returns:    >= 0 number of bytes still available in the beacon
- *              < 0 errno code on error.
- *
- * As per the protocol we obtain the host controller device lock to access
- * bandwidth structures.
- */
-int uwb_rc_send_all_drp_ie(struct uwb_rc *rc)
-{
-	int result;
-
-	mutex_lock(&rc->uwb_dev.mutex);
-	result = uwb_rc_gen_send_drp_ie(rc);
-	mutex_unlock(&rc->uwb_dev.mutex);
-	return result;
 }
 
 void uwb_drp_handle_timeout(struct uwb_rsv *rsv)
diff --git a/drivers/uwb/lc-rc.c b/drivers/uwb/lc-rc.c
index f00633d334dd..9cf21e6bb624 100644
--- a/drivers/uwb/lc-rc.c
+++ b/drivers/uwb/lc-rc.c
@@ -189,9 +189,9 @@ static int uwb_rc_setup(struct uwb_rc *rc)
 	int result;
 	struct device *dev = &rc->uwb_dev.dev;
 
-	result = uwb_rc_reset(rc);
+	result = uwb_radio_setup(rc);
 	if (result < 0) {
-		dev_err(dev, "cannot reset UWB radio: %d\n", result);
+		dev_err(dev, "cannot setup UWB radio: %d\n", result);
 		goto error;
 	}
 	result = uwb_rc_mac_addr_setup(rc);
@@ -311,12 +311,7 @@ void uwb_rc_rm(struct uwb_rc *rc)
 
 	uwb_dbg_del_rc(rc);
 	uwb_rsv_remove_all(rc);
-	uwb_rc_ie_rm(rc, UWB_IDENTIFICATION_IE);
-	if (rc->beaconing >= 0)
-		uwb_rc_beacon(rc, -1, 0);
-	if (rc->scan_type != UWB_SCAN_DISABLED)
-		uwb_rc_scan(rc, rc->scanning, UWB_SCAN_DISABLED, 0);
-	uwb_rc_reset(rc);
+	uwb_radio_shutdown(rc);
 
 	rc->stop(rc);
 
diff --git a/drivers/uwb/pal.c b/drivers/uwb/pal.c
index 1afb38eacb9a..605765124f5b 100644
--- a/drivers/uwb/pal.c
+++ b/drivers/uwb/pal.c
@@ -32,13 +32,13 @@ EXPORT_SYMBOL_GPL(uwb_pal_init);
 
 /**
  * uwb_pal_register - register a UWB PAL
- * @rc: the radio controller the PAL will be using
  * @pal: the PAL
  *
  * The PAL must be initialized with uwb_pal_init().
  */
-int uwb_pal_register(struct uwb_rc *rc, struct uwb_pal *pal)
+int uwb_pal_register(struct uwb_pal *pal)
 {
+	struct uwb_rc *rc = pal->rc;
 	int ret;
 
 	if (pal->device) {
@@ -54,9 +54,9 @@ int uwb_pal_register(struct uwb_rc *rc, struct uwb_pal *pal)
 		}
 	}
 
-	spin_lock(&rc->pal_lock);
+	mutex_lock(&rc->uwb_dev.mutex);
 	list_add(&pal->node, &rc->pals);
-	spin_unlock(&rc->pal_lock);
+	mutex_unlock(&rc->uwb_dev.mutex);
 
 	return 0;
 }
@@ -64,14 +64,17 @@ EXPORT_SYMBOL_GPL(uwb_pal_register);
 
 /**
  * uwb_pal_register - unregister a UWB PAL
- * @rc: the radio controller the PAL was using
  * @pal: the PAL
  */
-void uwb_pal_unregister(struct uwb_rc *rc, struct uwb_pal *pal)
+void uwb_pal_unregister(struct uwb_pal *pal)
 {
-	spin_lock(&rc->pal_lock);
+	struct uwb_rc *rc = pal->rc;
+
+	uwb_radio_stop(pal);
+
+	mutex_lock(&rc->uwb_dev.mutex);
 	list_del(&pal->node);
-	spin_unlock(&rc->pal_lock);
+	mutex_unlock(&rc->uwb_dev.mutex);
 
 	if (pal->device) {
 		sysfs_remove_link(&rc->uwb_dev.dev.kobj, pal->name);
@@ -86,6 +89,5 @@ EXPORT_SYMBOL_GPL(uwb_pal_unregister);
  */
 void uwb_rc_pal_init(struct uwb_rc *rc)
 {
-	spin_lock_init(&rc->pal_lock);
 	INIT_LIST_HEAD(&rc->pals);
 }
diff --git a/drivers/uwb/radio.c b/drivers/uwb/radio.c
new file mode 100644
index 000000000000..f0d55495f5e9
--- /dev/null
+++ b/drivers/uwb/radio.c
@@ -0,0 +1,202 @@
+/*
+ * UWB radio (channel) management.
+ *
+ * Copyright (C) 2008 Cambridge Silicon Radio Ltd.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+#include <linux/kernel.h>
+#include <linux/uwb.h>
+
+#include "uwb-internal.h"
+
+
+static int uwb_radio_select_channel(struct uwb_rc *rc)
+{
+	/*
+	 * Default to channel 9 (BG1, TFC1) unless the user has
+	 * selected a specific channel or there are no active PALs.
+	 */
+	if (rc->active_pals == 0)
+		return -1;
+	if (rc->beaconing_forced)
+		return rc->beaconing_forced;
+	return 9;
+}
+
+
+/*
+ * Notify all active PALs that the channel has changed.
+ */
+static void uwb_radio_channel_changed(struct uwb_rc *rc, int channel)
+{
+	struct uwb_pal *pal;
+
+	list_for_each_entry(pal, &rc->pals, node) {
+		if (pal->channel && channel != pal->channel) {
+			pal->channel = channel;
+			if (pal->channel_changed)
+				pal->channel_changed(pal, pal->channel);
+		}
+	}
+}
+
+/*
+ * Change to a new channel and notify any active PALs of the new
+ * channel.
+ *
+ * When stopping the radio, PALs need to be notified first so they can
+ * terminate any active reservations.
+ */
+static int uwb_radio_change_channel(struct uwb_rc *rc, int channel)
+{
+	int ret = 0;
+
+	if (channel == -1)
+		uwb_radio_channel_changed(rc, channel);
+
+	if (channel != rc->beaconing) {
+		if (rc->beaconing != -1 && channel != -1) {
+			/*
+			 * FIXME: should signal the channel change
+			 * with a Channel Change IE.
+			 */
+			ret = uwb_radio_change_channel(rc, -1);
+			if (ret < 0)
+				return ret;
+		}
+		ret = uwb_rc_beacon(rc, channel, 0);
+	}
+
+	if (channel != -1)
+		uwb_radio_channel_changed(rc, rc->beaconing);
+
+	return ret;
+}
+
+/**
+ * uwb_radio_start - request that the radio be started
+ * @pal: the PAL making the request.
+ *
+ * If the radio is not already active, aa suitable channel is selected
+ * and beacons are started.
+ */
+int uwb_radio_start(struct uwb_pal *pal)
+{
+	struct uwb_rc *rc = pal->rc;
+	int ret = 0;
+
+	mutex_lock(&rc->uwb_dev.mutex);
+
+	if (!pal->channel) {
+		pal->channel = -1;
+		rc->active_pals++;
+		ret = uwb_radio_change_channel(rc, uwb_radio_select_channel(rc));
+	}
+
+	mutex_unlock(&rc->uwb_dev.mutex);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(uwb_radio_start);
+
+/**
+ * uwb_radio_stop - request tha the radio be stopped.
+ * @pal: the PAL making the request.
+ *
+ * Stops the radio if no other PAL is making use of it.
+ */
+void uwb_radio_stop(struct uwb_pal *pal)
+{
+	struct uwb_rc *rc = pal->rc;
+
+	mutex_lock(&rc->uwb_dev.mutex);
+
+	if (pal->channel) {
+		rc->active_pals--;
+		uwb_radio_change_channel(rc, uwb_radio_select_channel(rc));
+		pal->channel = 0;
+	}
+
+	mutex_unlock(&rc->uwb_dev.mutex);
+}
+EXPORT_SYMBOL_GPL(uwb_radio_stop);
+
+/*
+ * uwb_radio_force_channel - force a specific channel to be used
+ * @rc: the radio controller.
+ * @channel: the channel to use; -1 to force the radio to stop; 0 to
+ *   use the default channel selection algorithm.
+ */
+int uwb_radio_force_channel(struct uwb_rc *rc, int channel)
+{
+	int ret = 0;
+
+	mutex_lock(&rc->uwb_dev.mutex);
+
+	rc->beaconing_forced = channel;
+	ret = uwb_radio_change_channel(rc, uwb_radio_select_channel(rc));
+
+	mutex_unlock(&rc->uwb_dev.mutex);
+	return ret;
+}
+
+/*
+ * uwb_radio_setup - setup the radio manager
+ * @rc: the radio controller.
+ *
+ * The radio controller is reset to ensure it's in a known state
+ * before it's used.
+ */
+int uwb_radio_setup(struct uwb_rc *rc)
+{
+	return uwb_rc_reset(rc);
+}
+
+/*
+ * uwb_radio_reset_state - reset any radio manager state
+ * @rc: the radio controller.
+ *
+ * All internal radio manager state is reset to values corresponding
+ * to a reset radio controller.
+ */
+void uwb_radio_reset_state(struct uwb_rc *rc)
+{
+	struct uwb_pal *pal;
+
+	mutex_lock(&rc->uwb_dev.mutex);
+
+	list_for_each_entry(pal, &rc->pals, node) {
+		if (pal->channel) {
+			pal->channel = -1;
+			if (pal->channel_changed)
+				pal->channel_changed(pal, -1);
+		}
+	}
+
+	rc->beaconing = -1;
+	rc->scanning = -1;
+
+	mutex_unlock(&rc->uwb_dev.mutex);
+}
+
+/*
+ * uwb_radio_shutdown - shutdown the radio manager
+ * @rc: the radio controller.
+ *
+ * The radio controller is reset.
+ */
+void uwb_radio_shutdown(struct uwb_rc *rc)
+{
+	uwb_radio_reset_state(rc);
+	uwb_rc_reset(rc);
+}
diff --git a/drivers/uwb/reset.c b/drivers/uwb/reset.c
index e39b32099af3..ce8283cc8098 100644
--- a/drivers/uwb/reset.c
+++ b/drivers/uwb/reset.c
@@ -365,11 +365,7 @@ void uwb_rc_pre_reset(struct uwb_rc *rc)
 	rc->stop(rc);
 	uwbd_flush(rc);
 
-	mutex_lock(&rc->uwb_dev.mutex);
-	rc->beaconing = -1;
-	rc->scanning = -1;
-	mutex_unlock(&rc->uwb_dev.mutex);
-
+	uwb_radio_reset_state(rc);
 	uwb_rsv_remove_all(rc);
 }
 EXPORT_SYMBOL_GPL(uwb_rc_pre_reset);
diff --git a/drivers/uwb/rsv.c b/drivers/uwb/rsv.c
index 935d5b536db7..1cd84f927540 100644
--- a/drivers/uwb/rsv.c
+++ b/drivers/uwb/rsv.c
@@ -555,14 +555,14 @@ static struct uwb_rsv *uwb_rsv_new_target(struct uwb_rc *rc,
 	 * deny the request.
 	 */
 	rsv->state = UWB_RSV_STATE_T_DENIED;
-	spin_lock(&rc->pal_lock);
+	mutex_lock(&rc->uwb_dev.mutex);
 	list_for_each_entry(pal, &rc->pals, node) {
 		if (pal->new_rsv)
 			pal->new_rsv(pal, rsv);
 		if (rsv->state == UWB_RSV_STATE_T_ACCEPTED)
 			break;
 	}
-	spin_unlock(&rc->pal_lock);
+	mutex_unlock(&rc->uwb_dev.mutex);
 
 	list_add_tail(&rsv->rc_node, &rc->reservations);
 	state = rsv->state;
diff --git a/drivers/uwb/uwb-debug.c b/drivers/uwb/uwb-debug.c
index 217ebaac128d..0e58071a232d 100644
--- a/drivers/uwb/uwb-debug.c
+++ b/drivers/uwb/uwb-debug.c
@@ -192,7 +192,7 @@ static ssize_t command_write(struct file *file, const char __user *buf,
 {
 	struct uwb_rc *rc = file->private_data;
 	struct uwb_dbg_cmd cmd;
-	int ret;
+	int ret = 0;
 
 	if (len != sizeof(struct uwb_dbg_cmd))
 		return -EINVAL;
@@ -213,6 +213,12 @@ static ssize_t command_write(struct file *file, const char __user *buf,
 	case UWB_DBG_CMD_IE_RM:
 		ret = cmd_ie_rm(rc, &cmd.ie_rm);
 		break;
+	case UWB_DBG_CMD_RADIO_START:
+		ret = uwb_radio_start(&rc->dbg->pal);
+		break;
+	case UWB_DBG_CMD_RADIO_STOP:
+		uwb_radio_stop(&rc->dbg->pal);
+		break;
 	default:
 		return -EINVAL;
 	}
@@ -306,6 +312,17 @@ static struct file_operations drp_avail_fops = {
 	.owner   = THIS_MODULE,
 };
 
+static void uwb_dbg_channel_changed(struct uwb_pal *pal, int channel)
+{
+	struct uwb_dbg *dbg = container_of(pal, struct uwb_dbg, pal);
+	struct device *dev = &pal->rc->uwb_dev.dev;
+
+	if (channel > 0)
+		dev_info(dev, "debug: channel %d started\n", channel);
+	else
+		dev_info(dev, "debug: channel stopped\n");
+}
+
 static void uwb_dbg_new_rsv(struct uwb_pal *pal, struct uwb_rsv *rsv)
 {
 	struct uwb_dbg *dbg = container_of(pal, struct uwb_dbg, pal);
@@ -329,8 +346,11 @@ void uwb_dbg_add_rc(struct uwb_rc *rc)
 	INIT_LIST_HEAD(&rc->dbg->rsvs);
 
 	uwb_pal_init(&rc->dbg->pal);
+	rc->dbg->pal.rc = rc;
+	rc->dbg->pal.channel_changed = uwb_dbg_channel_changed;
 	rc->dbg->pal.new_rsv = uwb_dbg_new_rsv;
-	uwb_pal_register(rc, &rc->dbg->pal);
+	uwb_pal_register(&rc->dbg->pal);
+
 	if (root_dir) {
 		rc->dbg->root_d = debugfs_create_dir(dev_name(&rc->uwb_dev.dev),
 						     root_dir);
@@ -364,7 +384,7 @@ void uwb_dbg_del_rc(struct uwb_rc *rc)
 		uwb_rsv_terminate(rsv);
 	}
 
-	uwb_pal_unregister(rc, &rc->dbg->pal);
+	uwb_pal_unregister(&rc->dbg->pal);
 
 	if (root_dir) {
 		debugfs_remove(rc->dbg->drp_avail_f);
diff --git a/drivers/uwb/uwb-internal.h b/drivers/uwb/uwb-internal.h
index af95541dabcd..9c0cdb4ded0c 100644
--- a/drivers/uwb/uwb-internal.h
+++ b/drivers/uwb/uwb-internal.h
@@ -238,6 +238,11 @@ struct uwb_dev *uwb_dev_get_by_devaddr(struct uwb_rc *rc,
 struct uwb_dev *uwb_dev_get_by_macaddr(struct uwb_rc *rc,
 				       const struct uwb_mac_addr *macaddr);
 
+int uwb_radio_setup(struct uwb_rc *rc);
+void uwb_radio_reset_state(struct uwb_rc *rc);
+void uwb_radio_shutdown(struct uwb_rc *rc);
+int uwb_radio_force_channel(struct uwb_rc *rc, int channel);
+
 /* -- UWB Sysfs representation */
 extern struct class uwb_rc_class;
 extern struct device_attribute dev_attr_mac_address;
diff --git a/drivers/uwb/wlp/wlp-lc.c b/drivers/uwb/wlp/wlp-lc.c
index 0799402e73fb..7e5eb49b03b8 100644
--- a/drivers/uwb/wlp/wlp-lc.c
+++ b/drivers/uwb/wlp/wlp-lc.c
@@ -543,7 +543,8 @@ int wlp_setup(struct wlp *wlp, struct uwb_rc *rc)
 	uwb_notifs_register(rc, &wlp->uwb_notifs_handler);
 
 	uwb_pal_init(&wlp->pal);
-	result = uwb_pal_register(rc, &wlp->pal);
+	wlp->pal.rc = rc;
+	result = uwb_pal_register(&wlp->pal);
 	if (result < 0)
 		uwb_notifs_deregister(wlp->rc, &wlp->uwb_notifs_handler);
 
@@ -557,7 +558,7 @@ void wlp_remove(struct wlp *wlp)
 	struct device *dev = &wlp->rc->uwb_dev.dev;
 	d_fnstart(6, dev, "wlp %p\n", wlp);
 	wlp_neighbors_release(wlp);
-	uwb_pal_unregister(wlp->rc, &wlp->pal);
+	uwb_pal_unregister(&wlp->pal);
 	uwb_notifs_deregister(wlp->rc, &wlp->uwb_notifs_handler);
 	wlp_eda_release(&wlp->eda);
 	mutex_lock(&wlp->mutex);
diff --git a/include/linux/uwb.h b/include/linux/uwb.h
index effd97998fd1..7d3ebf046f9a 100644
--- a/include/linux/uwb.h
+++ b/include/linux/uwb.h
@@ -355,6 +355,7 @@ struct uwb_rc {
 	u8 ctx_roll;
 
 	int beaconing;			/* Beaconing state [channel number] */
+	int beaconing_forced;
 	int scanning;
 	enum uwb_scan_type scan_type:3;
 	unsigned ready:1;
@@ -373,8 +374,8 @@ struct uwb_rc {
 	struct uwb_rc_cmd_set_ie *ies;
 	size_t ies_capacity;
 
-	spinlock_t pal_lock;
 	struct list_head pals;
+	int active_pals;
 
 	struct uwb_dbg *dbg;
 };
@@ -382,11 +383,17 @@ struct uwb_rc {
 
 /**
  * struct uwb_pal - a UWB PAL
- * @name:    descriptive name for this PAL (wushc, wlp, etc.).
+ * @name:    descriptive name for this PAL (wusbhc, wlp, etc.).
  * @device:  a device for the PAL.  Used to link the PAL and the radio
  *           controller in sysfs.
+ * @rc:      the radio controller the PAL uses.
+ * @channel_changed: called when the channel used by the radio changes.
+ *           A channel of -1 means the channel has been stopped.
  * @new_rsv: called when a peer requests a reservation (may be NULL if
  *           the PAL cannot accept reservation requests).
+ * @channel: channel being used by the PAL; 0 if the PAL isn't using
+ *           the radio; -1 if the PAL wishes to use the radio but
+ *           cannot.
  *
  * A Protocol Adaptation Layer (PAL) is a user of the WiMedia UWB
  * radio platform (e.g., WUSB, WLP or Bluetooth UWB AMP).
@@ -405,12 +412,20 @@ struct uwb_pal {
 	struct list_head node;
 	const char *name;
 	struct device *device;
+	struct uwb_rc *rc;
+
+	void (*channel_changed)(struct uwb_pal *pal, int channel);
 	void (*new_rsv)(struct uwb_pal *pal, struct uwb_rsv *rsv);
+
+	int channel;
 };
 
 void uwb_pal_init(struct uwb_pal *pal);
-int uwb_pal_register(struct uwb_rc *rc, struct uwb_pal *pal);
-void uwb_pal_unregister(struct uwb_rc *rc, struct uwb_pal *pal);
+int uwb_pal_register(struct uwb_pal *pal);
+void uwb_pal_unregister(struct uwb_pal *pal);
+
+int uwb_radio_start(struct uwb_pal *pal);
+void uwb_radio_stop(struct uwb_pal *pal);
 
 /*
  * General public API
diff --git a/include/linux/uwb/debug-cmd.h b/include/linux/uwb/debug-cmd.h
index 6a16566f0221..07efbe17db53 100644
--- a/include/linux/uwb/debug-cmd.h
+++ b/include/linux/uwb/debug-cmd.h
@@ -34,6 +34,8 @@ enum uwb_dbg_cmd_type {
 	UWB_DBG_CMD_RSV_TERMINATE = 2,
 	UWB_DBG_CMD_IE_ADD = 3,
 	UWB_DBG_CMD_IE_RM = 4,
+	UWB_DBG_CMD_RADIO_START = 5,
+	UWB_DBG_CMD_RADIO_STOP = 6,
 };
 
 struct uwb_dbg_cmd_rsv_establish {
-- 
cgit v1.2.3


From e8e1594c8126b1b773988fa2e3bfec76cff88336 Mon Sep 17 00:00:00 2001
From: David Vrabel <david.vrabel@csr.com>
Date: Mon, 17 Nov 2008 16:16:51 +0000
Subject: wlp: start/stop radio on network interface up/down

Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/uwb/i1480/i1480u-wlp/lc.c     |  2 +-
 drivers/uwb/i1480/i1480u-wlp/netdev.c | 51 ++++++++---------------------------
 drivers/uwb/uwb-debug.c               |  3 ---
 drivers/uwb/wlp/wlp-lc.c              | 14 +++++++++-
 include/linux/wlp.h                   |  3 ++-
 5 files changed, 27 insertions(+), 46 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/uwb/i1480/i1480u-wlp/lc.c b/drivers/uwb/i1480/i1480u-wlp/lc.c
index 384306c11bbe..488b2e30a0a8 100644
--- a/drivers/uwb/i1480/i1480u-wlp/lc.c
+++ b/drivers/uwb/i1480/i1480u-wlp/lc.c
@@ -206,7 +206,7 @@ int i1480u_add(struct i1480u *i1480u, struct usb_interface *iface)
 	wlp->fill_device_info = i1480u_fill_device_info;
 	wlp->stop_queue = i1480u_stop_queue;
 	wlp->start_queue = i1480u_start_queue;
-	result = wlp_setup(wlp, rc);
+	result = wlp_setup(wlp, rc, net_dev);
 	if (result < 0) {
 		dev_err(&iface->dev, "Cannot setup WLP\n");
 		goto error_wlp_setup;
diff --git a/drivers/uwb/i1480/i1480u-wlp/netdev.c b/drivers/uwb/i1480/i1480u-wlp/netdev.c
index 8802ac43d872..2eafb973cd05 100644
--- a/drivers/uwb/i1480/i1480u-wlp/netdev.c
+++ b/drivers/uwb/i1480/i1480u-wlp/netdev.c
@@ -207,6 +207,11 @@ int i1480u_open(struct net_device *net_dev)
 	result = i1480u_rx_setup(i1480u);		/* Alloc RX stuff */
 	if (result < 0)
 		goto error_rx_setup;
+
+	result = uwb_radio_start(&wlp->pal);
+	if (result < 0)
+		goto error_radio_start;
+
 	netif_wake_queue(net_dev);
 #ifdef i1480u_FLOW_CONTROL
 	result = usb_submit_urb(i1480u->notif_urb, GFP_KERNEL);;
@@ -215,25 +220,20 @@ int i1480u_open(struct net_device *net_dev)
 		goto error_notif_urb_submit;
 	}
 #endif
-	i1480u->uwb_notifs_handler.cb = i1480u_uwb_notifs_cb;
-	i1480u->uwb_notifs_handler.data = i1480u;
-	if (uwb_bg_joined(rc))
-		netif_carrier_on(net_dev);
-	else
-		netif_carrier_off(net_dev);
-	uwb_notifs_register(rc, &i1480u->uwb_notifs_handler);
 	/* Interface is up with an address, now we can create WSS */
 	result = wlp_wss_setup(net_dev, &wlp->wss);
 	if (result < 0) {
 		dev_err(dev, "Can't create WSS: %d. \n", result);
-		goto error_notif_deregister;
+		goto error_wss_setup;
 	}
 	return 0;
-error_notif_deregister:
-	uwb_notifs_deregister(rc, &i1480u->uwb_notifs_handler);
+error_wss_setup:
 #ifdef i1480u_FLOW_CONTROL
+	usb_kill_urb(i1480u->notif_urb);
 error_notif_urb_submit:
 #endif
+	uwb_radio_stop(&wlp->pal);
+error_radio_start:
 	netif_stop_queue(net_dev);
 	i1480u_rx_release(i1480u);
 error_rx_setup:
@@ -248,16 +248,15 @@ int i1480u_stop(struct net_device *net_dev)
 {
 	struct i1480u *i1480u = netdev_priv(net_dev);
 	struct wlp *wlp = &i1480u->wlp;
-	struct uwb_rc *rc = wlp->rc;
 
 	BUG_ON(wlp->rc == NULL);
 	wlp_wss_remove(&wlp->wss);
-	uwb_notifs_deregister(rc, &i1480u->uwb_notifs_handler);
 	netif_carrier_off(net_dev);
 #ifdef i1480u_FLOW_CONTROL
 	usb_kill_urb(i1480u->notif_urb);
 #endif
 	netif_stop_queue(net_dev);
+	uwb_radio_stop(&wlp->pal);
 	i1480u_rx_release(i1480u);
 	i1480u_tx_release(i1480u);
 	return 0;
@@ -303,34 +302,6 @@ int i1480u_change_mtu(struct net_device *net_dev, int mtu)
 	return 0;
 }
 
-
-/**
- * Callback function to handle events from UWB
- * When we see other devices we know the carrier is ok,
- * if we are the only device in the beacon group we set the carrier
- * state to off.
- * */
-void i1480u_uwb_notifs_cb(void *data, struct uwb_dev *uwb_dev,
-			  enum uwb_notifs event)
-{
-	struct i1480u *i1480u = data;
-	struct net_device *net_dev = i1480u->net_dev;
-	struct device *dev = &i1480u->usb_iface->dev;
-	switch (event) {
-	case UWB_NOTIF_BG_JOIN:
-		netif_carrier_on(net_dev);
-		dev_info(dev, "Link is up\n");
-		break;
-	case UWB_NOTIF_BG_LEAVE:
-		netif_carrier_off(net_dev);
-		dev_info(dev, "Link is down\n");
-		break;
-	default:
-		dev_err(dev, "don't know how to handle event %d from uwb\n",
-				event);
-	}
-}
-
 /**
  * Stop the network queue
  *
diff --git a/drivers/uwb/uwb-debug.c b/drivers/uwb/uwb-debug.c
index 0e58071a232d..e02fb83469d5 100644
--- a/drivers/uwb/uwb-debug.c
+++ b/drivers/uwb/uwb-debug.c
@@ -33,8 +33,6 @@
 #include <linux/seq_file.h>
 
 #include <linux/uwb/debug-cmd.h>
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
 
 #include "uwb-internal.h"
 
@@ -314,7 +312,6 @@ static struct file_operations drp_avail_fops = {
 
 static void uwb_dbg_channel_changed(struct uwb_pal *pal, int channel)
 {
-	struct uwb_dbg *dbg = container_of(pal, struct uwb_dbg, pal);
 	struct device *dev = &pal->rc->uwb_dev.dev;
 
 	if (channel > 0)
diff --git a/drivers/uwb/wlp/wlp-lc.c b/drivers/uwb/wlp/wlp-lc.c
index 7e5eb49b03b8..e531093c4162 100644
--- a/drivers/uwb/wlp/wlp-lc.c
+++ b/drivers/uwb/wlp/wlp-lc.c
@@ -526,7 +526,17 @@ void wlp_uwb_notifs_cb(void *_wlp, struct uwb_dev *uwb_dev,
 	}
 }
 
-int wlp_setup(struct wlp *wlp, struct uwb_rc *rc)
+static void wlp_channel_changed(struct uwb_pal *pal, int channel)
+{
+	struct wlp *wlp = container_of(pal, struct wlp, pal);
+
+	if (channel < 0)
+		netif_carrier_off(wlp->ndev);
+	else
+		netif_carrier_on(wlp->ndev);
+}
+
+int wlp_setup(struct wlp *wlp, struct uwb_rc *rc, struct net_device *ndev)
 {
 	struct device *dev = &rc->uwb_dev.dev;
 	int result;
@@ -537,6 +547,7 @@ int wlp_setup(struct wlp *wlp, struct uwb_rc *rc)
 	BUG_ON(wlp->stop_queue == NULL);
 	BUG_ON(wlp->start_queue == NULL);
 	wlp->rc = rc;
+	wlp->ndev = ndev;
 	wlp_eda_init(&wlp->eda);/* Set up address cache */
 	wlp->uwb_notifs_handler.cb = wlp_uwb_notifs_cb;
 	wlp->uwb_notifs_handler.data = wlp;
@@ -544,6 +555,7 @@ int wlp_setup(struct wlp *wlp, struct uwb_rc *rc)
 
 	uwb_pal_init(&wlp->pal);
 	wlp->pal.rc = rc;
+	wlp->pal.channel_changed = wlp_channel_changed;
 	result = uwb_pal_register(&wlp->pal);
 	if (result < 0)
 		uwb_notifs_deregister(wlp->rc, &wlp->uwb_notifs_handler);
diff --git a/include/linux/wlp.h b/include/linux/wlp.h
index 033545e145c7..ac95ce6606ac 100644
--- a/include/linux/wlp.h
+++ b/include/linux/wlp.h
@@ -646,6 +646,7 @@ struct wlp_wss {
 struct wlp {
 	struct mutex mutex;
 	struct uwb_rc *rc;		/* UWB radio controller */
+	struct net_device *ndev;
 	struct uwb_pal pal;
 	struct wlp_eda eda;
 	struct wlp_uuid uuid;
@@ -675,7 +676,7 @@ struct wlp_wss_attribute {
 static struct wlp_wss_attribute wss_attr_##_name = __ATTR(_name, _mode,	\
 							  _show, _store)
 
-extern int wlp_setup(struct wlp *, struct uwb_rc *);
+extern int wlp_setup(struct wlp *, struct uwb_rc *, struct net_device *ndev);
 extern void wlp_remove(struct wlp *);
 extern ssize_t wlp_neighborhood_show(struct wlp *, char *);
 extern int wlp_wss_setup(struct net_device *, struct wlp_wss *);
-- 
cgit v1.2.3


From 0996e6382482ce9014787693d3884e9468153a5c Mon Sep 17 00:00:00 2001
From: David Vrabel <david.vrabel@csr.com>
Date: Mon, 17 Nov 2008 16:23:22 +0000
Subject: uwb: remove unused beacon group join/leave events

The UWB_NOTIF_BG_JOIN/UWB_NOTIF_BG_LEAVE events have been
superceeded by the channel_changed callback in struct uwb_pal.

Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/uwb/beacon.c | 17 +----------------
 include/linux/uwb.h  |  7 +++----
 2 files changed, 4 insertions(+), 20 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/uwb/beacon.c b/drivers/uwb/beacon.c
index 247956098afa..d9c60cb94993 100644
--- a/drivers/uwb/beacon.c
+++ b/drivers/uwb/beacon.c
@@ -140,10 +140,8 @@ int uwb_rc_beacon(struct uwb_rc *rc, int channel, unsigned bpst_offset)
 		}
 	}
 
-	if (result >= 0) {
+	if (result >= 0)
 		rc->beaconing = channel;
-		uwb_notify(rc, NULL, uwb_bg_joined(rc) ? UWB_NOTIF_BG_JOIN : UWB_NOTIF_BG_LEAVE);
-	}
 	return result;
 }
 
@@ -581,19 +579,6 @@ error:
 	return result;
 }
 
-/**
- * uwb_bg_joined - is the RC in a beacon group?
- * @rc: the radio controller
- *
- * Returns true if the radio controller is in a beacon group (even if
- * it's the sole member).
- */
-int uwb_bg_joined(struct uwb_rc *rc)
-{
-	return rc->beaconing != -1;
-}
-EXPORT_SYMBOL_GPL(uwb_bg_joined);
-
 /*
  * Print beaconing state.
  */
diff --git a/include/linux/uwb.h b/include/linux/uwb.h
index 7d3ebf046f9a..1719709d60ca 100644
--- a/include/linux/uwb.h
+++ b/include/linux/uwb.h
@@ -479,7 +479,6 @@ ssize_t uwb_rc_vcmd(struct uwb_rc *rc, const char *cmd_name,
 		    struct uwb_rccb *cmd, size_t cmd_size,
 		    u8 expected_type, u16 expected_event,
 		    struct uwb_rceb **preply);
-int uwb_bg_joined(struct uwb_rc *rc);
 
 size_t __uwb_addr_print(char *, size_t, const unsigned char *, int);
 
@@ -568,7 +567,9 @@ static inline bool uwb_rsv_is_owner(struct uwb_rsv *rsv)
 }
 
 /**
- * Events generated by UWB that can be passed to any listeners
+ * enum uwb_notifs - UWB events that can be passed to any listeners
+ * @UWB_NOTIF_ONAIR: a new neighbour has joined the beacon group.
+ * @UWB_NOTIF_OFFAIR: a neighbour has left the beacon group.
  *
  * Higher layers can register callback functions with the radio
  * controller using uwb_notifs_register(). The radio controller
@@ -576,8 +577,6 @@ static inline bool uwb_rsv_is_owner(struct uwb_rsv *rsv)
  * nodes when an event occurs.
  */
 enum uwb_notifs {
-	UWB_NOTIF_BG_JOIN = 0,	/* radio controller joined a beacon group */
-	UWB_NOTIF_BG_LEAVE = 1,	/* radio controller left a beacon group */
 	UWB_NOTIF_ONAIR,
 	UWB_NOTIF_OFFAIR,
 };
-- 
cgit v1.2.3


From 3eb1aa43ef5cb871ba3fb2f08633675eca374d2e Mon Sep 17 00:00:00 2001
From: Jaya Kumar <jayakumar.lkml@gmail.com>
Date: Wed, 19 Nov 2008 16:58:50 -0500
Subject: Input: add support for Wacom W8001 penabled serial touchscreen

The Wacom W8001 sensor is a sensor device (uses electromagnetic
resonance) and it is interfaced via its serial microcontroller
to the host.

Signed-off-by: Jaya Kumar <jayakumar.lkml@gmail.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/touchscreen/Kconfig       |  13 ++
 drivers/input/touchscreen/Makefile      |   1 +
 drivers/input/touchscreen/wacom_w8001.c | 325 ++++++++++++++++++++++++++++++++
 include/linux/serio.h                   |   1 +
 4 files changed, 340 insertions(+)
 create mode 100644 drivers/input/touchscreen/wacom_w8001.c

(limited to 'include/linux')

diff --git a/drivers/input/touchscreen/Kconfig b/drivers/input/touchscreen/Kconfig
index 3d1ab8fa9acc..20eb52ed176d 100644
--- a/drivers/input/touchscreen/Kconfig
+++ b/drivers/input/touchscreen/Kconfig
@@ -95,6 +95,19 @@ config TOUCHSCREEN_ELO
 	  To compile this driver as a module, choose M here: the
 	  module will be called elo.
 
+config TOUCHSCREEN_WACOM_W8001
+	tristate "Wacom W8001 penabled serial touchscreen"
+	select SERIO
+	help
+	  Say Y here if you have an Wacom W8001 penabled serial touchscreen
+	  connected to your system.
+
+	  If unsure, say N.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called wacom_w8001.
+
+
 config TOUCHSCREEN_MTOUCH
 	tristate "MicroTouch serial touchscreens"
 	select SERIO
diff --git a/drivers/input/touchscreen/Makefile b/drivers/input/touchscreen/Makefile
index 15cf29079489..3dc84d3846c1 100644
--- a/drivers/input/touchscreen/Makefile
+++ b/drivers/input/touchscreen/Makefile
@@ -26,6 +26,7 @@ obj-$(CONFIG_TOUCHSCREEN_TOUCHIT213)	+= touchit213.o
 obj-$(CONFIG_TOUCHSCREEN_TOUCHRIGHT)	+= touchright.o
 obj-$(CONFIG_TOUCHSCREEN_TOUCHWIN)	+= touchwin.o
 obj-$(CONFIG_TOUCHSCREEN_UCB1400)	+= ucb1400_ts.o
+obj-$(CONFIG_TOUCHSCREEN_WACOM_W8001)	+= wacom_w8001.o
 obj-$(CONFIG_TOUCHSCREEN_WM97XX)	+= wm97xx-ts.o
 wm97xx-ts-$(CONFIG_TOUCHSCREEN_WM9705)	+= wm9705.o
 wm97xx-ts-$(CONFIG_TOUCHSCREEN_WM9712)	+= wm9712.o
diff --git a/drivers/input/touchscreen/wacom_w8001.c b/drivers/input/touchscreen/wacom_w8001.c
new file mode 100644
index 000000000000..2f33a0167644
--- /dev/null
+++ b/drivers/input/touchscreen/wacom_w8001.c
@@ -0,0 +1,325 @@
+/*
+ * Wacom W8001 penabled serial touchscreen driver
+ *
+ * Copyright (c) 2008 Jaya Kumar
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file COPYING in the main directory of this archive for
+ * more details.
+ *
+ * Layout based on Elo serial touchscreen driver by Vojtech Pavlik
+ */
+
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/input.h>
+#include <linux/serio.h>
+#include <linux/init.h>
+#include <linux/ctype.h>
+
+#define DRIVER_DESC	"Wacom W8001 serial touchscreen driver"
+
+MODULE_AUTHOR("Jaya Kumar <jayakumar.lkml@gmail.com>");
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
+
+/*
+ * Definitions & global arrays.
+ */
+
+#define W8001_MAX_LENGTH	11
+#define W8001_PACKET_LEN	11
+#define W8001_LEAD_MASK 0x80
+#define W8001_LEAD_BYTE 0x80
+#define W8001_TAB_MASK 0x40
+#define W8001_TAB_BYTE 0x40
+
+#define W8001_QUERY_PACKET 0x20
+
+struct w8001_coord {
+	u8 rdy;
+	u8 tsw;
+	u8 f1;
+	u8 f2;
+	u16 x;
+	u16 y;
+	u16 pen_pressure;
+	u8 tilt_x;
+	u8 tilt_y;
+};
+
+/*
+ * Per-touchscreen data.
+ */
+
+struct w8001 {
+	struct input_dev *dev;
+	struct serio *serio;
+	struct mutex cmd_mutex;
+	struct completion cmd_done;
+	int id;
+	int idx;
+	unsigned char expected_packet;
+	unsigned char data[W8001_MAX_LENGTH];
+	unsigned char response[W8001_PACKET_LEN];
+	char phys[32];
+};
+
+static int parse_data(u8 *data, struct w8001_coord *coord)
+{
+	coord->rdy = data[0] & 0x20;
+	coord->tsw = data[0] & 0x01;
+	coord->f1 = data[0] & 0x02;
+	coord->f2 = data[0] & 0x04;
+
+	coord->x = (data[1] & 0x7F) << 9;
+	coord->x |= (data[2] & 0x7F) << 2;
+	coord->x |= (data[6] & 0x60) >> 5;
+
+	coord->y = (data[3] & 0x7F) << 9;
+	coord->y |= (data[4] & 0x7F) << 2;
+	coord->y |= (data[6] & 0x18) >> 3;
+
+	coord->pen_pressure = data[5] & 0x7F;
+	coord->pen_pressure |= (data[6] & 0x07) << 7 ;
+
+	coord->tilt_x = data[7] & 0x7F;
+	coord->tilt_y = data[8] & 0x7F;
+
+	return 0;
+}
+
+static void w8001_process_data(struct w8001 *w8001, unsigned char data)
+{
+	struct input_dev *dev = w8001->dev;
+	u8 tmp;
+	struct w8001_coord coord;
+
+	w8001->data[w8001->idx] = data;
+	switch (w8001->idx++) {
+	case 0:
+		if ((data & W8001_LEAD_MASK) != W8001_LEAD_BYTE) {
+			pr_debug("w8001: unsynchronized data: 0x%02x\n", data);
+			w8001->idx = 0;
+		}
+		break;
+	case 8:
+		tmp = w8001->data[0] & W8001_TAB_MASK;
+		if (unlikely(tmp == W8001_TAB_BYTE))
+			break;
+		w8001->idx = 0;
+		memset(&coord, 0, sizeof(coord));
+		parse_data(w8001->data, &coord);
+		input_report_abs(dev, ABS_X, coord.x);
+		input_report_abs(dev, ABS_Y, coord.y);
+		input_report_abs(dev, ABS_PRESSURE, coord.pen_pressure);
+		input_report_key(dev, BTN_TOUCH, coord.tsw);
+		input_sync(dev);
+		break;
+	case 10:
+		w8001->idx = 0;
+		memcpy(w8001->response, &w8001->data, W8001_PACKET_LEN);
+		w8001->expected_packet = W8001_QUERY_PACKET;
+		complete(&w8001->cmd_done);
+		break;
+	}
+}
+
+
+static irqreturn_t w8001_interrupt(struct serio *serio,
+		unsigned char data, unsigned int flags)
+{
+	struct w8001 *w8001 = serio_get_drvdata(serio);
+
+	w8001_process_data(w8001, data);
+
+	return IRQ_HANDLED;
+}
+
+static int w8001_async_command(struct w8001 *w8001, unsigned char *packet,
+					int len)
+{
+	int rc = -1;
+	int i;
+
+	mutex_lock(&w8001->cmd_mutex);
+
+	for (i = 0; i < len; i++) {
+		if (serio_write(w8001->serio, packet[i]))
+			goto out;
+	}
+	rc = 0;
+
+out:
+	mutex_unlock(&w8001->cmd_mutex);
+	return rc;
+}
+
+static int w8001_command(struct w8001 *w8001, unsigned char *packet, int len)
+{
+	int rc = -1;
+	int i;
+
+	mutex_lock(&w8001->cmd_mutex);
+
+	serio_pause_rx(w8001->serio);
+	init_completion(&w8001->cmd_done);
+	serio_continue_rx(w8001->serio);
+
+	for (i = 0; i < len; i++) {
+		if (serio_write(w8001->serio, packet[i]))
+			goto out;
+	}
+
+	wait_for_completion_timeout(&w8001->cmd_done, HZ);
+
+	if (w8001->expected_packet == W8001_QUERY_PACKET) {
+		/* We are back in reporting mode, the query was ACKed */
+		memcpy(packet, w8001->response, W8001_PACKET_LEN);
+		rc = 0;
+	}
+
+out:
+	mutex_unlock(&w8001->cmd_mutex);
+	return rc;
+}
+
+static int w8001_setup(struct w8001 *w8001)
+{
+	struct w8001_coord coord;
+	struct input_dev *dev = w8001->dev;
+	unsigned char start[1] = { '1' };
+	unsigned char query[11] = { '*' };
+
+	if (w8001_command(w8001, query, 1))
+		return -1;
+
+	memset(&coord, 0, sizeof(coord));
+	parse_data(query, &coord);
+
+	input_set_abs_params(dev, ABS_X, 0, coord.x, 0, 0);
+	input_set_abs_params(dev, ABS_Y, 0, coord.y, 0, 0);
+	input_set_abs_params(dev, ABS_PRESSURE, 0, coord.pen_pressure, 0, 0);
+	input_set_abs_params(dev, ABS_TILT_X, 0, coord.tilt_x, 0, 0);
+	input_set_abs_params(dev, ABS_TILT_Y, 0, coord.tilt_y, 0, 0);
+
+	if (w8001_async_command(w8001, start, 1))
+		return -1;
+
+	return 0;
+}
+
+/*
+ * w8001_disconnect() is the opposite of w8001_connect()
+ */
+
+static void w8001_disconnect(struct serio *serio)
+{
+	struct w8001 *w8001 = serio_get_drvdata(serio);
+
+	input_get_device(w8001->dev);
+	input_unregister_device(w8001->dev);
+	serio_close(serio);
+	serio_set_drvdata(serio, NULL);
+	input_put_device(w8001->dev);
+	kfree(w8001);
+}
+
+/*
+ * w8001_connect() is the routine that is called when someone adds a
+ * new serio device that supports the w8001 protocol and registers it as
+ * an input device.
+ */
+
+static int w8001_connect(struct serio *serio, struct serio_driver *drv)
+{
+	struct w8001 *w8001;
+	struct input_dev *input_dev;
+	int err;
+
+	w8001 = kzalloc(sizeof(struct w8001), GFP_KERNEL);
+	input_dev = input_allocate_device();
+	if (!w8001 || !input_dev) {
+		err = -ENOMEM;
+		goto fail1;
+	}
+
+	w8001->serio = serio;
+	w8001->id = serio->id.id;
+	w8001->dev = input_dev;
+	mutex_init(&w8001->cmd_mutex);
+	init_completion(&w8001->cmd_done);
+	snprintf(w8001->phys, sizeof(w8001->phys), "%s/input0", serio->phys);
+
+	input_dev->name = "Wacom W8001 Penabled Serial TouchScreen";
+	input_dev->phys = w8001->phys;
+	input_dev->id.bustype = BUS_RS232;
+	input_dev->id.vendor = SERIO_W8001;
+	input_dev->id.product = w8001->id;
+	input_dev->id.version = 0x0100;
+	input_dev->dev.parent = &serio->dev;
+
+	input_dev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_ABS);
+	input_dev->keybit[BIT_WORD(BTN_TOUCH)] = BIT_MASK(BTN_TOUCH);
+
+	serio_set_drvdata(serio, w8001);
+	err = serio_open(serio, drv);
+	if (err)
+		goto fail2;
+
+	if (w8001_setup(w8001))
+		goto fail3;
+
+	err = input_register_device(w8001->dev);
+	if (err)
+		goto fail3;
+
+	return 0;
+
+fail3:
+	serio_close(serio);
+fail2:
+	serio_set_drvdata(serio, NULL);
+fail1:
+	input_free_device(input_dev);
+	kfree(w8001);
+	return err;
+}
+
+static struct serio_device_id w8001_serio_ids[] = {
+	{
+		.type	= SERIO_RS232,
+		.proto	= SERIO_W8001,
+		.id	= SERIO_ANY,
+		.extra	= SERIO_ANY,
+	},
+	{ 0 }
+};
+
+MODULE_DEVICE_TABLE(serio, w8001_serio_ids);
+
+static struct serio_driver w8001_drv = {
+	.driver		= {
+		.name	= "w8001",
+	},
+	.description	= DRIVER_DESC,
+	.id_table	= w8001_serio_ids,
+	.interrupt	= w8001_interrupt,
+	.connect	= w8001_connect,
+	.disconnect	= w8001_disconnect,
+};
+
+static int __init w8001_init(void)
+{
+	return serio_register_driver(&w8001_drv);
+}
+
+static void __exit w8001_exit(void)
+{
+	serio_unregister_driver(&w8001_drv);
+}
+
+module_init(w8001_init);
+module_exit(w8001_exit);
diff --git a/include/linux/serio.h b/include/linux/serio.h
index 25641d9e0ea8..1bcb357a01a1 100644
--- a/include/linux/serio.h
+++ b/include/linux/serio.h
@@ -213,5 +213,6 @@ static inline void serio_unpin_driver(struct serio *serio)
 #define SERIO_ZHENHUA	0x36
 #define SERIO_INEXIO	0x37
 #define SERIO_TOUCHIT213	0x37
+#define SERIO_W8001	0x39
 
 #endif
-- 
cgit v1.2.3


From a2d781fc8d9b16113dd9440107d73c0f21d7cbef Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Date: Wed, 19 Nov 2008 17:02:24 -0500
Subject: Input: libps2 - handle 0xfc responses from devices

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/serio/libps2.c | 20 ++++++++++++++++----
 include/linux/libps2.h       |  2 ++
 2 files changed, 18 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/input/serio/libps2.c b/drivers/input/serio/libps2.c
index 2b304c22c200..67248c31e19a 100644
--- a/drivers/input/serio/libps2.c
+++ b/drivers/input/serio/libps2.c
@@ -262,9 +262,17 @@ int ps2_handle_ack(struct ps2dev *ps2dev, unsigned char data)
 			break;
 
 		case PS2_RET_NAK:
-			ps2dev->nak = 1;
+			ps2dev->flags |= PS2_FLAG_NAK;
+			ps2dev->nak = PS2_RET_NAK;
 			break;
 
+		case PS2_RET_ERR:
+			if (ps2dev->flags & PS2_FLAG_NAK) {
+				ps2dev->flags &= ~PS2_FLAG_NAK;
+				ps2dev->nak = PS2_RET_ERR;
+				break;
+			}
+
 		/*
 		 * Workaround for mice which don't ACK the Get ID command.
 		 * These are valid mouse IDs that we recognize.
@@ -282,8 +290,11 @@ int ps2_handle_ack(struct ps2dev *ps2dev, unsigned char data)
 	}
 
 
-	if (!ps2dev->nak && ps2dev->cmdcnt)
-		ps2dev->flags |= PS2_FLAG_CMD | PS2_FLAG_CMD1;
+	if (!ps2dev->nak) {
+		ps2dev->flags &= ~PS2_FLAG_NAK;
+		if (ps2dev->cmdcnt)
+			ps2dev->flags |= PS2_FLAG_CMD | PS2_FLAG_CMD1;
+	}
 
 	ps2dev->flags &= ~PS2_FLAG_ACK;
 	wake_up(&ps2dev->wait);
@@ -329,6 +340,7 @@ void ps2_cmd_aborted(struct ps2dev *ps2dev)
 	if (ps2dev->flags & (PS2_FLAG_ACK | PS2_FLAG_CMD))
 		wake_up(&ps2dev->wait);
 
-	ps2dev->flags = 0;
+	/* reset all flags except last nack */
+	ps2dev->flags &= PS2_FLAG_NAK;
 }
 EXPORT_SYMBOL(ps2_cmd_aborted);
diff --git a/include/linux/libps2.h b/include/linux/libps2.h
index afc413369101..b94534b7e266 100644
--- a/include/linux/libps2.h
+++ b/include/linux/libps2.h
@@ -18,11 +18,13 @@
 #define PS2_RET_ID		0x00
 #define PS2_RET_ACK		0xfa
 #define PS2_RET_NAK		0xfe
+#define PS2_RET_ERR		0xfc
 
 #define PS2_FLAG_ACK		1	/* Waiting for ACK/NAK */
 #define PS2_FLAG_CMD		2	/* Waiting for command to finish */
 #define PS2_FLAG_CMD1		4	/* Waiting for the first byte of command response */
 #define PS2_FLAG_WAITID		8	/* Command execiting is GET ID */
+#define PS2_FLAG_NAK		16	/* Last transmission was NAKed */
 
 struct ps2dev {
 	struct serio *serio;
-- 
cgit v1.2.3


From 758b2cdc6f6a22c702bd8f2344382fb1270b2161 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 25 Nov 2008 02:35:04 +1030
Subject: sched: wrap sched_group and sched_domain cpumask accesses.

Impact: trivial wrap of member accesses

This eases the transition in the next patch.

We also get rid of a temporary cpumask in find_idlest_cpu() thanks to
for_each_cpu_and, and sched_balance_self() due to getting weight before
setting sd to NULL.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |  10 +++++
 kernel/sched.c        | 114 ++++++++++++++++++++++++--------------------------
 kernel/sched_fair.c   |  10 ++---
 kernel/sched_rt.c     |   3 +-
 kernel/sched_stats.h  |   3 +-
 5 files changed, 73 insertions(+), 67 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4ce5c603c51a..2b95aa9f779b 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -786,6 +786,11 @@ struct sched_group {
 	u32 reciprocal_cpu_power;
 };
 
+static inline struct cpumask *sched_group_cpus(struct sched_group *sg)
+{
+	return &sg->cpumask;
+}
+
 enum sched_domain_level {
 	SD_LV_NONE = 0,
 	SD_LV_SIBLING,
@@ -866,6 +871,11 @@ struct sched_domain {
 #endif
 };
 
+static inline struct cpumask *sched_domain_span(struct sched_domain *sd)
+{
+	return &sd->span;
+}
+
 extern void partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
 				    struct sched_domain_attr *dattr_new);
 extern int arch_reinit_sched_domains(void);
diff --git a/kernel/sched.c b/kernel/sched.c
index a2de33d05340..575f38acf4da 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1501,7 +1501,7 @@ static int tg_shares_up(struct task_group *tg, void *data)
 	struct sched_domain *sd = data;
 	int i;
 
-	for_each_cpu_mask(i, sd->span) {
+	for_each_cpu(i, sched_domain_span(sd)) {
 		/*
 		 * If there are currently no tasks on the cpu pretend there
 		 * is one of average load so that when a new task gets to
@@ -1522,7 +1522,7 @@ static int tg_shares_up(struct task_group *tg, void *data)
 	if (!sd->parent || !(sd->parent->flags & SD_LOAD_BALANCE))
 		shares = tg->shares;
 
-	for_each_cpu_mask(i, sd->span)
+	for_each_cpu(i, sched_domain_span(sd))
 		update_group_shares_cpu(tg, i, shares, rq_weight);
 
 	return 0;
@@ -2053,15 +2053,17 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)
 		int i;
 
 		/* Skip over this group if it has no CPUs allowed */
-		if (!cpus_intersects(group->cpumask, p->cpus_allowed))
+		if (!cpumask_intersects(sched_group_cpus(group),
+					&p->cpus_allowed))
 			continue;
 
-		local_group = cpu_isset(this_cpu, group->cpumask);
+		local_group = cpumask_test_cpu(this_cpu,
+					       sched_group_cpus(group));
 
 		/* Tally up the load of all CPUs in the group */
 		avg_load = 0;
 
-		for_each_cpu(i, &group->cpumask) {
+		for_each_cpu(i, sched_group_cpus(group)) {
 			/* Bias balancing toward cpus of our domain */
 			if (local_group)
 				load = source_load(i, load_idx);
@@ -2093,17 +2095,14 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)
  * find_idlest_cpu - find the idlest cpu among the cpus in group.
  */
 static int
-find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu,
-		cpumask_t *tmp)
+find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
 {
 	unsigned long load, min_load = ULONG_MAX;
 	int idlest = -1;
 	int i;
 
 	/* Traverse only the allowed CPUs */
-	cpus_and(*tmp, group->cpumask, p->cpus_allowed);
-
-	for_each_cpu(i, tmp) {
+	for_each_cpu_and(i, sched_group_cpus(group), &p->cpus_allowed) {
 		load = weighted_cpuload(i);
 
 		if (load < min_load || (load == min_load && i == this_cpu)) {
@@ -2145,7 +2144,6 @@ static int sched_balance_self(int cpu, int flag)
 		update_shares(sd);
 
 	while (sd) {
-		cpumask_t span, tmpmask;
 		struct sched_group *group;
 		int new_cpu, weight;
 
@@ -2154,14 +2152,13 @@ static int sched_balance_self(int cpu, int flag)
 			continue;
 		}
 
-		span = sd->span;
 		group = find_idlest_group(sd, t, cpu);
 		if (!group) {
 			sd = sd->child;
 			continue;
 		}
 
-		new_cpu = find_idlest_cpu(group, t, cpu, &tmpmask);
+		new_cpu = find_idlest_cpu(group, t, cpu);
 		if (new_cpu == -1 || new_cpu == cpu) {
 			/* Now try balancing at a lower domain level of cpu */
 			sd = sd->child;
@@ -2170,10 +2167,10 @@ static int sched_balance_self(int cpu, int flag)
 
 		/* Now try balancing at a lower domain level of new_cpu */
 		cpu = new_cpu;
+		weight = cpumask_weight(sched_domain_span(sd));
 		sd = NULL;
-		weight = cpus_weight(span);
 		for_each_domain(cpu, tmp) {
-			if (weight <= cpus_weight(tmp->span))
+			if (weight <= cpumask_weight(sched_domain_span(tmp)))
 				break;
 			if (tmp->flags & flag)
 				sd = tmp;
@@ -2218,7 +2215,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 		cpu = task_cpu(p);
 
 		for_each_domain(this_cpu, sd) {
-			if (cpu_isset(cpu, sd->span)) {
+			if (cpumask_test_cpu(cpu, sched_domain_span(sd))) {
 				update_shares(sd);
 				break;
 			}
@@ -2266,7 +2263,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 	else {
 		struct sched_domain *sd;
 		for_each_domain(this_cpu, sd) {
-			if (cpu_isset(cpu, sd->span)) {
+			if (cpumask_test_cpu(cpu, sched_domain_span(sd))) {
 				schedstat_inc(sd, ttwu_wake_remote);
 				break;
 			}
@@ -3109,10 +3106,11 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		unsigned long sum_avg_load_per_task;
 		unsigned long avg_load_per_task;
 
-		local_group = cpu_isset(this_cpu, group->cpumask);
+		local_group = cpumask_test_cpu(this_cpu,
+					       sched_group_cpus(group));
 
 		if (local_group)
-			balance_cpu = first_cpu(group->cpumask);
+			balance_cpu = cpumask_first(sched_group_cpus(group));
 
 		/* Tally up the load of all CPUs in the group */
 		sum_weighted_load = sum_nr_running = avg_load = 0;
@@ -3121,13 +3119,8 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		max_cpu_load = 0;
 		min_cpu_load = ~0UL;
 
-		for_each_cpu(i, &group->cpumask) {
-			struct rq *rq;
-
-			if (!cpu_isset(i, *cpus))
-				continue;
-
-			rq = cpu_rq(i);
+		for_each_cpu_and(i, sched_group_cpus(group), cpus) {
+			struct rq *rq = cpu_rq(i);
 
 			if (*sd_idle && rq->nr_running)
 				*sd_idle = 0;
@@ -3238,8 +3231,8 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		 */
 		if ((sum_nr_running < min_nr_running) ||
 		    (sum_nr_running == min_nr_running &&
-		     first_cpu(group->cpumask) <
-		     first_cpu(group_min->cpumask))) {
+		     cpumask_first(sched_group_cpus(group)) <
+		     cpumask_first(sched_group_cpus(group_min)))) {
 			group_min = group;
 			min_nr_running = sum_nr_running;
 			min_load_per_task = sum_weighted_load /
@@ -3254,8 +3247,8 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		if (sum_nr_running <= group_capacity - 1) {
 			if (sum_nr_running > leader_nr_running ||
 			    (sum_nr_running == leader_nr_running &&
-			     first_cpu(group->cpumask) >
-			      first_cpu(group_leader->cpumask))) {
+			     cpumask_first(sched_group_cpus(group)) >
+			     cpumask_first(sched_group_cpus(group_leader)))) {
 				group_leader = group;
 				leader_nr_running = sum_nr_running;
 			}
@@ -3400,7 +3393,7 @@ find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle,
 	unsigned long max_load = 0;
 	int i;
 
-	for_each_cpu(i, &group->cpumask) {
+	for_each_cpu(i, sched_group_cpus(group)) {
 		unsigned long wl;
 
 		if (!cpu_isset(i, *cpus))
@@ -3746,7 +3739,7 @@ static void active_load_balance(struct rq *busiest_rq, int busiest_cpu)
 	/* Search for an sd spanning us and the target CPU. */
 	for_each_domain(target_cpu, sd) {
 		if ((sd->flags & SD_LOAD_BALANCE) &&
-		    cpu_isset(busiest_cpu, sd->span))
+		    cpumask_test_cpu(busiest_cpu, sched_domain_span(sd)))
 				break;
 	}
 
@@ -6618,7 +6611,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
 	struct sched_group *group = sd->groups;
 	char str[256];
 
-	cpulist_scnprintf(str, sizeof(str), sd->span);
+	cpulist_scnprintf(str, sizeof(str), *sched_domain_span(sd));
 	cpus_clear(*groupmask);
 
 	printk(KERN_DEBUG "%*s domain %d: ", level, "", level);
@@ -6633,11 +6626,11 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
 
 	printk(KERN_CONT "span %s level %s\n", str, sd->name);
 
-	if (!cpu_isset(cpu, sd->span)) {
+	if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) {
 		printk(KERN_ERR "ERROR: domain->span does not contain "
 				"CPU%d\n", cpu);
 	}
-	if (!cpu_isset(cpu, group->cpumask)) {
+	if (!cpumask_test_cpu(cpu, sched_group_cpus(group))) {
 		printk(KERN_ERR "ERROR: domain->groups does not contain"
 				" CPU%d\n", cpu);
 	}
@@ -6657,31 +6650,32 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
 			break;
 		}
 
-		if (!cpus_weight(group->cpumask)) {
+		if (!cpumask_weight(sched_group_cpus(group))) {
 			printk(KERN_CONT "\n");
 			printk(KERN_ERR "ERROR: empty group\n");
 			break;
 		}
 
-		if (cpus_intersects(*groupmask, group->cpumask)) {
+		if (cpumask_intersects(groupmask, sched_group_cpus(group))) {
 			printk(KERN_CONT "\n");
 			printk(KERN_ERR "ERROR: repeated CPUs\n");
 			break;
 		}
 
-		cpus_or(*groupmask, *groupmask, group->cpumask);
+		cpumask_or(groupmask, groupmask, sched_group_cpus(group));
 
-		cpulist_scnprintf(str, sizeof(str), group->cpumask);
+		cpulist_scnprintf(str, sizeof(str), *sched_group_cpus(group));
 		printk(KERN_CONT " %s", str);
 
 		group = group->next;
 	} while (group != sd->groups);
 	printk(KERN_CONT "\n");
 
-	if (!cpus_equal(sd->span, *groupmask))
+	if (!cpumask_equal(sched_domain_span(sd), groupmask))
 		printk(KERN_ERR "ERROR: groups don't span domain->span\n");
 
-	if (sd->parent && !cpus_subset(*groupmask, sd->parent->span))
+	if (sd->parent &&
+	    !cpumask_subset(groupmask, sched_domain_span(sd->parent)))
 		printk(KERN_ERR "ERROR: parent span is not a superset "
 			"of domain->span\n");
 	return 0;
@@ -6721,7 +6715,7 @@ static void sched_domain_debug(struct sched_domain *sd, int cpu)
 
 static int sd_degenerate(struct sched_domain *sd)
 {
-	if (cpus_weight(sd->span) == 1)
+	if (cpumask_weight(sched_domain_span(sd)) == 1)
 		return 1;
 
 	/* Following flags need at least 2 groups */
@@ -6752,7 +6746,7 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent)
 	if (sd_degenerate(parent))
 		return 1;
 
-	if (!cpus_equal(sd->span, parent->span))
+	if (!cpumask_equal(sched_domain_span(sd), sched_domain_span(parent)))
 		return 0;
 
 	/* Does parent contain flags not in child? */
@@ -6913,10 +6907,10 @@ init_sched_build_groups(const cpumask_t *span, const cpumask_t *cpu_map,
 		int group = group_fn(i, cpu_map, &sg, tmpmask);
 		int j;
 
-		if (cpu_isset(i, *covered))
+		if (cpumask_test_cpu(i, covered))
 			continue;
 
-		cpus_clear(sg->cpumask);
+		cpumask_clear(sched_group_cpus(sg));
 		sg->__cpu_power = 0;
 
 		for_each_cpu(j, span) {
@@ -6924,7 +6918,7 @@ init_sched_build_groups(const cpumask_t *span, const cpumask_t *cpu_map,
 				continue;
 
 			cpu_set(j, *covered);
-			cpu_set(j, sg->cpumask);
+			cpumask_set_cpu(j, sched_group_cpus(sg));
 		}
 		if (!first)
 			first = sg;
@@ -7119,11 +7113,11 @@ static void init_numa_sched_groups_power(struct sched_group *group_head)
 	if (!sg)
 		return;
 	do {
-		for_each_cpu(j, &sg->cpumask) {
+		for_each_cpu(j, sched_group_cpus(sg)) {
 			struct sched_domain *sd;
 
 			sd = &per_cpu(phys_domains, j);
-			if (j != first_cpu(sd->groups->cpumask)) {
+			if (j != cpumask_first(sched_group_cpus(sd->groups))) {
 				/*
 				 * Only add "power" once for each
 				 * physical package.
@@ -7200,7 +7194,7 @@ static void init_sched_groups_power(int cpu, struct sched_domain *sd)
 
 	WARN_ON(!sd || !sd->groups);
 
-	if (cpu != first_cpu(sd->groups->cpumask))
+	if (cpu != cpumask_first(sched_group_cpus(sd->groups)))
 		return;
 
 	child = sd->child;
@@ -7372,7 +7366,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 			sd = &per_cpu(allnodes_domains, i);
 			SD_INIT(sd, ALLNODES);
 			set_domain_attribute(sd, attr);
-			sd->span = *cpu_map;
+			cpumask_copy(sched_domain_span(sd), cpu_map);
 			cpu_to_allnodes_group(i, cpu_map, &sd->groups, tmpmask);
 			p = sd;
 			sd_allnodes = 1;
@@ -7382,18 +7376,19 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		sd = &per_cpu(node_domains, i);
 		SD_INIT(sd, NODE);
 		set_domain_attribute(sd, attr);
-		sched_domain_node_span(cpu_to_node(i), &sd->span);
+		sched_domain_node_span(cpu_to_node(i), sched_domain_span(sd));
 		sd->parent = p;
 		if (p)
 			p->child = sd;
-		cpus_and(sd->span, sd->span, *cpu_map);
+		cpumask_and(sched_domain_span(sd),
+			    sched_domain_span(sd), cpu_map);
 #endif
 
 		p = sd;
 		sd = &per_cpu(phys_domains, i);
 		SD_INIT(sd, CPU);
 		set_domain_attribute(sd, attr);
-		sd->span = *nodemask;
+		cpumask_copy(sched_domain_span(sd), nodemask);
 		sd->parent = p;
 		if (p)
 			p->child = sd;
@@ -7404,8 +7399,9 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		sd = &per_cpu(core_domains, i);
 		SD_INIT(sd, MC);
 		set_domain_attribute(sd, attr);
-		sd->span = cpu_coregroup_map(i);
-		cpus_and(sd->span, sd->span, *cpu_map);
+		*sched_domain_span(sd) = cpu_coregroup_map(i);
+		cpumask_and(sched_domain_span(sd),
+			    sched_domain_span(sd), cpu_map);
 		sd->parent = p;
 		p->child = sd;
 		cpu_to_core_group(i, cpu_map, &sd->groups, tmpmask);
@@ -7416,8 +7412,8 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		sd = &per_cpu(cpu_domains, i);
 		SD_INIT(sd, SIBLING);
 		set_domain_attribute(sd, attr);
-		sd->span = per_cpu(cpu_sibling_map, i);
-		cpus_and(sd->span, sd->span, *cpu_map);
+		cpumask_and(sched_domain_span(sd),
+			    &per_cpu(cpu_sibling_map, i), cpu_map);
 		sd->parent = p;
 		p->child = sd;
 		cpu_to_cpu_group(i, cpu_map, &sd->groups, tmpmask);
@@ -7503,7 +7499,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 			sd->groups = sg;
 		}
 		sg->__cpu_power = 0;
-		sg->cpumask = *nodemask;
+		cpumask_copy(sched_group_cpus(sg), nodemask);
 		sg->next = sg;
 		cpus_or(*covered, *covered, *nodemask);
 		prev = sg;
@@ -7530,7 +7526,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 				goto error;
 			}
 			sg->__cpu_power = 0;
-			sg->cpumask = *tmpmask;
+			cpumask_copy(sched_group_cpus(sg), tmpmask);
 			sg->next = prev->next;
 			cpus_or(*covered, *covered, *tmpmask);
 			prev->next = sg;
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 98345e45b059..bba00402ed90 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1024,7 +1024,6 @@ static void yield_task_fair(struct rq *rq)
 #if defined(ARCH_HAS_SCHED_WAKE_IDLE)
 static int wake_idle(int cpu, struct task_struct *p)
 {
-	cpumask_t tmp;
 	struct sched_domain *sd;
 	int i;
 
@@ -1044,10 +1043,9 @@ static int wake_idle(int cpu, struct task_struct *p)
 		if ((sd->flags & SD_WAKE_IDLE)
 		    || ((sd->flags & SD_WAKE_IDLE_FAR)
 			&& !task_hot(p, task_rq(p)->clock, sd))) {
-			cpus_and(tmp, sd->span, p->cpus_allowed);
-			cpus_and(tmp, tmp, cpu_active_map);
-			for_each_cpu_mask_nr(i, tmp) {
-				if (idle_cpu(i)) {
+			for_each_cpu_and(i, sched_domain_span(sd),
+					 &p->cpus_allowed) {
+				if (cpu_active(i) && idle_cpu(i)) {
 					if (i != task_cpu(p)) {
 						schedstat_inc(p,
 						       se.nr_wakeups_idle);
@@ -1240,7 +1238,7 @@ static int select_task_rq_fair(struct task_struct *p, int sync)
 	 * this_cpu and prev_cpu are present in:
 	 */
 	for_each_domain(this_cpu, sd) {
-		if (cpu_isset(prev_cpu, sd->span)) {
+		if (cpumask_test_cpu(prev_cpu, sched_domain_span(sd))) {
 			this_sd = sd;
 			break;
 		}
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 2bdd44423599..4cd813abc23a 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -1017,7 +1017,8 @@ static int find_lowest_rq(struct task_struct *task)
 			cpumask_t domain_mask;
 			int       best_cpu;
 
-			cpus_and(domain_mask, sd->span, *lowest_mask);
+			cpumask_and(&domain_mask, sched_domain_span(sd),
+				    lowest_mask);
 
 			best_cpu = pick_optimal_cpu(this_cpu,
 						    &domain_mask);
diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h
index 7dbf72a2b02c..ce340835d055 100644
--- a/kernel/sched_stats.h
+++ b/kernel/sched_stats.h
@@ -42,7 +42,8 @@ static int show_schedstat(struct seq_file *seq, void *v)
 		for_each_domain(cpu, sd) {
 			enum cpu_idle_type itype;
 
-			cpumask_scnprintf(mask_str, mask_len, sd->span);
+			cpumask_scnprintf(mask_str, mask_len,
+					  *sched_domain_span(sd));
 			seq_printf(seq, "domain%d %s", dcount++, mask_str);
 			for (itype = CPU_IDLE; itype < CPU_MAX_IDLE_TYPES;
 					itype++) {
-- 
cgit v1.2.3


From 6c99e9ad47d9c082bd096f42fb49e397b05d58a8 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 25 Nov 2008 02:35:04 +1030
Subject: sched: convert struct sched_group/sched_domain cpumask_ts to variable
 bitmaps

Impact: (future) size reduction for large NR_CPUS.

We move the 'cpumask' member of sched_group to the end, so when we
kmalloc it we can do a minimal allocation: saves space for small
nr_cpu_ids but big CONFIG_NR_CPUS.  Similar trick for 'span' in
sched_domain.

This isn't quite as good as converting to a cpumask_var_t, as some
sched_groups are actually static, but it's safer: we don't have to
figure out where to call alloc_cpumask_var/free_cpumask_var.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h | 11 +++++----
 kernel/sched.c        | 65 ++++++++++++++++++++++++++++++++-------------------
 2 files changed, 48 insertions(+), 28 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2b95aa9f779b..c5be6c6bc741 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -771,7 +771,6 @@ enum cpu_idle_type {
 
 struct sched_group {
 	struct sched_group *next;	/* Must be a circular list */
-	cpumask_t cpumask;
 
 	/*
 	 * CPU power of this group, SCHED_LOAD_SCALE being max power for a
@@ -784,11 +783,13 @@ struct sched_group {
 	 * (see include/linux/reciprocal_div.h)
 	 */
 	u32 reciprocal_cpu_power;
+
+	unsigned long cpumask[];
 };
 
 static inline struct cpumask *sched_group_cpus(struct sched_group *sg)
 {
-	return &sg->cpumask;
+	return to_cpumask(sg->cpumask);
 }
 
 enum sched_domain_level {
@@ -814,7 +815,6 @@ struct sched_domain {
 	struct sched_domain *parent;	/* top domain must be null terminated */
 	struct sched_domain *child;	/* bottom domain must be null terminated */
 	struct sched_group *groups;	/* the balancing groups of the domain */
-	cpumask_t span;			/* span of all CPUs in this domain */
 	unsigned long min_interval;	/* Minimum balance interval ms */
 	unsigned long max_interval;	/* Maximum balance interval ms */
 	unsigned int busy_factor;	/* less balancing by factor if busy */
@@ -869,11 +869,14 @@ struct sched_domain {
 #ifdef CONFIG_SCHED_DEBUG
 	char *name;
 #endif
+
+	/* span of all CPUs in this domain */
+	unsigned long span[];
 };
 
 static inline struct cpumask *sched_domain_span(struct sched_domain *sd)
 {
-	return &sd->span;
+	return to_cpumask(sd->span);
 }
 
 extern void partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
diff --git a/kernel/sched.c b/kernel/sched.c
index 575f38acf4da..6b9606a6cabf 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -7005,19 +7005,34 @@ static void sched_domain_node_span(int node, cpumask_t *span)
 
 int sched_smt_power_savings = 0, sched_mc_power_savings = 0;
 
+/*
+ * The cpus mask in sched_group and sched_domain hangs off the end.
+ * FIXME: use cpumask_var_t or dynamic percpu alloc to avoid wasting space
+ * for nr_cpu_ids < CONFIG_NR_CPUS.
+ */
+struct static_sched_group {
+	struct sched_group sg;
+	DECLARE_BITMAP(cpus, CONFIG_NR_CPUS);
+};
+
+struct static_sched_domain {
+	struct sched_domain sd;
+	DECLARE_BITMAP(span, CONFIG_NR_CPUS);
+};
+
 /*
  * SMT sched-domains:
  */
 #ifdef CONFIG_SCHED_SMT
-static DEFINE_PER_CPU(struct sched_domain, cpu_domains);
-static DEFINE_PER_CPU(struct sched_group, sched_group_cpus);
+static DEFINE_PER_CPU(struct static_sched_domain, cpu_domains);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_cpus);
 
 static int
 cpu_to_cpu_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
 		 cpumask_t *unused)
 {
 	if (sg)
-		*sg = &per_cpu(sched_group_cpus, cpu);
+		*sg = &per_cpu(sched_group_cpus, cpu).sg;
 	return cpu;
 }
 #endif /* CONFIG_SCHED_SMT */
@@ -7026,8 +7041,8 @@ cpu_to_cpu_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
  * multi-core sched-domains:
  */
 #ifdef CONFIG_SCHED_MC
-static DEFINE_PER_CPU(struct sched_domain, core_domains);
-static DEFINE_PER_CPU(struct sched_group, sched_group_core);
+static DEFINE_PER_CPU(struct static_sched_domain, core_domains);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_core);
 #endif /* CONFIG_SCHED_MC */
 
 #if defined(CONFIG_SCHED_MC) && defined(CONFIG_SCHED_SMT)
@@ -7041,7 +7056,7 @@ cpu_to_core_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
 	cpus_and(*mask, *mask, *cpu_map);
 	group = first_cpu(*mask);
 	if (sg)
-		*sg = &per_cpu(sched_group_core, group);
+		*sg = &per_cpu(sched_group_core, group).sg;
 	return group;
 }
 #elif defined(CONFIG_SCHED_MC)
@@ -7050,13 +7065,13 @@ cpu_to_core_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
 		  cpumask_t *unused)
 {
 	if (sg)
-		*sg = &per_cpu(sched_group_core, cpu);
+		*sg = &per_cpu(sched_group_core, cpu).sg;
 	return cpu;
 }
 #endif
 
-static DEFINE_PER_CPU(struct sched_domain, phys_domains);
-static DEFINE_PER_CPU(struct sched_group, sched_group_phys);
+static DEFINE_PER_CPU(struct static_sched_domain, phys_domains);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_phys);
 
 static int
 cpu_to_phys_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
@@ -7075,7 +7090,7 @@ cpu_to_phys_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
 	group = cpu;
 #endif
 	if (sg)
-		*sg = &per_cpu(sched_group_phys, group);
+		*sg = &per_cpu(sched_group_phys, group).sg;
 	return group;
 }
 
@@ -7089,7 +7104,7 @@ static DEFINE_PER_CPU(struct sched_domain, node_domains);
 static struct sched_group ***sched_group_nodes_bycpu;
 
 static DEFINE_PER_CPU(struct sched_domain, allnodes_domains);
-static DEFINE_PER_CPU(struct sched_group, sched_group_allnodes);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_allnodes);
 
 static int cpu_to_allnodes_group(int cpu, const cpumask_t *cpu_map,
 				 struct sched_group **sg, cpumask_t *nodemask)
@@ -7101,7 +7116,7 @@ static int cpu_to_allnodes_group(int cpu, const cpumask_t *cpu_map,
 	group = first_cpu(*nodemask);
 
 	if (sg)
-		*sg = &per_cpu(sched_group_allnodes, group);
+		*sg = &per_cpu(sched_group_allnodes, group).sg;
 	return group;
 }
 
@@ -7116,7 +7131,7 @@ static void init_numa_sched_groups_power(struct sched_group *group_head)
 		for_each_cpu(j, sched_group_cpus(sg)) {
 			struct sched_domain *sd;
 
-			sd = &per_cpu(phys_domains, j);
+			sd = &per_cpu(phys_domains, j).sd;
 			if (j != cpumask_first(sched_group_cpus(sd->groups))) {
 				/*
 				 * Only add "power" once for each
@@ -7385,7 +7400,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 #endif
 
 		p = sd;
-		sd = &per_cpu(phys_domains, i);
+		sd = &per_cpu(phys_domains, i).sd;
 		SD_INIT(sd, CPU);
 		set_domain_attribute(sd, attr);
 		cpumask_copy(sched_domain_span(sd), nodemask);
@@ -7396,7 +7411,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 
 #ifdef CONFIG_SCHED_MC
 		p = sd;
-		sd = &per_cpu(core_domains, i);
+		sd = &per_cpu(core_domains, i).sd;
 		SD_INIT(sd, MC);
 		set_domain_attribute(sd, attr);
 		*sched_domain_span(sd) = cpu_coregroup_map(i);
@@ -7409,7 +7424,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 
 #ifdef CONFIG_SCHED_SMT
 		p = sd;
-		sd = &per_cpu(cpu_domains, i);
+		sd = &per_cpu(cpu_domains, i).sd;
 		SD_INIT(sd, SIBLING);
 		set_domain_attribute(sd, attr);
 		cpumask_and(sched_domain_span(sd),
@@ -7485,7 +7500,8 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		sched_domain_node_span(i, domainspan);
 		cpus_and(*domainspan, *domainspan, *cpu_map);
 
-		sg = kmalloc_node(sizeof(struct sched_group), GFP_KERNEL, i);
+		sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(),
+				  GFP_KERNEL, i);
 		if (!sg) {
 			printk(KERN_WARNING "Can not alloc domain group for "
 				"node %d\n", i);
@@ -7518,7 +7534,8 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 			if (cpus_empty(*tmpmask))
 				continue;
 
-			sg = kmalloc_node(sizeof(struct sched_group),
+			sg = kmalloc_node(sizeof(struct sched_group) +
+					  cpumask_size(),
 					  GFP_KERNEL, i);
 			if (!sg) {
 				printk(KERN_WARNING
@@ -7538,21 +7555,21 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 	/* Calculate CPU power for physical packages and nodes */
 #ifdef CONFIG_SCHED_SMT
 	for_each_cpu(i, cpu_map) {
-		struct sched_domain *sd = &per_cpu(cpu_domains, i);
+		struct sched_domain *sd = &per_cpu(cpu_domains, i).sd;
 
 		init_sched_groups_power(i, sd);
 	}
 #endif
 #ifdef CONFIG_SCHED_MC
 	for_each_cpu(i, cpu_map) {
-		struct sched_domain *sd = &per_cpu(core_domains, i);
+		struct sched_domain *sd = &per_cpu(core_domains, i).sd;
 
 		init_sched_groups_power(i, sd);
 	}
 #endif
 
 	for_each_cpu(i, cpu_map) {
-		struct sched_domain *sd = &per_cpu(phys_domains, i);
+		struct sched_domain *sd = &per_cpu(phys_domains, i).sd;
 
 		init_sched_groups_power(i, sd);
 	}
@@ -7574,11 +7591,11 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 	for_each_cpu(i, cpu_map) {
 		struct sched_domain *sd;
 #ifdef CONFIG_SCHED_SMT
-		sd = &per_cpu(cpu_domains, i);
+		sd = &per_cpu(cpu_domains, i).sd;
 #elif defined(CONFIG_SCHED_MC)
-		sd = &per_cpu(core_domains, i);
+		sd = &per_cpu(core_domains, i).sd;
 #else
-		sd = &per_cpu(phys_domains, i);
+		sd = &per_cpu(phys_domains, i).sd;
 #endif
 		cpu_attach_domain(sd, rd, i);
 	}
-- 
cgit v1.2.3


From 6a7b3dc3440f7b5a9b67594af01ed562cdeb41e4 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 25 Nov 2008 02:35:04 +1030
Subject: sched: convert nohz_cpu_mask to cpumask_var_t.

Impact: (future) size reduction for large NR_CPUS.

Dynamically allocating cpumasks (when CONFIG_CPUMASK_OFFSTACK) saves
space for small nr_cpu_ids but big CONFIG_NR_CPUS.  cpumask_var_t
is just a struct cpumask for !CONFIG_CPUMASK_OFFSTACK.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h    |  2 +-
 kernel/rcuclassic.c      |  2 +-
 kernel/sched.c           |  7 +++++--
 kernel/time/tick-sched.c | 10 +++++-----
 4 files changed, 12 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index c5be6c6bc741..1e33e2cb7f8c 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -249,7 +249,7 @@ extern void init_idle_bootup_task(struct task_struct *idle);
 extern int runqueue_is_locked(void);
 extern void task_rq_unlock_wait(struct task_struct *p);
 
-extern cpumask_t nohz_cpu_mask;
+extern cpumask_var_t nohz_cpu_mask;
 #if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ)
 extern int select_nohz_load_balancer(int cpu);
 #else
diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c
index e503a002f330..c03ca3e61919 100644
--- a/kernel/rcuclassic.c
+++ b/kernel/rcuclassic.c
@@ -393,7 +393,7 @@ static void rcu_start_batch(struct rcu_ctrlblk *rcp)
 		 * unnecessarily.
 		 */
 		smp_mb();
-		cpus_andnot(rcp->cpumask, cpu_online_map, nohz_cpu_mask);
+		cpumask_andnot(&rcp->cpumask, cpu_online_mask, nohz_cpu_mask);
 
 		rcp->signaled = 0;
 	}
diff --git a/kernel/sched.c b/kernel/sched.c
index 6b9606a6cabf..2723d7a4a42d 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -5870,9 +5870,9 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu)
  * indicates which cpus entered this state. This is used
  * in the rcu update to wait only for active cpus. For system
  * which do not switch off the HZ timer nohz_cpu_mask should
- * always be CPU_MASK_NONE.
+ * always be CPU_BITS_NONE.
  */
-cpumask_t nohz_cpu_mask = CPU_MASK_NONE;
+cpumask_var_t nohz_cpu_mask;
 
 /*
  * Increase the granularity value when there are more CPUs,
@@ -8274,6 +8274,9 @@ void __init sched_init(void)
 	 */
 	current->sched_class = &fair_sched_class;
 
+	/* Allocate the nohz_cpu_mask if CONFIG_CPUMASK_OFFSTACK */
+	alloc_bootmem_cpumask_var(&nohz_cpu_mask);
+
 	scheduler_running = 1;
 }
 
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 342fc9ccab46..70f872c71f4e 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -144,7 +144,7 @@ void tick_nohz_update_jiffies(void)
 	if (!ts->tick_stopped)
 		return;
 
-	cpu_clear(cpu, nohz_cpu_mask);
+	cpumask_clear_cpu(cpu, nohz_cpu_mask);
 	now = ktime_get();
 	ts->idle_waketime = now;
 
@@ -283,7 +283,7 @@ void tick_nohz_stop_sched_tick(int inidle)
 	if ((long)delta_jiffies >= 1) {
 
 		if (delta_jiffies > 1)
-			cpu_set(cpu, nohz_cpu_mask);
+			cpumask_set_cpu(cpu, nohz_cpu_mask);
 		/*
 		 * nohz_stop_sched_tick can be called several times before
 		 * the nohz_restart_sched_tick is called. This happens when
@@ -296,7 +296,7 @@ void tick_nohz_stop_sched_tick(int inidle)
 				/*
 				 * sched tick not stopped!
 				 */
-				cpu_clear(cpu, nohz_cpu_mask);
+				cpumask_clear_cpu(cpu, nohz_cpu_mask);
 				goto out;
 			}
 
@@ -354,7 +354,7 @@ void tick_nohz_stop_sched_tick(int inidle)
 		 * softirq.
 		 */
 		tick_do_update_jiffies64(ktime_get());
-		cpu_clear(cpu, nohz_cpu_mask);
+		cpumask_clear_cpu(cpu, nohz_cpu_mask);
 	}
 	raise_softirq_irqoff(TIMER_SOFTIRQ);
 out:
@@ -432,7 +432,7 @@ void tick_nohz_restart_sched_tick(void)
 	select_nohz_load_balancer(0);
 	now = ktime_get();
 	tick_do_update_jiffies64(now);
-	cpu_clear(cpu, nohz_cpu_mask);
+	cpumask_clear_cpu(cpu, nohz_cpu_mask);
 
 	/*
 	 * We stopped the tick in idle. Update process times would miss the
-- 
cgit v1.2.3


From 96f874e26428ab5d2db681c100210c254775e154 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 25 Nov 2008 02:35:14 +1030
Subject: sched: convert remaining old-style cpumask operators

Impact: Trivial API conversion

  NR_CPUS -> nr_cpu_ids
  cpumask_t -> struct cpumask
  sizeof(cpumask_t) -> cpumask_size()
  cpumask_a = cpumask_b -> cpumask_copy(&cpumask_a, &cpumask_b)

  cpu_set() -> cpumask_set_cpu()
  first_cpu() -> cpumask_first()
  cpumask_of_cpu() -> cpumask_of()
  cpus_* -> cpumask_*

There are some FIXMEs where we all archs to complete infrastructure
(patches have been sent):

  cpu_coregroup_map -> cpu_coregroup_mask
  node_to_cpumask* -> cpumask_of_node

There is also one FIXME where we pass an array of cpumasks to
partition_sched_domains(): this implies knowing the definition of
'struct cpumask' and the size of a cpumask.  This will be fixed in a
future patch.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |  16 ++--
 kernel/sched.c        | 212 +++++++++++++++++++++++++++-----------------------
 kernel/sched_fair.c   |   4 +-
 kernel/sched_rt.c     |  18 ++---
 4 files changed, 132 insertions(+), 118 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 1e33e2cb7f8c..4b7b0187374c 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -879,7 +879,7 @@ static inline struct cpumask *sched_domain_span(struct sched_domain *sd)
 	return to_cpumask(sd->span);
 }
 
-extern void partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
+extern void partition_sched_domains(int ndoms_new, struct cpumask *doms_new,
 				    struct sched_domain_attr *dattr_new);
 extern int arch_reinit_sched_domains(void);
 
@@ -888,7 +888,7 @@ extern int arch_reinit_sched_domains(void);
 struct sched_domain_attr;
 
 static inline void
-partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
+partition_sched_domains(int ndoms_new, struct cpumask *doms_new,
 			struct sched_domain_attr *dattr_new)
 {
 }
@@ -970,7 +970,7 @@ struct sched_class {
 	void (*task_wake_up) (struct rq *this_rq, struct task_struct *task);
 
 	void (*set_cpus_allowed)(struct task_struct *p,
-				 const cpumask_t *newmask);
+				 const struct cpumask *newmask);
 
 	void (*rq_online)(struct rq *rq);
 	void (*rq_offline)(struct rq *rq);
@@ -1612,12 +1612,12 @@ extern cputime_t task_gtime(struct task_struct *p);
 
 #ifdef CONFIG_SMP
 extern int set_cpus_allowed_ptr(struct task_struct *p,
-				const cpumask_t *new_mask);
+				const struct cpumask *new_mask);
 #else
 static inline int set_cpus_allowed_ptr(struct task_struct *p,
-				       const cpumask_t *new_mask)
+				       const struct cpumask *new_mask)
 {
-	if (!cpu_isset(0, *new_mask))
+	if (!cpumask_test_cpu(0, new_mask))
 		return -EINVAL;
 	return 0;
 }
@@ -2230,8 +2230,8 @@ __trace_special(void *__tr, void *__data,
 }
 #endif
 
-extern long sched_setaffinity(pid_t pid, const cpumask_t *new_mask);
-extern long sched_getaffinity(pid_t pid, cpumask_t *mask);
+extern long sched_setaffinity(pid_t pid, const struct cpumask *new_mask);
+extern long sched_getaffinity(pid_t pid, struct cpumask *mask);
 
 extern int sched_mc_power_savings, sched_smt_power_savings;
 
diff --git a/kernel/sched.c b/kernel/sched.c
index f2be61870030..eba6a156d334 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -2829,7 +2829,7 @@ static void sched_migrate_task(struct task_struct *p, int dest_cpu)
 	struct rq *rq;
 
 	rq = task_rq_lock(p, &flags);
-	if (!cpu_isset(dest_cpu, p->cpus_allowed)
+	if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed)
 	    || unlikely(!cpu_active(dest_cpu)))
 		goto out;
 
@@ -2895,7 +2895,7 @@ int can_migrate_task(struct task_struct *p, struct rq *rq, int this_cpu,
 	 * 2) cannot be migrated to this CPU due to cpus_allowed, or
 	 * 3) are cache-hot on their current CPU.
 	 */
-	if (!cpu_isset(this_cpu, p->cpus_allowed)) {
+	if (!cpumask_test_cpu(this_cpu, &p->cpus_allowed)) {
 		schedstat_inc(p, se.nr_failed_migrations_affine);
 		return 0;
 	}
@@ -3070,7 +3070,7 @@ static int move_one_task(struct rq *this_rq, int this_cpu, struct rq *busiest,
 static struct sched_group *
 find_busiest_group(struct sched_domain *sd, int this_cpu,
 		   unsigned long *imbalance, enum cpu_idle_type idle,
-		   int *sd_idle, const cpumask_t *cpus, int *balance)
+		   int *sd_idle, const struct cpumask *cpus, int *balance)
 {
 	struct sched_group *busiest = NULL, *this = NULL, *group = sd->groups;
 	unsigned long max_load, avg_load, total_load, this_load, total_pwr;
@@ -3387,7 +3387,7 @@ ret:
  */
 static struct rq *
 find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle,
-		   unsigned long imbalance, const cpumask_t *cpus)
+		   unsigned long imbalance, const struct cpumask *cpus)
 {
 	struct rq *busiest = NULL, *rq;
 	unsigned long max_load = 0;
@@ -3396,7 +3396,7 @@ find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle,
 	for_each_cpu(i, sched_group_cpus(group)) {
 		unsigned long wl;
 
-		if (!cpu_isset(i, *cpus))
+		if (!cpumask_test_cpu(i, cpus))
 			continue;
 
 		rq = cpu_rq(i);
@@ -3426,7 +3426,7 @@ find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle,
  */
 static int load_balance(int this_cpu, struct rq *this_rq,
 			struct sched_domain *sd, enum cpu_idle_type idle,
-			int *balance, cpumask_t *cpus)
+			int *balance, struct cpumask *cpus)
 {
 	int ld_moved, all_pinned = 0, active_balance = 0, sd_idle = 0;
 	struct sched_group *group;
@@ -3434,7 +3434,7 @@ static int load_balance(int this_cpu, struct rq *this_rq,
 	struct rq *busiest;
 	unsigned long flags;
 
-	cpus_setall(*cpus);
+	cpumask_setall(cpus);
 
 	/*
 	 * When power savings policy is enabled for the parent domain, idle
@@ -3494,8 +3494,8 @@ redo:
 
 		/* All tasks on this runqueue were pinned by CPU affinity */
 		if (unlikely(all_pinned)) {
-			cpu_clear(cpu_of(busiest), *cpus);
-			if (!cpus_empty(*cpus))
+			cpumask_clear_cpu(cpu_of(busiest), cpus);
+			if (!cpumask_empty(cpus))
 				goto redo;
 			goto out_balanced;
 		}
@@ -3512,7 +3512,8 @@ redo:
 			/* don't kick the migration_thread, if the curr
 			 * task on busiest cpu can't be moved to this_cpu
 			 */
-			if (!cpu_isset(this_cpu, busiest->curr->cpus_allowed)) {
+			if (!cpumask_test_cpu(this_cpu,
+					      &busiest->curr->cpus_allowed)) {
 				spin_unlock_irqrestore(&busiest->lock, flags);
 				all_pinned = 1;
 				goto out_one_pinned;
@@ -3587,7 +3588,7 @@ out:
  */
 static int
 load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd,
-			cpumask_t *cpus)
+			struct cpumask *cpus)
 {
 	struct sched_group *group;
 	struct rq *busiest = NULL;
@@ -3596,7 +3597,7 @@ load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd,
 	int sd_idle = 0;
 	int all_pinned = 0;
 
-	cpus_setall(*cpus);
+	cpumask_setall(cpus);
 
 	/*
 	 * When power savings policy is enabled for the parent domain, idle
@@ -3640,8 +3641,8 @@ redo:
 		double_unlock_balance(this_rq, busiest);
 
 		if (unlikely(all_pinned)) {
-			cpu_clear(cpu_of(busiest), *cpus);
-			if (!cpus_empty(*cpus))
+			cpumask_clear_cpu(cpu_of(busiest), cpus);
+			if (!cpumask_empty(cpus))
 				goto redo;
 		}
 	}
@@ -5376,7 +5377,7 @@ out_unlock:
 	return retval;
 }
 
-long sched_setaffinity(pid_t pid, const cpumask_t *in_mask)
+long sched_setaffinity(pid_t pid, const struct cpumask *in_mask)
 {
 	cpumask_var_t cpus_allowed, new_mask;
 	struct task_struct *p;
@@ -5445,13 +5446,13 @@ out_put_task:
 }
 
 static int get_user_cpu_mask(unsigned long __user *user_mask_ptr, unsigned len,
-			     cpumask_t *new_mask)
+			     struct cpumask *new_mask)
 {
-	if (len < sizeof(cpumask_t)) {
-		memset(new_mask, 0, sizeof(cpumask_t));
-	} else if (len > sizeof(cpumask_t)) {
-		len = sizeof(cpumask_t);
-	}
+	if (len < cpumask_size())
+		cpumask_clear(new_mask);
+	else if (len > cpumask_size())
+		len = cpumask_size();
+
 	return copy_from_user(new_mask, user_mask_ptr, len) ? -EFAULT : 0;
 }
 
@@ -5477,7 +5478,7 @@ asmlinkage long sys_sched_setaffinity(pid_t pid, unsigned int len,
 	return retval;
 }
 
-long sched_getaffinity(pid_t pid, cpumask_t *mask)
+long sched_getaffinity(pid_t pid, struct cpumask *mask)
 {
 	struct task_struct *p;
 	int retval;
@@ -5494,7 +5495,7 @@ long sched_getaffinity(pid_t pid, cpumask_t *mask)
 	if (retval)
 		goto out_unlock;
 
-	cpus_and(*mask, p->cpus_allowed, cpu_online_map);
+	cpumask_and(mask, &p->cpus_allowed, cpu_online_mask);
 
 out_unlock:
 	read_unlock(&tasklist_lock);
@@ -5872,7 +5873,7 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu)
 	idle->se.exec_start = sched_clock();
 
 	idle->prio = idle->normal_prio = MAX_PRIO;
-	idle->cpus_allowed = cpumask_of_cpu(cpu);
+	cpumask_copy(&idle->cpus_allowed, cpumask_of(cpu));
 	__set_task_cpu(idle, cpu);
 
 	rq->curr = rq->idle = idle;
@@ -5956,7 +5957,7 @@ static inline void sched_init_granularity(void)
  * task must not exit() & deallocate itself prematurely. The
  * call is not atomic; no spinlocks may be held.
  */
-int set_cpus_allowed_ptr(struct task_struct *p, const cpumask_t *new_mask)
+int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask)
 {
 	struct migration_req req;
 	unsigned long flags;
@@ -5964,13 +5965,13 @@ int set_cpus_allowed_ptr(struct task_struct *p, const cpumask_t *new_mask)
 	int ret = 0;
 
 	rq = task_rq_lock(p, &flags);
-	if (!cpus_intersects(*new_mask, cpu_online_map)) {
+	if (!cpumask_intersects(new_mask, cpu_online_mask)) {
 		ret = -EINVAL;
 		goto out;
 	}
 
 	if (unlikely((p->flags & PF_THREAD_BOUND) && p != current &&
-		     !cpus_equal(p->cpus_allowed, *new_mask))) {
+		     !cpumask_equal(&p->cpus_allowed, new_mask))) {
 		ret = -EINVAL;
 		goto out;
 	}
@@ -5978,12 +5979,12 @@ int set_cpus_allowed_ptr(struct task_struct *p, const cpumask_t *new_mask)
 	if (p->sched_class->set_cpus_allowed)
 		p->sched_class->set_cpus_allowed(p, new_mask);
 	else {
-		p->cpus_allowed = *new_mask;
-		p->rt.nr_cpus_allowed = cpus_weight(*new_mask);
+		cpumask_copy(&p->cpus_allowed, new_mask);
+		p->rt.nr_cpus_allowed = cpumask_weight(new_mask);
 	}
 
 	/* Can the task run on the task's current CPU? If so, we're done */
-	if (cpu_isset(task_cpu(p), *new_mask))
+	if (cpumask_test_cpu(task_cpu(p), new_mask))
 		goto out;
 
 	if (migrate_task(p, cpumask_any_and(cpu_online_mask, new_mask), &req)) {
@@ -6028,7 +6029,7 @@ static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu)
 	if (task_cpu(p) != src_cpu)
 		goto done;
 	/* Affinity changed (again). */
-	if (!cpu_isset(dest_cpu, p->cpus_allowed))
+	if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed))
 		goto fail;
 
 	on_rq = p->se.on_rq;
@@ -6629,13 +6630,13 @@ early_initcall(migration_init);
 #ifdef CONFIG_SCHED_DEBUG
 
 static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
-				  cpumask_t *groupmask)
+				  struct cpumask *groupmask)
 {
 	struct sched_group *group = sd->groups;
 	char str[256];
 
 	cpulist_scnprintf(str, sizeof(str), *sched_domain_span(sd));
-	cpus_clear(*groupmask);
+	cpumask_clear(groupmask);
 
 	printk(KERN_DEBUG "%*s domain %d: ", level, "", level);
 
@@ -6936,24 +6937,25 @@ __setup("isolcpus=", isolated_cpu_setup);
 /*
  * init_sched_build_groups takes the cpumask we wish to span, and a pointer
  * to a function which identifies what group(along with sched group) a CPU
- * belongs to. The return value of group_fn must be a >= 0 and < NR_CPUS
- * (due to the fact that we keep track of groups covered with a cpumask_t).
+ * belongs to. The return value of group_fn must be a >= 0 and < nr_cpu_ids
+ * (due to the fact that we keep track of groups covered with a struct cpumask).
  *
  * init_sched_build_groups will build a circular linked list of the groups
  * covered by the given span, and will set each group's ->cpumask correctly,
  * and ->cpu_power to 0.
  */
 static void
-init_sched_build_groups(const cpumask_t *span, const cpumask_t *cpu_map,
-			int (*group_fn)(int cpu, const cpumask_t *cpu_map,
+init_sched_build_groups(const struct cpumask *span,
+			const struct cpumask *cpu_map,
+			int (*group_fn)(int cpu, const struct cpumask *cpu_map,
 					struct sched_group **sg,
-					cpumask_t *tmpmask),
-			cpumask_t *covered, cpumask_t *tmpmask)
+					struct cpumask *tmpmask),
+			struct cpumask *covered, struct cpumask *tmpmask)
 {
 	struct sched_group *first = NULL, *last = NULL;
 	int i;
 
-	cpus_clear(*covered);
+	cpumask_clear(covered);
 
 	for_each_cpu(i, span) {
 		struct sched_group *sg;
@@ -6970,7 +6972,7 @@ init_sched_build_groups(const cpumask_t *span, const cpumask_t *cpu_map,
 			if (group_fn(j, cpu_map, NULL, tmpmask) != group)
 				continue;
 
-			cpu_set(j, *covered);
+			cpumask_set_cpu(j, covered);
 			cpumask_set_cpu(j, sched_group_cpus(sg));
 		}
 		if (!first)
@@ -7035,9 +7037,10 @@ static int find_next_best_node(int node, nodemask_t *used_nodes)
  * should be one that prevents unnecessary balancing, but also spreads tasks
  * out optimally.
  */
-static void sched_domain_node_span(int node, cpumask_t *span)
+static void sched_domain_node_span(int node, struct cpumask *span)
 {
 	nodemask_t used_nodes;
+	/* FIXME: use cpumask_of_node() */
 	node_to_cpumask_ptr(nodemask, node);
 	int i;
 
@@ -7081,8 +7084,8 @@ static DEFINE_PER_CPU(struct static_sched_domain, cpu_domains);
 static DEFINE_PER_CPU(struct static_sched_group, sched_group_cpus);
 
 static int
-cpu_to_cpu_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
-		 cpumask_t *unused)
+cpu_to_cpu_group(int cpu, const struct cpumask *cpu_map,
+		 struct sched_group **sg, struct cpumask *unused)
 {
 	if (sg)
 		*sg = &per_cpu(sched_group_cpus, cpu).sg;
@@ -7100,22 +7103,21 @@ static DEFINE_PER_CPU(struct static_sched_group, sched_group_core);
 
 #if defined(CONFIG_SCHED_MC) && defined(CONFIG_SCHED_SMT)
 static int
-cpu_to_core_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
-		  cpumask_t *mask)
+cpu_to_core_group(int cpu, const struct cpumask *cpu_map,
+		  struct sched_group **sg, struct cpumask *mask)
 {
 	int group;
 
-	*mask = per_cpu(cpu_sibling_map, cpu);
-	cpus_and(*mask, *mask, *cpu_map);
-	group = first_cpu(*mask);
+	cpumask_and(mask, &per_cpu(cpu_sibling_map, cpu), cpu_map);
+	group = cpumask_first(mask);
 	if (sg)
 		*sg = &per_cpu(sched_group_core, group).sg;
 	return group;
 }
 #elif defined(CONFIG_SCHED_MC)
 static int
-cpu_to_core_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
-		  cpumask_t *unused)
+cpu_to_core_group(int cpu, const struct cpumask *cpu_map,
+		  struct sched_group **sg, struct cpumask *unused)
 {
 	if (sg)
 		*sg = &per_cpu(sched_group_core, cpu).sg;
@@ -7127,18 +7129,18 @@ static DEFINE_PER_CPU(struct static_sched_domain, phys_domains);
 static DEFINE_PER_CPU(struct static_sched_group, sched_group_phys);
 
 static int
-cpu_to_phys_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg,
-		  cpumask_t *mask)
+cpu_to_phys_group(int cpu, const struct cpumask *cpu_map,
+		  struct sched_group **sg, struct cpumask *mask)
 {
 	int group;
 #ifdef CONFIG_SCHED_MC
+	/* FIXME: Use cpu_coregroup_mask. */
 	*mask = cpu_coregroup_map(cpu);
 	cpus_and(*mask, *mask, *cpu_map);
-	group = first_cpu(*mask);
+	group = cpumask_first(mask);
 #elif defined(CONFIG_SCHED_SMT)
-	*mask = per_cpu(cpu_sibling_map, cpu);
-	cpus_and(*mask, *mask, *cpu_map);
-	group = first_cpu(*mask);
+	cpumask_and(mask, &per_cpu(cpu_sibling_map, cpu), cpu_map);
+	group = cpumask_first(mask);
 #else
 	group = cpu;
 #endif
@@ -7159,14 +7161,16 @@ static struct sched_group ***sched_group_nodes_bycpu;
 static DEFINE_PER_CPU(struct sched_domain, allnodes_domains);
 static DEFINE_PER_CPU(struct static_sched_group, sched_group_allnodes);
 
-static int cpu_to_allnodes_group(int cpu, const cpumask_t *cpu_map,
-				 struct sched_group **sg, cpumask_t *nodemask)
+static int cpu_to_allnodes_group(int cpu, const struct cpumask *cpu_map,
+				 struct sched_group **sg,
+				 struct cpumask *nodemask)
 {
 	int group;
+	/* FIXME: use cpumask_of_node */
 	node_to_cpumask_ptr(pnodemask, cpu_to_node(cpu));
 
-	cpus_and(*nodemask, *pnodemask, *cpu_map);
-	group = first_cpu(*nodemask);
+	cpumask_and(nodemask, pnodemask, cpu_map);
+	group = cpumask_first(nodemask);
 
 	if (sg)
 		*sg = &per_cpu(sched_group_allnodes, group).sg;
@@ -7202,7 +7206,8 @@ static void init_numa_sched_groups_power(struct sched_group *group_head)
 
 #ifdef CONFIG_NUMA
 /* Free memory allocated for various sched_group structures */
-static void free_sched_groups(const cpumask_t *cpu_map, cpumask_t *nodemask)
+static void free_sched_groups(const struct cpumask *cpu_map,
+			      struct cpumask *nodemask)
 {
 	int cpu, i;
 
@@ -7215,10 +7220,11 @@ static void free_sched_groups(const cpumask_t *cpu_map, cpumask_t *nodemask)
 
 		for (i = 0; i < nr_node_ids; i++) {
 			struct sched_group *oldsg, *sg = sched_group_nodes[i];
+			/* FIXME: Use cpumask_of_node */
 			node_to_cpumask_ptr(pnodemask, i);
 
 			cpus_and(*nodemask, *pnodemask, *cpu_map);
-			if (cpus_empty(*nodemask))
+			if (cpumask_empty(nodemask))
 				continue;
 
 			if (sg == NULL)
@@ -7236,7 +7242,8 @@ next_sg:
 	}
 }
 #else /* !CONFIG_NUMA */
-static void free_sched_groups(const cpumask_t *cpu_map, cpumask_t *nodemask)
+static void free_sched_groups(const struct cpumask *cpu_map,
+			      struct cpumask *nodemask)
 {
 }
 #endif /* CONFIG_NUMA */
@@ -7366,7 +7373,7 @@ static void set_domain_attribute(struct sched_domain *sd,
  * Build sched domains for a given set of cpus and attach the sched domains
  * to the individual cpus
  */
-static int __build_sched_domains(const cpumask_t *cpu_map,
+static int __build_sched_domains(const struct cpumask *cpu_map,
 				 struct sched_domain_attr *attr)
 {
 	int i, err = -ENOMEM;
@@ -7416,7 +7423,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 	}
 
 #ifdef CONFIG_NUMA
-	sched_group_nodes_bycpu[first_cpu(*cpu_map)] = sched_group_nodes;
+	sched_group_nodes_bycpu[cpumask_first(cpu_map)] = sched_group_nodes;
 #endif
 
 	/*
@@ -7425,12 +7432,13 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 	for_each_cpu(i, cpu_map) {
 		struct sched_domain *sd = NULL, *p;
 
+		/* FIXME: use cpumask_of_node */
 		*nodemask = node_to_cpumask(cpu_to_node(i));
 		cpus_and(*nodemask, *nodemask, *cpu_map);
 
 #ifdef CONFIG_NUMA
-		if (cpus_weight(*cpu_map) >
-				SD_NODES_PER_DOMAIN*cpus_weight(*nodemask)) {
+		if (cpumask_weight(cpu_map) >
+				SD_NODES_PER_DOMAIN*cpumask_weight(nodemask)) {
 			sd = &per_cpu(allnodes_domains, i);
 			SD_INIT(sd, ALLNODES);
 			set_domain_attribute(sd, attr);
@@ -7491,9 +7499,9 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 #ifdef CONFIG_SCHED_SMT
 	/* Set up CPU (sibling) groups */
 	for_each_cpu(i, cpu_map) {
-		*this_sibling_map = per_cpu(cpu_sibling_map, i);
-		cpus_and(*this_sibling_map, *this_sibling_map, *cpu_map);
-		if (i != first_cpu(*this_sibling_map))
+		cpumask_and(this_sibling_map,
+			    &per_cpu(cpu_sibling_map, i), cpu_map);
+		if (i != cpumask_first(this_sibling_map))
 			continue;
 
 		init_sched_build_groups(this_sibling_map, cpu_map,
@@ -7505,9 +7513,10 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 #ifdef CONFIG_SCHED_MC
 	/* Set up multi-core groups */
 	for_each_cpu(i, cpu_map) {
+		/* FIXME: Use cpu_coregroup_mask */
 		*this_core_map = cpu_coregroup_map(i);
 		cpus_and(*this_core_map, *this_core_map, *cpu_map);
-		if (i != first_cpu(*this_core_map))
+		if (i != cpumask_first(this_core_map))
 			continue;
 
 		init_sched_build_groups(this_core_map, cpu_map,
@@ -7518,9 +7527,10 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 
 	/* Set up physical groups */
 	for (i = 0; i < nr_node_ids; i++) {
+		/* FIXME: Use cpumask_of_node */
 		*nodemask = node_to_cpumask(i);
 		cpus_and(*nodemask, *nodemask, *cpu_map);
-		if (cpus_empty(*nodemask))
+		if (cpumask_empty(nodemask))
 			continue;
 
 		init_sched_build_groups(nodemask, cpu_map,
@@ -7541,17 +7551,18 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		struct sched_group *sg, *prev;
 		int j;
 
+		/* FIXME: Use cpumask_of_node */
 		*nodemask = node_to_cpumask(i);
-		cpus_clear(*covered);
+		cpumask_clear(covered);
 
 		cpus_and(*nodemask, *nodemask, *cpu_map);
-		if (cpus_empty(*nodemask)) {
+		if (cpumask_empty(nodemask)) {
 			sched_group_nodes[i] = NULL;
 			continue;
 		}
 
 		sched_domain_node_span(i, domainspan);
-		cpus_and(*domainspan, *domainspan, *cpu_map);
+		cpumask_and(domainspan, domainspan, cpu_map);
 
 		sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(),
 				  GFP_KERNEL, i);
@@ -7570,21 +7581,22 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 		sg->__cpu_power = 0;
 		cpumask_copy(sched_group_cpus(sg), nodemask);
 		sg->next = sg;
-		cpus_or(*covered, *covered, *nodemask);
+		cpumask_or(covered, covered, nodemask);
 		prev = sg;
 
 		for (j = 0; j < nr_node_ids; j++) {
 			int n = (i + j) % nr_node_ids;
+			/* FIXME: Use cpumask_of_node */
 			node_to_cpumask_ptr(pnodemask, n);
 
-			cpus_complement(*notcovered, *covered);
-			cpus_and(*tmpmask, *notcovered, *cpu_map);
-			cpus_and(*tmpmask, *tmpmask, *domainspan);
-			if (cpus_empty(*tmpmask))
+			cpumask_complement(notcovered, covered);
+			cpumask_and(tmpmask, notcovered, cpu_map);
+			cpumask_and(tmpmask, tmpmask, domainspan);
+			if (cpumask_empty(tmpmask))
 				break;
 
-			cpus_and(*tmpmask, *tmpmask, *pnodemask);
-			if (cpus_empty(*tmpmask))
+			cpumask_and(tmpmask, tmpmask, pnodemask);
+			if (cpumask_empty(tmpmask))
 				continue;
 
 			sg = kmalloc_node(sizeof(struct sched_group) +
@@ -7598,7 +7610,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 			sg->__cpu_power = 0;
 			cpumask_copy(sched_group_cpus(sg), tmpmask);
 			sg->next = prev->next;
-			cpus_or(*covered, *covered, *tmpmask);
+			cpumask_or(covered, covered, tmpmask);
 			prev->next = sg;
 			prev = sg;
 		}
@@ -7634,7 +7646,7 @@ static int __build_sched_domains(const cpumask_t *cpu_map,
 	if (sd_allnodes) {
 		struct sched_group *sg;
 
-		cpu_to_allnodes_group(first_cpu(*cpu_map), cpu_map, &sg,
+		cpu_to_allnodes_group(cpumask_first(cpu_map), cpu_map, &sg,
 								tmpmask);
 		init_numa_sched_groups_power(sg);
 	}
@@ -7690,12 +7702,12 @@ error:
 #endif
 }
 
-static int build_sched_domains(const cpumask_t *cpu_map)
+static int build_sched_domains(const struct cpumask *cpu_map)
 {
 	return __build_sched_domains(cpu_map, NULL);
 }
 
-static cpumask_t *doms_cur;	/* current sched domains */
+static struct cpumask *doms_cur;	/* current sched domains */
 static int ndoms_cur;		/* number of sched domains in 'doms_cur' */
 static struct sched_domain_attr *dattr_cur;
 				/* attribues of custom domains in 'doms_cur' */
@@ -7716,13 +7728,13 @@ void __attribute__((weak)) arch_update_cpu_topology(void)
  * For now this just excludes isolated cpus, but could be used to
  * exclude other special cases in the future.
  */
-static int arch_init_sched_domains(const cpumask_t *cpu_map)
+static int arch_init_sched_domains(const struct cpumask *cpu_map)
 {
 	int err;
 
 	arch_update_cpu_topology();
 	ndoms_cur = 1;
-	doms_cur = kmalloc(sizeof(cpumask_t), GFP_KERNEL);
+	doms_cur = kmalloc(cpumask_size(), GFP_KERNEL);
 	if (!doms_cur)
 		doms_cur = fallback_doms;
 	cpumask_andnot(doms_cur, cpu_map, cpu_isolated_map);
@@ -7733,8 +7745,8 @@ static int arch_init_sched_domains(const cpumask_t *cpu_map)
 	return err;
 }
 
-static void arch_destroy_sched_domains(const cpumask_t *cpu_map,
-				       cpumask_t *tmpmask)
+static void arch_destroy_sched_domains(const struct cpumask *cpu_map,
+				       struct cpumask *tmpmask)
 {
 	free_sched_groups(cpu_map, tmpmask);
 }
@@ -7743,15 +7755,16 @@ static void arch_destroy_sched_domains(const cpumask_t *cpu_map,
  * Detach sched domains from a group of cpus specified in cpu_map
  * These cpus will now be attached to the NULL domain
  */
-static void detach_destroy_domains(const cpumask_t *cpu_map)
+static void detach_destroy_domains(const struct cpumask *cpu_map)
 {
-	cpumask_t tmpmask;
+	/* Save because hotplug lock held. */
+	static DECLARE_BITMAP(tmpmask, CONFIG_NR_CPUS);
 	int i;
 
 	for_each_cpu(i, cpu_map)
 		cpu_attach_domain(NULL, &def_root_domain, i);
 	synchronize_sched();
-	arch_destroy_sched_domains(cpu_map, &tmpmask);
+	arch_destroy_sched_domains(cpu_map, to_cpumask(tmpmask));
 }
 
 /* handle null as "default" */
@@ -7776,7 +7789,7 @@ static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur,
  * doms_new[] to the current sched domain partitioning, doms_cur[].
  * It destroys each deleted domain and builds each new domain.
  *
- * 'doms_new' is an array of cpumask_t's of length 'ndoms_new'.
+ * 'doms_new' is an array of cpumask's of length 'ndoms_new'.
  * The masks don't intersect (don't overlap.) We should setup one
  * sched domain for each mask. CPUs not in any of the cpumasks will
  * not be load balanced. If the same cpumask appears both in the
@@ -7790,13 +7803,14 @@ static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur,
  * the single partition 'fallback_doms', it also forces the domains
  * to be rebuilt.
  *
- * If doms_new == NULL it will be replaced with cpu_online_map.
+ * If doms_new == NULL it will be replaced with cpu_online_mask.
  * ndoms_new == 0 is a special case for destroying existing domains,
  * and it will not create the default domain.
  *
  * Call with hotplug lock held
  */
-void partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
+/* FIXME: Change to struct cpumask *doms_new[] */
+void partition_sched_domains(int ndoms_new, struct cpumask *doms_new,
 			     struct sched_domain_attr *dattr_new)
 {
 	int i, j, n;
@@ -7811,7 +7825,7 @@ void partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
 	/* Destroy deleted domains */
 	for (i = 0; i < ndoms_cur; i++) {
 		for (j = 0; j < n; j++) {
-			if (cpus_equal(doms_cur[i], doms_new[j])
+			if (cpumask_equal(&doms_cur[i], &doms_new[j])
 			    && dattrs_equal(dattr_cur, i, dattr_new, j))
 				goto match1;
 		}
@@ -7831,7 +7845,7 @@ match1:
 	/* Build new domains */
 	for (i = 0; i < ndoms_new; i++) {
 		for (j = 0; j < ndoms_cur; j++) {
-			if (cpus_equal(doms_new[i], doms_cur[j])
+			if (cpumask_equal(&doms_new[i], &doms_cur[j])
 			    && dattrs_equal(dattr_new, i, dattr_cur, j))
 				goto match2;
 		}
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index bba00402ed90..08ffffd4a410 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1017,7 +1017,7 @@ static void yield_task_fair(struct rq *rq)
  * search starts with cpus closest then further out as needed,
  * so we always favor a closer, idle cpu.
  * Domains may include CPUs that are not usable for migration,
- * hence we need to mask them out (cpu_active_map)
+ * hence we need to mask them out (cpu_active_mask)
  *
  * Returns the CPU we should wake onto.
  */
@@ -1244,7 +1244,7 @@ static int select_task_rq_fair(struct task_struct *p, int sync)
 		}
 	}
 
-	if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed)))
+	if (unlikely(!cpumask_test_cpu(this_cpu, &p->cpus_allowed)))
 		goto out;
 
 	/*
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 1f0e99d1a8ce..fb3964579a8a 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -923,7 +923,7 @@ static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep);
 static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu)
 {
 	if (!task_running(rq, p) &&
-	    (cpu < 0 || cpu_isset(cpu, p->cpus_allowed)) &&
+	    (cpu < 0 || cpumask_test_cpu(cpu, &p->cpus_allowed)) &&
 	    (p->rt.nr_cpus_allowed > 1))
 		return 1;
 	return 0;
@@ -982,7 +982,7 @@ static inline int pick_optimal_cpu(int this_cpu, cpumask_t *mask)
 static int find_lowest_rq(struct task_struct *task)
 {
 	struct sched_domain *sd;
-	cpumask_t *lowest_mask = __get_cpu_var(local_cpu_mask);
+	struct cpumask *lowest_mask = __get_cpu_var(local_cpu_mask);
 	int this_cpu = smp_processor_id();
 	int cpu      = task_cpu(task);
 
@@ -997,7 +997,7 @@ static int find_lowest_rq(struct task_struct *task)
 	 * I guess we might want to change cpupri_find() to ignore those
 	 * in the first place.
 	 */
-	cpus_and(*lowest_mask, *lowest_mask, cpu_active_map);
+	cpumask_and(lowest_mask, lowest_mask, cpu_active_mask);
 
 	/*
 	 * At this point we have built a mask of cpus representing the
@@ -1007,7 +1007,7 @@ static int find_lowest_rq(struct task_struct *task)
 	 * We prioritize the last cpu that the task executed on since
 	 * it is most likely cache-hot in that location.
 	 */
-	if (cpu_isset(cpu, *lowest_mask))
+	if (cpumask_test_cpu(cpu, lowest_mask))
 		return cpu;
 
 	/*
@@ -1064,8 +1064,8 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq)
 			 * Also make sure that it wasn't scheduled on its rq.
 			 */
 			if (unlikely(task_rq(task) != rq ||
-				     !cpu_isset(lowest_rq->cpu,
-						task->cpus_allowed) ||
+				     !cpumask_test_cpu(lowest_rq->cpu,
+						       &task->cpus_allowed) ||
 				     task_running(rq, task) ||
 				     !task->se.on_rq)) {
 
@@ -1315,9 +1315,9 @@ move_one_task_rt(struct rq *this_rq, int this_cpu, struct rq *busiest,
 }
 
 static void set_cpus_allowed_rt(struct task_struct *p,
-				const cpumask_t *new_mask)
+				const struct cpumask *new_mask)
 {
-	int weight = cpus_weight(*new_mask);
+	int weight = cpumask_weight(new_mask);
 
 	BUG_ON(!rt_task(p));
 
@@ -1338,7 +1338,7 @@ static void set_cpus_allowed_rt(struct task_struct *p,
 		update_rt_migration(rq);
 	}
 
-	p->cpus_allowed    = *new_mask;
+	cpumask_copy(&p->cpus_allowed, new_mask);
 	p->rt.nr_cpus_allowed = weight;
 }
 
-- 
cgit v1.2.3


From 193da6092764ab693da7170c5badbf60d7758c1d Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Wed, 26 Nov 2008 12:03:54 +0100
Subject: fuse: move FUSE_MINOR to miscdevice.h

Move FUSE_MINOR to miscdevice.h.  While at it, de-uglify the file.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
---
 include/linux/fuse.h       |  3 ---
 include/linux/miscdevice.h | 42 +++++++++++++++++++++---------------------
 2 files changed, 21 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/fuse.h b/include/linux/fuse.h
index 350fe9767bbc..7caa473306e4 100644
--- a/include/linux/fuse.h
+++ b/include/linux/fuse.h
@@ -40,9 +40,6 @@
 /** The major number of the fuse character device */
 #define FUSE_MAJOR MISC_MAJOR
 
-/** The minor number of the fuse character device */
-#define FUSE_MINOR 229
-
 /* Make sure all structures are padded to 64bit boundary, so 32bit
    userspace works under 64bit kernels */
 
diff --git a/include/linux/miscdevice.h b/include/linux/miscdevice.h
index 26433ec520b3..a820f816a49e 100644
--- a/include/linux/miscdevice.h
+++ b/include/linux/miscdevice.h
@@ -3,33 +3,33 @@
 #include <linux/module.h>
 #include <linux/major.h>
 
-#define PSMOUSE_MINOR  1
-#define MS_BUSMOUSE_MINOR 2
-#define ATIXL_BUSMOUSE_MINOR 3
-/*#define AMIGAMOUSE_MINOR 4	FIXME OBSOLETE */
-#define ATARIMOUSE_MINOR 5
-#define SUN_MOUSE_MINOR 6
-#define APOLLO_MOUSE_MINOR 7
-#define PC110PAD_MINOR 9
-/*#define ADB_MOUSE_MINOR 10	FIXME OBSOLETE */
+#define PSMOUSE_MINOR		1
+#define MS_BUSMOUSE_MINOR	2
+#define ATIXL_BUSMOUSE_MINOR	3
+/*#define AMIGAMOUSE_MINOR	4	FIXME OBSOLETE */
+#define ATARIMOUSE_MINOR	5
+#define SUN_MOUSE_MINOR		6
+#define APOLLO_MOUSE_MINOR	7
+#define PC110PAD_MINOR		9
+/*#define ADB_MOUSE_MINOR	10	FIXME OBSOLETE */
 #define WATCHDOG_MINOR		130	/* Watchdog timer     */
 #define TEMP_MINOR		131	/* Temperature Sensor */
-#define RTC_MINOR 135
+#define RTC_MINOR		135
 #define EFI_RTC_MINOR		136	/* EFI Time services */
-#define SUN_OPENPROM_MINOR 139
+#define SUN_OPENPROM_MINOR	139
 #define DMAPI_MINOR		140	/* DMAPI */
-#define NVRAM_MINOR 144
-#define SGI_MMTIMER        153
+#define NVRAM_MINOR		144
+#define SGI_MMTIMER		153
 #define STORE_QUEUE_MINOR	155
-#define I2O_MINOR 166
+#define I2O_MINOR		166
 #define MICROCODE_MINOR		184
-#define MWAVE_MINOR	219		/* ACP/Mwave Modem */
-#define MPT_MINOR	220
-#define MISC_DYNAMIC_MINOR 255
-
-#define TUN_MINOR	     200
-#define	HPET_MINOR	     228
-#define KVM_MINOR            232
+#define TUN_MINOR		200
+#define MWAVE_MINOR		219	/* ACP/Mwave Modem */
+#define MPT_MINOR		220
+#define HPET_MINOR		228
+#define FUSE_MINOR		229
+#define KVM_MINOR		232
+#define MISC_DYNAMIC_MINOR	255
 
 struct device;
 
-- 
cgit v1.2.3


From 59efec7b903987dcb60b9ebc85c7acd4443a11a1 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Wed, 26 Nov 2008 12:03:55 +0100
Subject: fuse: implement ioctl support

Generic ioctl support is tricky to implement because only the ioctl
implementation itself knows which memory regions need to be read
and/or written.  To support this, fuse client can request retry of
ioctl specifying memory regions to read and write.  Deep copying
(nested pointers) can be implemented by retrying multiple times
resolving one depth of dereference at a time.

For security and cleanliness considerations, ioctl implementation has
restricted mode where the kernel determines data transfer directions
and sizes using the _IOC_*() macros on the ioctl command.  In this
mode, retry is not allowed.

For all FUSE servers, restricted mode is enforced.  Unrestricted ioctl
will be used by CUSE.

Plese read the comment on top of fs/fuse/file.c::fuse_file_do_ioctl()
for more information.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
---
 fs/fuse/file.c       | 280 +++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/fuse.h |  32 ++++++
 2 files changed, 312 insertions(+)

(limited to 'include/linux')

diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 617269803913..baed06ea7622 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -1469,6 +1469,282 @@ static loff_t fuse_file_llseek(struct file *file, loff_t offset, int origin)
 	return retval;
 }
 
+static int fuse_ioctl_copy_user(struct page **pages, struct iovec *iov,
+			unsigned int nr_segs, size_t bytes, bool to_user)
+{
+	struct iov_iter ii;
+	int page_idx = 0;
+
+	if (!bytes)
+		return 0;
+
+	iov_iter_init(&ii, iov, nr_segs, bytes, 0);
+
+	while (iov_iter_count(&ii)) {
+		struct page *page = pages[page_idx++];
+		size_t todo = min_t(size_t, PAGE_SIZE, iov_iter_count(&ii));
+		void *kaddr, *map;
+
+		kaddr = map = kmap(page);
+
+		while (todo) {
+			char __user *uaddr = ii.iov->iov_base + ii.iov_offset;
+			size_t iov_len = ii.iov->iov_len - ii.iov_offset;
+			size_t copy = min(todo, iov_len);
+			size_t left;
+
+			if (!to_user)
+				left = copy_from_user(kaddr, uaddr, copy);
+			else
+				left = copy_to_user(uaddr, kaddr, copy);
+
+			if (unlikely(left))
+				return -EFAULT;
+
+			iov_iter_advance(&ii, copy);
+			todo -= copy;
+			kaddr += copy;
+		}
+
+		kunmap(map);
+	}
+
+	return 0;
+}
+
+/*
+ * For ioctls, there is no generic way to determine how much memory
+ * needs to be read and/or written.  Furthermore, ioctls are allowed
+ * to dereference the passed pointer, so the parameter requires deep
+ * copying but FUSE has no idea whatsoever about what to copy in or
+ * out.
+ *
+ * This is solved by allowing FUSE server to retry ioctl with
+ * necessary in/out iovecs.  Let's assume the ioctl implementation
+ * needs to read in the following structure.
+ *
+ * struct a {
+ *	char	*buf;
+ *	size_t	buflen;
+ * }
+ *
+ * On the first callout to FUSE server, inarg->in_size and
+ * inarg->out_size will be NULL; then, the server completes the ioctl
+ * with FUSE_IOCTL_RETRY set in out->flags, out->in_iovs set to 1 and
+ * the actual iov array to
+ *
+ * { { .iov_base = inarg.arg,	.iov_len = sizeof(struct a) } }
+ *
+ * which tells FUSE to copy in the requested area and retry the ioctl.
+ * On the second round, the server has access to the structure and
+ * from that it can tell what to look for next, so on the invocation,
+ * it sets FUSE_IOCTL_RETRY, out->in_iovs to 2 and iov array to
+ *
+ * { { .iov_base = inarg.arg,	.iov_len = sizeof(struct a)	},
+ *   { .iov_base = a.buf,	.iov_len = a.buflen		} }
+ *
+ * FUSE will copy both struct a and the pointed buffer from the
+ * process doing the ioctl and retry ioctl with both struct a and the
+ * buffer.
+ *
+ * This time, FUSE server has everything it needs and completes ioctl
+ * without FUSE_IOCTL_RETRY which finishes the ioctl call.
+ *
+ * Copying data out works the same way.
+ *
+ * Note that if FUSE_IOCTL_UNRESTRICTED is clear, the kernel
+ * automatically initializes in and out iovs by decoding @cmd with
+ * _IOC_* macros and the server is not allowed to request RETRY.  This
+ * limits ioctl data transfers to well-formed ioctls and is the forced
+ * behavior for all FUSE servers.
+ */
+static long fuse_file_do_ioctl(struct file *file, unsigned int cmd,
+			       unsigned long arg, unsigned int flags)
+{
+	struct inode *inode = file->f_dentry->d_inode;
+	struct fuse_file *ff = file->private_data;
+	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_ioctl_in inarg = {
+		.fh = ff->fh,
+		.cmd = cmd,
+		.arg = arg,
+		.flags = flags
+	};
+	struct fuse_ioctl_out outarg;
+	struct fuse_req *req = NULL;
+	struct page **pages = NULL;
+	struct page *iov_page = NULL;
+	struct iovec *in_iov = NULL, *out_iov = NULL;
+	unsigned int in_iovs = 0, out_iovs = 0, num_pages = 0, max_pages;
+	size_t in_size, out_size, transferred;
+	int err;
+
+	/* assume all the iovs returned by client always fits in a page */
+	BUILD_BUG_ON(sizeof(struct iovec) * FUSE_IOCTL_MAX_IOV > PAGE_SIZE);
+
+	if (!fuse_allow_task(fc, current))
+		return -EACCES;
+
+	err = -EIO;
+	if (is_bad_inode(inode))
+		goto out;
+
+	err = -ENOMEM;
+	pages = kzalloc(sizeof(pages[0]) * FUSE_MAX_PAGES_PER_REQ, GFP_KERNEL);
+	iov_page = alloc_page(GFP_KERNEL);
+	if (!pages || !iov_page)
+		goto out;
+
+	/*
+	 * If restricted, initialize IO parameters as encoded in @cmd.
+	 * RETRY from server is not allowed.
+	 */
+	if (!(flags & FUSE_IOCTL_UNRESTRICTED)) {
+		struct iovec *iov = page_address(iov_page);
+
+		iov->iov_base = (void *)arg;
+		iov->iov_len = _IOC_SIZE(cmd);
+
+		if (_IOC_DIR(cmd) & _IOC_WRITE) {
+			in_iov = iov;
+			in_iovs = 1;
+		}
+
+		if (_IOC_DIR(cmd) & _IOC_READ) {
+			out_iov = iov;
+			out_iovs = 1;
+		}
+	}
+
+ retry:
+	inarg.in_size = in_size = iov_length(in_iov, in_iovs);
+	inarg.out_size = out_size = iov_length(out_iov, out_iovs);
+
+	/*
+	 * Out data can be used either for actual out data or iovs,
+	 * make sure there always is at least one page.
+	 */
+	out_size = max_t(size_t, out_size, PAGE_SIZE);
+	max_pages = DIV_ROUND_UP(max(in_size, out_size), PAGE_SIZE);
+
+	/* make sure there are enough buffer pages and init request with them */
+	err = -ENOMEM;
+	if (max_pages > FUSE_MAX_PAGES_PER_REQ)
+		goto out;
+	while (num_pages < max_pages) {
+		pages[num_pages] = alloc_page(GFP_KERNEL | __GFP_HIGHMEM);
+		if (!pages[num_pages])
+			goto out;
+		num_pages++;
+	}
+
+	req = fuse_get_req(fc);
+	if (IS_ERR(req)) {
+		err = PTR_ERR(req);
+		req = NULL;
+		goto out;
+	}
+	memcpy(req->pages, pages, sizeof(req->pages[0]) * num_pages);
+	req->num_pages = num_pages;
+
+	/* okay, let's send it to the client */
+	req->in.h.opcode = FUSE_IOCTL;
+	req->in.h.nodeid = get_node_id(inode);
+	req->in.numargs = 1;
+	req->in.args[0].size = sizeof(inarg);
+	req->in.args[0].value = &inarg;
+	if (in_size) {
+		req->in.numargs++;
+		req->in.args[1].size = in_size;
+		req->in.argpages = 1;
+
+		err = fuse_ioctl_copy_user(pages, in_iov, in_iovs, in_size,
+					   false);
+		if (err)
+			goto out;
+	}
+
+	req->out.numargs = 2;
+	req->out.args[0].size = sizeof(outarg);
+	req->out.args[0].value = &outarg;
+	req->out.args[1].size = out_size;
+	req->out.argpages = 1;
+	req->out.argvar = 1;
+
+	request_send(fc, req);
+	err = req->out.h.error;
+	transferred = req->out.args[1].size;
+	fuse_put_request(fc, req);
+	req = NULL;
+	if (err)
+		goto out;
+
+	/* did it ask for retry? */
+	if (outarg.flags & FUSE_IOCTL_RETRY) {
+		char *vaddr;
+
+		/* no retry if in restricted mode */
+		err = -EIO;
+		if (!(flags & FUSE_IOCTL_UNRESTRICTED))
+			goto out;
+
+		in_iovs = outarg.in_iovs;
+		out_iovs = outarg.out_iovs;
+
+		/*
+		 * Make sure things are in boundary, separate checks
+		 * are to protect against overflow.
+		 */
+		err = -ENOMEM;
+		if (in_iovs > FUSE_IOCTL_MAX_IOV ||
+		    out_iovs > FUSE_IOCTL_MAX_IOV ||
+		    in_iovs + out_iovs > FUSE_IOCTL_MAX_IOV)
+			goto out;
+
+		err = -EIO;
+		if ((in_iovs + out_iovs) * sizeof(struct iovec) != transferred)
+			goto out;
+
+		/* okay, copy in iovs and retry */
+		vaddr = kmap_atomic(pages[0], KM_USER0);
+		memcpy(page_address(iov_page), vaddr, transferred);
+		kunmap_atomic(vaddr, KM_USER0);
+
+		in_iov = page_address(iov_page);
+		out_iov = in_iov + in_iovs;
+
+		goto retry;
+	}
+
+	err = -EIO;
+	if (transferred > inarg.out_size)
+		goto out;
+
+	err = fuse_ioctl_copy_user(pages, out_iov, out_iovs, transferred, true);
+ out:
+	if (req)
+		fuse_put_request(fc, req);
+	if (iov_page)
+		__free_page(iov_page);
+	while (num_pages)
+		__free_page(pages[--num_pages]);
+	kfree(pages);
+
+	return err ? err : outarg.result;
+}
+
+static long fuse_file_ioctl(struct file *file, unsigned int cmd,
+			    unsigned long arg)
+{
+	return fuse_file_do_ioctl(file, cmd, arg, 0);
+}
+
+static long fuse_file_compat_ioctl(struct file *file, unsigned int cmd,
+				   unsigned long arg)
+{
+	return fuse_file_do_ioctl(file, cmd, arg, FUSE_IOCTL_COMPAT);
+}
+
 static const struct file_operations fuse_file_operations = {
 	.llseek		= fuse_file_llseek,
 	.read		= do_sync_read,
@@ -1483,6 +1759,8 @@ static const struct file_operations fuse_file_operations = {
 	.lock		= fuse_file_lock,
 	.flock		= fuse_file_flock,
 	.splice_read	= generic_file_splice_read,
+	.unlocked_ioctl	= fuse_file_ioctl,
+	.compat_ioctl	= fuse_file_compat_ioctl,
 };
 
 static const struct file_operations fuse_direct_io_file_operations = {
@@ -1495,6 +1773,8 @@ static const struct file_operations fuse_direct_io_file_operations = {
 	.fsync		= fuse_fsync,
 	.lock		= fuse_file_lock,
 	.flock		= fuse_file_flock,
+	.unlocked_ioctl	= fuse_file_ioctl,
+	.compat_ioctl	= fuse_file_compat_ioctl,
 	/* no mmap and splice_read */
 };
 
diff --git a/include/linux/fuse.h b/include/linux/fuse.h
index 7caa473306e4..608e300ae883 100644
--- a/include/linux/fuse.h
+++ b/include/linux/fuse.h
@@ -148,6 +148,21 @@ struct fuse_file_lock {
  */
 #define FUSE_READ_LOCKOWNER	(1 << 1)
 
+/**
+ * Ioctl flags
+ *
+ * FUSE_IOCTL_COMPAT: 32bit compat ioctl on 64bit machine
+ * FUSE_IOCTL_UNRESTRICTED: not restricted to well-formed ioctls, retry allowed
+ * FUSE_IOCTL_RETRY: retry with new iovecs
+ *
+ * FUSE_IOCTL_MAX_IOV: maximum of in_iovecs + out_iovecs
+ */
+#define FUSE_IOCTL_COMPAT	(1 << 0)
+#define FUSE_IOCTL_UNRESTRICTED	(1 << 1)
+#define FUSE_IOCTL_RETRY	(1 << 2)
+
+#define FUSE_IOCTL_MAX_IOV	256
+
 enum fuse_opcode {
 	FUSE_LOOKUP	   = 1,
 	FUSE_FORGET	   = 2,  /* no reply */
@@ -185,6 +200,7 @@ enum fuse_opcode {
 	FUSE_INTERRUPT     = 36,
 	FUSE_BMAP          = 37,
 	FUSE_DESTROY       = 38,
+	FUSE_IOCTL         = 39,
 };
 
 /* The read buffer is required to be at least 8k, but may be much larger */
@@ -385,6 +401,22 @@ struct fuse_bmap_out {
 	__u64	block;
 };
 
+struct fuse_ioctl_in {
+	__u64	fh;
+	__u32	flags;
+	__u32	cmd;
+	__u64	arg;
+	__u32	in_size;
+	__u32	out_size;
+};
+
+struct fuse_ioctl_out {
+	__s32	result;
+	__u32	flags;
+	__u32	in_iovs;
+	__u32	out_iovs;
+};
+
 struct fuse_in_header {
 	__u32	len;
 	__u32	opcode;
-- 
cgit v1.2.3


From 8599396b5062bf6bd2a0b433503849e2322df1c2 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Wed, 26 Nov 2008 12:03:55 +0100
Subject: fuse: implement unsolicited notification

Clients always used to write only in response to read requests.  To
implement poll efficiently, clients should be able to issue
unsolicited notifications.  This patch implements basic notification
support.

Zero fuse_out_header.unique is now accepted and considered unsolicited
notification and the error field contains notification code.  This
patch doesn't implement any actual notification.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
---
 fs/fuse/dev.c        | 27 +++++++++++++++++++++++++--
 include/linux/fuse.h |  4 ++++
 2 files changed, 29 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index 225388f54ae7..ffd670bb8c8c 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -816,6 +816,15 @@ static ssize_t fuse_dev_read(struct kiocb *iocb, const struct iovec *iov,
 	return err;
 }
 
+static int fuse_notify(struct fuse_conn *fc, enum fuse_notify_code code,
+		       unsigned int size, struct fuse_copy_state *cs)
+{
+	switch (code) {
+	default:
+		return -EINVAL;
+	}
+}
+
 /* Look up request on processing list by unique ID */
 static struct fuse_req *request_find(struct fuse_conn *fc, u64 unique)
 {
@@ -879,9 +888,23 @@ static ssize_t fuse_dev_write(struct kiocb *iocb, const struct iovec *iov,
 	err = fuse_copy_one(&cs, &oh, sizeof(oh));
 	if (err)
 		goto err_finish;
+
+	err = -EINVAL;
+	if (oh.len != nbytes)
+		goto err_finish;
+
+	/*
+	 * Zero oh.unique indicates unsolicited notification message
+	 * and error contains notification code.
+	 */
+	if (!oh.unique) {
+		err = fuse_notify(fc, oh.error, nbytes - sizeof(oh), &cs);
+		fuse_copy_finish(&cs);
+		return err ? err : nbytes;
+	}
+
 	err = -EINVAL;
-	if (!oh.unique || oh.error <= -1000 || oh.error > 0 ||
-	    oh.len != nbytes)
+	if (oh.error <= -1000 || oh.error > 0)
 		goto err_finish;
 
 	spin_lock(&fc->lock);
diff --git a/include/linux/fuse.h b/include/linux/fuse.h
index 608e300ae883..abde9949e2c0 100644
--- a/include/linux/fuse.h
+++ b/include/linux/fuse.h
@@ -203,6 +203,10 @@ enum fuse_opcode {
 	FUSE_IOCTL         = 39,
 };
 
+enum fuse_notify_code {
+	FUSE_NOTIFY_CODE_MAX,
+};
+
 /* The read buffer is required to be at least 8k, but may be much larger */
 #define FUSE_MIN_READ_BUFFER 8192
 
-- 
cgit v1.2.3


From 95668a69a4bb862063c4d28a746e55107dee7b98 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Wed, 26 Nov 2008 12:03:55 +0100
Subject: fuse: implement poll support

Implement poll support.  Polled files are indexed using kh in a RB
tree rooted at fuse_conn->polled_files.

Client should send FUSE_NOTIFY_POLL notification once after processing
FUSE_POLL which has FUSE_POLL_SCHEDULE_NOTIFY set.  Sending
notification unconditionally after the latest poll or everytime file
content might have changed is inefficient but won't cause malfunction.

fuse_file_poll() can sleep and requires patches from the following
thread which allows f_op->poll() to sleep.

  http://thread.gmane.org/gmane.linux.kernel/726176

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
---
 fs/fuse/dev.c        |  19 ++++++++
 fs/fuse/file.c       | 132 +++++++++++++++++++++++++++++++++++++++++++++++++++
 fs/fuse/fuse_i.h     |  20 ++++++++
 fs/fuse/inode.c      |   1 +
 include/linux/fuse.h |  25 ++++++++++
 5 files changed, 197 insertions(+)

(limited to 'include/linux')

diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index ffd670bb8c8c..6176e444c76e 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -816,10 +816,29 @@ static ssize_t fuse_dev_read(struct kiocb *iocb, const struct iovec *iov,
 	return err;
 }
 
+static int fuse_notify_poll(struct fuse_conn *fc, unsigned int size,
+			    struct fuse_copy_state *cs)
+{
+	struct fuse_notify_poll_wakeup_out outarg;
+	int err;
+
+	if (size != sizeof(outarg))
+		return -EINVAL;
+
+	err = fuse_copy_one(cs, &outarg, sizeof(outarg));
+	if (err)
+		return err;
+
+	return fuse_notify_poll_wakeup(fc, &outarg);
+}
+
 static int fuse_notify(struct fuse_conn *fc, enum fuse_notify_code code,
 		       unsigned int size, struct fuse_copy_state *cs)
 {
 	switch (code) {
+	case FUSE_NOTIFY_POLL:
+		return fuse_notify_poll(fc, size, cs);
+
 	default:
 		return -EINVAL;
 	}
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index a28ced678d38..b3a944e4bb9c 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -62,6 +62,8 @@ struct fuse_file *fuse_file_alloc(struct fuse_conn *fc)
 			ff->kh = ++fc->khctr;
 			spin_unlock(&fc->lock);
 		}
+		RB_CLEAR_NODE(&ff->polled_node);
+		init_waitqueue_head(&ff->poll_wait);
 	}
 	return ff;
 }
@@ -170,7 +172,11 @@ int fuse_release_common(struct inode *inode, struct file *file, int isdir)
 
 		spin_lock(&fc->lock);
 		list_del(&ff->write_entry);
+		if (!RB_EMPTY_NODE(&ff->polled_node))
+			rb_erase(&ff->polled_node, &fc->polled_files);
 		spin_unlock(&fc->lock);
+
+		wake_up_interruptible_sync(&ff->poll_wait);
 		/*
 		 * Normally this will send the RELEASE request,
 		 * however if some asynchronous READ or WRITE requests
@@ -1749,6 +1755,130 @@ static long fuse_file_compat_ioctl(struct file *file, unsigned int cmd,
 	return fuse_file_do_ioctl(file, cmd, arg, FUSE_IOCTL_COMPAT);
 }
 
+/*
+ * All files which have been polled are linked to RB tree
+ * fuse_conn->polled_files which is indexed by kh.  Walk the tree and
+ * find the matching one.
+ */
+static struct rb_node **fuse_find_polled_node(struct fuse_conn *fc, u64 kh,
+					      struct rb_node **parent_out)
+{
+	struct rb_node **link = &fc->polled_files.rb_node;
+	struct rb_node *last = NULL;
+
+	while (*link) {
+		struct fuse_file *ff;
+
+		last = *link;
+		ff = rb_entry(last, struct fuse_file, polled_node);
+
+		if (kh < ff->kh)
+			link = &last->rb_left;
+		else if (kh > ff->kh)
+			link = &last->rb_right;
+		else
+			return link;
+	}
+
+	if (parent_out)
+		*parent_out = last;
+	return link;
+}
+
+/*
+ * The file is about to be polled.  Make sure it's on the polled_files
+ * RB tree.  Note that files once added to the polled_files tree are
+ * not removed before the file is released.  This is because a file
+ * polled once is likely to be polled again.
+ */
+static void fuse_register_polled_file(struct fuse_conn *fc,
+				      struct fuse_file *ff)
+{
+	spin_lock(&fc->lock);
+	if (RB_EMPTY_NODE(&ff->polled_node)) {
+		struct rb_node **link, *parent;
+
+		link = fuse_find_polled_node(fc, ff->kh, &parent);
+		BUG_ON(*link);
+		rb_link_node(&ff->polled_node, parent, link);
+		rb_insert_color(&ff->polled_node, &fc->polled_files);
+	}
+	spin_unlock(&fc->lock);
+}
+
+static unsigned fuse_file_poll(struct file *file, poll_table *wait)
+{
+	struct inode *inode = file->f_dentry->d_inode;
+	struct fuse_file *ff = file->private_data;
+	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_poll_in inarg = { .fh = ff->fh, .kh = ff->kh };
+	struct fuse_poll_out outarg;
+	struct fuse_req *req;
+	int err;
+
+	if (fc->no_poll)
+		return DEFAULT_POLLMASK;
+
+	poll_wait(file, &ff->poll_wait, wait);
+
+	/*
+	 * Ask for notification iff there's someone waiting for it.
+	 * The client may ignore the flag and always notify.
+	 */
+	if (waitqueue_active(&ff->poll_wait)) {
+		inarg.flags |= FUSE_POLL_SCHEDULE_NOTIFY;
+		fuse_register_polled_file(fc, ff);
+	}
+
+	req = fuse_get_req(fc);
+	if (IS_ERR(req))
+		return PTR_ERR(req);
+
+	req->in.h.opcode = FUSE_POLL;
+	req->in.h.nodeid = get_node_id(inode);
+	req->in.numargs = 1;
+	req->in.args[0].size = sizeof(inarg);
+	req->in.args[0].value = &inarg;
+	req->out.numargs = 1;
+	req->out.args[0].size = sizeof(outarg);
+	req->out.args[0].value = &outarg;
+	request_send(fc, req);
+	err = req->out.h.error;
+	fuse_put_request(fc, req);
+
+	if (!err)
+		return outarg.revents;
+	if (err == -ENOSYS) {
+		fc->no_poll = 1;
+		return DEFAULT_POLLMASK;
+	}
+	return POLLERR;
+}
+
+/*
+ * This is called from fuse_handle_notify() on FUSE_NOTIFY_POLL and
+ * wakes up the poll waiters.
+ */
+int fuse_notify_poll_wakeup(struct fuse_conn *fc,
+			    struct fuse_notify_poll_wakeup_out *outarg)
+{
+	u64 kh = outarg->kh;
+	struct rb_node **link;
+
+	spin_lock(&fc->lock);
+
+	link = fuse_find_polled_node(fc, kh, NULL);
+	if (*link) {
+		struct fuse_file *ff;
+
+		ff = rb_entry(*link, struct fuse_file, polled_node);
+		wake_up_interruptible_sync(&ff->poll_wait);
+	}
+
+	spin_unlock(&fc->lock);
+	return 0;
+}
+
 static const struct file_operations fuse_file_operations = {
 	.llseek		= fuse_file_llseek,
 	.read		= do_sync_read,
@@ -1765,6 +1895,7 @@ static const struct file_operations fuse_file_operations = {
 	.splice_read	= generic_file_splice_read,
 	.unlocked_ioctl	= fuse_file_ioctl,
 	.compat_ioctl	= fuse_file_compat_ioctl,
+	.poll		= fuse_file_poll,
 };
 
 static const struct file_operations fuse_direct_io_file_operations = {
@@ -1779,6 +1910,7 @@ static const struct file_operations fuse_direct_io_file_operations = {
 	.flock		= fuse_file_flock,
 	.unlocked_ioctl	= fuse_file_ioctl,
 	.compat_ioctl	= fuse_file_compat_ioctl,
+	.poll		= fuse_file_poll,
 	/* no mmap and splice_read */
 };
 
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index 86f013303828..986fbd4c1ff5 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -19,6 +19,8 @@
 #include <linux/backing-dev.h>
 #include <linux/mutex.h>
 #include <linux/rwsem.h>
+#include <linux/rbtree.h>
+#include <linux/poll.h>
 
 /** Max number of pages that can be used in a single read request */
 #define FUSE_MAX_PAGES_PER_REQ 32
@@ -111,6 +113,12 @@ struct fuse_file {
 
 	/** Entry on inode's write_files list */
 	struct list_head write_entry;
+
+	/** RB node to be linked on fuse_conn->polled_files */
+	struct rb_node polled_node;
+
+	/** Wait queue head for poll */
+	wait_queue_head_t poll_wait;
 };
 
 /** One input argument of a request */
@@ -328,6 +336,9 @@ struct fuse_conn {
 	/** The next unique kernel file handle */
 	u64 khctr;
 
+	/** rbtree of fuse_files waiting for poll events indexed by ph */
+	struct rb_root polled_files;
+
 	/** Number of requests currently in the background */
 	unsigned num_background;
 
@@ -416,6 +427,9 @@ struct fuse_conn {
 	/** Is bmap not implemented by fs? */
 	unsigned no_bmap:1;
 
+	/** Is poll not implemented by fs? */
+	unsigned no_poll:1;
+
 	/** Do multi-page cached writes */
 	unsigned big_writes:1;
 
@@ -524,6 +538,12 @@ int fuse_release_common(struct inode *inode, struct file *file, int isdir);
 int fuse_fsync_common(struct file *file, struct dentry *de, int datasync,
 		      int isdir);
 
+/**
+ * Notify poll wakeup
+ */
+int fuse_notify_poll_wakeup(struct fuse_conn *fc,
+			    struct fuse_notify_poll_wakeup_out *outarg);
+
 /**
  * Initialize file operations on a regular file
  */
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 0e15bc221d23..ba7256128084 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -486,6 +486,7 @@ static struct fuse_conn *new_conn(struct super_block *sb)
 		/* fuse does it's own writeback accounting */
 		fc->bdi.capabilities = BDI_CAP_NO_ACCT_WB;
 		fc->khctr = 0;
+		fc->polled_files = RB_ROOT;
 		fc->dev = sb->s_dev;
 		err = bdi_init(&fc->bdi);
 		if (err)
diff --git a/include/linux/fuse.h b/include/linux/fuse.h
index abde9949e2c0..5650cf033e73 100644
--- a/include/linux/fuse.h
+++ b/include/linux/fuse.h
@@ -163,6 +163,13 @@ struct fuse_file_lock {
 
 #define FUSE_IOCTL_MAX_IOV	256
 
+/**
+ * Poll flags
+ *
+ * FUSE_POLL_SCHEDULE_NOTIFY: request poll notify
+ */
+#define FUSE_POLL_SCHEDULE_NOTIFY (1 << 0)
+
 enum fuse_opcode {
 	FUSE_LOOKUP	   = 1,
 	FUSE_FORGET	   = 2,  /* no reply */
@@ -201,9 +208,11 @@ enum fuse_opcode {
 	FUSE_BMAP          = 37,
 	FUSE_DESTROY       = 38,
 	FUSE_IOCTL         = 39,
+	FUSE_POLL          = 40,
 };
 
 enum fuse_notify_code {
+	FUSE_NOTIFY_POLL   = 1,
 	FUSE_NOTIFY_CODE_MAX,
 };
 
@@ -421,6 +430,22 @@ struct fuse_ioctl_out {
 	__u32	out_iovs;
 };
 
+struct fuse_poll_in {
+	__u64	fh;
+	__u64	kh;
+	__u32	flags;
+	__u32   padding;
+};
+
+struct fuse_poll_out {
+	__u32	revents;
+	__u32	padding;
+};
+
+struct fuse_notify_poll_wakeup_out {
+	__u64	kh;
+};
+
 struct fuse_in_header {
 	__u32	len;
 	__u32	opcode;
-- 
cgit v1.2.3


From dcc7461eef7341e84e2f7274f904ce01a43b2506 Mon Sep 17 00:00:00 2001
From: David Vrabel <david.vrabel@csr.com>
Date: Wed, 26 Nov 2008 13:36:59 +0000
Subject: wusb: add debug files for ASL, PZL and DI to the whci-hcd driver

Add asl, pzl and di debugfs files to uwb/uwbN/wusbhc for WHCI host
controller.  These dump the current ASL, PZL and DI buffer.

Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/usb/host/whci/Kbuild  |   1 +
 drivers/usb/host/whci/asl.c   |  25 ------
 drivers/usb/host/whci/debug.c | 189 ++++++++++++++++++++++++++++++++++++++++++
 drivers/usb/host/whci/hcd.c   |   3 +
 drivers/usb/host/whci/pzl.c   |  28 -------
 drivers/usb/host/whci/qset.c  |  40 ---------
 drivers/usb/host/whci/whcd.h  |   9 +-
 drivers/usb/host/whci/wusb.c  |  27 ------
 drivers/uwb/pal.c             |   5 ++
 drivers/uwb/uwb-debug.c       |  13 +++
 drivers/uwb/uwb-internal.h    |   3 +-
 include/linux/uwb.h           |   3 +
 12 files changed, 223 insertions(+), 123 deletions(-)
 create mode 100644 drivers/usb/host/whci/debug.c

(limited to 'include/linux')

diff --git a/drivers/usb/host/whci/Kbuild b/drivers/usb/host/whci/Kbuild
index 26a3871ea0f9..11e5040b8337 100644
--- a/drivers/usb/host/whci/Kbuild
+++ b/drivers/usb/host/whci/Kbuild
@@ -2,6 +2,7 @@ obj-$(CONFIG_USB_WHCI_HCD) += whci-hcd.o
 
 whci-hcd-y := \
 	asl.o	\
+	debug.o \
 	hcd.o 	\
 	hw.o	\
 	init.o	\
diff --git a/drivers/usb/host/whci/asl.c b/drivers/usb/host/whci/asl.c
index ba99a7a3f81a..577c0d29849d 100644
--- a/drivers/usb/host/whci/asl.c
+++ b/drivers/usb/host/whci/asl.c
@@ -19,32 +19,11 @@
 #include <linux/dma-mapping.h>
 #include <linux/uwb/umc.h>
 #include <linux/usb.h>
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
 
 #include "../../wusbcore/wusbhc.h"
 
 #include "whcd.h"
 
-#if D_LOCAL >= 4
-static void dump_asl(struct whc *whc, const char *tag)
-{
-	struct device *dev = &whc->umc->dev;
-	struct whc_qset *qset;
-
-	d_printf(4, dev, "ASL %s\n", tag);
-
-	list_for_each_entry(qset, &whc->async_list, list_node) {
-		dump_qset(qset, dev);
-	}
-}
-#else
-static inline void dump_asl(struct whc *whc, const char *tag)
-{
-}
-#endif
-
-
 static void qset_get_next_prev(struct whc *whc, struct whc_qset *qset,
 			       struct whc_qset **next, struct whc_qset **prev)
 {
@@ -217,8 +196,6 @@ void scan_async_work(struct work_struct *work)
 
 	spin_lock_irq(&whc->lock);
 
-	dump_asl(whc, "before processing");
-
 	/*
 	 * Transerve the software list backwards so new qsets can be
 	 * safely inserted into the ASL without making it non-circular.
@@ -232,8 +209,6 @@ void scan_async_work(struct work_struct *work)
 		update |= process_qset(whc, qset);
 	}
 
-	dump_asl(whc, "after processing");
-
 	spin_unlock_irq(&whc->lock);
 
 	if (update) {
diff --git a/drivers/usb/host/whci/debug.c b/drivers/usb/host/whci/debug.c
new file mode 100644
index 000000000000..cf2d45946c57
--- /dev/null
+++ b/drivers/usb/host/whci/debug.c
@@ -0,0 +1,189 @@
+/*
+ * Wireless Host Controller (WHC) debug.
+ *
+ * Copyright (C) 2008 Cambridge Silicon Radio Ltd.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+#include <linux/kernel.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+
+#include "../../wusbcore/wusbhc.h"
+
+#include "whcd.h"
+
+struct whc_dbg {
+	struct dentry *di_f;
+	struct dentry *asl_f;
+	struct dentry *pzl_f;
+};
+
+void qset_print(struct seq_file *s, struct whc_qset *qset)
+{
+	struct whc_std *std;
+	struct urb *urb = NULL;
+	int i;
+
+	seq_printf(s, "qset %08x\n", (u32)qset->qset_dma);
+	seq_printf(s, "  -> %08x\n", (u32)qset->qh.link);
+	seq_printf(s, "  info: %08x %08x %08x\n",
+		qset->qh.info1, qset->qh.info2,  qset->qh.info3);
+	seq_printf(s, "  sts: %04x errs: %d\n", qset->qh.status, qset->qh.err_count);
+	seq_printf(s, "  TD: sts: %08x opts: %08x\n",
+		qset->qh.overlay.qtd.status, qset->qh.overlay.qtd.options);
+
+	for (i = 0; i < WHCI_QSET_TD_MAX; i++) {
+		seq_printf(s, "  %c%c TD[%d]: sts: %08x opts: %08x ptr: %08x\n",
+			i == qset->td_start ? 'S' : ' ',
+			i == qset->td_end ? 'E' : ' ',
+			i, qset->qtd[i].status, qset->qtd[i].options,
+			(u32)qset->qtd[i].page_list_ptr);
+	}
+	seq_printf(s, "  ntds: %d\n", qset->ntds);
+	list_for_each_entry(std, &qset->stds, list_node) {
+		if (urb != std->urb) {
+			urb = std->urb;
+			seq_printf(s, "  urb %p transferred: %d bytes\n", urb,
+				urb->actual_length);
+		}
+		if (std->qtd)
+			seq_printf(s, "    sTD[%td]: %zu bytes @ %08x\n",
+				std->qtd - &qset->qtd[0],
+				std->len, std->num_pointers ?
+				(u32)(std->pl_virt[0].buf_ptr) : (u32)std->dma_addr);
+		else
+			seq_printf(s, "    sTD[-]: %zd bytes @ %08x\n",
+				std->len, std->num_pointers ?
+				(u32)(std->pl_virt[0].buf_ptr) : (u32)std->dma_addr);
+	}
+}
+
+static int di_print(struct seq_file *s, void *p)
+{
+	struct whc *whc = s->private;
+	char buf[72];
+	int d;
+
+	for (d = 0; d < whc->n_devices; d++) {
+		struct di_buf_entry *di = &whc->di_buf[d];
+
+		bitmap_scnprintf(buf, sizeof(buf),
+				 (unsigned long *)di->availability_info, UWB_NUM_MAS);
+
+		seq_printf(s, "DI[%d]\n", d);
+		seq_printf(s, "  availability: %s\n", buf);
+		seq_printf(s, "  %c%c key idx: %d dev addr: %d\n",
+			   (di->addr_sec_info & WHC_DI_SECURE) ? 'S' : ' ',
+			   (di->addr_sec_info & WHC_DI_DISABLE) ? 'D' : ' ',
+			   (di->addr_sec_info & WHC_DI_KEY_IDX_MASK) >> 8,
+			   (di->addr_sec_info & WHC_DI_DEV_ADDR_MASK));
+	}
+	return 0;
+}
+
+static int asl_print(struct seq_file *s, void *p)
+{
+	struct whc *whc = s->private;
+	struct whc_qset *qset;
+
+	list_for_each_entry(qset, &whc->async_list, list_node) {
+		qset_print(s, qset);
+	}
+
+	return 0;
+}
+
+static int pzl_print(struct seq_file *s, void *p)
+{
+	struct whc *whc = s->private;
+	struct whc_qset *qset;
+	int period;
+
+	for (period = 0; period < 5; period++) {
+		seq_printf(s, "Period %d\n", period);
+		list_for_each_entry(qset, &whc->periodic_list[period], list_node) {
+			qset_print(s, qset);
+		}
+	}
+	return 0;
+}
+
+static int di_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, di_print, inode->i_private);
+}
+
+static int asl_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, asl_print, inode->i_private);
+}
+
+static int pzl_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, pzl_print, inode->i_private);
+}
+
+static struct file_operations di_fops = {
+	.open    = di_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = single_release,
+	.owner   = THIS_MODULE,
+};
+
+static struct file_operations asl_fops = {
+	.open    = asl_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = single_release,
+	.owner   = THIS_MODULE,
+};
+
+static struct file_operations pzl_fops = {
+	.open    = pzl_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = single_release,
+	.owner   = THIS_MODULE,
+};
+
+void whc_dbg_init(struct whc *whc)
+{
+	if (whc->wusbhc.pal.debugfs_dir == NULL)
+		return;
+
+	whc->dbg = kzalloc(sizeof(struct whc_dbg), GFP_KERNEL);
+	if (whc->dbg == NULL)
+		return;
+
+	whc->dbg->di_f = debugfs_create_file("di", 0444,
+					      whc->wusbhc.pal.debugfs_dir, whc,
+					      &di_fops);
+	whc->dbg->asl_f = debugfs_create_file("asl", 0444,
+					      whc->wusbhc.pal.debugfs_dir, whc,
+					      &asl_fops);
+	whc->dbg->pzl_f = debugfs_create_file("pzl", 0444,
+					      whc->wusbhc.pal.debugfs_dir, whc,
+					      &pzl_fops);
+}
+
+void whc_dbg_clean_up(struct whc *whc)
+{
+	if (whc->dbg) {
+		debugfs_remove(whc->dbg->pzl_f);
+		debugfs_remove(whc->dbg->asl_f);
+		debugfs_remove(whc->dbg->di_f);
+		kfree(whc->dbg);
+	}
+}
diff --git a/drivers/usb/host/whci/hcd.c b/drivers/usb/host/whci/hcd.c
index f599f89d3be1..1569afd6245b 100644
--- a/drivers/usb/host/whci/hcd.c
+++ b/drivers/usb/host/whci/hcd.c
@@ -273,6 +273,8 @@ static int whc_probe(struct umc_dev *umc)
 		goto error_wusbhc_b_create;
 	}
 
+	whc_dbg_init(whc);
+
 	return 0;
 
 error_wusbhc_b_create:
@@ -296,6 +298,7 @@ static void whc_remove(struct umc_dev *umc)
 	struct whc *whc = wusbhc_to_whc(wusbhc);
 
 	if (usb_hcd) {
+		whc_dbg_clean_up(whc);
 		wusbhc_b_destroy(wusbhc);
 		usb_remove_hcd(usb_hcd);
 		wusbhc_destroy(wusbhc);
diff --git a/drivers/usb/host/whci/pzl.c b/drivers/usb/host/whci/pzl.c
index 34d3a0aeab2b..2ae5abf69a6a 100644
--- a/drivers/usb/host/whci/pzl.c
+++ b/drivers/usb/host/whci/pzl.c
@@ -19,35 +19,11 @@
 #include <linux/dma-mapping.h>
 #include <linux/uwb/umc.h>
 #include <linux/usb.h>
-#define D_LOCAL 0
-#include <linux/uwb/debug.h>
 
 #include "../../wusbcore/wusbhc.h"
 
 #include "whcd.h"
 
-#if D_LOCAL >= 4
-static void dump_pzl(struct whc *whc, const char *tag)
-{
-	struct device *dev = &whc->umc->dev;
-	struct whc_qset *qset;
-	int period = 0;
-
-	d_printf(4, dev, "PZL %s\n", tag);
-
-	for (period = 0; period < 5; period++) {
-		d_printf(4, dev, "Period %d\n", period);
-		list_for_each_entry(qset, &whc->periodic_list[period], list_node) {
-			dump_qset(qset, dev);
-		}
-	}
-}
-#else
-static inline void dump_pzl(struct whc *whc, const char *tag)
-{
-}
-#endif
-
 static void update_pzl_pointers(struct whc *whc, int period, u64 addr)
 {
 	switch (period) {
@@ -250,8 +226,6 @@ void scan_periodic_work(struct work_struct *work)
 
 	spin_lock_irq(&whc->lock);
 
-	dump_pzl(whc, "before processing");
-
 	for (period = 4; period >= 0; period--) {
 		list_for_each_entry_safe(qset, t, &whc->periodic_list[period], list_node) {
 			if (!qset->in_hw_list)
@@ -263,8 +237,6 @@ void scan_periodic_work(struct work_struct *work)
 	if (update & (WHC_UPDATE_ADDED | WHC_UPDATE_REMOVED))
 		update_pzl_hw_view(whc);
 
-	dump_pzl(whc, "after processing");
-
 	spin_unlock_irq(&whc->lock);
 
 	if (update) {
diff --git a/drivers/usb/host/whci/qset.c b/drivers/usb/host/whci/qset.c
index 0420037d2e18..7be74314ee12 100644
--- a/drivers/usb/host/whci/qset.c
+++ b/drivers/usb/host/whci/qset.c
@@ -24,46 +24,6 @@
 
 #include "whcd.h"
 
-void dump_qset(struct whc_qset *qset, struct device *dev)
-{
-	struct whc_std *std;
-	struct urb *urb = NULL;
-	int i;
-
-	dev_dbg(dev, "qset %08x\n", (u32)qset->qset_dma);
-	dev_dbg(dev, "  -> %08x\n", (u32)qset->qh.link);
-	dev_dbg(dev, "  info: %08x %08x %08x\n",
-		qset->qh.info1, qset->qh.info2,  qset->qh.info3);
-	dev_dbg(dev, "  sts: %04x errs: %d\n", qset->qh.status, qset->qh.err_count);
-	dev_dbg(dev, "  TD: sts: %08x opts: %08x\n",
-		qset->qh.overlay.qtd.status, qset->qh.overlay.qtd.options);
-
-	for (i = 0; i < WHCI_QSET_TD_MAX; i++) {
-		dev_dbg(dev, "  %c%c TD[%d]: sts: %08x opts: %08x ptr: %08x\n",
-			i == qset->td_start ? 'S' : ' ',
-			i == qset->td_end ? 'E' : ' ',
-			i, qset->qtd[i].status, qset->qtd[i].options,
-			(u32)qset->qtd[i].page_list_ptr);
-	}
-	dev_dbg(dev, "  ntds: %d\n", qset->ntds);
-	list_for_each_entry(std, &qset->stds, list_node) {
-		if (urb != std->urb) {
-			urb = std->urb;
-			dev_dbg(dev, "  urb %p transferred: %d bytes\n", urb,
-				urb->actual_length);
-		}
-		if (std->qtd)
-			dev_dbg(dev, "    sTD[%td]: %zu bytes @ %08x\n",
-				std->qtd - &qset->qtd[0],
-				std->len, std->num_pointers ?
-				(u32)(std->pl_virt[0].buf_ptr) : (u32)std->dma_addr);
-		else
-			dev_dbg(dev, "    sTD[-]: %zd bytes @ %08x\n",
-				std->len, std->num_pointers ?
-				(u32)(std->pl_virt[0].buf_ptr) : (u32)std->dma_addr);
-	}
-}
-
 struct whc_qset *qset_alloc(struct whc *whc, gfp_t mem_flags)
 {
 	struct whc_qset *qset;
diff --git a/drivers/usb/host/whci/whcd.h b/drivers/usb/host/whci/whcd.h
index 1bbb8cb6bf80..0f3540f04f53 100644
--- a/drivers/usb/host/whci/whcd.h
+++ b/drivers/usb/host/whci/whcd.h
@@ -21,6 +21,7 @@
 #define __WHCD_H
 
 #include <linux/uwb/whci.h>
+#include <linux/uwb/umc.h>
 #include <linux/workqueue.h>
 
 #include "whci-hc.h"
@@ -28,6 +29,7 @@
 /* Generic command timeout. */
 #define WHC_GENCMD_TIMEOUT_MS 100
 
+struct whc_dbg;
 
 struct whc {
 	struct wusbhc wusbhc;
@@ -69,6 +71,8 @@ struct whc {
 	struct list_head periodic_removed_list;
 	wait_queue_head_t periodic_list_wq;
 	struct work_struct periodic_work;
+
+	struct whc_dbg *dbg;
 };
 
 #define wusbhc_to_whc(w) (container_of((w), struct whc, wusbhc))
@@ -190,8 +194,11 @@ void process_inactive_qtd(struct whc *whc, struct whc_qset *qset,
 				 struct whc_qtd *qtd);
 enum whc_update qset_add_qtds(struct whc *whc, struct whc_qset *qset);
 void qset_remove_complete(struct whc *whc, struct whc_qset *qset);
-void dump_qset(struct whc_qset *qset, struct device *dev);
 void pzl_update(struct whc *whc, uint32_t wusbcmd);
 void asl_update(struct whc *whc, uint32_t wusbcmd);
 
+/* debug.c */
+void whc_dbg_init(struct whc *whc);
+void whc_dbg_clean_up(struct whc *whc);
+
 #endif /* #ifndef __WHCD_H */
diff --git a/drivers/usb/host/whci/wusb.c b/drivers/usb/host/whci/wusb.c
index 540021a0971e..f24efdebad17 100644
--- a/drivers/usb/host/whci/wusb.c
+++ b/drivers/usb/host/whci/wusb.c
@@ -18,43 +18,16 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/uwb/umc.h>
-#define D_LOCAL 1
-#include <linux/uwb/debug.h>
 
 #include "../../wusbcore/wusbhc.h"
 
 #include "whcd.h"
 
-#if D_LOCAL >= 1
-static void dump_di(struct whc *whc, int idx)
-{
-	struct di_buf_entry *di = &whc->di_buf[idx];
-	struct device *dev = &whc->umc->dev;
-	char buf[128];
-
-	bitmap_scnprintf(buf, sizeof(buf), (unsigned long *)di->availability_info, UWB_NUM_MAS);
-
-	d_printf(1, dev, "DI[%d]\n", idx);
-	d_printf(1, dev, "  availability: %s\n", buf);
-	d_printf(1, dev, "  %c%c key idx: %d dev addr: %d\n",
-		 (di->addr_sec_info & WHC_DI_SECURE) ? 'S' : ' ',
-		 (di->addr_sec_info & WHC_DI_DISABLE) ? 'D' : ' ',
-		 (di->addr_sec_info & WHC_DI_KEY_IDX_MASK) >> 8,
-		 (di->addr_sec_info & WHC_DI_DEV_ADDR_MASK));
-}
-#else
-static inline void dump_di(struct whc *whc, int idx)
-{
-}
-#endif
-
 static int whc_update_di(struct whc *whc, int idx)
 {
 	int offset = idx / 32;
 	u32 bit = 1 << (idx % 32);
 
-	dump_di(whc, idx);
-
 	le_writel(bit, whc->base + WUSBDIBUPDATED + offset);
 
 	return whci_wait_for(&whc->umc->dev,
diff --git a/drivers/uwb/pal.c b/drivers/uwb/pal.c
index 605765124f5b..99a19c199095 100644
--- a/drivers/uwb/pal.c
+++ b/drivers/uwb/pal.c
@@ -16,6 +16,7 @@
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 #include <linux/kernel.h>
+#include <linux/debugfs.h>
 #include <linux/uwb.h>
 
 #include "uwb-internal.h"
@@ -54,6 +55,8 @@ int uwb_pal_register(struct uwb_pal *pal)
 		}
 	}
 
+	pal->debugfs_dir = uwb_dbg_create_pal_dir(pal);
+
 	mutex_lock(&rc->uwb_dev.mutex);
 	list_add(&pal->node, &rc->pals);
 	mutex_unlock(&rc->uwb_dev.mutex);
@@ -76,6 +79,8 @@ void uwb_pal_unregister(struct uwb_pal *pal)
 	list_del(&pal->node);
 	mutex_unlock(&rc->uwb_dev.mutex);
 
+	debugfs_remove(pal->debugfs_dir);
+
 	if (pal->device) {
 		sysfs_remove_link(&rc->uwb_dev.dev.kobj, pal->name);
 		sysfs_remove_link(&pal->device->kobj, "uwb_rc");
diff --git a/drivers/uwb/uwb-debug.c b/drivers/uwb/uwb-debug.c
index ec1b7a403ff3..a6debb9baf38 100644
--- a/drivers/uwb/uwb-debug.c
+++ b/drivers/uwb/uwb-debug.c
@@ -407,3 +407,16 @@ void uwb_dbg_exit(void)
 {
 	debugfs_remove(root_dir);
 }
+
+/**
+ * uwb_dbg_create_pal_dir - create a debugfs directory for a PAL
+ * @pal: The PAL.
+ */
+struct dentry *uwb_dbg_create_pal_dir(struct uwb_pal *pal)
+{
+	struct uwb_rc *rc = pal->rc;
+
+	if (root_dir && rc->dbg && rc->dbg->root_d && pal->name)
+		return debugfs_create_dir(pal->name, rc->dbg->root_d);
+	return NULL;
+}
diff --git a/drivers/uwb/uwb-internal.h b/drivers/uwb/uwb-internal.h
index 9c0cdb4ded0c..f0f21f406bf0 100644
--- a/drivers/uwb/uwb-internal.h
+++ b/drivers/uwb/uwb-internal.h
@@ -284,8 +284,7 @@ void uwb_dbg_init(void);
 void uwb_dbg_exit(void);
 void uwb_dbg_add_rc(struct uwb_rc *rc);
 void uwb_dbg_del_rc(struct uwb_rc *rc);
-
-/* Workarounds for version specific stuff */
+struct dentry *uwb_dbg_create_pal_dir(struct uwb_pal *pal);
 
 static inline void uwb_dev_lock(struct uwb_dev *uwb_dev)
 {
diff --git a/include/linux/uwb.h b/include/linux/uwb.h
index 1719709d60ca..d7ed5201ade6 100644
--- a/include/linux/uwb.h
+++ b/include/linux/uwb.h
@@ -394,6 +394,8 @@ struct uwb_rc {
  * @channel: channel being used by the PAL; 0 if the PAL isn't using
  *           the radio; -1 if the PAL wishes to use the radio but
  *           cannot.
+ * @debugfs_dir: a debugfs directory which the PAL can use for its own
+ *           debugfs files.
  *
  * A Protocol Adaptation Layer (PAL) is a user of the WiMedia UWB
  * radio platform (e.g., WUSB, WLP or Bluetooth UWB AMP).
@@ -418,6 +420,7 @@ struct uwb_pal {
 	void (*new_rsv)(struct uwb_pal *pal, struct uwb_rsv *rsv);
 
 	int channel;
+	struct dentry *debugfs_dir;
 };
 
 void uwb_pal_init(struct uwb_pal *pal);
-- 
cgit v1.2.3


From 1f55ed06cf0c361b293b32e5947d35d173eff2aa Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Mon, 1 Dec 2008 19:14:02 +0100
Subject: fuse: update interface version

Change interface version to 7.11 after adding the IOCTL and POLL
messages.

Also clean up the <linux/fuse.h> header a bit:
  - update copyright date to 2008
  - fix checkpatch warning:
      WARNING: Use #include <linux/types.h> instead of <asm/types.h>
  - remove FUSE_MAJOR define, which is not being used any more

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
---
 include/linux/fuse.h | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/fuse.h b/include/linux/fuse.h
index 5650cf033e73..162e5defe683 100644
--- a/include/linux/fuse.h
+++ b/include/linux/fuse.h
@@ -1,6 +1,6 @@
 /*
     FUSE: Filesystem in Userspace
-    Copyright (C) 2001-2006  Miklos Szeredi <miklos@szeredi.hu>
+    Copyright (C) 2001-2008  Miklos Szeredi <miklos@szeredi.hu>
 
     This program can be distributed under the terms of the GNU GPL.
     See the file COPYING.
@@ -20,26 +20,27 @@
  *
  * 7.10
  *  - add nonseekable open flag
+ *
+ * 7.11
+ *  - add IOCTL message
+ *  - add unsolicited notification support
+ *  - add POLL message and NOTIFY_POLL notification
  */
 
 #ifndef _LINUX_FUSE_H
 #define _LINUX_FUSE_H
 
-#include <asm/types.h>
-#include <linux/major.h>
+#include <linux/types.h>
 
 /** Version number of this interface */
 #define FUSE_KERNEL_VERSION 7
 
 /** Minor version number of this interface */
-#define FUSE_KERNEL_MINOR_VERSION 10
+#define FUSE_KERNEL_MINOR_VERSION 11
 
 /** The node ID of the root inode */
 #define FUSE_ROOT_ID 1
 
-/** The major number of the fuse character device */
-#define FUSE_MAJOR MISC_MAJOR
-
 /* Make sure all structures are padded to 64bit boundary, so 32bit
    userspace works under 64bit kernels */
 
-- 
cgit v1.2.3


From e088e4c9cdb618675874becb91b2fd581ee707e6 Mon Sep 17 00:00:00 2001
From: Matthew Garrett <mjg@redhat.com>
Date: Tue, 25 Nov 2008 13:29:47 -0500
Subject: [CPUFREQ] Disable sysfs ui for p4-clockmod.

p4-clockmod has a long history of abuse.   It pretends to be a CPU
frequency scaling driver, even though it doesn't actually change
the CPU frequency, but instead just modulates the frequency with
wait-states.
The biggest misconception is that when running at the lower 'frequency'
p4-clockmod is saving power.  This isn't the case, as workloads running
slower take longer to complete, preventing the CPU from entering deep C states.

However p4-clockmod does have a purpose.  It can prevent overheating.
Having it hooked up to the cpufreq interfaces is the wrong way to achieve
cooling however. It should instead be hooked up to ACPI.

This diff introduces a means for a cpufreq driver to register with the
cpufreq core, but not present a sysfs interface.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 arch/x86/kernel/cpu/cpufreq/p4-clockmod.c |  1 +
 drivers/cpufreq/cpufreq.c                 | 51 ++++++++++++++++++++-----------
 include/linux/cpufreq.h                   |  1 +
 3 files changed, 35 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
index ba3a94a997c3..0c43b2240517 100644
--- a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
+++ b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
@@ -276,6 +276,7 @@ static struct cpufreq_driver p4clockmod_driver = {
 	.name		= "p4-clockmod",
 	.owner		= THIS_MODULE,
 	.attr		= p4clockmod_attr,
+	.hide_interface	= 1,
 };
 
 
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 31d6f535a79d..9044b911d8ae 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -754,6 +754,11 @@ static struct kobj_type ktype_cpufreq = {
 	.release	= cpufreq_sysfs_release,
 };
 
+static struct kobj_type ktype_empty_cpufreq = {
+	.sysfs_ops	= &sysfs_ops,
+	.release	= cpufreq_sysfs_release,
+};
+
 
 /**
  * cpufreq_add_dev - add a CPU device
@@ -876,26 +881,36 @@ static int cpufreq_add_dev(struct sys_device *sys_dev)
 	memcpy(&new_policy, policy, sizeof(struct cpufreq_policy));
 
 	/* prepare interface data */
-	ret = kobject_init_and_add(&policy->kobj, &ktype_cpufreq, &sys_dev->kobj,
-				   "cpufreq");
-	if (ret)
-		goto err_out_driver_exit;
-
-	/* set up files for this cpu device */
-	drv_attr = cpufreq_driver->attr;
-	while ((drv_attr) && (*drv_attr)) {
-		ret = sysfs_create_file(&policy->kobj, &((*drv_attr)->attr));
+	if (!cpufreq_driver->hide_interface) {
+		ret = kobject_init_and_add(&policy->kobj, &ktype_cpufreq,
+					   &sys_dev->kobj, "cpufreq");
 		if (ret)
 			goto err_out_driver_exit;
-		drv_attr++;
-	}
-	if (cpufreq_driver->get) {
-		ret = sysfs_create_file(&policy->kobj, &cpuinfo_cur_freq.attr);
-		if (ret)
-			goto err_out_driver_exit;
-	}
-	if (cpufreq_driver->target) {
-		ret = sysfs_create_file(&policy->kobj, &scaling_cur_freq.attr);
+
+		/* set up files for this cpu device */
+		drv_attr = cpufreq_driver->attr;
+		while ((drv_attr) && (*drv_attr)) {
+			ret = sysfs_create_file(&policy->kobj,
+						&((*drv_attr)->attr));
+			if (ret)
+				goto err_out_driver_exit;
+			drv_attr++;
+		}
+		if (cpufreq_driver->get) {
+			ret = sysfs_create_file(&policy->kobj,
+						&cpuinfo_cur_freq.attr);
+			if (ret)
+				goto err_out_driver_exit;
+		}
+		if (cpufreq_driver->target) {
+			ret = sysfs_create_file(&policy->kobj,
+						&scaling_cur_freq.attr);
+			if (ret)
+				goto err_out_driver_exit;
+		}
+	} else {
+		ret = kobject_init_and_add(&policy->kobj, &ktype_empty_cpufreq,
+					   &sys_dev->kobj, "cpufreq");
 		if (ret)
 			goto err_out_driver_exit;
 	}
diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
index 1ee608fd7b77..484b3abf61bb 100644
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -234,6 +234,7 @@ struct cpufreq_driver {
 	int	(*suspend)	(struct cpufreq_policy *policy, pm_message_t pmsg);
 	int	(*resume)	(struct cpufreq_policy *policy);
 	struct freq_attr	**attr;
+	bool			hide_interface;
 };
 
 /* flags */
-- 
cgit v1.2.3


From 69423d99fc182a81f3c5db3eb5c140acc6fc64be Mon Sep 17 00:00:00 2001
From: Adrian Hunter <ext-adrian.hunter@nokia.com>
Date: Wed, 10 Dec 2008 13:37:21 +0000
Subject: [MTD] update internal API to support 64-bit device size

MTD internal API presently uses 32-bit values to represent
device size.  This patch updates them to 64-bits but leaves
the external API unchanged.  Extending the external API
is a separate issue for several reasons.  First, no one
needs it at the moment.  Secondly, whether the implementation
is done with IOCTLs, sysfs or both is still debated.  Thirdly
external API changes require the internal API to be accepted
first.

Note that although the MTD API will be able to support 64-bit
device sizes, existing drivers do not and are not required
to do so, although NAND base has been updated.

In general, changing from 32-bit to 64-bit values cause little
or no changes to the majority of the code with the following
exceptions:
    	- printk message formats
    	- division and modulus of 64-bit values
    	- NAND base support
	- 32-bit local variables used by mtdpart and mtdconcat
	- naughtily assuming one structure maps to another
	in MEMERASE ioctl

Signed-off-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/chips/cfi_cmdset_0001.c | 12 ++++----
 drivers/mtd/chips/cfi_cmdset_0002.c |  8 +++---
 drivers/mtd/chips/cfi_cmdset_0020.c | 14 ++++-----
 drivers/mtd/chips/fwh_lock.h        |  4 +--
 drivers/mtd/inftlcore.c             |  2 +-
 drivers/mtd/inftlmount.c            |  4 +--
 drivers/mtd/maps/amd76xrom.c        |  4 +--
 drivers/mtd/maps/ck804xrom.c        |  4 +--
 drivers/mtd/maps/esb2rom.c          |  4 +--
 drivers/mtd/maps/ichxrom.c          |  4 +--
 drivers/mtd/maps/nettel.c           |  2 +-
 drivers/mtd/maps/scb2_flash.c       |  8 ++++--
 drivers/mtd/mtdchar.c               |  6 +++-
 drivers/mtd/mtdconcat.c             | 33 ++++++++++++---------
 drivers/mtd/mtdcore.c               | 16 ++++++++++-
 drivers/mtd/mtdoops.c               |  9 ++++--
 drivers/mtd/mtdpart.c               | 34 +++++++++++-----------
 drivers/mtd/nand/nand_base.c        | 24 ++++++++++------
 drivers/mtd/nand/nand_bbt.c         | 31 ++++++++++----------
 drivers/mtd/nftlcore.c              |  2 +-
 drivers/mtd/nftlmount.c             |  4 +--
 drivers/mtd/onenand/onenand_base.c  |  8 +++---
 drivers/mtd/rfd_ftl.c               | 23 ++++++++-------
 drivers/mtd/ssfdc.c                 |  7 +++--
 drivers/mtd/ubi/build.c             |  2 +-
 drivers/mtd/ubi/gluebi.c            | 17 +++++------
 fs/jffs2/erase.c                    |  5 ++--
 include/linux/mtd/mtd.h             | 57 +++++++++++++++++++++++++++++++------
 include/linux/mtd/nand.h            |  2 +-
 include/linux/mtd/partitions.h      |  4 +--
 30 files changed, 216 insertions(+), 138 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mtd/chips/cfi_cmdset_0001.c b/drivers/mtd/chips/cfi_cmdset_0001.c
index c93a8be5d5f1..f5ab6fa1057b 100644
--- a/drivers/mtd/chips/cfi_cmdset_0001.c
+++ b/drivers/mtd/chips/cfi_cmdset_0001.c
@@ -58,8 +58,8 @@ static int cfi_intelext_write_buffers(struct mtd_info *, loff_t, size_t, size_t
 static int cfi_intelext_writev(struct mtd_info *, const struct kvec *, unsigned long, loff_t, size_t *);
 static int cfi_intelext_erase_varsize(struct mtd_info *, struct erase_info *);
 static void cfi_intelext_sync (struct mtd_info *);
-static int cfi_intelext_lock(struct mtd_info *mtd, loff_t ofs, size_t len);
-static int cfi_intelext_unlock(struct mtd_info *mtd, loff_t ofs, size_t len);
+static int cfi_intelext_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len);
+static int cfi_intelext_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len);
 #ifdef CONFIG_MTD_OTP
 static int cfi_intelext_read_fact_prot_reg (struct mtd_info *, loff_t, size_t, size_t *, u_char *);
 static int cfi_intelext_read_user_prot_reg (struct mtd_info *, loff_t, size_t, size_t *, u_char *);
@@ -558,8 +558,8 @@ static struct mtd_info *cfi_intelext_setup(struct mtd_info *mtd)
 	}
 
 	for (i=0; i<mtd->numeraseregions;i++){
-		printk(KERN_DEBUG "erase region %d: offset=0x%x,size=0x%x,blocks=%d\n",
-		       i,mtd->eraseregions[i].offset,
+		printk(KERN_DEBUG "erase region %d: offset=0x%llx,size=0x%x,blocks=%d\n",
+		       i,(unsigned long long)mtd->eraseregions[i].offset,
 		       mtd->eraseregions[i].erasesize,
 		       mtd->eraseregions[i].numblocks);
 	}
@@ -2058,7 +2058,7 @@ out:	put_chip(map, chip, adr);
 	return ret;
 }
 
-static int cfi_intelext_lock(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int cfi_intelext_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	int ret;
 
@@ -2082,7 +2082,7 @@ static int cfi_intelext_lock(struct mtd_info *mtd, loff_t ofs, size_t len)
 	return ret;
 }
 
-static int cfi_intelext_unlock(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int cfi_intelext_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	int ret;
 
diff --git a/drivers/mtd/chips/cfi_cmdset_0002.c b/drivers/mtd/chips/cfi_cmdset_0002.c
index d74ec46aa032..f9c435a42670 100644
--- a/drivers/mtd/chips/cfi_cmdset_0002.c
+++ b/drivers/mtd/chips/cfi_cmdset_0002.c
@@ -71,8 +71,8 @@ static int get_chip(struct map_info *map, struct flchip *chip, unsigned long adr
 static void put_chip(struct map_info *map, struct flchip *chip, unsigned long adr);
 #include "fwh_lock.h"
 
-static int cfi_atmel_lock(struct mtd_info *mtd, loff_t ofs, size_t len);
-static int cfi_atmel_unlock(struct mtd_info *mtd, loff_t ofs, size_t len);
+static int cfi_atmel_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len);
+static int cfi_atmel_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len);
 
 static struct mtd_chip_driver cfi_amdstd_chipdrv = {
 	.probe		= NULL, /* Not usable directly */
@@ -1774,12 +1774,12 @@ out_unlock:
 	return ret;
 }
 
-static int cfi_atmel_lock(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int cfi_atmel_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	return cfi_varsize_frob(mtd, do_atmel_lock, ofs, len, NULL);
 }
 
-static int cfi_atmel_unlock(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int cfi_atmel_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	return cfi_varsize_frob(mtd, do_atmel_unlock, ofs, len, NULL);
 }
diff --git a/drivers/mtd/chips/cfi_cmdset_0020.c b/drivers/mtd/chips/cfi_cmdset_0020.c
index d4714dd9f7ab..6c740f346f91 100644
--- a/drivers/mtd/chips/cfi_cmdset_0020.c
+++ b/drivers/mtd/chips/cfi_cmdset_0020.c
@@ -42,8 +42,8 @@ static int cfi_staa_writev(struct mtd_info *mtd, const struct kvec *vecs,
 		unsigned long count, loff_t to, size_t *retlen);
 static int cfi_staa_erase_varsize(struct mtd_info *, struct erase_info *);
 static void cfi_staa_sync (struct mtd_info *);
-static int cfi_staa_lock(struct mtd_info *mtd, loff_t ofs, size_t len);
-static int cfi_staa_unlock(struct mtd_info *mtd, loff_t ofs, size_t len);
+static int cfi_staa_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len);
+static int cfi_staa_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len);
 static int cfi_staa_suspend (struct mtd_info *);
 static void cfi_staa_resume (struct mtd_info *);
 
@@ -221,8 +221,8 @@ static struct mtd_info *cfi_staa_setup(struct map_info *map)
 		}
 
 		for (i=0; i<mtd->numeraseregions;i++){
-			printk(KERN_DEBUG "%d: offset=0x%x,size=0x%x,blocks=%d\n",
-			       i,mtd->eraseregions[i].offset,
+			printk(KERN_DEBUG "%d: offset=0x%llx,size=0x%x,blocks=%d\n",
+			       i, (unsigned long long)mtd->eraseregions[i].offset,
 			       mtd->eraseregions[i].erasesize,
 			       mtd->eraseregions[i].numblocks);
 		}
@@ -964,7 +964,7 @@ static int cfi_staa_erase_varsize(struct mtd_info *mtd,
 		adr += regions[i].erasesize;
 		len -= regions[i].erasesize;
 
-		if (adr % (1<< cfi->chipshift) == ((regions[i].offset + (regions[i].erasesize * regions[i].numblocks)) %( 1<< cfi->chipshift)))
+		if (adr % (1<< cfi->chipshift) == (((unsigned long)regions[i].offset + (regions[i].erasesize * regions[i].numblocks)) %( 1<< cfi->chipshift)))
 			i++;
 
 		if (adr >> cfi->chipshift) {
@@ -1135,7 +1135,7 @@ retry:
 	spin_unlock_bh(chip->mutex);
 	return 0;
 }
-static int cfi_staa_lock(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int cfi_staa_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	struct map_info *map = mtd->priv;
 	struct cfi_private *cfi = map->fldrv_priv;
@@ -1284,7 +1284,7 @@ retry:
 	spin_unlock_bh(chip->mutex);
 	return 0;
 }
-static int cfi_staa_unlock(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int cfi_staa_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	struct map_info *map = mtd->priv;
 	struct cfi_private *cfi = map->fldrv_priv;
diff --git a/drivers/mtd/chips/fwh_lock.h b/drivers/mtd/chips/fwh_lock.h
index ab44f2b996f8..57e0e4e921f9 100644
--- a/drivers/mtd/chips/fwh_lock.h
+++ b/drivers/mtd/chips/fwh_lock.h
@@ -77,7 +77,7 @@ static int fwh_xxlock_oneblock(struct map_info *map, struct flchip *chip,
 }
 
 
-static int fwh_lock_varsize(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int fwh_lock_varsize(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	int ret;
 
@@ -88,7 +88,7 @@ static int fwh_lock_varsize(struct mtd_info *mtd, loff_t ofs, size_t len)
 }
 
 
-static int fwh_unlock_varsize(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int fwh_unlock_varsize(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	int ret;
 
diff --git a/drivers/mtd/inftlcore.c b/drivers/mtd/inftlcore.c
index 50ce13887f63..73f05227dc8c 100644
--- a/drivers/mtd/inftlcore.c
+++ b/drivers/mtd/inftlcore.c
@@ -50,7 +50,7 @@ static void inftl_add_mtd(struct mtd_blktrans_ops *tr, struct mtd_info *mtd)
 	struct INFTLrecord *inftl;
 	unsigned long temp;
 
-	if (mtd->type != MTD_NANDFLASH)
+	if (mtd->type != MTD_NANDFLASH || mtd->size > UINT_MAX)
 		return;
 	/* OK, this is moderately ugly.  But probably safe.  Alternatives? */
 	if (memcmp(mtd->name, "DiskOnChip", 10))
diff --git a/drivers/mtd/inftlmount.c b/drivers/mtd/inftlmount.c
index 9113628ed1ef..f751dd97c549 100644
--- a/drivers/mtd/inftlmount.c
+++ b/drivers/mtd/inftlmount.c
@@ -63,7 +63,7 @@ static int find_boot_record(struct INFTLrecord *inftl)
 	 * otherwise.
 	 */
 	inftl->EraseSize = inftl->mbd.mtd->erasesize;
-        inftl->nb_blocks = inftl->mbd.mtd->size / inftl->EraseSize;
+        inftl->nb_blocks = (u32)inftl->mbd.mtd->size / inftl->EraseSize;
 
 	inftl->MediaUnit = BLOCK_NIL;
 
@@ -187,7 +187,7 @@ static int find_boot_record(struct INFTLrecord *inftl)
 				mh->BlockMultiplierBits);
 			inftl->EraseSize = inftl->mbd.mtd->erasesize <<
 				mh->BlockMultiplierBits;
-			inftl->nb_blocks = inftl->mbd.mtd->size / inftl->EraseSize;
+			inftl->nb_blocks = (u32)inftl->mbd.mtd->size / inftl->EraseSize;
 			block >>= mh->BlockMultiplierBits;
 		}
 
diff --git a/drivers/mtd/maps/amd76xrom.c b/drivers/mtd/maps/amd76xrom.c
index d1eec7d3243f..237733d094c4 100644
--- a/drivers/mtd/maps/amd76xrom.c
+++ b/drivers/mtd/maps/amd76xrom.c
@@ -232,8 +232,8 @@ static int __devinit amd76xrom_init_one (struct pci_dev *pdev,
 		/* Trim the size if we are larger than the map */
 		if (map->mtd->size > map->map.size) {
 			printk(KERN_WARNING MOD_NAME
-				" rom(%u) larger than window(%lu). fixing...\n",
-				map->mtd->size, map->map.size);
+				" rom(%llu) larger than window(%lu). fixing...\n",
+				(unsigned long long)map->mtd->size, map->map.size);
 			map->mtd->size = map->map.size;
 		}
 		if (window->rsrc.parent) {
diff --git a/drivers/mtd/maps/ck804xrom.c b/drivers/mtd/maps/ck804xrom.c
index 1a6feb4474de..5f7a245ed132 100644
--- a/drivers/mtd/maps/ck804xrom.c
+++ b/drivers/mtd/maps/ck804xrom.c
@@ -263,8 +263,8 @@ static int __devinit ck804xrom_init_one (struct pci_dev *pdev,
 		/* Trim the size if we are larger than the map */
 		if (map->mtd->size > map->map.size) {
 			printk(KERN_WARNING MOD_NAME
-				" rom(%u) larger than window(%lu). fixing...\n",
-				map->mtd->size, map->map.size);
+				" rom(%llu) larger than window(%lu). fixing...\n",
+				(unsigned long long)map->mtd->size, map->map.size);
 			map->mtd->size = map->map.size;
 		}
 		if (window->rsrc.parent) {
diff --git a/drivers/mtd/maps/esb2rom.c b/drivers/mtd/maps/esb2rom.c
index bbbcdd4c8d13..11a2f57df9cf 100644
--- a/drivers/mtd/maps/esb2rom.c
+++ b/drivers/mtd/maps/esb2rom.c
@@ -324,8 +324,8 @@ static int __devinit esb2rom_init_one(struct pci_dev *pdev,
 		/* Trim the size if we are larger than the map */
 		if (map->mtd->size > map->map.size) {
 			printk(KERN_WARNING MOD_NAME
-				" rom(%u) larger than window(%lu). fixing...\n",
-				map->mtd->size, map->map.size);
+				" rom(%llu) larger than window(%lu). fixing...\n",
+				(unsigned long long)map->mtd->size, map->map.size);
 			map->mtd->size = map->map.size;
 		}
 		if (window->rsrc.parent) {
diff --git a/drivers/mtd/maps/ichxrom.c b/drivers/mtd/maps/ichxrom.c
index aeb6c916e23f..c32bc28920b3 100644
--- a/drivers/mtd/maps/ichxrom.c
+++ b/drivers/mtd/maps/ichxrom.c
@@ -258,8 +258,8 @@ static int __devinit ichxrom_init_one (struct pci_dev *pdev,
 		/* Trim the size if we are larger than the map */
 		if (map->mtd->size > map->map.size) {
 			printk(KERN_WARNING MOD_NAME
-				" rom(%u) larger than window(%lu). fixing...\n",
-				map->mtd->size, map->map.size);
+				" rom(%llu) larger than window(%lu). fixing...\n",
+				(unsigned long long)map->mtd->size, map->map.size);
 			map->mtd->size = map->map.size;
 		}
 		if (window->rsrc.parent) {
diff --git a/drivers/mtd/maps/nettel.c b/drivers/mtd/maps/nettel.c
index 965e6c6d6ab0..a8723a6b7457 100644
--- a/drivers/mtd/maps/nettel.c
+++ b/drivers/mtd/maps/nettel.c
@@ -226,7 +226,7 @@ static int __init nettel_init(void)
 
 	if ((amd_mtd = do_map_probe("jedec_probe", &nettel_amd_map))) {
 		printk(KERN_NOTICE "SNAPGEAR: AMD flash device size = %dK\n",
-			amd_mtd->size>>10);
+			(int)(amd_mtd->size>>10));
 
 		amd_mtd->owner = THIS_MODULE;
 
diff --git a/drivers/mtd/maps/scb2_flash.c b/drivers/mtd/maps/scb2_flash.c
index 21169e6d646c..7e329f09a548 100644
--- a/drivers/mtd/maps/scb2_flash.c
+++ b/drivers/mtd/maps/scb2_flash.c
@@ -118,7 +118,8 @@ scb2_fixup_mtd(struct mtd_info *mtd)
 		struct mtd_erase_region_info *region = &mtd->eraseregions[i];
 
 		if (region->numblocks * region->erasesize > mtd->size) {
-			region->numblocks = (mtd->size / region->erasesize);
+			region->numblocks = ((unsigned long)mtd->size /
+						region->erasesize);
 			done = 1;
 		} else {
 			region->numblocks = 0;
@@ -187,8 +188,9 @@ scb2_flash_probe(struct pci_dev *dev, const struct pci_device_id *ent)
 		return -ENODEV;
 	}
 
-	printk(KERN_NOTICE MODNAME ": chip size 0x%x at offset 0x%x\n",
-	       scb2_mtd->size, SCB2_WINDOW - scb2_mtd->size);
+	printk(KERN_NOTICE MODNAME ": chip size 0x%llx at offset 0x%llx\n",
+	       (unsigned long long)scb2_mtd->size,
+	       (unsigned long long)(SCB2_WINDOW - scb2_mtd->size));
 
 	add_mtd_device(scb2_mtd);
 
diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c
index bcffeda2df3d..e9ec59e9a566 100644
--- a/drivers/mtd/mtdchar.c
+++ b/drivers/mtd/mtdchar.c
@@ -450,16 +450,20 @@ static int mtd_ioctl(struct inode *inode, struct file *file,
 		if (!erase)
 			ret = -ENOMEM;
 		else {
+			struct erase_info_user einfo;
+
 			wait_queue_head_t waitq;
 			DECLARE_WAITQUEUE(wait, current);
 
 			init_waitqueue_head(&waitq);
 
-			if (copy_from_user(&erase->addr, argp,
+			if (copy_from_user(&einfo, argp,
 				    sizeof(struct erase_info_user))) {
 				kfree(erase);
 				return -EFAULT;
 			}
+			erase->addr = einfo.start;
+			erase->len = einfo.length;
 			erase->mtd = mtd;
 			erase->callback = mtdchar_erase_callback;
 			erase->priv = (unsigned long)&waitq;
diff --git a/drivers/mtd/mtdconcat.c b/drivers/mtd/mtdconcat.c
index 789842d0e6f2..ff8c14bcac68 100644
--- a/drivers/mtd/mtdconcat.c
+++ b/drivers/mtd/mtdconcat.c
@@ -197,7 +197,7 @@ concat_writev(struct mtd_info *mtd, const struct kvec *vecs,
 			continue;
 		}
 
-		size = min(total_len, (size_t)(subdev->size - to));
+		size = min_t(uint64_t, total_len, subdev->size - to);
 		wsize = size; /* store for future use */
 
 		entry_high = entry_low;
@@ -385,7 +385,7 @@ static int concat_erase(struct mtd_info *mtd, struct erase_info *instr)
 	struct mtd_concat *concat = CONCAT(mtd);
 	struct mtd_info *subdev;
 	int i, err;
-	u_int32_t length, offset = 0;
+	uint64_t length, offset = 0;
 	struct erase_info *erase;
 
 	if (!(mtd->flags & MTD_WRITEABLE))
@@ -518,7 +518,7 @@ static int concat_erase(struct mtd_info *mtd, struct erase_info *instr)
 	return 0;
 }
 
-static int concat_lock(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int concat_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	struct mtd_concat *concat = CONCAT(mtd);
 	int i, err = -EINVAL;
@@ -528,7 +528,7 @@ static int concat_lock(struct mtd_info *mtd, loff_t ofs, size_t len)
 
 	for (i = 0; i < concat->num_subdev; i++) {
 		struct mtd_info *subdev = concat->subdev[i];
-		size_t size;
+		uint64_t size;
 
 		if (ofs >= subdev->size) {
 			size = 0;
@@ -556,7 +556,7 @@ static int concat_lock(struct mtd_info *mtd, loff_t ofs, size_t len)
 	return err;
 }
 
-static int concat_unlock(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int concat_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	struct mtd_concat *concat = CONCAT(mtd);
 	int i, err = 0;
@@ -566,7 +566,7 @@ static int concat_unlock(struct mtd_info *mtd, loff_t ofs, size_t len)
 
 	for (i = 0; i < concat->num_subdev; i++) {
 		struct mtd_info *subdev = concat->subdev[i];
-		size_t size;
+		uint64_t size;
 
 		if (ofs >= subdev->size) {
 			size = 0;
@@ -842,12 +842,14 @@ struct mtd_info *mtd_concat_create(struct mtd_info *subdev[],	/* subdevices to c
 		concat->mtd.erasesize = curr_erasesize;
 		concat->mtd.numeraseregions = 0;
 	} else {
+		uint64_t tmp64;
+
 		/*
 		 * erase block size varies across the subdevices: allocate
 		 * space to store the data describing the variable erase regions
 		 */
 		struct mtd_erase_region_info *erase_region_p;
-		u_int32_t begin, position;
+		uint64_t begin, position;
 
 		concat->mtd.erasesize = max_erasesize;
 		concat->mtd.numeraseregions = num_erase_region;
@@ -879,8 +881,9 @@ struct mtd_info *mtd_concat_create(struct mtd_info *subdev[],	/* subdevices to c
 					erase_region_p->offset = begin;
 					erase_region_p->erasesize =
 					    curr_erasesize;
-					erase_region_p->numblocks =
-					    (position - begin) / curr_erasesize;
+					tmp64 = position - begin;
+					do_div(tmp64, curr_erasesize);
+					erase_region_p->numblocks = tmp64;
 					begin = position;
 
 					curr_erasesize = subdev[i]->erasesize;
@@ -897,9 +900,9 @@ struct mtd_info *mtd_concat_create(struct mtd_info *subdev[],	/* subdevices to c
 						erase_region_p->offset = begin;
 						erase_region_p->erasesize =
 						    curr_erasesize;
-						erase_region_p->numblocks =
-						    (position -
-						     begin) / curr_erasesize;
+						tmp64 = position - begin;
+						do_div(tmp64, curr_erasesize);
+						erase_region_p->numblocks = tmp64;
 						begin = position;
 
 						curr_erasesize =
@@ -909,14 +912,16 @@ struct mtd_info *mtd_concat_create(struct mtd_info *subdev[],	/* subdevices to c
 					}
 					position +=
 					    subdev[i]->eraseregions[j].
-					    numblocks * curr_erasesize;
+					    numblocks * (uint64_t)curr_erasesize;
 				}
 			}
 		}
 		/* Now write the final entry */
 		erase_region_p->offset = begin;
 		erase_region_p->erasesize = curr_erasesize;
-		erase_region_p->numblocks = (position - begin) / curr_erasesize;
+		tmp64 = position - begin;
+		do_div(tmp64, curr_erasesize);
+		erase_region_p->numblocks = tmp64;
 	}
 
 	return &concat->mtd;
diff --git a/drivers/mtd/mtdcore.c b/drivers/mtd/mtdcore.c
index a9d246949820..76fe0a1e7a5e 100644
--- a/drivers/mtd/mtdcore.c
+++ b/drivers/mtd/mtdcore.c
@@ -57,6 +57,19 @@ int add_mtd_device(struct mtd_info *mtd)
 			mtd->index = i;
 			mtd->usecount = 0;
 
+			if (is_power_of_2(mtd->erasesize))
+				mtd->erasesize_shift = ffs(mtd->erasesize) - 1;
+			else
+				mtd->erasesize_shift = 0;
+
+			if (is_power_of_2(mtd->writesize))
+				mtd->writesize_shift = ffs(mtd->writesize) - 1;
+			else
+				mtd->writesize_shift = 0;
+
+			mtd->erasesize_mask = (1 << mtd->erasesize_shift) - 1;
+			mtd->writesize_mask = (1 << mtd->writesize_shift) - 1;
+
 			/* Some chips always power up locked. Unlock them now */
 			if ((mtd->flags & MTD_WRITEABLE)
 			    && (mtd->flags & MTD_POWERUP_LOCK) && mtd->unlock) {
@@ -344,7 +357,8 @@ static inline int mtd_proc_info (char *buf, int i)
 	if (!this)
 		return 0;
 
-	return sprintf(buf, "mtd%d: %8.8x %8.8x \"%s\"\n", i, this->size,
+	return sprintf(buf, "mtd%d: %8.8llx %8.8x \"%s\"\n", i,
+		       (unsigned long long)this->size,
 		       this->erasesize, this->name);
 }
 
diff --git a/drivers/mtd/mtdoops.c b/drivers/mtd/mtdoops.c
index aebb3b27edbd..1a6b3beabe8d 100644
--- a/drivers/mtd/mtdoops.c
+++ b/drivers/mtd/mtdoops.c
@@ -80,9 +80,9 @@ static int mtdoops_erase_block(struct mtd_info *mtd, int offset)
 	if (ret) {
 		set_current_state(TASK_RUNNING);
 		remove_wait_queue(&wait_q, &wait);
-		printk (KERN_WARNING "mtdoops: erase of region [0x%x, 0x%x] "
+		printk (KERN_WARNING "mtdoops: erase of region [0x%llx, 0x%llx] "
 				     "on \"%s\" failed\n",
-			erase.addr, erase.len, mtd->name);
+			(unsigned long long)erase.addr, (unsigned long long)erase.len, mtd->name);
 		return ret;
 	}
 
@@ -289,7 +289,10 @@ static void mtdoops_notify_add(struct mtd_info *mtd)
 	}
 
 	cxt->mtd = mtd;
-	cxt->oops_pages = mtd->size / OOPS_PAGE_SIZE;
+	if (mtd->size > INT_MAX)
+		cxt->oops_pages = INT_MAX / OOPS_PAGE_SIZE;
+	else
+		cxt->oops_pages = (int)mtd->size / OOPS_PAGE_SIZE;
 
 	find_next_position(cxt);
 
diff --git a/drivers/mtd/mtdpart.c b/drivers/mtd/mtdpart.c
index 3728913fa5fa..144e6b613a77 100644
--- a/drivers/mtd/mtdpart.c
+++ b/drivers/mtd/mtdpart.c
@@ -26,7 +26,7 @@ static LIST_HEAD(mtd_partitions);
 struct mtd_part {
 	struct mtd_info mtd;
 	struct mtd_info *master;
-	u_int32_t offset;
+	uint64_t offset;
 	int index;
 	struct list_head list;
 	int registered;
@@ -235,7 +235,7 @@ void mtd_erase_callback(struct erase_info *instr)
 }
 EXPORT_SYMBOL_GPL(mtd_erase_callback);
 
-static int part_lock(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int part_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	struct mtd_part *part = PART(mtd);
 	if ((len + ofs) > mtd->size)
@@ -243,7 +243,7 @@ static int part_lock(struct mtd_info *mtd, loff_t ofs, size_t len)
 	return part->master->lock(part->master, ofs + part->offset, len);
 }
 
-static int part_unlock(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int part_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	struct mtd_part *part = PART(mtd);
 	if ((len + ofs) > mtd->size)
@@ -317,7 +317,7 @@ EXPORT_SYMBOL(del_mtd_partitions);
 
 static struct mtd_part *add_one_partition(struct mtd_info *master,
 		const struct mtd_partition *part, int partno,
-		u_int32_t cur_offset)
+		uint64_t cur_offset)
 {
 	struct mtd_part *slave;
 
@@ -395,19 +395,19 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 		slave->offset = cur_offset;
 	if (slave->offset == MTDPART_OFS_NXTBLK) {
 		slave->offset = cur_offset;
-		if ((cur_offset % master->erasesize) != 0) {
+		if (mtd_mod_by_eb(cur_offset, master) != 0) {
 			/* Round up to next erasesize */
-			slave->offset = ((cur_offset / master->erasesize) + 1) * master->erasesize;
+			slave->offset = (mtd_div_by_eb(cur_offset, master) + 1) * master->erasesize;
 			printk(KERN_NOTICE "Moving partition %d: "
-			       "0x%08x -> 0x%08x\n", partno,
-			       cur_offset, slave->offset);
+			       "0x%012llx -> 0x%012llx\n", partno,
+			       (unsigned long long)cur_offset, (unsigned long long)slave->offset);
 		}
 	}
 	if (slave->mtd.size == MTDPART_SIZ_FULL)
 		slave->mtd.size = master->size - slave->offset;
 
-	printk(KERN_NOTICE "0x%08x-0x%08x : \"%s\"\n", slave->offset,
-		slave->offset + slave->mtd.size, slave->mtd.name);
+	printk(KERN_NOTICE "0x%012llx-0x%012llx : \"%s\"\n", (unsigned long long)slave->offset,
+		(unsigned long long)(slave->offset + slave->mtd.size), slave->mtd.name);
 
 	/* let's do some sanity checks */
 	if (slave->offset >= master->size) {
@@ -420,13 +420,13 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 	}
 	if (slave->offset + slave->mtd.size > master->size) {
 		slave->mtd.size = master->size - slave->offset;
-		printk(KERN_WARNING"mtd: partition \"%s\" extends beyond the end of device \"%s\" -- size truncated to %#x\n",
-			part->name, master->name, slave->mtd.size);
+		printk(KERN_WARNING"mtd: partition \"%s\" extends beyond the end of device \"%s\" -- size truncated to %#llx\n",
+			part->name, master->name, (unsigned long long)slave->mtd.size);
 	}
 	if (master->numeraseregions > 1) {
 		/* Deal with variable erase size stuff */
 		int i, max = master->numeraseregions;
-		u32 end = slave->offset + slave->mtd.size;
+		u64 end = slave->offset + slave->mtd.size;
 		struct mtd_erase_region_info *regions = master->eraseregions;
 
 		/* Find the first erase regions which is part of this
@@ -449,7 +449,7 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 	}
 
 	if ((slave->mtd.flags & MTD_WRITEABLE) &&
-	    (slave->offset % slave->mtd.erasesize)) {
+	    mtd_mod_by_eb(slave->offset, &slave->mtd)) {
 		/* Doesn't start on a boundary of major erase size */
 		/* FIXME: Let it be writable if it is on a boundary of
 		 * _minor_ erase size though */
@@ -458,7 +458,7 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 			part->name);
 	}
 	if ((slave->mtd.flags & MTD_WRITEABLE) &&
-	    (slave->mtd.size % slave->mtd.erasesize)) {
+	    mtd_mod_by_eb(slave->mtd.size, &slave->mtd)) {
 		slave->mtd.flags &= ~MTD_WRITEABLE;
 		printk(KERN_WARNING"mtd: partition \"%s\" doesn't end on an erase block -- force read-only\n",
 			part->name);
@@ -466,7 +466,7 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 
 	slave->mtd.ecclayout = master->ecclayout;
 	if (master->block_isbad) {
-		uint32_t offs = 0;
+		uint64_t offs = 0;
 
 		while (offs < slave->mtd.size) {
 			if (master->block_isbad(master,
@@ -501,7 +501,7 @@ int add_mtd_partitions(struct mtd_info *master,
 		       int nbparts)
 {
 	struct mtd_part *slave;
-	u_int32_t cur_offset = 0;
+	uint64_t cur_offset = 0;
 	int i;
 
 	printk(KERN_NOTICE "Creating %d MTD partitions on \"%s\":\n", nbparts, master->name);
diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c
index 0a9c9cd33f96..ff2d33e4d6d6 100644
--- a/drivers/mtd/nand/nand_base.c
+++ b/drivers/mtd/nand/nand_base.c
@@ -2014,13 +2014,14 @@ static int nand_erase(struct mtd_info *mtd, struct erase_info *instr)
 int nand_erase_nand(struct mtd_info *mtd, struct erase_info *instr,
 		    int allowbbt)
 {
-	int page, len, status, pages_per_block, ret, chipnr;
+	int page, status, pages_per_block, ret, chipnr;
 	struct nand_chip *chip = mtd->priv;
-	int rewrite_bbt[NAND_MAX_CHIPS]={0};
+	loff_t rewrite_bbt[NAND_MAX_CHIPS]={0};
 	unsigned int bbt_masked_page = 0xffffffff;
+	loff_t len;
 
-	DEBUG(MTD_DEBUG_LEVEL3, "nand_erase: start = 0x%08x, len = %i\n",
-	      (unsigned int)instr->addr, (unsigned int)instr->len);
+	DEBUG(MTD_DEBUG_LEVEL3, "nand_erase: start = 0x%012llx, len = %llu\n",
+	      (unsigned long long)instr->addr, (unsigned long long)instr->len);
 
 	/* Start address must align on block boundary */
 	if (instr->addr & ((1 << chip->phys_erase_shift) - 1)) {
@@ -2116,7 +2117,8 @@ int nand_erase_nand(struct mtd_info *mtd, struct erase_info *instr,
 			DEBUG(MTD_DEBUG_LEVEL0, "nand_erase: "
 			      "Failed erase, page 0x%08x\n", page);
 			instr->state = MTD_ERASE_FAILED;
-			instr->fail_addr = (page << chip->page_shift);
+			instr->fail_addr =
+				((loff_t)page << chip->page_shift);
 			goto erase_exit;
 		}
 
@@ -2126,7 +2128,8 @@ int nand_erase_nand(struct mtd_info *mtd, struct erase_info *instr,
 		 */
 		if (bbt_masked_page != 0xffffffff &&
 		    (page & BBT_PAGE_MASK) == bbt_masked_page)
-			    rewrite_bbt[chipnr] = (page << chip->page_shift);
+			    rewrite_bbt[chipnr] =
+					((loff_t)page << chip->page_shift);
 
 		/* Increment page address and decrement length */
 		len -= (1 << chip->phys_erase_shift);
@@ -2173,7 +2176,7 @@ int nand_erase_nand(struct mtd_info *mtd, struct erase_info *instr,
 			continue;
 		/* update the BBT for chip */
 		DEBUG(MTD_DEBUG_LEVEL0, "nand_erase_nand: nand_update_bbt "
-		      "(%d:0x%0x 0x%0x)\n", chipnr, rewrite_bbt[chipnr],
+		      "(%d:0x%0llx 0x%0x)\n", chipnr, rewrite_bbt[chipnr],
 		      chip->bbt_td->pages[chipnr]);
 		nand_update_bbt(mtd, rewrite_bbt[chipnr]);
 	}
@@ -2365,7 +2368,7 @@ static struct nand_flash_dev *nand_get_flash_type(struct mtd_info *mtd,
 	if (!mtd->name)
 		mtd->name = type->name;
 
-	chip->chipsize = type->chipsize << 20;
+	chip->chipsize = (uint64_t)type->chipsize << 20;
 
 	/* Newer devices have all the information in additional id bytes */
 	if (!type->pagesize) {
@@ -2423,7 +2426,10 @@ static struct nand_flash_dev *nand_get_flash_type(struct mtd_info *mtd,
 
 	chip->bbt_erase_shift = chip->phys_erase_shift =
 		ffs(mtd->erasesize) - 1;
-	chip->chip_shift = ffs(chip->chipsize) - 1;
+	if (chip->chipsize & 0xffffffff)
+		chip->chip_shift = ffs((unsigned)chip->chipsize) - 1;
+	else
+		chip->chip_shift = ffs((unsigned)(chip->chipsize >> 32)) + 32 - 1;
 
 	/* Set the bad block position */
 	chip->badblockpos = mtd->writesize > 512 ?
diff --git a/drivers/mtd/nand/nand_bbt.c b/drivers/mtd/nand/nand_bbt.c
index 0b1c48595f12..55c23e5cd210 100644
--- a/drivers/mtd/nand/nand_bbt.c
+++ b/drivers/mtd/nand/nand_bbt.c
@@ -171,16 +171,16 @@ static int read_bbt(struct mtd_info *mtd, uint8_t *buf, int page, int num,
 				if (tmp == msk)
 					continue;
 				if (reserved_block_code && (tmp == reserved_block_code)) {
-					printk(KERN_DEBUG "nand_read_bbt: Reserved block at 0x%08x\n",
-					       ((offs << 2) + (act >> 1)) << this->bbt_erase_shift);
+					printk(KERN_DEBUG "nand_read_bbt: Reserved block at 0x%012llx\n",
+					       (loff_t)((offs << 2) + (act >> 1)) << this->bbt_erase_shift);
 					this->bbt[offs + (act >> 3)] |= 0x2 << (act & 0x06);
 					mtd->ecc_stats.bbtblocks++;
 					continue;
 				}
 				/* Leave it for now, if its matured we can move this
 				 * message to MTD_DEBUG_LEVEL0 */
-				printk(KERN_DEBUG "nand_read_bbt: Bad block at 0x%08x\n",
-				       ((offs << 2) + (act >> 1)) << this->bbt_erase_shift);
+				printk(KERN_DEBUG "nand_read_bbt: Bad block at 0x%012llx\n",
+				       (loff_t)((offs << 2) + (act >> 1)) << this->bbt_erase_shift);
 				/* Factory marked bad or worn out ? */
 				if (tmp == 0)
 					this->bbt[offs + (act >> 3)] |= 0x3 << (act & 0x06);
@@ -284,7 +284,7 @@ static int read_abs_bbts(struct mtd_info *mtd, uint8_t *buf,
 
 	/* Read the primary version, if available */
 	if (td->options & NAND_BBT_VERSION) {
-		scan_read_raw(mtd, buf, td->pages[0] << this->page_shift,
+		scan_read_raw(mtd, buf, (loff_t)td->pages[0] << this->page_shift,
 			      mtd->writesize);
 		td->version[0] = buf[mtd->writesize + td->veroffs];
 		printk(KERN_DEBUG "Bad block table at page %d, version 0x%02X\n",
@@ -293,7 +293,7 @@ static int read_abs_bbts(struct mtd_info *mtd, uint8_t *buf,
 
 	/* Read the mirror version, if available */
 	if (md && (md->options & NAND_BBT_VERSION)) {
-		scan_read_raw(mtd, buf, md->pages[0] << this->page_shift,
+		scan_read_raw(mtd, buf, (loff_t)md->pages[0] << this->page_shift,
 			      mtd->writesize);
 		md->version[0] = buf[mtd->writesize + md->veroffs];
 		printk(KERN_DEBUG "Bad block table at page %d, version 0x%02X\n",
@@ -411,7 +411,7 @@ static int create_bbt(struct mtd_info *mtd, uint8_t *buf,
 		numblocks = this->chipsize >> (this->bbt_erase_shift - 1);
 		startblock = chip * numblocks;
 		numblocks += startblock;
-		from = startblock << (this->bbt_erase_shift - 1);
+		from = (loff_t)startblock << (this->bbt_erase_shift - 1);
 	}
 
 	for (i = startblock; i < numblocks;) {
@@ -428,8 +428,8 @@ static int create_bbt(struct mtd_info *mtd, uint8_t *buf,
 
 		if (ret) {
 			this->bbt[i >> 3] |= 0x03 << (i & 0x6);
-			printk(KERN_WARNING "Bad eraseblock %d at 0x%08x\n",
-			       i >> 1, (unsigned int)from);
+			printk(KERN_WARNING "Bad eraseblock %d at 0x%012llx\n",
+			       i >> 1, (unsigned long long)from);
 			mtd->ecc_stats.badblocks++;
 		}
 
@@ -495,7 +495,7 @@ static int search_bbt(struct mtd_info *mtd, uint8_t *buf, struct nand_bbt_descr
 		for (block = 0; block < td->maxblocks; block++) {
 
 			int actblock = startblock + dir * block;
-			loff_t offs = actblock << this->bbt_erase_shift;
+			loff_t offs = (loff_t)actblock << this->bbt_erase_shift;
 
 			/* Read first page */
 			scan_read_raw(mtd, buf, offs, mtd->writesize);
@@ -719,7 +719,7 @@ static int write_bbt(struct mtd_info *mtd, uint8_t *buf,
 
 		memset(&einfo, 0, sizeof(einfo));
 		einfo.mtd = mtd;
-		einfo.addr = (unsigned long)to;
+		einfo.addr = to;
 		einfo.len = 1 << this->bbt_erase_shift;
 		res = nand_erase_nand(mtd, &einfo, 1);
 		if (res < 0)
@@ -729,8 +729,8 @@ static int write_bbt(struct mtd_info *mtd, uint8_t *buf,
 		if (res < 0)
 			goto outerr;
 
-		printk(KERN_DEBUG "Bad block table written to 0x%08x, version "
-		       "0x%02X\n", (unsigned int)to, td->version[chip]);
+		printk(KERN_DEBUG "Bad block table written to 0x%012llx, version "
+		       "0x%02X\n", (unsigned long long)to, td->version[chip]);
 
 		/* Mark it as used */
 		td->pages[chip] = page;
@@ -910,7 +910,7 @@ static void mark_bbt_region(struct mtd_info *mtd, struct nand_bbt_descr *td)
 			newval = oldval | (0x2 << (block & 0x06));
 			this->bbt[(block >> 3)] = newval;
 			if ((oldval != newval) && td->reserved_block_code)
-				nand_update_bbt(mtd, block << (this->bbt_erase_shift - 1));
+				nand_update_bbt(mtd, (loff_t)block << (this->bbt_erase_shift - 1));
 			continue;
 		}
 		update = 0;
@@ -931,7 +931,7 @@ static void mark_bbt_region(struct mtd_info *mtd, struct nand_bbt_descr *td)
 		   new ones have been marked, then we need to update the stored
 		   bbts.  This should only happen once. */
 		if (update && td->reserved_block_code)
-			nand_update_bbt(mtd, (block - 2) << (this->bbt_erase_shift - 1));
+			nand_update_bbt(mtd, (loff_t)(block - 2) << (this->bbt_erase_shift - 1));
 	}
 }
 
@@ -1027,7 +1027,6 @@ int nand_update_bbt(struct mtd_info *mtd, loff_t offs)
 	if (!this->bbt || !td)
 		return -EINVAL;
 
-	len = mtd->size >> (this->bbt_erase_shift + 2);
 	/* Allocate a temporary buffer for one eraseblock incl. oob */
 	len = (1 << this->bbt_erase_shift);
 	len += (len >> this->page_shift) * mtd->oobsize;
diff --git a/drivers/mtd/nftlcore.c b/drivers/mtd/nftlcore.c
index 320b929abe79..d1c4546513f7 100644
--- a/drivers/mtd/nftlcore.c
+++ b/drivers/mtd/nftlcore.c
@@ -39,7 +39,7 @@ static void nftl_add_mtd(struct mtd_blktrans_ops *tr, struct mtd_info *mtd)
 	struct NFTLrecord *nftl;
 	unsigned long temp;
 
-	if (mtd->type != MTD_NANDFLASH)
+	if (mtd->type != MTD_NANDFLASH || mtd->size > UINT_MAX)
 		return;
 	/* OK, this is moderately ugly.  But probably safe.  Alternatives? */
 	if (memcmp(mtd->name, "DiskOnChip", 10))
diff --git a/drivers/mtd/nftlmount.c b/drivers/mtd/nftlmount.c
index ccc4f209fbb5..8b22b1836e9f 100644
--- a/drivers/mtd/nftlmount.c
+++ b/drivers/mtd/nftlmount.c
@@ -51,7 +51,7 @@ static int find_boot_record(struct NFTLrecord *nftl)
 	   the mtd device accordingly.  We could even get rid of
 	   nftl->EraseSize if there were any point in doing so. */
 	nftl->EraseSize = nftl->mbd.mtd->erasesize;
-        nftl->nb_blocks = nftl->mbd.mtd->size / nftl->EraseSize;
+        nftl->nb_blocks = (u32)nftl->mbd.mtd->size / nftl->EraseSize;
 
 	nftl->MediaUnit = BLOCK_NIL;
 	nftl->SpareMediaUnit = BLOCK_NIL;
@@ -168,7 +168,7 @@ device is already correct.
 			printk(KERN_NOTICE "WARNING: Support for NFTL with UnitSizeFactor 0x%02x is experimental\n",
 			       mh->UnitSizeFactor);
 			nftl->EraseSize = nftl->mbd.mtd->erasesize << (0xff - mh->UnitSizeFactor);
-			nftl->nb_blocks = nftl->mbd.mtd->size / nftl->EraseSize;
+			nftl->nb_blocks = (u32)nftl->mbd.mtd->size / nftl->EraseSize;
 		}
 #endif
 		nftl->nb_boot_blocks = le16_to_cpu(mh->FirstPhysicalEUN);
diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
index 90ed319f26e6..529af271db17 100644
--- a/drivers/mtd/onenand/onenand_base.c
+++ b/drivers/mtd/onenand/onenand_base.c
@@ -1772,7 +1772,7 @@ static int onenand_erase(struct mtd_info *mtd, struct erase_info *instr)
 	int len;
 	int ret = 0;
 
-	DEBUG(MTD_DEBUG_LEVEL3, "onenand_erase: start = 0x%08x, len = %i\n", (unsigned int) instr->addr, (unsigned int) instr->len);
+	DEBUG(MTD_DEBUG_LEVEL3, "onenand_erase: start = 0x%012llx, len = %llu\n", (unsigned long long) instr->addr, (unsigned long long) instr->len);
 
 	block_size = (1 << this->erase_shift);
 
@@ -1810,7 +1810,7 @@ static int onenand_erase(struct mtd_info *mtd, struct erase_info *instr)
 
 		/* Check if we have a bad block, we do not erase bad blocks */
 		if (onenand_block_isbad_nolock(mtd, addr, 0)) {
-			printk (KERN_WARNING "onenand_erase: attempt to erase a bad block at addr 0x%08x\n", (unsigned int) addr);
+			printk (KERN_WARNING "onenand_erase: attempt to erase a bad block at addr 0x%012llx\n", (unsigned long long) addr);
 			instr->state = MTD_ERASE_FAILED;
 			goto erase_exit;
 		}
@@ -2029,7 +2029,7 @@ static int onenand_do_lock_cmd(struct mtd_info *mtd, loff_t ofs, size_t len, int
  *
  * Lock one or more blocks
  */
-static int onenand_lock(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int onenand_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	int ret;
 
@@ -2047,7 +2047,7 @@ static int onenand_lock(struct mtd_info *mtd, loff_t ofs, size_t len)
  *
  * Unlock one or more blocks
  */
-static int onenand_unlock(struct mtd_info *mtd, loff_t ofs, size_t len)
+static int onenand_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	int ret;
 
diff --git a/drivers/mtd/rfd_ftl.c b/drivers/mtd/rfd_ftl.c
index e538c0a72abb..490b4742ce3a 100644
--- a/drivers/mtd/rfd_ftl.c
+++ b/drivers/mtd/rfd_ftl.c
@@ -156,7 +156,7 @@ static int scan_header(struct partition *part)
 	size_t retlen;
 
 	sectors_per_block = part->block_size / SECTOR_SIZE;
-	part->total_blocks = part->mbd.mtd->size / part->block_size;
+	part->total_blocks = (u32)part->mbd.mtd->size / part->block_size;
 
 	if (part->total_blocks < 2)
 		return -ENOENT;
@@ -276,16 +276,17 @@ static void erase_callback(struct erase_info *erase)
 
 	part = (struct partition*)erase->priv;
 
-	i = erase->addr / part->block_size;
-	if (i >= part->total_blocks || part->blocks[i].offset != erase->addr) {
-		printk(KERN_ERR PREFIX "erase callback for unknown offset %x "
-				"on '%s'\n", erase->addr, part->mbd.mtd->name);
+	i = (u32)erase->addr / part->block_size;
+	if (i >= part->total_blocks || part->blocks[i].offset != erase->addr ||
+	    erase->addr > UINT_MAX) {
+		printk(KERN_ERR PREFIX "erase callback for unknown offset %llx "
+				"on '%s'\n", (unsigned long long)erase->addr, part->mbd.mtd->name);
 		return;
 	}
 
 	if (erase->state != MTD_ERASE_DONE) {
-		printk(KERN_WARNING PREFIX "erase failed at 0x%x on '%s', "
-				"state %d\n", erase->addr,
+		printk(KERN_WARNING PREFIX "erase failed at 0x%llx on '%s', "
+				"state %d\n", (unsigned long long)erase->addr,
 				part->mbd.mtd->name, erase->state);
 
 		part->blocks[i].state = BLOCK_FAILED;
@@ -345,9 +346,9 @@ static int erase_block(struct partition *part, int block)
 	rc = part->mbd.mtd->erase(part->mbd.mtd, erase);
 
 	if (rc) {
-		printk(KERN_ERR PREFIX "erase of region %x,%x on '%s' "
-				"failed\n", erase->addr, erase->len,
-				part->mbd.mtd->name);
+		printk(KERN_ERR PREFIX "erase of region %llx,%llx on '%s' "
+				"failed\n", (unsigned long long)erase->addr,
+				(unsigned long long)erase->len, part->mbd.mtd->name);
 		kfree(erase);
 	}
 
@@ -763,7 +764,7 @@ static void rfd_ftl_add_mtd(struct mtd_blktrans_ops *tr, struct mtd_info *mtd)
 {
 	struct partition *part;
 
-	if (mtd->type != MTD_NORFLASH)
+	if (mtd->type != MTD_NORFLASH || mtd->size > UINT_MAX)
 		return;
 
 	part = kzalloc(sizeof(struct partition), GFP_KERNEL);
diff --git a/drivers/mtd/ssfdc.c b/drivers/mtd/ssfdc.c
index 33a5d6ed6f18..3f67e00d98e0 100644
--- a/drivers/mtd/ssfdc.c
+++ b/drivers/mtd/ssfdc.c
@@ -294,7 +294,8 @@ static void ssfdcr_add_mtd(struct mtd_blktrans_ops *tr, struct mtd_info *mtd)
 	int cis_sector;
 
 	/* Check for small page NAND flash */
-	if (mtd->type != MTD_NANDFLASH || mtd->oobsize != OOB_SIZE)
+	if (mtd->type != MTD_NANDFLASH || mtd->oobsize != OOB_SIZE ||
+	    mtd->size > UINT_MAX)
 		return;
 
 	/* Check for SSDFC format by reading CIS/IDI sector */
@@ -316,7 +317,7 @@ static void ssfdcr_add_mtd(struct mtd_blktrans_ops *tr, struct mtd_info *mtd)
 
 	ssfdc->cis_block = cis_sector / (mtd->erasesize >> SECTOR_SHIFT);
 	ssfdc->erase_size = mtd->erasesize;
-	ssfdc->map_len = mtd->size / mtd->erasesize;
+	ssfdc->map_len = (u32)mtd->size / mtd->erasesize;
 
 	DEBUG(MTD_DEBUG_LEVEL1,
 		"SSFDC_RO: cis_block=%d,erase_size=%d,map_len=%d,n_zones=%d\n",
@@ -327,7 +328,7 @@ static void ssfdcr_add_mtd(struct mtd_blktrans_ops *tr, struct mtd_info *mtd)
 	ssfdc->heads = 16;
 	ssfdc->sectors = 32;
 	get_chs(mtd->size, NULL, &ssfdc->heads, &ssfdc->sectors);
-	ssfdc->cylinders = (unsigned short)((mtd->size >> SECTOR_SHIFT) /
+	ssfdc->cylinders = (unsigned short)(((u32)mtd->size >> SECTOR_SHIFT) /
 			((long)ssfdc->sectors * (long)ssfdc->heads));
 
 	DEBUG(MTD_DEBUG_LEVEL1, "SSFDC_RO: using C:%d H:%d S:%d == %ld sects\n",
diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index c7630a228310..634e2e86525f 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -561,7 +561,7 @@ static int io_init(struct ubi_device *ubi)
 	 */
 
 	ubi->peb_size   = ubi->mtd->erasesize;
-	ubi->peb_count  = ubi->mtd->size / ubi->mtd->erasesize;
+	ubi->peb_count  = mtd_div_by_eb(ubi->mtd->size, ubi->mtd);
 	ubi->flash_size = ubi->mtd->size;
 
 	if (ubi->mtd->block_isbad && ubi->mtd->block_markbad)
diff --git a/drivers/mtd/ubi/gluebi.c b/drivers/mtd/ubi/gluebi.c
index 605812bb0b1a..6dd4f5e77f82 100644
--- a/drivers/mtd/ubi/gluebi.c
+++ b/drivers/mtd/ubi/gluebi.c
@@ -215,7 +215,8 @@ static int gluebi_erase(struct mtd_info *mtd, struct erase_info *instr)
 	struct ubi_volume *vol;
 	struct ubi_device *ubi;
 
-	dbg_gen("erase %u bytes at offset %u", instr->len, instr->addr);
+	dbg_gen("erase %llu bytes at offset %llu", (unsigned long long)instr->len,
+		 (unsigned long long)instr->addr);
 
 	if (instr->addr < 0 || instr->addr > mtd->size - mtd->erasesize)
 		return -EINVAL;
@@ -223,11 +224,11 @@ static int gluebi_erase(struct mtd_info *mtd, struct erase_info *instr)
 	if (instr->len < 0 || instr->addr + instr->len > mtd->size)
 		return -EINVAL;
 
-	if (instr->addr % mtd->writesize || instr->len % mtd->writesize)
+	if (mtd_mod_by_ws(instr->addr, mtd) || mtd_mod_by_ws(instr->len, mtd))
 		return -EINVAL;
 
-	lnum = instr->addr / mtd->erasesize;
-	count = instr->len / mtd->erasesize;
+	lnum = mtd_div_by_eb(instr->addr, mtd);
+	count = mtd_div_by_eb(instr->len, mtd);
 
 	vol = container_of(mtd, struct ubi_volume, gluebi_mtd);
 	ubi = vol->ubi;
@@ -255,7 +256,7 @@ static int gluebi_erase(struct mtd_info *mtd, struct erase_info *instr)
 
 out_err:
 	instr->state = MTD_ERASE_FAILED;
-	instr->fail_addr = lnum * mtd->erasesize;
+	instr->fail_addr = (long long)lnum * mtd->erasesize;
 	return err;
 }
 
@@ -294,7 +295,7 @@ int ubi_create_gluebi(struct ubi_device *ubi, struct ubi_volume *vol)
 	 * bytes.
 	 */
 	if (vol->vol_type == UBI_DYNAMIC_VOLUME)
-		mtd->size = vol->usable_leb_size * vol->reserved_pebs;
+		mtd->size = (long long)vol->usable_leb_size * vol->reserved_pebs;
 	else
 		mtd->size = vol->used_bytes;
 
@@ -304,8 +305,8 @@ int ubi_create_gluebi(struct ubi_device *ubi, struct ubi_volume *vol)
 		return -ENFILE;
 	}
 
-	dbg_gen("added mtd%d (\"%s\"), size %u, EB size %u",
-		mtd->index, mtd->name, mtd->size, mtd->erasesize);
+	dbg_gen("added mtd%d (\"%s\"), size %llu, EB size %u",
+		mtd->index, mtd->name, (unsigned long long)mtd->size, mtd->erasesize);
 	return 0;
 }
 
diff --git a/fs/jffs2/erase.c b/fs/jffs2/erase.c
index 259461b910af..c32b4a1ad6cf 100644
--- a/fs/jffs2/erase.c
+++ b/fs/jffs2/erase.c
@@ -175,7 +175,7 @@ static void jffs2_erase_failed(struct jffs2_sb_info *c, struct jffs2_eraseblock
 {
 	/* For NAND, if the failure did not occur at the device level for a
 	   specific physical page, don't bother updating the bad block table. */
-	if (jffs2_cleanmarker_oob(c) && (bad_offset != MTD_FAIL_ADDR_UNKNOWN)) {
+	if (jffs2_cleanmarker_oob(c) && (bad_offset != (uint32_t)MTD_FAIL_ADDR_UNKNOWN)) {
 		/* We had a device-level failure to erase.  Let's see if we've
 		   failed too many times. */
 		if (!jffs2_write_nand_badblock(c, jeb, bad_offset)) {
@@ -209,7 +209,8 @@ static void jffs2_erase_callback(struct erase_info *instr)
 	struct erase_priv_struct *priv = (void *)instr->priv;
 
 	if(instr->state != MTD_ERASE_DONE) {
-		printk(KERN_WARNING "Erase at 0x%08x finished, but state != MTD_ERASE_DONE. State is 0x%x instead.\n", instr->addr, instr->state);
+		printk(KERN_WARNING "Erase at 0x%08llx finished, but state != MTD_ERASE_DONE. State is 0x%x instead.\n",
+			(unsigned long long)instr->addr, instr->state);
 		jffs2_erase_failed(priv->c, priv->jeb, instr->fail_addr);
 	} else {
 		jffs2_erase_succeeded(priv->c, priv->jeb);
diff --git a/include/linux/mtd/mtd.h b/include/linux/mtd/mtd.h
index eae26bb6430a..95e585ecc297 100644
--- a/include/linux/mtd/mtd.h
+++ b/include/linux/mtd/mtd.h
@@ -15,6 +15,8 @@
 #include <linux/mtd/compatmac.h>
 #include <mtd/mtd-abi.h>
 
+#include <asm/div64.h>
+
 #define MTD_CHAR_MAJOR 90
 #define MTD_BLOCK_MAJOR 31
 #define MAX_MTD_DEVICES 32
@@ -25,16 +27,16 @@
 #define MTD_ERASE_DONE          0x08
 #define MTD_ERASE_FAILED        0x10
 
-#define MTD_FAIL_ADDR_UNKNOWN 0xffffffff
+#define MTD_FAIL_ADDR_UNKNOWN -1LL
 
 /* If the erase fails, fail_addr might indicate exactly which block failed.  If
    fail_addr = MTD_FAIL_ADDR_UNKNOWN, the failure was not at the device level or was not
    specific to any particular block. */
 struct erase_info {
 	struct mtd_info *mtd;
-	u_int32_t addr;
-	u_int32_t len;
-	u_int32_t fail_addr;
+	uint64_t addr;
+	uint64_t len;
+	uint64_t fail_addr;
 	u_long time;
 	u_long retries;
 	u_int dev;
@@ -46,7 +48,7 @@ struct erase_info {
 };
 
 struct mtd_erase_region_info {
-	u_int32_t offset;			/* At which this region starts, from the beginning of the MTD */
+	uint64_t offset;			/* At which this region starts, from the beginning of the MTD */
 	u_int32_t erasesize;		/* For this region */
 	u_int32_t numblocks;		/* Number of blocks of erasesize in this region */
 	unsigned long *lockmap;		/* If keeping bitmap of locks */
@@ -101,7 +103,7 @@ struct mtd_oob_ops {
 struct mtd_info {
 	u_char type;
 	u_int32_t flags;
-	u_int32_t size;	 // Total size of the MTD
+	uint64_t size;	 // Total size of the MTD
 
 	/* "Major" erase size for the device. Naïve users may take this
 	 * to be the only erase size available, or may use the more detailed
@@ -120,6 +122,16 @@ struct mtd_info {
 	u_int32_t oobsize;   // Amount of OOB data per block (e.g. 16)
 	u_int32_t oobavail;  // Available OOB bytes per block
 
+	/*
+	 * If erasesize is a power of 2 then the shift is stored in
+	 * erasesize_shift otherwise erasesize_shift is zero. Ditto writesize.
+	 */
+	unsigned int erasesize_shift;
+	unsigned int writesize_shift;
+	/* Masks based on erasesize_shift and writesize_shift */
+	unsigned int erasesize_mask;
+	unsigned int writesize_mask;
+
 	// Kernel-only stuff starts here.
 	const char *name;
 	int index;
@@ -190,8 +202,8 @@ struct mtd_info {
 	void (*sync) (struct mtd_info *mtd);
 
 	/* Chip-supported device locking */
-	int (*lock) (struct mtd_info *mtd, loff_t ofs, size_t len);
-	int (*unlock) (struct mtd_info *mtd, loff_t ofs, size_t len);
+	int (*lock) (struct mtd_info *mtd, loff_t ofs, uint64_t len);
+	int (*unlock) (struct mtd_info *mtd, loff_t ofs, uint64_t len);
 
 	/* Power Management functions */
 	int (*suspend) (struct mtd_info *mtd);
@@ -221,6 +233,35 @@ struct mtd_info {
 	void (*put_device) (struct mtd_info *mtd);
 };
 
+static inline u_int32_t mtd_div_by_eb(uint64_t sz, struct mtd_info *mtd)
+{
+	if (mtd->erasesize_shift)
+		return sz >> mtd->erasesize_shift;
+	do_div(sz, mtd->erasesize);
+	return sz;
+}
+
+static inline u_int32_t mtd_mod_by_eb(uint64_t sz, struct mtd_info *mtd)
+{
+	if (mtd->erasesize_shift)
+		return sz & mtd->erasesize_mask;
+	return do_div(sz, mtd->erasesize);
+}
+
+static inline u_int32_t mtd_div_by_ws(uint64_t sz, struct mtd_info *mtd)
+{
+	if (mtd->writesize_shift)
+		return sz >> mtd->writesize_shift;
+	do_div(sz, mtd->writesize);
+	return sz;
+}
+
+static inline u_int32_t mtd_mod_by_ws(uint64_t sz, struct mtd_info *mtd)
+{
+	if (mtd->writesize_shift)
+		return sz & mtd->writesize_mask;
+	return do_div(sz, mtd->writesize);
+}
 
 	/* Kernel-side ioctl definitions */
 
diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h
index 733d3f3b4eb8..c0677b8082be 100644
--- a/include/linux/mtd/nand.h
+++ b/include/linux/mtd/nand.h
@@ -399,7 +399,7 @@ struct nand_chip {
 	int		bbt_erase_shift;
 	int		chip_shift;
 	int		numchips;
-	unsigned long	chipsize;
+	uint64_t	chipsize;
 	int		pagemask;
 	int		pagebuf;
 	int		subpagesize;
diff --git a/include/linux/mtd/partitions.h b/include/linux/mtd/partitions.h
index c92b4d439609..164c7d78687d 100644
--- a/include/linux/mtd/partitions.h
+++ b/include/linux/mtd/partitions.h
@@ -36,8 +36,8 @@
 
 struct mtd_partition {
 	char *name;			/* identifier string */
-	u_int32_t size;			/* partition size */
-	u_int32_t offset;		/* offset within the master MTD space */
+	uint64_t size;			/* partition size */
+	uint64_t offset;		/* offset within the master MTD space */
 	u_int32_t mask_flags;		/* master MTD flags to mask out for this partition */
 	struct nand_ecclayout *ecclayout;	/* out of band layout for this partition (NAND only)*/
 	struct mtd_info **mtdp;		/* pointer to store the MTD object */
-- 
cgit v1.2.3


From 3854be7712f7b4bdcaed14664fc7c7124b3fef0d Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Wed, 10 Dec 2008 14:06:42 +0000
Subject: [MTD] Remove strange u_int32_t types from FTL

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/ftl.c       | 100 ++++++++++++++++++++++++------------------------
 include/linux/mtd/ftl.h |  38 +++++++++---------
 2 files changed, 69 insertions(+), 69 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mtd/ftl.c b/drivers/mtd/ftl.c
index 9bf581c4f740..a790c062af1f 100644
--- a/drivers/mtd/ftl.c
+++ b/drivers/mtd/ftl.c
@@ -109,25 +109,25 @@ module_param(shuffle_freq, int, 0);
 /* Each memory region corresponds to a minor device */
 typedef struct partition_t {
     struct mtd_blktrans_dev mbd;
-    u_int32_t		state;
-    u_int32_t		*VirtualBlockMap;
-    u_int32_t		*VirtualPageMap;
-    u_int32_t		FreeTotal;
+    uint32_t		state;
+    uint32_t		*VirtualBlockMap;
+    uint32_t		*VirtualPageMap;
+    uint32_t		FreeTotal;
     struct eun_info_t {
-	u_int32_t		Offset;
-	u_int32_t		EraseCount;
-	u_int32_t		Free;
-	u_int32_t		Deleted;
+	uint32_t		Offset;
+	uint32_t		EraseCount;
+	uint32_t		Free;
+	uint32_t		Deleted;
     } *EUNInfo;
     struct xfer_info_t {
-	u_int32_t		Offset;
-	u_int32_t		EraseCount;
-	u_int16_t		state;
+	uint32_t		Offset;
+	uint32_t		EraseCount;
+	uint16_t		state;
     } *XferInfo;
-    u_int16_t		bam_index;
-    u_int32_t		*bam_cache;
-    u_int16_t		DataUnits;
-    u_int32_t		BlocksPerUnit;
+    uint16_t		bam_index;
+    uint32_t		*bam_cache;
+    uint16_t		DataUnits;
+    uint32_t		BlocksPerUnit;
     erase_unit_header_t	header;
 } partition_t;
 
@@ -199,8 +199,8 @@ static int scan_header(partition_t *part)
 static int build_maps(partition_t *part)
 {
     erase_unit_header_t header;
-    u_int16_t xvalid, xtrans, i;
-    u_int blocks, j;
+    uint16_t xvalid, xtrans, i;
+    unsigned blocks, j;
     int hdr_ok, ret = -1;
     ssize_t retval;
     loff_t offset;
@@ -269,14 +269,14 @@ static int build_maps(partition_t *part)
 
     /* Set up virtual page map */
     blocks = le32_to_cpu(header.FormattedSize) >> header.BlockSize;
-    part->VirtualBlockMap = vmalloc(blocks * sizeof(u_int32_t));
+    part->VirtualBlockMap = vmalloc(blocks * sizeof(uint32_t));
     if (!part->VirtualBlockMap)
 	    goto out_XferInfo;
 
-    memset(part->VirtualBlockMap, 0xff, blocks * sizeof(u_int32_t));
+    memset(part->VirtualBlockMap, 0xff, blocks * sizeof(uint32_t));
     part->BlocksPerUnit = (1 << header.EraseUnitSize) >> header.BlockSize;
 
-    part->bam_cache = kmalloc(part->BlocksPerUnit * sizeof(u_int32_t),
+    part->bam_cache = kmalloc(part->BlocksPerUnit * sizeof(uint32_t),
 			      GFP_KERNEL);
     if (!part->bam_cache)
 	    goto out_VirtualBlockMap;
@@ -290,7 +290,7 @@ static int build_maps(partition_t *part)
 	offset = part->EUNInfo[i].Offset + le32_to_cpu(header.BAMOffset);
 
 	ret = part->mbd.mtd->read(part->mbd.mtd, offset,
-			      part->BlocksPerUnit * sizeof(u_int32_t), &retval,
+			      part->BlocksPerUnit * sizeof(uint32_t), &retval,
 			      (unsigned char *)part->bam_cache);
 
 	if (ret)
@@ -332,7 +332,7 @@ out:
 ======================================================================*/
 
 static int erase_xfer(partition_t *part,
-		      u_int16_t xfernum)
+		      uint16_t xfernum)
 {
     int ret;
     struct xfer_info_t *xfer;
@@ -408,7 +408,7 @@ static int prepare_xfer(partition_t *part, int i)
     erase_unit_header_t header;
     struct xfer_info_t *xfer;
     int nbam, ret;
-    u_int32_t ctl;
+    uint32_t ctl;
     ssize_t retlen;
     loff_t offset;
 
@@ -430,15 +430,15 @@ static int prepare_xfer(partition_t *part, int i)
     }
 
     /* Write the BAM stub */
-    nbam = (part->BlocksPerUnit * sizeof(u_int32_t) +
+    nbam = (part->BlocksPerUnit * sizeof(uint32_t) +
 	    le32_to_cpu(part->header.BAMOffset) + SECTOR_SIZE - 1) / SECTOR_SIZE;
 
     offset = xfer->Offset + le32_to_cpu(part->header.BAMOffset);
     ctl = cpu_to_le32(BLOCK_CONTROL);
 
-    for (i = 0; i < nbam; i++, offset += sizeof(u_int32_t)) {
+    for (i = 0; i < nbam; i++, offset += sizeof(uint32_t)) {
 
-	ret = part->mbd.mtd->write(part->mbd.mtd, offset, sizeof(u_int32_t),
+	ret = part->mbd.mtd->write(part->mbd.mtd, offset, sizeof(uint32_t),
 			       &retlen, (u_char *)&ctl);
 
 	if (ret)
@@ -461,18 +461,18 @@ static int prepare_xfer(partition_t *part, int i)
 
 ======================================================================*/
 
-static int copy_erase_unit(partition_t *part, u_int16_t srcunit,
-			   u_int16_t xferunit)
+static int copy_erase_unit(partition_t *part, uint16_t srcunit,
+			   uint16_t xferunit)
 {
     u_char buf[SECTOR_SIZE];
     struct eun_info_t *eun;
     struct xfer_info_t *xfer;
-    u_int32_t src, dest, free, i;
-    u_int16_t unit;
+    uint32_t src, dest, free, i;
+    uint16_t unit;
     int ret;
     ssize_t retlen;
     loff_t offset;
-    u_int16_t srcunitswap = cpu_to_le16(srcunit);
+    uint16_t srcunitswap = cpu_to_le16(srcunit);
 
     eun = &part->EUNInfo[srcunit];
     xfer = &part->XferInfo[xferunit];
@@ -486,7 +486,7 @@ static int copy_erase_unit(partition_t *part, u_int16_t srcunit,
 	offset = eun->Offset + le32_to_cpu(part->header.BAMOffset);
 
 	ret = part->mbd.mtd->read(part->mbd.mtd, offset,
-			      part->BlocksPerUnit * sizeof(u_int32_t),
+			      part->BlocksPerUnit * sizeof(uint32_t),
 			      &retlen, (u_char *) (part->bam_cache));
 
 	/* mark the cache bad, in case we get an error later */
@@ -503,7 +503,7 @@ static int copy_erase_unit(partition_t *part, u_int16_t srcunit,
     offset = xfer->Offset + 20; /* Bad! */
     unit = cpu_to_le16(0x7fff);
 
-    ret = part->mbd.mtd->write(part->mbd.mtd, offset, sizeof(u_int16_t),
+    ret = part->mbd.mtd->write(part->mbd.mtd, offset, sizeof(uint16_t),
 			   &retlen, (u_char *) &unit);
 
     if (ret) {
@@ -560,7 +560,7 @@ static int copy_erase_unit(partition_t *part, u_int16_t srcunit,
 
 
     /* All clear? Then update the LogicalEUN again */
-    ret = part->mbd.mtd->write(part->mbd.mtd, xfer->Offset + 20, sizeof(u_int16_t),
+    ret = part->mbd.mtd->write(part->mbd.mtd, xfer->Offset + 20, sizeof(uint16_t),
 			   &retlen, (u_char *)&srcunitswap);
 
     if (ret) {
@@ -605,8 +605,8 @@ static int copy_erase_unit(partition_t *part, u_int16_t srcunit,
 
 static int reclaim_block(partition_t *part)
 {
-    u_int16_t i, eun, xfer;
-    u_int32_t best;
+    uint16_t i, eun, xfer;
+    uint32_t best;
     int queued, ret;
 
     DEBUG(0, "ftl_cs: reclaiming space...\n");
@@ -723,10 +723,10 @@ static void dump_lists(partition_t *part)
 }
 #endif
 
-static u_int32_t find_free(partition_t *part)
+static uint32_t find_free(partition_t *part)
 {
-    u_int16_t stop, eun;
-    u_int32_t blk;
+    uint16_t stop, eun;
+    uint32_t blk;
     size_t retlen;
     int ret;
 
@@ -749,7 +749,7 @@ static u_int32_t find_free(partition_t *part)
 
 	ret = part->mbd.mtd->read(part->mbd.mtd,
 		       part->EUNInfo[eun].Offset + le32_to_cpu(part->header.BAMOffset),
-		       part->BlocksPerUnit * sizeof(u_int32_t),
+		       part->BlocksPerUnit * sizeof(uint32_t),
 		       &retlen, (u_char *) (part->bam_cache));
 
 	if (ret) {
@@ -786,7 +786,7 @@ static u_int32_t find_free(partition_t *part)
 static int ftl_read(partition_t *part, caddr_t buffer,
 		    u_long sector, u_long nblocks)
 {
-    u_int32_t log_addr, bsize;
+    uint32_t log_addr, bsize;
     u_long i;
     int ret;
     size_t offset, retlen;
@@ -829,14 +829,14 @@ static int ftl_read(partition_t *part, caddr_t buffer,
 
 ======================================================================*/
 
-static int set_bam_entry(partition_t *part, u_int32_t log_addr,
-			 u_int32_t virt_addr)
+static int set_bam_entry(partition_t *part, uint32_t log_addr,
+			 uint32_t virt_addr)
 {
-    u_int32_t bsize, blk, le_virt_addr;
+    uint32_t bsize, blk, le_virt_addr;
 #ifdef PSYCHO_DEBUG
-    u_int32_t old_addr;
+    uint32_t old_addr;
 #endif
-    u_int16_t eun;
+    uint16_t eun;
     int ret;
     size_t retlen, offset;
 
@@ -845,11 +845,11 @@ static int set_bam_entry(partition_t *part, u_int32_t log_addr,
     bsize = 1 << part->header.EraseUnitSize;
     eun = log_addr / bsize;
     blk = (log_addr % bsize) / SECTOR_SIZE;
-    offset = (part->EUNInfo[eun].Offset + blk * sizeof(u_int32_t) +
+    offset = (part->EUNInfo[eun].Offset + blk * sizeof(uint32_t) +
 		  le32_to_cpu(part->header.BAMOffset));
 
 #ifdef PSYCHO_DEBUG
-    ret = part->mbd.mtd->read(part->mbd.mtd, offset, sizeof(u_int32_t),
+    ret = part->mbd.mtd->read(part->mbd.mtd, offset, sizeof(uint32_t),
                         &retlen, (u_char *)&old_addr);
     if (ret) {
 	printk(KERN_WARNING"ftl: Error reading old_addr in set_bam_entry: %d\n",ret);
@@ -886,7 +886,7 @@ static int set_bam_entry(partition_t *part, u_int32_t log_addr,
 #endif
 	part->bam_cache[blk] = le_virt_addr;
     }
-    ret = part->mbd.mtd->write(part->mbd.mtd, offset, sizeof(u_int32_t),
+    ret = part->mbd.mtd->write(part->mbd.mtd, offset, sizeof(uint32_t),
                             &retlen, (u_char *)&le_virt_addr);
 
     if (ret) {
@@ -900,7 +900,7 @@ static int set_bam_entry(partition_t *part, u_int32_t log_addr,
 static int ftl_write(partition_t *part, caddr_t buffer,
 		     u_long sector, u_long nblocks)
 {
-    u_int32_t bsize, log_addr, virt_addr, old_addr, blk;
+    uint32_t bsize, log_addr, virt_addr, old_addr, blk;
     u_long i;
     int ret;
     size_t retlen, offset;
diff --git a/include/linux/mtd/ftl.h b/include/linux/mtd/ftl.h
index 0be442f881dd..0555f7a0b9ed 100644
--- a/include/linux/mtd/ftl.h
+++ b/include/linux/mtd/ftl.h
@@ -32,25 +32,25 @@
 #define _LINUX_FTL_H
 
 typedef struct erase_unit_header_t {
-    u_int8_t	LinkTargetTuple[5];
-    u_int8_t	DataOrgTuple[10];
-    u_int8_t	NumTransferUnits;
-    u_int32_t	EraseCount;
-    u_int16_t	LogicalEUN;
-    u_int8_t	BlockSize;
-    u_int8_t	EraseUnitSize;
-    u_int16_t	FirstPhysicalEUN;
-    u_int16_t	NumEraseUnits;
-    u_int32_t	FormattedSize;
-    u_int32_t	FirstVMAddress;
-    u_int16_t	NumVMPages;
-    u_int8_t	Flags;
-    u_int8_t	Code;
-    u_int32_t	SerialNumber;
-    u_int32_t	AltEUHOffset;
-    u_int32_t	BAMOffset;
-    u_int8_t	Reserved[12];
-    u_int8_t	EndTuple[2];
+    uint8_t	LinkTargetTuple[5];
+    uint8_t	DataOrgTuple[10];
+    uint8_t	NumTransferUnits;
+    uint32_t	EraseCount;
+    uint16_t	LogicalEUN;
+    uint8_t	BlockSize;
+    uint8_t	EraseUnitSize;
+    uint16_t	FirstPhysicalEUN;
+    uint16_t	NumEraseUnits;
+    uint32_t	FormattedSize;
+    uint32_t	FirstVMAddress;
+    uint16_t	NumVMPages;
+    uint8_t	Flags;
+    uint8_t	Code;
+    uint32_t	SerialNumber;
+    uint32_t	AltEUHOffset;
+    uint32_t	BAMOffset;
+    uint8_t	Reserved[12];
+    uint8_t	EndTuple[2];
 } erase_unit_header_t;
 
 /* Flags in erase_unit_header_t */
-- 
cgit v1.2.3


From 26cdb67c74aedc22367e6d0271f7f955220cca65 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Wed, 10 Dec 2008 14:08:12 +0000
Subject: [MTD] Remove more strange u_intxx_t types

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdconcat.c        |  2 +-
 include/linux/mtd/mtd.h        | 26 +++++++++++++-------------
 include/linux/mtd/partitions.h |  2 +-
 3 files changed, 15 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mtd/mtdconcat.c b/drivers/mtd/mtdconcat.c
index ff8c14bcac68..c26dd528d094 100644
--- a/drivers/mtd/mtdconcat.c
+++ b/drivers/mtd/mtdconcat.c
@@ -696,7 +696,7 @@ struct mtd_info *mtd_concat_create(struct mtd_info *subdev[],	/* subdevices to c
 	int i;
 	size_t size;
 	struct mtd_concat *concat;
-	u_int32_t max_erasesize, curr_erasesize;
+	uint32_t max_erasesize, curr_erasesize;
 	int num_erase_region;
 
 	printk(KERN_NOTICE "Concatenating MTD devices:\n");
diff --git a/include/linux/mtd/mtd.h b/include/linux/mtd/mtd.h
index 95e585ecc297..adef674855f3 100644
--- a/include/linux/mtd/mtd.h
+++ b/include/linux/mtd/mtd.h
@@ -39,8 +39,8 @@ struct erase_info {
 	uint64_t fail_addr;
 	u_long time;
 	u_long retries;
-	u_int dev;
-	u_int cell;
+	unsigned dev;
+	unsigned cell;
 	void (*callback) (struct erase_info *self);
 	u_long priv;
 	u_char state;
@@ -49,8 +49,8 @@ struct erase_info {
 
 struct mtd_erase_region_info {
 	uint64_t offset;			/* At which this region starts, from the beginning of the MTD */
-	u_int32_t erasesize;		/* For this region */
-	u_int32_t numblocks;		/* Number of blocks of erasesize in this region */
+	uint32_t erasesize;		/* For this region */
+	uint32_t numblocks;		/* Number of blocks of erasesize in this region */
 	unsigned long *lockmap;		/* If keeping bitmap of locks */
 };
 
@@ -102,14 +102,14 @@ struct mtd_oob_ops {
 
 struct mtd_info {
 	u_char type;
-	u_int32_t flags;
+	uint32_t flags;
 	uint64_t size;	 // Total size of the MTD
 
 	/* "Major" erase size for the device. Naïve users may take this
 	 * to be the only erase size available, or may use the more detailed
 	 * information below if they desire
 	 */
-	u_int32_t erasesize;
+	uint32_t erasesize;
 	/* Minimal writable flash unit size. In case of NOR flash it is 1 (even
 	 * though individual bits can be cleared), in case of NAND flash it is
 	 * one NAND page (or half, or one-fourths of it), in case of ECC-ed NOR
@@ -117,10 +117,10 @@ struct mtd_info {
 	 * Any driver registering a struct mtd_info must ensure a writesize of
 	 * 1 or larger.
 	 */
-	u_int32_t writesize;
+	uint32_t writesize;
 
-	u_int32_t oobsize;   // Amount of OOB data per block (e.g. 16)
-	u_int32_t oobavail;  // Available OOB bytes per block
+	uint32_t oobsize;   // Amount of OOB data per block (e.g. 16)
+	uint32_t oobavail;  // Available OOB bytes per block
 
 	/*
 	 * If erasesize is a power of 2 then the shift is stored in
@@ -233,7 +233,7 @@ struct mtd_info {
 	void (*put_device) (struct mtd_info *mtd);
 };
 
-static inline u_int32_t mtd_div_by_eb(uint64_t sz, struct mtd_info *mtd)
+static inline uint32_t mtd_div_by_eb(uint64_t sz, struct mtd_info *mtd)
 {
 	if (mtd->erasesize_shift)
 		return sz >> mtd->erasesize_shift;
@@ -241,14 +241,14 @@ static inline u_int32_t mtd_div_by_eb(uint64_t sz, struct mtd_info *mtd)
 	return sz;
 }
 
-static inline u_int32_t mtd_mod_by_eb(uint64_t sz, struct mtd_info *mtd)
+static inline uint32_t mtd_mod_by_eb(uint64_t sz, struct mtd_info *mtd)
 {
 	if (mtd->erasesize_shift)
 		return sz & mtd->erasesize_mask;
 	return do_div(sz, mtd->erasesize);
 }
 
-static inline u_int32_t mtd_div_by_ws(uint64_t sz, struct mtd_info *mtd)
+static inline uint32_t mtd_div_by_ws(uint64_t sz, struct mtd_info *mtd)
 {
 	if (mtd->writesize_shift)
 		return sz >> mtd->writesize_shift;
@@ -256,7 +256,7 @@ static inline u_int32_t mtd_div_by_ws(uint64_t sz, struct mtd_info *mtd)
 	return sz;
 }
 
-static inline u_int32_t mtd_mod_by_ws(uint64_t sz, struct mtd_info *mtd)
+static inline uint32_t mtd_mod_by_ws(uint64_t sz, struct mtd_info *mtd)
 {
 	if (mtd->writesize_shift)
 		return sz & mtd->writesize_mask;
diff --git a/include/linux/mtd/partitions.h b/include/linux/mtd/partitions.h
index 164c7d78687d..a45dd831b3f8 100644
--- a/include/linux/mtd/partitions.h
+++ b/include/linux/mtd/partitions.h
@@ -38,7 +38,7 @@ struct mtd_partition {
 	char *name;			/* identifier string */
 	uint64_t size;			/* partition size */
 	uint64_t offset;		/* offset within the master MTD space */
-	u_int32_t mask_flags;		/* master MTD flags to mask out for this partition */
+	uint32_t mask_flags;		/* master MTD flags to mask out for this partition */
 	struct nand_ecclayout *ecclayout;	/* out of band layout for this partition (NAND only)*/
 	struct mtd_info **mtdp;		/* pointer to store the MTD object */
 };
-- 
cgit v1.2.3


From d3af0f048c114dd53713d5920c54f6d5b6b12139 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Mon, 1 Dec 2008 14:23:38 -0800
Subject: [MTD] [NAND] remove excess kernel-doc notation

Delete extra kernel-doc notation for struct fields and function
parameters that don't exist:

Warning(include/linux/mtd/nand.h:428): Excess struct/union/enum/typedef member 'wq' description in 'nand_chip'
Warning(include/linux/mtd/nand.h:428): Excess struct/union/enum/typedef member 'datbuf' description in 'nand_chip'
Warning(include/linux/mtd/nand.h:428): Excess struct/union/enum/typedef member 'oobbuf' description in 'nand_chip'
Warning(include/linux/mtd/nand.h:428): Excess struct/union/enum/typedef member 'oobdirty' description in 'nand_chip'
Warning(include/linux/mtd/nand.h:428): Excess struct/union/enum/typedef member 'data_poi' description in 'nand_chip'
Warning(drivers/mtd/nand/nand_base.c:2527): Excess function parameter 'maxchips' description in 'nand_scan_tail'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/nand_base.c | 1 -
 include/linux/mtd/nand.h     | 5 -----
 2 files changed, 6 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c
index ff2d33e4d6d6..0c3afccde8a2 100644
--- a/drivers/mtd/nand/nand_base.c
+++ b/drivers/mtd/nand/nand_base.c
@@ -2523,7 +2523,6 @@ int nand_scan_ident(struct mtd_info *mtd, int maxchips)
 /**
  * nand_scan_tail - [NAND Interface] Scan for the NAND device
  * @mtd:	    MTD device structure
- * @maxchips:	    Number of chips to scan for
  *
  * This is the second phase of the normal nand_scan() function. It
  * fills out all the uninitialized function pointers with the defaults
diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h
index c0677b8082be..db5b63da2a7e 100644
--- a/include/linux/mtd/nand.h
+++ b/include/linux/mtd/nand.h
@@ -335,17 +335,12 @@ struct nand_buffers {
  * @erase_cmd:		[INTERN] erase command write function, selectable due to AND support
  * @scan_bbt:		[REPLACEABLE] function to scan bad block table
  * @chip_delay:		[BOARDSPECIFIC] chip dependent delay for transfering data from array to read regs (tR)
- * @wq:			[INTERN] wait queue to sleep on if a NAND operation is in progress
  * @state:		[INTERN] the current state of the NAND device
  * @oob_poi:		poison value buffer
  * @page_shift:		[INTERN] number of address bits in a page (column address bits)
  * @phys_erase_shift:	[INTERN] number of address bits in a physical eraseblock
  * @bbt_erase_shift:	[INTERN] number of address bits in a bbt entry
  * @chip_shift:		[INTERN] number of address bits in one chip
- * @datbuf:		[INTERN] internal buffer for one page + oob
- * @oobbuf:		[INTERN] oob buffer for one eraseblock
- * @oobdirty:		[INTERN] indicates that oob_buf must be reinitialized
- * @data_poi:		[INTERN] pointer to a data buffer
  * @options:		[BOARDSPECIFIC] various chip options. They can partly be set to inform nand_scan about
  *			special functionality. See the defines for further explanation
  * @badblockpos:	[INTERN] position of the bad block marker in the oob area
-- 
cgit v1.2.3


From 5b37717a23b8e40f6cf7ad85a26ddcf41c171e2c Mon Sep 17 00:00:00 2001
From: Stefano Panella <stefano.panella@csr.com>
Date: Fri, 12 Dec 2008 13:00:06 +0000
Subject: uwb: improved MAS allocator and reservation conflict handling

Greatly enhance the MAS allocator:
  - Handle row and column reservations.
  - Permit all the available MAS to be allocated.
  - Follows the WiMedia rules on MAS selection.

Take appropriate action when reservation conflicts are detected.
  - Correctly identify which reservation wins the conflict.
  - Protect alien BP reservations.
  - If an owned reservation loses, resize/move it.
  - Follow the backoff procedure before requesting additional MAS.

When reservations are terminated, move the remaining reservations (if
necessary) so they keep following the MAS allocation rules.

Signed-off-by: Stefano Panella <stefano.panella@csr.com>
Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/usb/wusbcore/reservation.c |  13 +-
 drivers/uwb/Makefile               |   1 +
 drivers/uwb/allocator.c            | 386 +++++++++++++++++++++
 drivers/uwb/drp-avail.c            |   4 +-
 drivers/uwb/drp-ie.c               | 160 +++++++--
 drivers/uwb/drp.c                  | 681 ++++++++++++++++++++++++++++---------
 drivers/uwb/rsv.c                  | 482 ++++++++++++++++++++------
 drivers/uwb/uwb-debug.c            |  49 +--
 drivers/uwb/uwb-internal.h         |  80 ++++-
 include/linux/uwb.h                |  47 ++-
 include/linux/uwb/debug-cmd.h      |   2 +-
 include/linux/uwb/spec.h           |  25 ++
 12 files changed, 1603 insertions(+), 327 deletions(-)
 create mode 100644 drivers/uwb/allocator.c

(limited to 'include/linux')

diff --git a/drivers/usb/wusbcore/reservation.c b/drivers/usb/wusbcore/reservation.c
index 7b6525dac2f1..c37e4f83e54a 100644
--- a/drivers/usb/wusbcore/reservation.c
+++ b/drivers/usb/wusbcore/reservation.c
@@ -48,13 +48,15 @@ static void wusbhc_rsv_complete_cb(struct uwb_rsv *rsv)
 {
 	struct wusbhc *wusbhc = rsv->pal_priv;
 	struct device *dev = wusbhc->dev;
+	struct uwb_mas_bm mas;
 	char buf[72];
 
 	switch (rsv->state) {
 	case UWB_RSV_STATE_O_ESTABLISHED:
-		bitmap_scnprintf(buf, sizeof(buf), rsv->mas.bm, UWB_NUM_MAS);
+		uwb_rsv_get_usable_mas(rsv, &mas);
+		bitmap_scnprintf(buf, sizeof(buf), mas.bm, UWB_NUM_MAS);
 		dev_dbg(dev, "established reservation: %s\n", buf);
-		wusbhc_bwa_set(wusbhc, rsv->stream, &rsv->mas);
+		wusbhc_bwa_set(wusbhc, rsv->stream, &mas);
 		break;
 	case UWB_RSV_STATE_NONE:
 		dev_dbg(dev, "removed reservation\n");
@@ -85,13 +87,12 @@ int wusbhc_rsv_establish(struct wusbhc *wusbhc)
 	bcid.data[0] = wusbhc->cluster_id;
 	bcid.data[1] = 0;
 
-	rsv->owner = &rc->uwb_dev;
 	rsv->target.type = UWB_RSV_TARGET_DEVADDR;
 	rsv->target.devaddr = bcid;
 	rsv->type = UWB_DRP_TYPE_PRIVATE;
-	rsv->max_mas = 256;
-	rsv->min_mas = 16;  /* one MAS per zone? */
-	rsv->sparsity = 16; /* at least one MAS in each zone? */
+	rsv->max_mas = 256; /* try to get as much as possible */
+	rsv->min_mas = 15;  /* one MAS per zone */
+	rsv->max_interval = 1; /* max latency is one zone */
 	rsv->is_multicast = true;
 
 	ret = uwb_rsv_establish(rsv);
diff --git a/drivers/uwb/Makefile b/drivers/uwb/Makefile
index ce21a95da04a..2f98d080fe78 100644
--- a/drivers/uwb/Makefile
+++ b/drivers/uwb/Makefile
@@ -6,6 +6,7 @@ obj-$(CONFIG_UWB_I1480U)	+= i1480/
 
 uwb-objs :=		\
 	address.o	\
+	allocator.o	\
 	beacon.o	\
 	driver.o	\
 	drp.o		\
diff --git a/drivers/uwb/allocator.c b/drivers/uwb/allocator.c
new file mode 100644
index 000000000000..c8185e6b0cd5
--- /dev/null
+++ b/drivers/uwb/allocator.c
@@ -0,0 +1,386 @@
+/*
+ * UWB reservation management.
+ *
+ * Copyright (C) 2008 Cambridge Silicon Radio Ltd.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+#include <linux/version.h>
+#include <linux/kernel.h>
+#include <linux/uwb.h>
+
+#include "uwb-internal.h"
+
+static void uwb_rsv_fill_column_alloc(struct uwb_rsv_alloc_info *ai)
+{
+	int col, mas, safe_mas, unsafe_mas;
+	unsigned char *bm = ai->bm;
+	struct uwb_rsv_col_info *ci = ai->ci;
+	unsigned char c;
+
+	for (col = ci->csi.start_col; col < UWB_NUM_ZONES; col += ci->csi.interval) {
+    
+		safe_mas   = ci->csi.safe_mas_per_col;
+		unsafe_mas = ci->csi.unsafe_mas_per_col;
+    
+		for (mas = 0; mas < UWB_MAS_PER_ZONE; mas++ ) {
+			if (bm[col * UWB_MAS_PER_ZONE + mas] == 0) {
+	
+				if (safe_mas > 0) {
+					safe_mas--;
+					c = UWB_RSV_MAS_SAFE;
+				} else if (unsafe_mas > 0) {
+					unsafe_mas--;
+					c = UWB_RSV_MAS_UNSAFE;
+				} else {
+					break;
+				}
+				bm[col * UWB_MAS_PER_ZONE + mas] = c;
+			}
+		}
+	}
+}
+
+static void uwb_rsv_fill_row_alloc(struct uwb_rsv_alloc_info *ai)
+{
+	int mas, col, rows;
+	unsigned char *bm = ai->bm;
+	struct uwb_rsv_row_info *ri = &ai->ri;
+	unsigned char c;
+
+	rows = 1;
+	c = UWB_RSV_MAS_SAFE;
+	for (mas = UWB_MAS_PER_ZONE - 1; mas >= 0; mas--) {
+		if (ri->avail[mas] == 1) {
+      
+			if (rows > ri->used_rows) {
+				break;
+			} else if (rows > 7) {
+				c = UWB_RSV_MAS_UNSAFE;
+			}
+
+			for (col = 0; col < UWB_NUM_ZONES; col++) {
+				if (bm[col * UWB_NUM_ZONES + mas] != UWB_RSV_MAS_NOT_AVAIL) {
+					bm[col * UWB_NUM_ZONES + mas] = c;
+					if(c == UWB_RSV_MAS_SAFE)
+						ai->safe_allocated_mases++;
+					else
+						ai->unsafe_allocated_mases++;
+				}
+			}
+			rows++;
+		}
+	}
+	ai->total_allocated_mases = ai->safe_allocated_mases + ai->unsafe_allocated_mases;
+}
+
+/*
+ * Find the best column set for a given availability, interval, num safe mas and
+ * num unsafe mas.
+ *
+ * The different sets are tried in order as shown below, depending on the interval.
+ *
+ * interval = 16
+ *	deep = 0
+ *		set 1 ->  {  8 }
+ *	deep = 1
+ *		set 1 ->  {  4 }
+ *		set 2 ->  { 12 }
+ *	deep = 2
+ *		set 1 ->  {  2 }
+ *		set 2 ->  {  6 }
+ *		set 3 ->  { 10 }
+ *		set 4 ->  { 14 }
+ *	deep = 3
+ *		set 1 ->  {  1 }
+ *		set 2 ->  {  3 }
+ *		set 3 ->  {  5 }
+ *		set 4 ->  {  7 }
+ *		set 5 ->  {  9 }
+ *		set 6 ->  { 11 }
+ *		set 7 ->  { 13 }
+ *		set 8 ->  { 15 }
+ *
+ * interval = 8
+ *	deep = 0
+ *		set 1 ->  {  4  12 }
+ *	deep = 1
+ *		set 1 ->  {  2  10 }
+ *		set 2 ->  {  6  14 }
+ *	deep = 2
+ *		set 1 ->  {  1   9 }
+ *		set 2 ->  {  3  11 }
+ *		set 3 ->  {  5  13 }
+ *		set 4 ->  {  7  15 }
+ *
+ * interval = 4
+ *	deep = 0
+ *		set 1 ->  {  2   6  10  14 }
+ *	deep = 1
+ *		set 1 ->  {  1   5   9  13 }
+ *		set 2 ->  {  3   7  11  15 }
+ *
+ * interval = 2
+ *	deep = 0
+ *		set 1 ->  {  1   3   5   7   9  11  13  15 }
+ */
+static int uwb_rsv_find_best_column_set(struct uwb_rsv_alloc_info *ai, int interval, 
+					int num_safe_mas, int num_unsafe_mas)
+{
+	struct uwb_rsv_col_info *ci = ai->ci;
+	struct uwb_rsv_col_set_info *csi = &ci->csi;
+	struct uwb_rsv_col_set_info tmp_csi;
+	int deep, set, col, start_col_deep, col_start_set;
+	int start_col, max_mas_in_set, lowest_max_mas_in_deep;
+	int n_mas;
+	int found = UWB_RSV_ALLOC_NOT_FOUND; 
+
+	tmp_csi.start_col = 0;
+	start_col_deep = interval;
+	n_mas = num_unsafe_mas + num_safe_mas;
+
+	for (deep = 0; ((interval >> deep) & 0x1) == 0; deep++) {
+		start_col_deep /= 2;
+		col_start_set = 0;
+		lowest_max_mas_in_deep = UWB_MAS_PER_ZONE;
+
+		for (set = 1; set <= (1 << deep); set++) {
+			max_mas_in_set = 0;
+			start_col = start_col_deep + col_start_set;
+			for (col = start_col; col < UWB_NUM_ZONES; col += interval) {
+                
+				if (ci[col].max_avail_safe >= num_safe_mas &&
+				    ci[col].max_avail_unsafe >= n_mas) {
+					if (ci[col].highest_mas[n_mas] > max_mas_in_set)
+						max_mas_in_set = ci[col].highest_mas[n_mas];
+				} else {
+					max_mas_in_set = 0;
+					break;
+				}
+			}
+			if ((lowest_max_mas_in_deep > max_mas_in_set) && max_mas_in_set) {
+				lowest_max_mas_in_deep = max_mas_in_set;
+
+				tmp_csi.start_col = start_col;
+			}
+			col_start_set += (interval >> deep);
+		}
+
+		if (lowest_max_mas_in_deep < 8) {
+			csi->start_col = tmp_csi.start_col;
+			found = UWB_RSV_ALLOC_FOUND;
+			break;
+		} else if ((lowest_max_mas_in_deep > 8) && 
+			   (lowest_max_mas_in_deep != UWB_MAS_PER_ZONE) &&
+			   (found == UWB_RSV_ALLOC_NOT_FOUND)) {
+			csi->start_col = tmp_csi.start_col;
+			found = UWB_RSV_ALLOC_FOUND;
+		}
+	}
+
+	if (found == UWB_RSV_ALLOC_FOUND) {
+		csi->interval = interval;
+		csi->safe_mas_per_col = num_safe_mas;
+		csi->unsafe_mas_per_col = num_unsafe_mas;
+
+		ai->safe_allocated_mases = (UWB_NUM_ZONES / interval) * num_safe_mas;
+		ai->unsafe_allocated_mases = (UWB_NUM_ZONES / interval) * num_unsafe_mas;
+		ai->total_allocated_mases = ai->safe_allocated_mases + ai->unsafe_allocated_mases;
+		ai->interval = interval;		
+	}
+	return found;
+}
+
+static void get_row_descriptors(struct uwb_rsv_alloc_info *ai)
+{
+	unsigned char *bm = ai->bm;
+	struct uwb_rsv_row_info *ri = &ai->ri;
+	int col, mas;
+  
+	ri->free_rows = 16;
+	for (mas = 0; mas < UWB_MAS_PER_ZONE; mas ++) {
+		ri->avail[mas] = 1;
+		for (col = 1; col < UWB_NUM_ZONES; col++) {
+			if (bm[col * UWB_NUM_ZONES + mas] == UWB_RSV_MAS_NOT_AVAIL) {
+				ri->free_rows--;
+				ri->avail[mas]=0;
+				break;
+			}
+		}
+	}
+}
+
+static void uwb_rsv_fill_column_info(unsigned char *bm, int column, struct uwb_rsv_col_info *rci)
+{
+	int mas;
+	int block_count = 0, start_block = 0; 
+	int previous_avail = 0;
+	int available = 0;
+	int safe_mas_in_row[UWB_MAS_PER_ZONE] = {
+		8, 7, 6, 5, 4, 4, 4, 4, 4, 4, 4, 4, 4, 3, 2, 1,
+	};
+
+	rci->max_avail_safe = 0;
+
+	for (mas = 0; mas < UWB_MAS_PER_ZONE; mas ++) {
+		if (!bm[column * UWB_NUM_ZONES + mas]) {
+			available++;
+			rci->max_avail_unsafe = available;
+
+			rci->highest_mas[available] = mas;
+
+			if (previous_avail) {
+				block_count++;
+				if ((block_count > safe_mas_in_row[start_block]) &&
+				    (!rci->max_avail_safe))
+					rci->max_avail_safe = available - 1;
+			} else {
+				previous_avail = 1;
+				start_block = mas;
+				block_count = 1;
+			}
+		} else {
+			previous_avail = 0;
+		}
+	}
+	if (!rci->max_avail_safe)
+		rci->max_avail_safe = rci->max_avail_unsafe;
+}
+
+static void get_column_descriptors(struct uwb_rsv_alloc_info *ai)
+{
+	unsigned char *bm = ai->bm;
+	struct uwb_rsv_col_info *ci = ai->ci;
+	int col;
+
+	for (col = 1; col < UWB_NUM_ZONES; col++) {
+		uwb_rsv_fill_column_info(bm, col, &ci[col]);
+	}
+}
+
+static int uwb_rsv_find_best_row_alloc(struct uwb_rsv_alloc_info *ai)
+{
+	int n_rows;
+	int max_rows = ai->max_mas / UWB_USABLE_MAS_PER_ROW;
+	int min_rows = ai->min_mas / UWB_USABLE_MAS_PER_ROW;
+	if (ai->min_mas % UWB_USABLE_MAS_PER_ROW)
+		min_rows++;
+	for (n_rows = max_rows; n_rows >= min_rows; n_rows--) {
+		if (n_rows <= ai->ri.free_rows) {
+			ai->ri.used_rows = n_rows;
+			ai->interval = 1; /* row reservation */
+			uwb_rsv_fill_row_alloc(ai);
+			return UWB_RSV_ALLOC_FOUND;
+		}
+	}  
+	return UWB_RSV_ALLOC_NOT_FOUND;
+}
+
+static int uwb_rsv_find_best_col_alloc(struct uwb_rsv_alloc_info *ai, int interval)
+{
+	int n_safe, n_unsafe, n_mas;  
+	int n_column = UWB_NUM_ZONES / interval;
+	int max_per_zone = ai->max_mas / n_column;
+	int min_per_zone = ai->min_mas / n_column;
+
+	if (ai->min_mas % n_column)
+		min_per_zone++;
+
+	if (min_per_zone > UWB_MAS_PER_ZONE) {
+		return UWB_RSV_ALLOC_NOT_FOUND;
+	}
+    
+	if (max_per_zone > UWB_MAS_PER_ZONE) {
+		max_per_zone = UWB_MAS_PER_ZONE;
+	}
+    
+	for (n_mas = max_per_zone; n_mas >= min_per_zone; n_mas--) {
+		if (uwb_rsv_find_best_column_set(ai, interval, 0, n_mas) == UWB_RSV_ALLOC_NOT_FOUND)
+			continue;
+		for (n_safe = n_mas; n_safe >= 0; n_safe--) {
+			n_unsafe = n_mas - n_safe;
+			if (uwb_rsv_find_best_column_set(ai, interval, n_safe, n_unsafe) == UWB_RSV_ALLOC_FOUND) {
+				uwb_rsv_fill_column_alloc(ai);
+				return UWB_RSV_ALLOC_FOUND;
+			}
+		}
+	}
+	return UWB_RSV_ALLOC_NOT_FOUND;
+}
+
+int uwb_rsv_find_best_allocation(struct uwb_rsv *rsv, struct uwb_mas_bm *available, 
+				 struct uwb_mas_bm *result)
+{
+	struct uwb_rsv_alloc_info *ai;
+	int interval;
+	int bit_index;
+
+	ai = kzalloc(sizeof(struct uwb_rsv_alloc_info), GFP_KERNEL);
+	
+	ai->min_mas = rsv->min_mas;
+	ai->max_mas = rsv->max_mas;
+	ai->max_interval = rsv->max_interval;
+
+
+	/* fill the not available vector from the available bm */
+	for (bit_index = 0; bit_index < UWB_NUM_MAS; bit_index++) {
+		if (!test_bit(bit_index, available->bm))
+			ai->bm[bit_index] = UWB_RSV_MAS_NOT_AVAIL;
+	}
+
+	if (ai->max_interval == 1) {
+		get_row_descriptors(ai);
+		if (uwb_rsv_find_best_row_alloc(ai) == UWB_RSV_ALLOC_FOUND)
+			goto alloc_found;
+		else
+			goto alloc_not_found;
+	}
+
+	get_column_descriptors(ai);
+        
+	for (interval = 16; interval >= 2; interval>>=1) {
+		if (interval > ai->max_interval)
+			continue;
+		if (uwb_rsv_find_best_col_alloc(ai, interval) == UWB_RSV_ALLOC_FOUND)
+			goto alloc_found;
+	}
+
+	/* try row reservation if no column is found */
+	get_row_descriptors(ai);
+	if (uwb_rsv_find_best_row_alloc(ai) == UWB_RSV_ALLOC_FOUND)
+		goto alloc_found;
+	else
+		goto alloc_not_found;
+
+  alloc_found:
+	bitmap_zero(result->bm, UWB_NUM_MAS);
+	bitmap_zero(result->unsafe_bm, UWB_NUM_MAS);
+	/* fill the safe and unsafe bitmaps */
+	for (bit_index = 0; bit_index < UWB_NUM_MAS; bit_index++) {
+		if (ai->bm[bit_index] == UWB_RSV_MAS_SAFE)
+			set_bit(bit_index, result->bm);
+		else if (ai->bm[bit_index] == UWB_RSV_MAS_UNSAFE)
+			set_bit(bit_index, result->unsafe_bm);
+	}
+	bitmap_or(result->bm, result->bm, result->unsafe_bm, UWB_NUM_MAS);
+
+	result->safe   = ai->safe_allocated_mases;
+	result->unsafe = ai->unsafe_allocated_mases;
+	
+	kfree(ai);		
+	return UWB_RSV_ALLOC_FOUND;
+  
+  alloc_not_found:
+	kfree(ai);
+	return UWB_RSV_ALLOC_NOT_FOUND;
+}
diff --git a/drivers/uwb/drp-avail.c b/drivers/uwb/drp-avail.c
index 3febd8552808..40a540a5a72e 100644
--- a/drivers/uwb/drp-avail.c
+++ b/drivers/uwb/drp-avail.c
@@ -58,7 +58,7 @@ void uwb_drp_avail_init(struct uwb_rc *rc)
  *
  * avail = global & local & pending
  */
-static void uwb_drp_available(struct uwb_rc *rc, struct uwb_mas_bm *avail)
+void uwb_drp_available(struct uwb_rc *rc, struct uwb_mas_bm *avail)
 {
 	bitmap_and(avail->bm, rc->drp_avail.global, rc->drp_avail.local, UWB_NUM_MAS);
 	bitmap_and(avail->bm, avail->bm, rc->drp_avail.pending, UWB_NUM_MAS);
@@ -105,6 +105,7 @@ void uwb_drp_avail_release(struct uwb_rc *rc, struct uwb_mas_bm *mas)
 	bitmap_or(rc->drp_avail.local, rc->drp_avail.local, mas->bm, UWB_NUM_MAS);
 	bitmap_or(rc->drp_avail.pending, rc->drp_avail.pending, mas->bm, UWB_NUM_MAS);
 	rc->drp_avail.ie_valid = false;
+	uwb_rsv_handle_drp_avail_change(rc);
 }
 
 /**
@@ -280,6 +281,7 @@ int uwbd_evt_handle_rc_drp_avail(struct uwb_event *evt)
 	mutex_lock(&rc->rsvs_mutex);
 	bitmap_copy(rc->drp_avail.global, bmp, UWB_NUM_MAS);
 	rc->drp_avail.ie_valid = false;
+	uwb_rsv_handle_drp_avail_change(rc);
 	mutex_unlock(&rc->rsvs_mutex);
 
 	uwb_rsv_sched_update(rc);
diff --git a/drivers/uwb/drp-ie.c b/drivers/uwb/drp-ie.c
index 75491d47806b..2840d7bf9e67 100644
--- a/drivers/uwb/drp-ie.c
+++ b/drivers/uwb/drp-ie.c
@@ -22,6 +22,96 @@
 
 #include "uwb-internal.h"
 
+
+/*
+ * Return the reason code for a reservations's DRP IE.
+ */
+int uwb_rsv_reason_code(struct uwb_rsv *rsv)
+{
+	static const int reason_codes[] = {
+		[UWB_RSV_STATE_O_INITIATED]          = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_O_PENDING]            = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_O_MODIFIED]           = UWB_DRP_REASON_MODIFIED,
+		[UWB_RSV_STATE_O_ESTABLISHED]        = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_O_TO_BE_MOVED]        = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_O_MOVE_COMBINING]     = UWB_DRP_REASON_MODIFIED,
+		[UWB_RSV_STATE_O_MOVE_REDUCING]      = UWB_DRP_REASON_MODIFIED,
+		[UWB_RSV_STATE_O_MOVE_EXPANDING]     = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_T_ACCEPTED]           = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_T_CONFLICT]           = UWB_DRP_REASON_CONFLICT,
+		[UWB_RSV_STATE_T_PENDING]            = UWB_DRP_REASON_PENDING,
+		[UWB_RSV_STATE_T_DENIED]             = UWB_DRP_REASON_DENIED,
+		[UWB_RSV_STATE_T_RESIZED]            = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_T_EXPANDING_ACCEPTED] = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_T_EXPANDING_CONFLICT] = UWB_DRP_REASON_CONFLICT,
+		[UWB_RSV_STATE_T_EXPANDING_PENDING]  = UWB_DRP_REASON_PENDING,
+		[UWB_RSV_STATE_T_EXPANDING_DENIED]   = UWB_DRP_REASON_DENIED,
+	};
+
+	return reason_codes[rsv->state];
+}
+
+/*
+ * Return the reason code for a reservations's companion DRP IE .
+ */
+int uwb_rsv_companion_reason_code(struct uwb_rsv *rsv)
+{
+	static const int companion_reason_codes[] = {
+		[UWB_RSV_STATE_O_MOVE_EXPANDING]     = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_T_EXPANDING_ACCEPTED] = UWB_DRP_REASON_ACCEPTED,
+		[UWB_RSV_STATE_T_EXPANDING_CONFLICT] = UWB_DRP_REASON_CONFLICT,
+		[UWB_RSV_STATE_T_EXPANDING_PENDING]  = UWB_DRP_REASON_PENDING,
+		[UWB_RSV_STATE_T_EXPANDING_DENIED]   = UWB_DRP_REASON_DENIED,
+	};
+
+	return companion_reason_codes[rsv->state];
+}
+
+/*
+ * Return the status bit for a reservations's DRP IE.
+ */
+int uwb_rsv_status(struct uwb_rsv *rsv)
+{
+	static const int statuses[] = {
+		[UWB_RSV_STATE_O_INITIATED]          = 0,
+		[UWB_RSV_STATE_O_PENDING]            = 0,
+		[UWB_RSV_STATE_O_MODIFIED]           = 1,
+		[UWB_RSV_STATE_O_ESTABLISHED]        = 1,
+		[UWB_RSV_STATE_O_TO_BE_MOVED]        = 0,
+		[UWB_RSV_STATE_O_MOVE_COMBINING]     = 1,
+		[UWB_RSV_STATE_O_MOVE_REDUCING]      = 1,
+		[UWB_RSV_STATE_O_MOVE_EXPANDING]     = 1,
+		[UWB_RSV_STATE_T_ACCEPTED]           = 1,
+		[UWB_RSV_STATE_T_CONFLICT]           = 0,
+		[UWB_RSV_STATE_T_PENDING]            = 0,
+		[UWB_RSV_STATE_T_DENIED]             = 0,
+		[UWB_RSV_STATE_T_RESIZED]            = 1,
+		[UWB_RSV_STATE_T_EXPANDING_ACCEPTED] = 1,
+		[UWB_RSV_STATE_T_EXPANDING_CONFLICT] = 1,
+		[UWB_RSV_STATE_T_EXPANDING_PENDING]  = 1,
+		[UWB_RSV_STATE_T_EXPANDING_DENIED]   = 1,
+
+	};
+
+	return statuses[rsv->state];
+}
+
+/*
+ * Return the status bit for a reservations's companion DRP IE .
+ */
+int uwb_rsv_companion_status(struct uwb_rsv *rsv)
+{
+	static const int companion_statuses[] = {
+		[UWB_RSV_STATE_O_MOVE_EXPANDING]     = 0,
+		[UWB_RSV_STATE_T_EXPANDING_ACCEPTED] = 1,
+		[UWB_RSV_STATE_T_EXPANDING_CONFLICT] = 0,
+		[UWB_RSV_STATE_T_EXPANDING_PENDING]  = 0,
+		[UWB_RSV_STATE_T_EXPANDING_DENIED]   = 0,
+	};
+
+	return companion_statuses[rsv->state];
+}
+
 /*
  * Allocate a DRP IE.
  *
@@ -33,16 +123,12 @@
 static struct uwb_ie_drp *uwb_drp_ie_alloc(void)
 {
 	struct uwb_ie_drp *drp_ie;
-	unsigned tiebreaker;
 
 	drp_ie = kzalloc(sizeof(struct uwb_ie_drp) +
 			UWB_NUM_ZONES * sizeof(struct uwb_drp_alloc),
 			GFP_KERNEL);
 	if (drp_ie) {
 		drp_ie->hdr.element_id = UWB_IE_DRP;
-
-		get_random_bytes(&tiebreaker, sizeof(unsigned));
-		uwb_ie_drp_set_tiebreaker(drp_ie, tiebreaker & 1);
 	}
 	return drp_ie;
 }
@@ -103,43 +189,17 @@ static void uwb_drp_ie_from_bm(struct uwb_ie_drp *drp_ie,
  */
 int uwb_drp_ie_update(struct uwb_rsv *rsv)
 {
-	struct device *dev = &rsv->rc->uwb_dev.dev;
 	struct uwb_ie_drp *drp_ie;
-	int reason_code, status;
+	struct uwb_rsv_move *mv;
+	int unsafe;
 
-	switch (rsv->state) {
-	case UWB_RSV_STATE_NONE:
+	if (rsv->state == UWB_RSV_STATE_NONE) {
 		kfree(rsv->drp_ie);
 		rsv->drp_ie = NULL;
 		return 0;
-	case UWB_RSV_STATE_O_INITIATED:
-		reason_code = UWB_DRP_REASON_ACCEPTED;
-		status = 0;
-		break;
-	case UWB_RSV_STATE_O_PENDING:
-		reason_code = UWB_DRP_REASON_ACCEPTED;
-		status = 0;
-		break;
-	case UWB_RSV_STATE_O_MODIFIED:
-		reason_code = UWB_DRP_REASON_MODIFIED;
-		status = 1;
-		break;
-	case UWB_RSV_STATE_O_ESTABLISHED:
-		reason_code = UWB_DRP_REASON_ACCEPTED;
-		status = 1;
-		break;
-	case UWB_RSV_STATE_T_ACCEPTED:
-		reason_code = UWB_DRP_REASON_ACCEPTED;
-		status = 1;
-		break;
-	case UWB_RSV_STATE_T_DENIED:
-		reason_code = UWB_DRP_REASON_DENIED;
-		status = 0;
-		break;
-	default:
-		dev_dbg(dev, "rsv with unhandled state (%d)\n", rsv->state);
-		return -EINVAL;
 	}
+	
+	unsafe = rsv->mas.unsafe ? 1 : 0;
 
 	if (rsv->drp_ie == NULL) {
 		rsv->drp_ie = uwb_drp_ie_alloc();
@@ -148,9 +208,11 @@ int uwb_drp_ie_update(struct uwb_rsv *rsv)
 	}
 	drp_ie = rsv->drp_ie;
 
+	uwb_ie_drp_set_unsafe(drp_ie,       unsafe);
+	uwb_ie_drp_set_tiebreaker(drp_ie,   rsv->tiebreaker);
 	uwb_ie_drp_set_owner(drp_ie,        uwb_rsv_is_owner(rsv));
-	uwb_ie_drp_set_status(drp_ie,       status);
-	uwb_ie_drp_set_reason_code(drp_ie,  reason_code);
+	uwb_ie_drp_set_status(drp_ie,       uwb_rsv_status(rsv));
+	uwb_ie_drp_set_reason_code(drp_ie,  uwb_rsv_reason_code(rsv));
 	uwb_ie_drp_set_stream_index(drp_ie, rsv->stream);
 	uwb_ie_drp_set_type(drp_ie,         rsv->type);
 
@@ -168,6 +230,27 @@ int uwb_drp_ie_update(struct uwb_rsv *rsv)
 
 	uwb_drp_ie_from_bm(drp_ie, &rsv->mas);
 
+	if (uwb_rsv_has_two_drp_ies(rsv)) {
+		mv = &rsv->mv; 
+		if (mv->companion_drp_ie == NULL) {
+			mv->companion_drp_ie = uwb_drp_ie_alloc();
+			if (mv->companion_drp_ie == NULL)
+				return -ENOMEM;
+		}
+		drp_ie = mv->companion_drp_ie;
+		
+		/* keep all the same configuration of the main drp_ie */
+		memcpy(drp_ie, rsv->drp_ie, sizeof(struct uwb_ie_drp));
+		
+
+		/* FIXME: handle properly the unsafe bit */
+		uwb_ie_drp_set_unsafe(drp_ie,       1);
+		uwb_ie_drp_set_status(drp_ie,       uwb_rsv_companion_status(rsv));
+		uwb_ie_drp_set_reason_code(drp_ie,  uwb_rsv_companion_reason_code(rsv));
+	
+		uwb_drp_ie_from_bm(drp_ie, &mv->companion_mas);
+	}
+
 	rsv->ie_valid = true;
 	return 0;
 }
@@ -218,6 +301,8 @@ void uwb_drp_ie_to_bm(struct uwb_mas_bm *bm, const struct uwb_ie_drp *drp_ie)
 	u8 zone;
 	u16 zone_mask;
 
+	bitmap_zero(bm->bm, UWB_NUM_MAS);
+
 	for (cnt = 0; cnt < numallocs; cnt++) {
 		alloc = &drp_ie->allocs[cnt];
 		zone_bm = le16_to_cpu(alloc->zone_bm);
@@ -229,3 +314,4 @@ void uwb_drp_ie_to_bm(struct uwb_mas_bm *bm, const struct uwb_ie_drp *drp_ie)
 		}
 	}
 }
+
diff --git a/drivers/uwb/drp.c b/drivers/uwb/drp.c
index fe328146adb7..2b4f9406789d 100644
--- a/drivers/uwb/drp.c
+++ b/drivers/uwb/drp.c
@@ -23,6 +23,59 @@
 #include <linux/delay.h>
 #include "uwb-internal.h"
 
+
+/* DRP Conflict Actions ([ECMA-368 2nd Edition] 17.4.6) */
+enum uwb_drp_conflict_action {
+	/* Reservation is mantained, no action needed */
+	UWB_DRP_CONFLICT_MANTAIN = 0,
+	
+	/* the device shall not transmit frames in conflicting MASs in
+	 * the following superframe. If the device is the reservation
+	 * target, it shall also set the Reason Code in its DRP IE to
+	 * Conflict in its beacon in the following superframe.
+	 */
+	UWB_DRP_CONFLICT_ACT1,
+	
+	/* the device shall not set the Reservation Status bit to ONE
+	 * and shall not transmit frames in conflicting MASs. If the
+	 * device is the reservation target, it shall also set the
+	 * Reason Code in its DRP IE to Conflict.
+	 */	
+	UWB_DRP_CONFLICT_ACT2,
+
+	/* the device shall not transmit frames in conflicting MASs in
+	 * the following superframe. It shall remove the conflicting
+	 * MASs from the reservation or set the Reservation Status to
+	 * ZERO in its beacon in the following superframe. If the
+	 * device is the reservation target, it shall also set the
+	 * Reason Code in its DRP IE to Conflict.
+	 */
+	UWB_DRP_CONFLICT_ACT3,
+};
+
+
+static void uwb_rc_set_drp_cmd_done(struct uwb_rc *rc, void *arg,
+				    struct uwb_rceb *reply, ssize_t reply_size)
+{
+	struct uwb_rc_evt_set_drp_ie *r = (struct uwb_rc_evt_set_drp_ie *)reply;
+
+	if (r != NULL) {
+		if (r->bResultCode != UWB_RC_RES_SUCCESS)
+			dev_err(&rc->uwb_dev.dev, "SET-DRP-IE failed: %s (%d)\n",
+				uwb_rc_strerror(r->bResultCode), r->bResultCode);
+	} else
+		dev_err(&rc->uwb_dev.dev, "SET-DRP-IE: timeout\n");
+
+	spin_lock(&rc->rsvs_lock);
+	if (rc->set_drp_ie_pending > 1) {
+		rc->set_drp_ie_pending = 0;
+		uwb_rsv_queue_update(rc);	
+	} else {
+		rc->set_drp_ie_pending = 0;	
+	}
+	spin_unlock(&rc->rsvs_lock);
+}
+
 /**
  * Construct and send the SET DRP IE
  *
@@ -46,18 +99,23 @@
 int uwb_rc_send_all_drp_ie(struct uwb_rc *rc)
 {
 	int result;
-	struct device *dev = &rc->uwb_dev.dev;
 	struct uwb_rc_cmd_set_drp_ie *cmd;
-	struct uwb_rc_evt_set_drp_ie reply;
 	struct uwb_rsv *rsv;
+	struct uwb_rsv_move *mv;
 	int num_bytes = 0;
 	u8 *IEDataptr;
 
 	result = -ENOMEM;
 	/* First traverse all reservations to determine memory needed. */
 	list_for_each_entry(rsv, &rc->reservations, rc_node) {
-		if (rsv->drp_ie != NULL)
+		if (rsv->drp_ie != NULL) {
 			num_bytes += rsv->drp_ie->hdr.length + 2;
+			if (uwb_rsv_has_two_drp_ies(rsv) &&
+				(rsv->mv.companion_drp_ie != NULL)) {
+				mv = &rsv->mv;
+				num_bytes += mv->companion_drp_ie->hdr.length + 2;	
+			}
+		}
 	}
 	num_bytes += sizeof(rc->drp_avail.ie);
 	cmd = kzalloc(sizeof(*cmd) + num_bytes, GFP_KERNEL);
@@ -68,109 +126,322 @@ int uwb_rc_send_all_drp_ie(struct uwb_rc *rc)
 	cmd->wIELength = num_bytes;
 	IEDataptr = (u8 *)&cmd->IEData[0];
 
+	/* FIXME: DRV avail IE is not always needed */
+	/* put DRP avail IE first */
+	memcpy(IEDataptr, &rc->drp_avail.ie, sizeof(rc->drp_avail.ie));
+	IEDataptr += sizeof(struct uwb_ie_drp_avail);
+
 	/* Next traverse all reservations to place IEs in allocated memory. */
 	list_for_each_entry(rsv, &rc->reservations, rc_node) {
 		if (rsv->drp_ie != NULL) {
 			memcpy(IEDataptr, rsv->drp_ie,
 			       rsv->drp_ie->hdr.length + 2);
 			IEDataptr += rsv->drp_ie->hdr.length + 2;
+			
+			if (uwb_rsv_has_two_drp_ies(rsv) &&
+				(rsv->mv.companion_drp_ie != NULL)) {
+				mv = &rsv->mv;
+				memcpy(IEDataptr, mv->companion_drp_ie,
+				       mv->companion_drp_ie->hdr.length + 2);
+				IEDataptr += mv->companion_drp_ie->hdr.length + 2;	
+			}
 		}
 	}
-	memcpy(IEDataptr, &rc->drp_avail.ie, sizeof(rc->drp_avail.ie));
 
-	reply.rceb.bEventType = UWB_RC_CET_GENERAL;
-	reply.rceb.wEvent = UWB_RC_CMD_SET_DRP_IE;
-	result = uwb_rc_cmd(rc, "SET-DRP-IE", &cmd->rccb,
-			sizeof(*cmd) + num_bytes, &reply.rceb,
-			sizeof(reply));
-	if (result < 0)
-		goto error_cmd;
-	result = le16_to_cpu(reply.wRemainingSpace);
-	if (reply.bResultCode != UWB_RC_RES_SUCCESS) {
-		dev_err(&rc->uwb_dev.dev, "SET-DRP-IE: command execution "
-				"failed: %s (%d). RemainingSpace in beacon "
-				"= %d\n", uwb_rc_strerror(reply.bResultCode),
-				reply.bResultCode, result);
-		result = -EIO;
-	} else {
-		dev_dbg(dev, "SET-DRP-IE sent. RemainingSpace in beacon "
-			     "= %d.\n", result);
-		result = 0;
-	}
-error_cmd:
+	result = uwb_rc_cmd_async(rc, "SET-DRP-IE", &cmd->rccb, sizeof(*cmd) + num_bytes,
+				  UWB_RC_CET_GENERAL, UWB_RC_CMD_SET_DRP_IE,
+				  uwb_rc_set_drp_cmd_done, NULL);
+	
+	rc->set_drp_ie_pending = 1;
+
 	kfree(cmd);
 error:
 	return result;
 }
 
-void uwb_drp_handle_timeout(struct uwb_rsv *rsv)
+/*
+ * Evaluate the action to perform using conflict resolution rules
+ *
+ * Return a uwb_drp_conflict_action.
+ */
+static int evaluate_conflict_action(struct uwb_ie_drp *ext_drp_ie, int ext_beacon_slot,
+				    struct uwb_rsv *rsv, int our_status)
 {
-	struct device *dev = &rsv->rc->uwb_dev.dev;
+	int our_tie_breaker = rsv->tiebreaker;
+	int our_type        = rsv->type;
+	int our_beacon_slot = rsv->rc->uwb_dev.beacon_slot;
+
+	int ext_tie_breaker = uwb_ie_drp_tiebreaker(ext_drp_ie);
+	int ext_status      = uwb_ie_drp_status(ext_drp_ie);
+	int ext_type        = uwb_ie_drp_type(ext_drp_ie);
+	
+	
+	/* [ECMA-368 2nd Edition] 17.4.6 */
+	if (ext_type == UWB_DRP_TYPE_PCA && our_type == UWB_DRP_TYPE_PCA) {
+		return UWB_DRP_CONFLICT_MANTAIN;
+	}
+
+	/* [ECMA-368 2nd Edition] 17.4.6-1 */
+	if (our_type == UWB_DRP_TYPE_ALIEN_BP) {
+		return UWB_DRP_CONFLICT_MANTAIN;
+	}
+	
+	/* [ECMA-368 2nd Edition] 17.4.6-2 */
+	if (ext_type == UWB_DRP_TYPE_ALIEN_BP) {
+		/* here we know our_type != UWB_DRP_TYPE_ALIEN_BP */
+		return UWB_DRP_CONFLICT_ACT1;
+	}
+
+	/* [ECMA-368 2nd Edition] 17.4.6-3 */
+	if (our_status == 0 && ext_status == 1) {
+		return UWB_DRP_CONFLICT_ACT2;
+	}
 
-	dev_dbg(dev, "reservation timeout in state %s (%d)\n",
-		uwb_rsv_state_str(rsv->state), rsv->state);
+	/* [ECMA-368 2nd Edition] 17.4.6-4 */
+	if (our_status == 1 && ext_status == 0) {
+		return UWB_DRP_CONFLICT_MANTAIN;
+	}
 
-	switch (rsv->state) {
-	case UWB_RSV_STATE_O_INITIATED:
-		if (rsv->is_multicast) {
-			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
-			return;
+	/* [ECMA-368 2nd Edition] 17.4.6-5a */
+	if (our_tie_breaker == ext_tie_breaker &&
+	    our_beacon_slot <  ext_beacon_slot) {
+		return UWB_DRP_CONFLICT_MANTAIN;
+	}
+
+	/* [ECMA-368 2nd Edition] 17.4.6-5b */
+	if (our_tie_breaker != ext_tie_breaker &&
+	    our_beacon_slot >  ext_beacon_slot) {
+		return UWB_DRP_CONFLICT_MANTAIN;
+	}
+	
+	if (our_status == 0) {
+		if (our_tie_breaker == ext_tie_breaker) {
+			/* [ECMA-368 2nd Edition] 17.4.6-6a */
+			if (our_beacon_slot > ext_beacon_slot) {
+				return UWB_DRP_CONFLICT_ACT2;
+			}
+		} else  {
+			/* [ECMA-368 2nd Edition] 17.4.6-6b */
+			if (our_beacon_slot < ext_beacon_slot) {
+				return UWB_DRP_CONFLICT_ACT2;
+			}
 		}
-		break;
-	case UWB_RSV_STATE_O_ESTABLISHED:
-		if (rsv->is_multicast)
-			return;
-		break;
-	default:
-		break;
+	} else {
+		if (our_tie_breaker == ext_tie_breaker) {
+			/* [ECMA-368 2nd Edition] 17.4.6-7a */
+			if (our_beacon_slot > ext_beacon_slot) {
+				return UWB_DRP_CONFLICT_ACT3;
+			}
+		} else {
+			/* [ECMA-368 2nd Edition] 17.4.6-7b */
+			if (our_beacon_slot < ext_beacon_slot) {
+				return UWB_DRP_CONFLICT_ACT3;
+			}
+		}
+	}
+	return UWB_DRP_CONFLICT_MANTAIN;
+}
+
+static void handle_conflict_normal(struct uwb_ie_drp *drp_ie, 
+				   int ext_beacon_slot, 
+				   struct uwb_rsv *rsv, 
+				   struct uwb_mas_bm *conflicting_mas)
+{
+	struct uwb_rc *rc = rsv->rc;
+	struct uwb_rsv_move *mv = &rsv->mv;
+	struct uwb_drp_backoff_win *bow = &rc->bow;
+	int action;
+
+	action = evaluate_conflict_action(drp_ie, ext_beacon_slot, rsv, uwb_rsv_status(rsv));
+
+	if (uwb_rsv_is_owner(rsv)) {
+		switch(action) {
+		case UWB_DRP_CONFLICT_ACT2:
+			/* try move */
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_TO_BE_MOVED);
+			if (bow->can_reserve_extra_mases == false)
+				uwb_rsv_backoff_win_increment(rc);
+			
+			break;
+		case UWB_DRP_CONFLICT_ACT3:
+			uwb_rsv_backoff_win_increment(rc);
+			/* drop some mases with reason modified */
+			/* put in the companion the mases to be dropped */
+			bitmap_and(mv->companion_mas.bm, rsv->mas.bm, conflicting_mas->bm, UWB_NUM_MAS);
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MODIFIED);
+		default:
+			break;
+		}
+	} else {
+		switch(action) {
+		case UWB_DRP_CONFLICT_ACT2:
+		case UWB_DRP_CONFLICT_ACT3:
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_CONFLICT);	
+		default:
+			break;
+		}
+
+	}
+	
+}
+
+static void handle_conflict_expanding(struct uwb_ie_drp *drp_ie, int ext_beacon_slot,
+				      struct uwb_rsv *rsv, bool companion_only,
+				      struct uwb_mas_bm *conflicting_mas)
+{
+	struct uwb_rc *rc = rsv->rc;
+	struct uwb_drp_backoff_win *bow = &rc->bow;
+	struct uwb_rsv_move *mv = &rsv->mv;
+	int action;
+	
+	if (companion_only) {
+		/* status of companion is 0 at this point */
+		action = evaluate_conflict_action(drp_ie, ext_beacon_slot, rsv, 0);
+		if (uwb_rsv_is_owner(rsv)) {
+			switch(action) {
+			case UWB_DRP_CONFLICT_ACT2:
+			case UWB_DRP_CONFLICT_ACT3:
+				uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
+				rsv->needs_release_companion_mas = false;
+				if (bow->can_reserve_extra_mases == false)
+					uwb_rsv_backoff_win_increment(rc);
+				uwb_drp_avail_release(rsv->rc, &rsv->mv.companion_mas);
+			}
+		} else { /* rsv is target */			
+			switch(action) {
+			case UWB_DRP_CONFLICT_ACT2:
+			case UWB_DRP_CONFLICT_ACT3:
+				uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_EXPANDING_CONFLICT);
+                                /* send_drp_avail_ie = true; */
+			}
+		}
+	} else { /* also base part of the reservation is conflicting */		
+		if (uwb_rsv_is_owner(rsv)) {
+			uwb_rsv_backoff_win_increment(rc);
+			/* remove companion part */
+			uwb_drp_avail_release(rsv->rc, &rsv->mv.companion_mas);
+
+			/* drop some mases with reason modified */
+
+			/* put in the companion the mases to be dropped */
+			bitmap_andnot(mv->companion_mas.bm, rsv->mas.bm, conflicting_mas->bm, UWB_NUM_MAS);
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MODIFIED);
+		} else { /* it is a target rsv */
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_CONFLICT);
+                        /* send_drp_avail_ie = true; */
+		}
+	}
+}
+
+static void uwb_drp_handle_conflict_rsv(struct uwb_rc *rc, struct uwb_rsv *rsv,
+					struct uwb_rc_evt_drp *drp_evt, 
+					struct uwb_ie_drp *drp_ie,
+					struct uwb_mas_bm *conflicting_mas)
+{
+	struct uwb_rsv_move *mv;
+
+	/* check if the conflicting reservation has two drp_ies */
+	if (uwb_rsv_has_two_drp_ies(rsv)) {
+		mv = &rsv->mv;
+		if (bitmap_intersects(rsv->mas.bm, conflicting_mas->bm, UWB_NUM_MAS)) {
+			handle_conflict_expanding(drp_ie, drp_evt->beacon_slot_number,
+						  rsv, false, conflicting_mas);
+		} else {
+			if (bitmap_intersects(mv->companion_mas.bm, conflicting_mas->bm, UWB_NUM_MAS)) {
+				handle_conflict_expanding(drp_ie, drp_evt->beacon_slot_number,
+							  rsv, true, conflicting_mas);	
+			}
+		}
+	} else if (bitmap_intersects(rsv->mas.bm, conflicting_mas->bm, UWB_NUM_MAS)) {
+		handle_conflict_normal(drp_ie, drp_evt->beacon_slot_number, rsv, conflicting_mas);
 	}
-	uwb_rsv_remove(rsv);
 }
 
+static void uwb_drp_handle_all_conflict_rsv(struct uwb_rc *rc,
+					    struct uwb_rc_evt_drp *drp_evt, 
+					    struct uwb_ie_drp *drp_ie,
+					    struct uwb_mas_bm *conflicting_mas)
+{
+	struct uwb_rsv *rsv;
+	
+	list_for_each_entry(rsv, &rc->reservations, rc_node) {
+		uwb_drp_handle_conflict_rsv(rc, rsv, drp_evt, drp_ie, conflicting_mas);	
+	}
+}
+	
 /*
  * Based on the DRP IE, transition a target reservation to a new
  * state.
  */
 static void uwb_drp_process_target(struct uwb_rc *rc, struct uwb_rsv *rsv,
-				   struct uwb_ie_drp *drp_ie)
+				   struct uwb_ie_drp *drp_ie, struct uwb_rc_evt_drp *drp_evt)
 {
 	struct device *dev = &rc->uwb_dev.dev;
+	struct uwb_rsv_move *mv = &rsv->mv;
 	int status;
 	enum uwb_drp_reason reason_code;
-
+	struct uwb_mas_bm mas;
+	
 	status = uwb_ie_drp_status(drp_ie);
 	reason_code = uwb_ie_drp_reason_code(drp_ie);
+	uwb_drp_ie_to_bm(&mas, drp_ie);
 
-	if (status) {
-		switch (reason_code) {
-		case UWB_DRP_REASON_ACCEPTED:
-			uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_ACCEPTED);
-			break;
-		case UWB_DRP_REASON_MODIFIED:
-			dev_err(dev, "FIXME: unhandled reason code (%d/%d)\n",
-				reason_code, status);
+	switch (reason_code) {
+	case UWB_DRP_REASON_ACCEPTED:
+
+		if (rsv->state == UWB_RSV_STATE_T_CONFLICT) {
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_CONFLICT);
 			break;
-		default:
-			dev_warn(dev, "ignoring invalid DRP IE state (%d/%d)\n",
-				 reason_code, status);
 		}
-	} else {
-		switch (reason_code) {
-		case UWB_DRP_REASON_ACCEPTED:
-			/* New reservations are handled in uwb_rsv_find(). */
-			break;
-		case UWB_DRP_REASON_DENIED:
-			uwb_rsv_set_state(rsv, UWB_RSV_STATE_NONE);
-			break;
-		case UWB_DRP_REASON_CONFLICT:
-		case UWB_DRP_REASON_MODIFIED:
-			dev_err(dev, "FIXME: unhandled reason code (%d/%d)\n",
-				reason_code, status);
+
+		if (rsv->state == UWB_RSV_STATE_T_EXPANDING_ACCEPTED) {
+			/* drp_ie is companion */
+			if (!bitmap_equal(rsv->mas.bm, mas.bm, UWB_NUM_MAS))
+				/* stroke companion */
+				uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_EXPANDING_ACCEPTED);	
+		} else {
+			if (!bitmap_equal(rsv->mas.bm, mas.bm, UWB_NUM_MAS)) {
+				if (uwb_drp_avail_reserve_pending(rc, &mas) == -EBUSY) {
+					/* FIXME: there is a conflict, find
+					 * the conflicting reservations and
+					 * take a sensible action. Consider
+					 * that in drp_ie there is the
+					 * "neighbour" */
+					uwb_drp_handle_all_conflict_rsv(rc, drp_evt, drp_ie, &mas);
+				} else {
+					/* accept the extra reservation */
+					bitmap_copy(mv->companion_mas.bm, mas.bm, UWB_NUM_MAS);
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_EXPANDING_ACCEPTED);
+				}
+			} else {
+				if (status) {
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_ACCEPTED);
+				}
+			}
+			
+		}
+		break;
+
+	case UWB_DRP_REASON_MODIFIED:
+		/* check to see if we have already modified the reservation */
+		if (bitmap_equal(rsv->mas.bm, mas.bm, UWB_NUM_MAS)) {
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_ACCEPTED);
 			break;
-		default:
-			dev_warn(dev, "ignoring invalid DRP IE state (%d/%d)\n",
-				 reason_code, status);
 		}
+
+		/* find if the owner wants to expand or reduce */
+		if (bitmap_subset(mas.bm, rsv->mas.bm, UWB_NUM_MAS)) {
+			/* owner is reducing */
+			bitmap_andnot(mv->companion_mas.bm, rsv->mas.bm, mas.bm, UWB_NUM_MAS);
+			uwb_drp_avail_release(rsv->rc, &mv->companion_mas);
+		}
+
+		bitmap_copy(rsv->mas.bm, mas.bm, UWB_NUM_MAS);
+		uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_RESIZED);
+		break;
+	default:
+		dev_warn(dev, "ignoring invalid DRP IE state (%d/%d)\n",
+			 reason_code, status);
 	}
 }
 
@@ -179,23 +450,60 @@ static void uwb_drp_process_target(struct uwb_rc *rc, struct uwb_rsv *rsv,
  * state.
  */
 static void uwb_drp_process_owner(struct uwb_rc *rc, struct uwb_rsv *rsv,
-				  struct uwb_ie_drp *drp_ie)
+				  struct uwb_dev *src, struct uwb_ie_drp *drp_ie,
+				  struct uwb_rc_evt_drp *drp_evt)
 {
 	struct device *dev = &rc->uwb_dev.dev;
+	struct uwb_rsv_move *mv = &rsv->mv;
 	int status;
 	enum uwb_drp_reason reason_code;
+	struct uwb_mas_bm mas;
 
 	status = uwb_ie_drp_status(drp_ie);
 	reason_code = uwb_ie_drp_reason_code(drp_ie);
+	uwb_drp_ie_to_bm(&mas, drp_ie);
 
 	if (status) {
 		switch (reason_code) {
 		case UWB_DRP_REASON_ACCEPTED:
-			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
-			break;
-		case UWB_DRP_REASON_MODIFIED:
-			dev_err(dev, "FIXME: unhandled reason code (%d/%d)\n",
-				reason_code, status);
+			switch (rsv->state) {
+			case UWB_RSV_STATE_O_PENDING:
+			case UWB_RSV_STATE_O_INITIATED:
+			case UWB_RSV_STATE_O_ESTABLISHED:
+				uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
+				break;
+			case UWB_RSV_STATE_O_MODIFIED:
+				if (bitmap_equal(mas.bm, rsv->mas.bm, UWB_NUM_MAS)) {
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
+				} else {
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MODIFIED);	
+				}
+				break;
+				
+			case UWB_RSV_STATE_O_MOVE_REDUCING: /* shouldn' t be a problem */
+				if (bitmap_equal(mas.bm, rsv->mas.bm, UWB_NUM_MAS)) {
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
+				} else {
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_REDUCING);	
+				}
+				break;
+			case UWB_RSV_STATE_O_MOVE_EXPANDING:
+				if (bitmap_equal(mas.bm, mv->companion_mas.bm, UWB_NUM_MAS)) {
+					/* Companion reservation accepted */
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_COMBINING);
+				} else {
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_EXPANDING);
+				}
+				break;
+			case UWB_RSV_STATE_O_MOVE_COMBINING:
+				if (bitmap_equal(mas.bm, rsv->mas.bm, UWB_NUM_MAS))
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_REDUCING);
+				else
+					uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_COMBINING);
+				break;
+			default:
+				break;	
+			}
 			break;
 		default:
 			dev_warn(dev, "ignoring invalid DRP IE state (%d/%d)\n",
@@ -210,9 +518,10 @@ static void uwb_drp_process_owner(struct uwb_rc *rc, struct uwb_rsv *rsv,
 			uwb_rsv_set_state(rsv, UWB_RSV_STATE_NONE);
 			break;
 		case UWB_DRP_REASON_CONFLICT:
-		case UWB_DRP_REASON_MODIFIED:
-			dev_err(dev, "FIXME: unhandled reason code (%d/%d)\n",
-				reason_code, status);
+			/* resolve the conflict */
+			bitmap_complement(mas.bm, src->last_availability_bm,
+					  UWB_NUM_MAS);
+			uwb_drp_handle_conflict_rsv(rc, rsv, drp_evt, drp_ie, &mas);
 			break;
 		default:
 			dev_warn(dev, "ignoring invalid DRP IE state (%d/%d)\n",
@@ -221,12 +530,110 @@ static void uwb_drp_process_owner(struct uwb_rc *rc, struct uwb_rsv *rsv,
 	}
 }
 
+static void uwb_cnflt_alien_stroke_timer(struct uwb_cnflt_alien *cnflt)
+{
+	unsigned timeout_us = UWB_MAX_LOST_BEACONS * UWB_SUPERFRAME_LENGTH_US;
+	mod_timer(&cnflt->timer, jiffies + usecs_to_jiffies(timeout_us));
+}
+
+static void uwb_cnflt_update_work(struct work_struct *work)
+{
+	struct uwb_cnflt_alien *cnflt = container_of(work,
+						     struct uwb_cnflt_alien,
+						     cnflt_update_work);
+	struct uwb_cnflt_alien *c;
+	struct uwb_rc *rc = cnflt->rc;
+	
+	unsigned long delay_us = UWB_MAS_LENGTH_US * UWB_MAS_PER_ZONE;
+	
+	mutex_lock(&rc->rsvs_mutex);
+
+	list_del(&cnflt->rc_node);
+
+	/* update rc global conflicting alien bitmap */
+	bitmap_zero(rc->cnflt_alien_bitmap.bm, UWB_NUM_MAS);
+
+	list_for_each_entry(c, &rc->cnflt_alien_list, rc_node) {
+		bitmap_or(rc->cnflt_alien_bitmap.bm, rc->cnflt_alien_bitmap.bm, c->mas.bm, UWB_NUM_MAS);			
+	}
+	
+	queue_delayed_work(rc->rsv_workq, &rc->rsv_alien_bp_work, usecs_to_jiffies(delay_us));
+
+	kfree(cnflt);
+	mutex_unlock(&rc->rsvs_mutex);
+}
+
+static void uwb_cnflt_timer(unsigned long arg)
+{
+	struct uwb_cnflt_alien *cnflt = (struct uwb_cnflt_alien *)arg;
+
+	queue_work(cnflt->rc->rsv_workq, &cnflt->cnflt_update_work);
+}
+
 /*
- * Process a received DRP IE, it's either for a reservation owned by
- * the RC or targeted at it (or it's for a WUSB cluster reservation).
+ * We have received an DRP_IE of type Alien BP and we need to make
+ * sure we do not transmit in conflicting MASs.
  */
-static void uwb_drp_process(struct uwb_rc *rc, struct uwb_dev *src,
-		     struct uwb_ie_drp *drp_ie)
+static void uwb_drp_handle_alien_drp(struct uwb_rc *rc, struct uwb_ie_drp *drp_ie)
+{
+	struct device *dev = &rc->uwb_dev.dev;
+	struct uwb_mas_bm mas;
+	struct uwb_cnflt_alien *cnflt;
+	char buf[72];
+	unsigned long delay_us = UWB_MAS_LENGTH_US * UWB_MAS_PER_ZONE;
+	
+	uwb_drp_ie_to_bm(&mas, drp_ie);
+	bitmap_scnprintf(buf, sizeof(buf), mas.bm, UWB_NUM_MAS);
+	
+	list_for_each_entry(cnflt, &rc->cnflt_alien_list, rc_node) {
+		if (bitmap_equal(cnflt->mas.bm, mas.bm, UWB_NUM_MAS)) {
+			/* Existing alien BP reservation conflicting
+			 * bitmap, just reset the timer */
+			uwb_cnflt_alien_stroke_timer(cnflt);
+			return;
+		}
+	}
+
+	/* New alien BP reservation conflicting bitmap */
+
+	/* alloc and initialize new uwb_cnflt_alien */
+	cnflt = kzalloc(sizeof(struct uwb_cnflt_alien), GFP_KERNEL);
+	if (!cnflt)
+		dev_err(dev, "failed to alloc uwb_cnflt_alien struct\n");
+	INIT_LIST_HEAD(&cnflt->rc_node);
+	init_timer(&cnflt->timer);
+	cnflt->timer.function = uwb_cnflt_timer;
+	cnflt->timer.data     = (unsigned long)cnflt;
+
+	cnflt->rc = rc;
+	INIT_WORK(&cnflt->cnflt_update_work, uwb_cnflt_update_work);
+	
+	bitmap_copy(cnflt->mas.bm, mas.bm, UWB_NUM_MAS);
+
+	list_add_tail(&cnflt->rc_node, &rc->cnflt_alien_list);
+
+	/* update rc global conflicting alien bitmap */
+	bitmap_or(rc->cnflt_alien_bitmap.bm, rc->cnflt_alien_bitmap.bm, mas.bm, UWB_NUM_MAS);
+
+	queue_delayed_work(rc->rsv_workq, &rc->rsv_alien_bp_work, usecs_to_jiffies(delay_us));
+	
+	/* start the timer */
+	uwb_cnflt_alien_stroke_timer(cnflt);
+}
+
+static void uwb_drp_process_not_involved(struct uwb_rc *rc,
+					 struct uwb_rc_evt_drp *drp_evt, 
+					 struct uwb_ie_drp *drp_ie)
+{
+	struct uwb_mas_bm mas;
+	
+	uwb_drp_ie_to_bm(&mas, drp_ie);
+	uwb_drp_handle_all_conflict_rsv(rc, drp_evt, drp_ie, &mas);
+}
+
+static void uwb_drp_process_involved(struct uwb_rc *rc, struct uwb_dev *src,
+				     struct uwb_rc_evt_drp *drp_evt,
+				     struct uwb_ie_drp *drp_ie)
 {
 	struct uwb_rsv *rsv;
 
@@ -239,7 +646,7 @@ static void uwb_drp_process(struct uwb_rc *rc, struct uwb_dev *src,
 		 */
 		return;
 	}
-
+	
 	/*
 	 * Do nothing with DRP IEs for reservations that have been
 	 * terminated.
@@ -248,13 +655,43 @@ static void uwb_drp_process(struct uwb_rc *rc, struct uwb_dev *src,
 		uwb_rsv_set_state(rsv, UWB_RSV_STATE_NONE);
 		return;
 	}
-
+			
 	if (uwb_ie_drp_owner(drp_ie))
-		uwb_drp_process_target(rc, rsv, drp_ie);
+		uwb_drp_process_target(rc, rsv, drp_ie, drp_evt);
+	else
+		uwb_drp_process_owner(rc, rsv, src, drp_ie, drp_evt);
+	
+}
+
+
+static bool uwb_drp_involves_us(struct uwb_rc *rc, struct uwb_ie_drp *drp_ie)
+{
+	return uwb_dev_addr_cmp(&rc->uwb_dev.dev_addr, &drp_ie->dev_addr) == 0;
+}
+
+/*
+ * Process a received DRP IE.
+ */
+static void uwb_drp_process(struct uwb_rc *rc, struct uwb_rc_evt_drp *drp_evt,
+			    struct uwb_dev *src, struct uwb_ie_drp *drp_ie)
+{
+	if (uwb_ie_drp_type(drp_ie) == UWB_DRP_TYPE_ALIEN_BP)
+		uwb_drp_handle_alien_drp(rc, drp_ie);
+	else if (uwb_drp_involves_us(rc, drp_ie))
+		uwb_drp_process_involved(rc, src, drp_evt, drp_ie);
 	else
-		uwb_drp_process_owner(rc, rsv, drp_ie);
+		uwb_drp_process_not_involved(rc, drp_evt, drp_ie);
 }
 
+/*
+ * Process a received DRP Availability IE
+ */
+static void uwb_drp_availability_process(struct uwb_rc *rc, struct uwb_dev *src,
+					 struct uwb_ie_drp_avail *drp_availability_ie)
+{
+	bitmap_copy(src->last_availability_bm,
+		    drp_availability_ie->bmp, UWB_NUM_MAS);
+}
 
 /*
  * Process all the DRP IEs (both DRP IEs and the DRP Availability IE)
@@ -276,10 +713,10 @@ void uwb_drp_process_all(struct uwb_rc *rc, struct uwb_rc_evt_drp *drp_evt,
 
 		switch (ie_hdr->element_id) {
 		case UWB_IE_DRP_AVAILABILITY:
-			/* FIXME: does something need to be done with this? */
+			uwb_drp_availability_process(rc, src_dev, (struct uwb_ie_drp_avail *)ie_hdr);
 			break;
 		case UWB_IE_DRP:
-			uwb_drp_process(rc, src_dev, (struct uwb_ie_drp *)ie_hdr);
+			uwb_drp_process(rc, drp_evt, src_dev, (struct uwb_ie_drp *)ie_hdr);
 			break;
 		default:
 			dev_warn(dev, "unexpected IE in DRP notification\n");
@@ -292,55 +729,6 @@ void uwb_drp_process_all(struct uwb_rc *rc, struct uwb_rc_evt_drp *drp_evt,
 			 (int)ielen);
 }
 
-
-/*
- * Go through all the DRP IEs and find the ones that conflict with our
- * reservations.
- *
- * FIXME: must resolve the conflict according the the rules in
- * [ECMA-368].
- */
-static
-void uwb_drp_process_conflict_all(struct uwb_rc *rc, struct uwb_rc_evt_drp *drp_evt,
-				  size_t ielen, struct uwb_dev *src_dev)
-{
-	struct device *dev = &rc->uwb_dev.dev;
-	struct uwb_ie_hdr *ie_hdr;
-	struct uwb_ie_drp *drp_ie;
-	void *ptr;
-
-	ptr = drp_evt->ie_data;
-	for (;;) {
-		ie_hdr = uwb_ie_next(&ptr, &ielen);
-		if (!ie_hdr)
-			break;
-
-		drp_ie = container_of(ie_hdr, struct uwb_ie_drp, hdr);
-
-		/* FIXME: check if this DRP IE conflicts. */
-	}
-
-	if (ielen > 0)
-		dev_warn(dev, "%d octets remaining in DRP notification\n",
-			 (int)ielen);
-}
-
-
-/*
- * Terminate all reservations owned by, or targeted at, 'uwb_dev'.
- */
-static void uwb_drp_terminate_all(struct uwb_rc *rc, struct uwb_dev *uwb_dev)
-{
-	struct uwb_rsv *rsv;
-
-	list_for_each_entry(rsv, &rc->reservations, rc_node) {
-		if (rsv->owner == uwb_dev
-		    || (rsv->target.type == UWB_RSV_TARGET_DEV && rsv->target.dev == uwb_dev))
-			uwb_rsv_remove(rsv);
-	}
-}
-
-
 /**
  * uwbd_evt_handle_rc_drp - handle a DRP_IE event
  * @evt: the DRP_IE event from the radio controller
@@ -381,7 +769,6 @@ int uwbd_evt_handle_rc_drp(struct uwb_event *evt)
 	size_t ielength, bytes_left;
 	struct uwb_dev_addr src_addr;
 	struct uwb_dev *src_dev;
-	int reason;
 
 	/* Is there enough data to decode the event (and any IEs in
 	   its payload)? */
@@ -417,22 +804,8 @@ int uwbd_evt_handle_rc_drp(struct uwb_event *evt)
 
 	mutex_lock(&rc->rsvs_mutex);
 
-	reason = uwb_rc_evt_drp_reason(drp_evt);
-
-	switch (reason) {
-	case UWB_DRP_NOTIF_DRP_IE_RCVD:
-		uwb_drp_process_all(rc, drp_evt, ielength, src_dev);
-		break;
-	case UWB_DRP_NOTIF_CONFLICT:
-		uwb_drp_process_conflict_all(rc, drp_evt, ielength, src_dev);
-		break;
-	case UWB_DRP_NOTIF_TERMINATE:
-		uwb_drp_terminate_all(rc, src_dev);
-		break;
-	default:
-		dev_warn(dev, "ignored DRP event with reason code: %d\n", reason);
-		break;
-	}
+	/* We do not distinguish from the reason */
+	uwb_drp_process_all(rc, drp_evt, ielength, src_dev);
 
 	mutex_unlock(&rc->rsvs_mutex);
 
diff --git a/drivers/uwb/rsv.c b/drivers/uwb/rsv.c
index 1cd84f927540..165aec6a8f97 100644
--- a/drivers/uwb/rsv.c
+++ b/drivers/uwb/rsv.c
@@ -17,20 +17,31 @@
  */
 #include <linux/kernel.h>
 #include <linux/uwb.h>
+#include <linux/random.h>
 
 #include "uwb-internal.h"
 
 static void uwb_rsv_timer(unsigned long arg);
 
 static const char *rsv_states[] = {
-	[UWB_RSV_STATE_NONE]          = "none",
-	[UWB_RSV_STATE_O_INITIATED]   = "initiated",
-	[UWB_RSV_STATE_O_PENDING]     = "pending",
-	[UWB_RSV_STATE_O_MODIFIED]    = "modified",
-	[UWB_RSV_STATE_O_ESTABLISHED] = "established",
-	[UWB_RSV_STATE_T_ACCEPTED]    = "accepted",
-	[UWB_RSV_STATE_T_DENIED]      = "denied",
-	[UWB_RSV_STATE_T_PENDING]     = "pending",
+	[UWB_RSV_STATE_NONE]                 = "none            ",
+	[UWB_RSV_STATE_O_INITIATED]          = "o initiated     ",
+	[UWB_RSV_STATE_O_PENDING]            = "o pending       ",
+	[UWB_RSV_STATE_O_MODIFIED]           = "o modified      ",
+	[UWB_RSV_STATE_O_ESTABLISHED]        = "o established   ",
+	[UWB_RSV_STATE_O_TO_BE_MOVED]        = "o to be moved   ",
+	[UWB_RSV_STATE_O_MOVE_EXPANDING]     = "o move expanding",
+	[UWB_RSV_STATE_O_MOVE_COMBINING]     = "o move combining",
+	[UWB_RSV_STATE_O_MOVE_REDUCING]      = "o move reducing ",
+	[UWB_RSV_STATE_T_ACCEPTED]           = "t accepted      ",
+	[UWB_RSV_STATE_T_CONFLICT]           = "t conflict      ",
+	[UWB_RSV_STATE_T_PENDING]            = "t pending       ",
+	[UWB_RSV_STATE_T_DENIED]             = "t denied        ",
+	[UWB_RSV_STATE_T_RESIZED]            = "t resized       ",
+	[UWB_RSV_STATE_T_EXPANDING_ACCEPTED] = "t expanding acc ",
+	[UWB_RSV_STATE_T_EXPANDING_CONFLICT] = "t expanding conf",
+	[UWB_RSV_STATE_T_EXPANDING_PENDING]  = "t expanding pend",
+	[UWB_RSV_STATE_T_EXPANDING_DENIED]   = "t expanding den ",
 };
 
 static const char *rsv_types[] = {
@@ -41,6 +52,31 @@ static const char *rsv_types[] = {
 	[UWB_DRP_TYPE_PCA]      = "pca",
 };
 
+bool uwb_rsv_has_two_drp_ies(struct uwb_rsv *rsv)
+{
+	static const bool has_two_drp_ies[] = {
+		[UWB_RSV_STATE_O_INITIATED]               = false,
+		[UWB_RSV_STATE_O_PENDING]                 = false,
+		[UWB_RSV_STATE_O_MODIFIED]                = false,
+		[UWB_RSV_STATE_O_ESTABLISHED]             = false,
+		[UWB_RSV_STATE_O_TO_BE_MOVED]             = false,
+		[UWB_RSV_STATE_O_MOVE_COMBINING]          = false,
+		[UWB_RSV_STATE_O_MOVE_REDUCING]           = false,
+		[UWB_RSV_STATE_O_MOVE_EXPANDING]          = true,
+		[UWB_RSV_STATE_T_ACCEPTED]                = false,
+		[UWB_RSV_STATE_T_CONFLICT]                = false,
+		[UWB_RSV_STATE_T_PENDING]                 = false,
+		[UWB_RSV_STATE_T_DENIED]                  = false,
+		[UWB_RSV_STATE_T_RESIZED]                 = false,
+		[UWB_RSV_STATE_T_EXPANDING_ACCEPTED]      = true,
+		[UWB_RSV_STATE_T_EXPANDING_CONFLICT]      = true,
+		[UWB_RSV_STATE_T_EXPANDING_PENDING]       = true,
+		[UWB_RSV_STATE_T_EXPANDING_DENIED]        = true,
+	};
+
+	return has_two_drp_ies[rsv->state];
+}
+
 /**
  * uwb_rsv_state_str - return a string for a reservation state
  * @state: the reservation state.
@@ -65,7 +101,7 @@ const char *uwb_rsv_type_str(enum uwb_drp_type type)
 }
 EXPORT_SYMBOL_GPL(uwb_rsv_type_str);
 
-static void uwb_rsv_dump(struct uwb_rsv *rsv)
+void uwb_rsv_dump(char *text, struct uwb_rsv *rsv)
 {
 	struct device *dev = &rsv->rc->uwb_dev.dev;
 	struct uwb_dev_addr devaddr;
@@ -88,12 +124,12 @@ static void uwb_rsv_release(struct kref *kref)
 	kfree(rsv);
 }
 
-static void uwb_rsv_get(struct uwb_rsv *rsv)
+void uwb_rsv_get(struct uwb_rsv *rsv)
 {
 	kref_get(&rsv->kref);
 }
 
-static void uwb_rsv_put(struct uwb_rsv *rsv)
+void uwb_rsv_put(struct uwb_rsv *rsv)
 {
 	kref_put(&rsv->kref, uwb_rsv_release);
 }
@@ -108,6 +144,7 @@ static void uwb_rsv_put(struct uwb_rsv *rsv)
 static int uwb_rsv_get_stream(struct uwb_rsv *rsv)
 {
 	struct uwb_rc *rc = rsv->rc;
+	struct device *dev = &rc->uwb_dev.dev;
 	unsigned long *streams_bm;
 	int stream;
 
@@ -129,12 +166,15 @@ static int uwb_rsv_get_stream(struct uwb_rsv *rsv)
 	rsv->stream = stream;
 	set_bit(stream, streams_bm);
 
+	dev_dbg(dev, "get stream %d\n", rsv->stream);
+
 	return 0;
 }
 
 static void uwb_rsv_put_stream(struct uwb_rsv *rsv)
 {
 	struct uwb_rc *rc = rsv->rc;
+	struct device *dev = &rc->uwb_dev.dev;
 	unsigned long *streams_bm;
 
 	switch (rsv->target.type) {
@@ -149,86 +189,52 @@ static void uwb_rsv_put_stream(struct uwb_rsv *rsv)
 	}
 
 	clear_bit(rsv->stream, streams_bm);
+
+	dev_dbg(dev, "put stream %d\n", rsv->stream);
 }
 
-/*
- * Generate a MAS allocation with a single row component.
- */
-static void uwb_rsv_gen_alloc_row(struct uwb_mas_bm *mas,
-				  int first_mas, int mas_per_zone,
-				  int zs, int ze)
+void uwb_rsv_backoff_win_timer(unsigned long arg)
 {
-	struct uwb_mas_bm col;
-	int z;
+	struct uwb_drp_backoff_win *bow = (struct uwb_drp_backoff_win *)arg;
+	struct uwb_rc *rc = container_of(bow, struct uwb_rc, bow);
+	struct device *dev = &rc->uwb_dev.dev;
 
-	bitmap_zero(mas->bm, UWB_NUM_MAS);
-	bitmap_zero(col.bm, UWB_NUM_MAS);
-	bitmap_fill(col.bm, mas_per_zone);
-	bitmap_shift_left(col.bm, col.bm, first_mas + zs * UWB_MAS_PER_ZONE, UWB_NUM_MAS);
-
-	for (z = zs; z <= ze; z++) {
-		bitmap_or(mas->bm, mas->bm, col.bm, UWB_NUM_MAS);
-		bitmap_shift_left(col.bm, col.bm, UWB_MAS_PER_ZONE, UWB_NUM_MAS);
+	bow->can_reserve_extra_mases = true;
+	if (bow->total_expired <= 4) {
+		bow->total_expired++;
+	} else {
+		/* after 4 backoff window has expired we can exit from
+		 * the backoff procedure */
+		bow->total_expired = 0;
+		bow->window = UWB_DRP_BACKOFF_WIN_MIN >> 1;
 	}
+	dev_dbg(dev, "backoff_win_timer total_expired=%d, n=%d\n: ", bow->total_expired, bow->n);
+
+	/* try to relocate all the "to be moved" relocations */
+	uwb_rsv_handle_drp_avail_change(rc);
 }
 
-/*
- * Allocate some MAS for this reservation based on current local
- * availability, the reservation parameters (max_mas, min_mas,
- * sparsity), and the WiMedia rules for MAS allocations.
- *
- * Returns -EBUSY is insufficient free MAS are available.
- *
- * FIXME: to simplify this, only safe reservations with a single row
- * component in zones 1 to 15 are tried (zone 0 is skipped to avoid
- * problems with the MAS reserved for the BP).
- *
- * [ECMA-368] section B.2.
- */
-static int uwb_rsv_alloc_mas(struct uwb_rsv *rsv)
+void uwb_rsv_backoff_win_increment(struct uwb_rc *rc)
 {
-	static const int safe_mas_in_row[UWB_NUM_ZONES] = {
-		8, 7, 6, 5, 4, 4, 4, 4, 4, 4, 4, 4, 4, 3, 2, 1,
-	};
-	int n, r;
-	struct uwb_mas_bm mas;
-	bool found = false;
+	struct uwb_drp_backoff_win *bow = &rc->bow;
+	struct device *dev = &rc->uwb_dev.dev;
+	unsigned timeout_us;
 
-	/*
-	 * Search all valid safe allocations until either: too few MAS
-	 * are available; or the smallest allocation with sufficient
-	 * MAS is found.
-	 *
-	 * The top of the zones are preferred, so space for larger
-	 * allocations is available in the bottom of the zone (e.g., a
-	 * 15 MAS allocation should start in row 14 leaving space for
-	 * a 120 MAS allocation at row 0).
-	 */
-	for (n = safe_mas_in_row[0]; n >= 1; n--) {
-		int num_mas;
+	dev_dbg(dev, "backoff_win_increment: window=%d\n", bow->window);
 
-		num_mas = n * (UWB_NUM_ZONES - 1);
-		if (num_mas < rsv->min_mas)
-			break;
-		if (found && num_mas < rsv->max_mas)
-			break;
+	bow->can_reserve_extra_mases = false;
 
-		for (r = UWB_MAS_PER_ZONE-1;  r >= 0; r--) {
-			if (safe_mas_in_row[r] < n)
-				continue;
-			uwb_rsv_gen_alloc_row(&mas, r, n, 1, UWB_NUM_ZONES);
-			if (uwb_drp_avail_reserve_pending(rsv->rc, &mas) == 0) {
-				found = true;
-				break;
-			}
-		}
-	}
+	if((bow->window << 1) == UWB_DRP_BACKOFF_WIN_MAX)
+		return;
 
-	if (!found)
-		return -EBUSY;
+	bow->window <<= 1;
+	bow->n = random32() & (bow->window - 1);
+	dev_dbg(dev, "new_window=%d, n=%d\n: ", bow->window, bow->n);
 
-	bitmap_copy(rsv->mas.bm, mas.bm, UWB_NUM_MAS);
-	return 0;
+	/* reset the timer associated variables */
+	timeout_us = bow->n * UWB_SUPERFRAME_LENGTH_US;
+	bow->total_expired = 0;
+	mod_timer(&bow->timer, jiffies + usecs_to_jiffies(timeout_us));		
 }
 
 static void uwb_rsv_stroke_timer(struct uwb_rsv *rsv)
@@ -241,13 +247,16 @@ static void uwb_rsv_stroke_timer(struct uwb_rsv *rsv)
 	 * received.
 	 */
 	if (rsv->is_multicast) {
-		if (rsv->state == UWB_RSV_STATE_O_INITIATED)
+		if (rsv->state == UWB_RSV_STATE_O_INITIATED
+		    || rsv->state == UWB_RSV_STATE_O_MOVE_EXPANDING
+		    || rsv->state == UWB_RSV_STATE_O_MOVE_COMBINING
+		    || rsv->state == UWB_RSV_STATE_O_MOVE_REDUCING)
 			sframes = 1;
 		if (rsv->state == UWB_RSV_STATE_O_ESTABLISHED)
 			sframes = 0;
+		
 	}
 
-	rsv->expired = false;
 	if (sframes > 0) {
 		/*
 		 * Add an additional 2 superframes to account for the
@@ -269,7 +278,7 @@ static void uwb_rsv_state_update(struct uwb_rsv *rsv,
 	rsv->state = new_state;
 	rsv->ie_valid = false;
 
-	uwb_rsv_dump(rsv);
+	uwb_rsv_dump("SU", rsv);
 
 	uwb_rsv_stroke_timer(rsv);
 	uwb_rsv_sched_update(rsv->rc);
@@ -283,10 +292,17 @@ static void uwb_rsv_callback(struct uwb_rsv *rsv)
 
 void uwb_rsv_set_state(struct uwb_rsv *rsv, enum uwb_rsv_state new_state)
 {
+	struct uwb_rsv_move *mv = &rsv->mv;
+
 	if (rsv->state == new_state) {
 		switch (rsv->state) {
 		case UWB_RSV_STATE_O_ESTABLISHED:
+		case UWB_RSV_STATE_O_MOVE_EXPANDING:
+		case UWB_RSV_STATE_O_MOVE_COMBINING:
+		case UWB_RSV_STATE_O_MOVE_REDUCING:
 		case UWB_RSV_STATE_T_ACCEPTED:
+		case UWB_RSV_STATE_T_EXPANDING_ACCEPTED:
+		case UWB_RSV_STATE_T_RESIZED:
 		case UWB_RSV_STATE_NONE:
 			uwb_rsv_stroke_timer(rsv);
 			break;
@@ -298,11 +314,10 @@ void uwb_rsv_set_state(struct uwb_rsv *rsv, enum uwb_rsv_state new_state)
 		return;
 	}
 
+	uwb_rsv_dump("SC", rsv);
+
 	switch (new_state) {
 	case UWB_RSV_STATE_NONE:
-		uwb_drp_avail_release(rsv->rc, &rsv->mas);
-		if (uwb_rsv_is_owner(rsv))
-			uwb_rsv_put_stream(rsv);
 		uwb_rsv_state_update(rsv, UWB_RSV_STATE_NONE);
 		uwb_rsv_callback(rsv);
 		break;
@@ -312,12 +327,45 @@ void uwb_rsv_set_state(struct uwb_rsv *rsv, enum uwb_rsv_state new_state)
 	case UWB_RSV_STATE_O_PENDING:
 		uwb_rsv_state_update(rsv, UWB_RSV_STATE_O_PENDING);
 		break;
+	case UWB_RSV_STATE_O_MODIFIED:
+		/* in the companion there are the MASes to drop */
+		bitmap_andnot(rsv->mas.bm, rsv->mas.bm, mv->companion_mas.bm, UWB_NUM_MAS);
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_O_MODIFIED);
+		break;
 	case UWB_RSV_STATE_O_ESTABLISHED:
+		if (rsv->state == UWB_RSV_STATE_O_MODIFIED
+		    || rsv->state == UWB_RSV_STATE_O_MOVE_REDUCING) {
+			uwb_drp_avail_release(rsv->rc, &mv->companion_mas);
+			rsv->needs_release_companion_mas = false;
+		}
 		uwb_drp_avail_reserve(rsv->rc, &rsv->mas);
 		uwb_rsv_state_update(rsv, UWB_RSV_STATE_O_ESTABLISHED);
 		uwb_rsv_callback(rsv);
 		break;
+	case UWB_RSV_STATE_O_MOVE_EXPANDING:
+		rsv->needs_release_companion_mas = true;
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_O_MOVE_EXPANDING);
+		break;
+	case UWB_RSV_STATE_O_MOVE_COMBINING:
+		rsv->needs_release_companion_mas = false;
+		uwb_drp_avail_reserve(rsv->rc, &mv->companion_mas);
+		bitmap_or(rsv->mas.bm, rsv->mas.bm, mv->companion_mas.bm, UWB_NUM_MAS);
+		rsv->mas.safe   += mv->companion_mas.safe;
+		rsv->mas.unsafe += mv->companion_mas.unsafe;
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_O_MOVE_COMBINING);
+		break;
+	case UWB_RSV_STATE_O_MOVE_REDUCING:
+		bitmap_andnot(mv->companion_mas.bm, rsv->mas.bm, mv->final_mas.bm, UWB_NUM_MAS);
+		rsv->needs_release_companion_mas = true;
+		rsv->mas.safe   = mv->final_mas.safe;
+		rsv->mas.unsafe = mv->final_mas.unsafe;
+		bitmap_copy(rsv->mas.bm, mv->final_mas.bm, UWB_NUM_MAS);
+		bitmap_copy(rsv->mas.unsafe_bm, mv->final_mas.unsafe_bm, UWB_NUM_MAS);
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_O_MOVE_REDUCING);
+		break;
 	case UWB_RSV_STATE_T_ACCEPTED:
+	case UWB_RSV_STATE_T_RESIZED:
+		rsv->needs_release_companion_mas = false;
 		uwb_drp_avail_reserve(rsv->rc, &rsv->mas);
 		uwb_rsv_state_update(rsv, UWB_RSV_STATE_T_ACCEPTED);
 		uwb_rsv_callback(rsv);
@@ -325,12 +373,82 @@ void uwb_rsv_set_state(struct uwb_rsv *rsv, enum uwb_rsv_state new_state)
 	case UWB_RSV_STATE_T_DENIED:
 		uwb_rsv_state_update(rsv, UWB_RSV_STATE_T_DENIED);
 		break;
+	case UWB_RSV_STATE_T_CONFLICT:
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_T_CONFLICT);
+		break;
+	case UWB_RSV_STATE_T_PENDING:
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_T_PENDING);
+		break;
+	case UWB_RSV_STATE_T_EXPANDING_ACCEPTED:
+		rsv->needs_release_companion_mas = true;
+		uwb_drp_avail_reserve(rsv->rc, &mv->companion_mas);
+		uwb_rsv_state_update(rsv, UWB_RSV_STATE_T_EXPANDING_ACCEPTED);
+		break;
 	default:
 		dev_err(&rsv->rc->uwb_dev.dev, "unhandled state: %s (%d)\n",
 			uwb_rsv_state_str(new_state), new_state);
 	}
 }
 
+static void uwb_rsv_handle_timeout_work(struct work_struct *work)
+{
+	struct uwb_rsv *rsv = container_of(work, struct uwb_rsv,
+					   handle_timeout_work);
+	struct uwb_rc *rc = rsv->rc;
+
+	mutex_lock(&rc->rsvs_mutex);
+
+	uwb_rsv_dump("TO", rsv);
+
+	switch (rsv->state) {
+	case UWB_RSV_STATE_O_INITIATED:
+		if (rsv->is_multicast) {
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
+			goto unlock;
+		}
+		break;
+	case UWB_RSV_STATE_O_MOVE_EXPANDING:
+		if (rsv->is_multicast) {
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_COMBINING);
+			goto unlock;
+		}
+		break;
+	case UWB_RSV_STATE_O_MOVE_COMBINING:
+		if (rsv->is_multicast) {
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_REDUCING);
+			goto unlock;
+		}
+		break;
+	case UWB_RSV_STATE_O_MOVE_REDUCING:
+		if (rsv->is_multicast) {
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_ESTABLISHED);
+			goto unlock;
+		}
+		break;
+	case UWB_RSV_STATE_O_ESTABLISHED:
+		if (rsv->is_multicast)
+			goto unlock;
+		break;
+	case UWB_RSV_STATE_T_EXPANDING_ACCEPTED:
+		/*
+		 * The time out could be for the main or of the
+		 * companion DRP, assume it's for the companion and
+		 * drop that first.  A further time out is required to
+		 * drop the main.
+		 */
+		uwb_rsv_set_state(rsv, UWB_RSV_STATE_T_ACCEPTED);
+		uwb_drp_avail_release(rsv->rc, &rsv->mv.companion_mas);
+		goto unlock;
+	default:
+		break;
+	}
+
+	uwb_rsv_remove(rsv);
+
+unlock:
+	mutex_unlock(&rc->rsvs_mutex);
+}
+
 static struct uwb_rsv *uwb_rsv_alloc(struct uwb_rc *rc)
 {
 	struct uwb_rsv *rsv;
@@ -347,6 +465,7 @@ static struct uwb_rsv *uwb_rsv_alloc(struct uwb_rc *rc)
 	rsv->timer.data     = (unsigned long)rsv;
 
 	rsv->rc = rc;
+	INIT_WORK(&rsv->handle_timeout_work, uwb_rsv_handle_timeout_work);
 
 	return rsv;
 }
@@ -381,8 +500,18 @@ EXPORT_SYMBOL_GPL(uwb_rsv_create);
 
 void uwb_rsv_remove(struct uwb_rsv *rsv)
 {
+	uwb_rsv_dump("RM", rsv);
+
 	if (rsv->state != UWB_RSV_STATE_NONE)
 		uwb_rsv_set_state(rsv, UWB_RSV_STATE_NONE);
+
+	if (rsv->needs_release_companion_mas)
+		uwb_drp_avail_release(rsv->rc, &rsv->mv.companion_mas);
+	uwb_drp_avail_release(rsv->rc, &rsv->mas);
+
+	if (uwb_rsv_is_owner(rsv))
+		uwb_rsv_put_stream(rsv);
+	
 	del_timer_sync(&rsv->timer);
 	uwb_dev_put(rsv->owner);
 	if (rsv->target.type == UWB_RSV_TARGET_DEV)
@@ -409,7 +538,7 @@ EXPORT_SYMBOL_GPL(uwb_rsv_destroy);
  * @rsv: the reservation
  *
  * The PAL should fill in @rsv's owner, target, type, max_mas,
- * min_mas, sparsity and is_multicast fields.  If the target is a
+ * min_mas, max_interval and is_multicast fields.  If the target is a
  * uwb_dev it must be referenced.
  *
  * The reservation's callback will be called when the reservation is
@@ -418,16 +547,27 @@ EXPORT_SYMBOL_GPL(uwb_rsv_destroy);
 int uwb_rsv_establish(struct uwb_rsv *rsv)
 {
 	struct uwb_rc *rc = rsv->rc;
+	struct uwb_mas_bm available;
 	int ret;
 
 	mutex_lock(&rc->rsvs_mutex);
-
 	ret = uwb_rsv_get_stream(rsv);
 	if (ret)
 		goto out;
 
-	ret = uwb_rsv_alloc_mas(rsv);
-	if (ret) {
+	rsv->tiebreaker = random32() & 1;
+	/* get available mas bitmap */
+	uwb_drp_available(rc, &available);
+
+	ret = uwb_rsv_find_best_allocation(rsv, &available, &rsv->mas);
+	if (ret == UWB_RSV_ALLOC_NOT_FOUND) {
+		ret = -EBUSY;
+		uwb_rsv_put_stream(rsv);
+		goto out;
+	}
+
+	ret = uwb_drp_avail_reserve_pending(rc, &rsv->mas);
+	if (ret != 0) {
 		uwb_rsv_put_stream(rsv);
 		goto out;
 	}
@@ -448,16 +588,71 @@ EXPORT_SYMBOL_GPL(uwb_rsv_establish);
  * @rsv: the reservation to modify
  * @max_mas: new maximum MAS to reserve
  * @min_mas: new minimum MAS to reserve
- * @sparsity: new sparsity to use
+ * @max_interval: new max_interval to use
  *
  * FIXME: implement this once there are PALs that use it.
  */
-int uwb_rsv_modify(struct uwb_rsv *rsv, int max_mas, int min_mas, int sparsity)
+int uwb_rsv_modify(struct uwb_rsv *rsv, int max_mas, int min_mas, int max_interval)
 {
 	return -ENOSYS;
 }
 EXPORT_SYMBOL_GPL(uwb_rsv_modify);
 
+/*
+ * move an already established reservation (rc->rsvs_mutex must to be
+ * taken when tis function is called)
+ */
+int uwb_rsv_try_move(struct uwb_rsv *rsv, struct uwb_mas_bm *available)
+{
+	struct uwb_rc *rc = rsv->rc;
+	struct uwb_drp_backoff_win *bow = &rc->bow;
+	struct device *dev = &rc->uwb_dev.dev;
+	struct uwb_rsv_move *mv;
+	int ret = 0;
+ 
+	if (bow->can_reserve_extra_mases == false)
+		return -EBUSY;
+
+	mv = &rsv->mv;
+
+	if (uwb_rsv_find_best_allocation(rsv, available, &mv->final_mas) == UWB_RSV_ALLOC_FOUND) {
+
+		if (!bitmap_equal(rsv->mas.bm, mv->final_mas.bm, UWB_NUM_MAS)) {
+			/* We want to move the reservation */
+			bitmap_andnot(mv->companion_mas.bm, mv->final_mas.bm, rsv->mas.bm, UWB_NUM_MAS);
+			uwb_drp_avail_reserve_pending(rc, &mv->companion_mas);
+			uwb_rsv_set_state(rsv, UWB_RSV_STATE_O_MOVE_EXPANDING);
+		}
+	} else {
+		dev_dbg(dev, "new allocation not found\n");
+	}
+	
+	return ret;
+}
+
+/* It will try to move every reservation in state O_ESTABLISHED giving
+ * to the MAS allocator algorithm an availability that is the real one
+ * plus the allocation already established from the reservation. */
+void uwb_rsv_handle_drp_avail_change(struct uwb_rc *rc)
+{
+	struct uwb_drp_backoff_win *bow = &rc->bow;
+	struct uwb_rsv *rsv;
+	struct uwb_mas_bm mas;
+	
+	if (bow->can_reserve_extra_mases == false)
+		return;
+
+	list_for_each_entry(rsv, &rc->reservations, rc_node) {
+		if (rsv->state == UWB_RSV_STATE_O_ESTABLISHED ||
+		    rsv->state == UWB_RSV_STATE_O_TO_BE_MOVED) {
+			uwb_drp_available(rc, &mas);
+			bitmap_or(mas.bm, mas.bm, rsv->mas.bm, UWB_NUM_MAS);
+			uwb_rsv_try_move(rsv, &mas);
+		}
+	}
+	
+}
+
 /**
  * uwb_rsv_terminate - terminate an established reservation
  * @rsv: the reservation to terminate
@@ -546,6 +741,7 @@ static struct uwb_rsv *uwb_rsv_new_target(struct uwb_rc *rc,
 	uwb_dev_get(rsv->owner);
 	rsv->target.type = UWB_RSV_TARGET_DEV;
 	rsv->target.dev  = &rc->uwb_dev;
+	uwb_dev_get(&rc->uwb_dev);
 	rsv->type        = uwb_ie_drp_type(drp_ie);
 	rsv->stream      = uwb_ie_drp_stream_index(drp_ie);
 	uwb_drp_ie_to_bm(&rsv->mas, drp_ie);
@@ -567,11 +763,33 @@ static struct uwb_rsv *uwb_rsv_new_target(struct uwb_rc *rc,
 	list_add_tail(&rsv->rc_node, &rc->reservations);
 	state = rsv->state;
 	rsv->state = UWB_RSV_STATE_NONE;
-	uwb_rsv_set_state(rsv, state);
+
+	/* FIXME: do something sensible here */
+	if (state == UWB_RSV_STATE_T_ACCEPTED
+	    && uwb_drp_avail_reserve_pending(rc, &rsv->mas) == -EBUSY) {
+		/* FIXME: do something sensible here */
+	} else {
+		uwb_rsv_set_state(rsv, state);
+	}
 
 	return rsv;
 }
 
+/**
+ * uwb_rsv_get_usable_mas - get the bitmap of the usable MAS of a reservations
+ * @rsv: the reservation.
+ * @mas: returns the available MAS.
+ *
+ * The usable MAS of a reservation may be less than the negotiated MAS
+ * if alien BPs are present.
+ */
+void uwb_rsv_get_usable_mas(struct uwb_rsv *rsv, struct uwb_mas_bm *mas)
+{
+	bitmap_zero(mas->bm, UWB_NUM_MAS);
+	bitmap_andnot(mas->bm, rsv->mas.bm, rsv->rc->cnflt_alien_bitmap.bm, UWB_NUM_MAS);
+}
+EXPORT_SYMBOL_GPL(uwb_rsv_get_usable_mas);
+
 /**
  * uwb_rsv_find - find a reservation for a received DRP IE.
  * @rc: the radio controller
@@ -611,8 +829,6 @@ static bool uwb_rsv_update_all(struct uwb_rc *rc)
 	bool ie_updated = false;
 
 	list_for_each_entry_safe(rsv, t, &rc->reservations, rc_node) {
-		if (rsv->expired)
-			uwb_drp_handle_timeout(rsv);
 		if (!rsv->ie_valid) {
 			uwb_drp_ie_update(rsv);
 			ie_updated = true;
@@ -622,9 +838,47 @@ static bool uwb_rsv_update_all(struct uwb_rc *rc)
 	return ie_updated;
 }
 
+void uwb_rsv_queue_update(struct uwb_rc *rc)
+{
+	unsigned long delay_us = UWB_MAS_LENGTH_US * UWB_MAS_PER_ZONE;
+
+	queue_delayed_work(rc->rsv_workq, &rc->rsv_update_work, usecs_to_jiffies(delay_us));
+}
+
+/**
+ * uwb_rsv_sched_update - schedule an update of the DRP IEs
+ * @rc: the radio controller.
+ *
+ * To improve performance and ensure correctness with [ECMA-368] the
+ * number of SET-DRP-IE commands that are done are limited.
+ *
+ * DRP IEs update come from two sources: DRP events from the hardware
+ * which all occur at the beginning of the superframe ('syncronous'
+ * events) and reservation establishment/termination requests from
+ * PALs or timers ('asynchronous' events).
+ *
+ * A delayed work ensures that all the synchronous events result in
+ * one SET-DRP-IE command.
+ *
+ * Additional logic (the set_drp_ie_pending and rsv_updated_postponed
+ * flags) will prevent an asynchrous event starting a SET-DRP-IE
+ * command if one is currently awaiting a response.
+ *
+ * FIXME: this does leave a window where an asynchrous event can delay
+ * the SET-DRP-IE for a synchronous event by one superframe.
+ */
 void uwb_rsv_sched_update(struct uwb_rc *rc)
 {
-	queue_work(rc->rsv_workq, &rc->rsv_update_work);
+	spin_lock(&rc->rsvs_lock);
+	if (!delayed_work_pending(&rc->rsv_update_work)) {
+		if (rc->set_drp_ie_pending > 0) {
+			rc->set_drp_ie_pending++;
+			goto unlock;
+		}
+		uwb_rsv_queue_update(rc);
+	}
+unlock:
+	spin_unlock(&rc->rsvs_lock);
 }
 
 /*
@@ -633,7 +887,8 @@ void uwb_rsv_sched_update(struct uwb_rc *rc)
  */
 static void uwb_rsv_update_work(struct work_struct *work)
 {
-	struct uwb_rc *rc = container_of(work, struct uwb_rc, rsv_update_work);
+	struct uwb_rc *rc = container_of(work, struct uwb_rc,
+					 rsv_update_work.work);
 	bool ie_updated;
 
 	mutex_lock(&rc->rsvs_mutex);
@@ -645,18 +900,34 @@ static void uwb_rsv_update_work(struct work_struct *work)
 		ie_updated = true;
 	}
 
-	if (ie_updated)
+	if (ie_updated && (rc->set_drp_ie_pending == 0))
 		uwb_rc_send_all_drp_ie(rc);
 
 	mutex_unlock(&rc->rsvs_mutex);
 }
 
+static void uwb_rsv_alien_bp_work(struct work_struct *work)
+{
+	struct uwb_rc *rc = container_of(work, struct uwb_rc,
+					 rsv_alien_bp_work.work);
+	struct uwb_rsv *rsv;
+
+	mutex_lock(&rc->rsvs_mutex);
+	
+	list_for_each_entry(rsv, &rc->reservations, rc_node) {
+		if (rsv->type != UWB_DRP_TYPE_ALIEN_BP) {
+			rsv->callback(rsv);
+		}
+	}
+
+	mutex_unlock(&rc->rsvs_mutex);
+}
+
 static void uwb_rsv_timer(unsigned long arg)
 {
 	struct uwb_rsv *rsv = (struct uwb_rsv *)arg;
 
-	rsv->expired = true;
-	uwb_rsv_sched_update(rsv->rc);
+	queue_work(rsv->rc->rsv_workq, &rsv->handle_timeout_work);
 }
 
 /**
@@ -673,16 +944,27 @@ void uwb_rsv_remove_all(struct uwb_rc *rc)
 	list_for_each_entry_safe(rsv, t, &rc->reservations, rc_node) {
 		uwb_rsv_remove(rsv);
 	}
+	/* Cancel any postponed update. */
+	rc->set_drp_ie_pending = 0;
 	mutex_unlock(&rc->rsvs_mutex);
 
-	cancel_work_sync(&rc->rsv_update_work);
+	cancel_delayed_work_sync(&rc->rsv_update_work);
 }
 
 void uwb_rsv_init(struct uwb_rc *rc)
 {
 	INIT_LIST_HEAD(&rc->reservations);
+	INIT_LIST_HEAD(&rc->cnflt_alien_list);
 	mutex_init(&rc->rsvs_mutex);
-	INIT_WORK(&rc->rsv_update_work, uwb_rsv_update_work);
+	spin_lock_init(&rc->rsvs_lock);
+	INIT_DELAYED_WORK(&rc->rsv_update_work, uwb_rsv_update_work);
+	INIT_DELAYED_WORK(&rc->rsv_alien_bp_work, uwb_rsv_alien_bp_work);
+	rc->bow.can_reserve_extra_mases = true;
+	rc->bow.total_expired = 0;
+	rc->bow.window = UWB_DRP_BACKOFF_WIN_MIN >> 1;
+	init_timer(&rc->bow.timer);
+	rc->bow.timer.function = uwb_rsv_backoff_win_timer;
+	rc->bow.timer.data     = (unsigned long)&rc->bow;
 
 	bitmap_complement(rc->uwb_dev.streams, rc->uwb_dev.streams, UWB_NUM_STREAMS);
 }
diff --git a/drivers/uwb/uwb-debug.c b/drivers/uwb/uwb-debug.c
index a6debb9baf38..89b2e6a7214c 100644
--- a/drivers/uwb/uwb-debug.c
+++ b/drivers/uwb/uwb-debug.c
@@ -82,29 +82,21 @@ struct uwb_dbg {
 	struct dentry *reservations_f;
 	struct dentry *accept_f;
 	struct dentry *drp_avail_f;
+	spinlock_t list_lock;
 };
 
 static struct dentry *root_dir;
 
 static void uwb_dbg_rsv_cb(struct uwb_rsv *rsv)
 {
-	struct uwb_rc *rc = rsv->rc;
-	struct device *dev = &rc->uwb_dev.dev;
-	struct uwb_dev_addr devaddr;
-	char owner[UWB_ADDR_STRSIZE], target[UWB_ADDR_STRSIZE];
-
-	uwb_dev_addr_print(owner, sizeof(owner), &rsv->owner->dev_addr);
-	if (rsv->target.type == UWB_RSV_TARGET_DEV)
-		devaddr = rsv->target.dev->dev_addr;
-	else
-		devaddr = rsv->target.devaddr;
-	uwb_dev_addr_print(target, sizeof(target), &devaddr);
+	struct uwb_dbg *dbg = rsv->pal_priv;
 
-	dev_dbg(dev, "debug: rsv %s -> %s: %s\n",
-		owner, target, uwb_rsv_state_str(rsv->state));
+	uwb_rsv_dump("debug", rsv);
 
 	if (rsv->state == UWB_RSV_STATE_NONE) {
+		spin_lock(&dbg->list_lock);
 		list_del(&rsv->pal_node);
+		spin_unlock(&dbg->list_lock);
 		uwb_rsv_destroy(rsv);
 	}
 }
@@ -128,20 +120,21 @@ static int cmd_rsv_establish(struct uwb_rc *rc,
 		return -ENOMEM;
 	}
 
-	rsv->owner       = &rc->uwb_dev;
-	rsv->target.type = UWB_RSV_TARGET_DEV;
-	rsv->target.dev  = target;
-	rsv->type        = cmd->type;
-	rsv->max_mas     = cmd->max_mas;
-	rsv->min_mas     = cmd->min_mas;
-	rsv->sparsity    = cmd->sparsity;
+	rsv->target.type  = UWB_RSV_TARGET_DEV;
+	rsv->target.dev   = target;
+	rsv->type         = cmd->type;
+	rsv->max_mas      = cmd->max_mas;
+	rsv->min_mas      = cmd->min_mas;
+	rsv->max_interval = cmd->max_interval;
 
 	ret = uwb_rsv_establish(rsv);
 	if (ret)
 		uwb_rsv_destroy(rsv);
-	else
+	else {
+		spin_lock(&(rc->dbg)->list_lock);
 		list_add_tail(&rsv->pal_node, &rc->dbg->rsvs);
-
+		spin_unlock(&(rc->dbg)->list_lock);
+	}
 	return ret;
 }
 
@@ -151,17 +144,24 @@ static int cmd_rsv_terminate(struct uwb_rc *rc,
 	struct uwb_rsv *rsv, *found = NULL;
 	int i = 0;
 
+	spin_lock(&(rc->dbg)->list_lock);
+
 	list_for_each_entry(rsv, &rc->dbg->rsvs, pal_node) {
 		if (i == cmd->index) {
 			found = rsv;
+			uwb_rsv_get(found);
 			break;
 		}
 		i++;
 	}
+
+	spin_unlock(&(rc->dbg)->list_lock);
+
 	if (!found)
 		return -EINVAL;
 
 	uwb_rsv_terminate(found);
+	uwb_rsv_put(found);
 
 	return 0;
 }
@@ -191,7 +191,7 @@ static ssize_t command_write(struct file *file, const char __user *buf,
 	struct uwb_rc *rc = file->private_data;
 	struct uwb_dbg_cmd cmd;
 	int ret = 0;
-
+	
 	if (len != sizeof(struct uwb_dbg_cmd))
 		return -EINVAL;
 
@@ -325,7 +325,9 @@ static void uwb_dbg_new_rsv(struct uwb_pal *pal, struct uwb_rsv *rsv)
 	struct uwb_dbg *dbg = container_of(pal, struct uwb_dbg, pal);
 
 	if (dbg->accept) {
+		spin_lock(&dbg->list_lock);
 		list_add_tail(&rsv->pal_node, &dbg->rsvs);
+		spin_unlock(&dbg->list_lock);
 		uwb_rsv_accept(rsv, uwb_dbg_rsv_cb, dbg);
 	}
 }
@@ -341,6 +343,7 @@ void uwb_dbg_add_rc(struct uwb_rc *rc)
 		return;
 
 	INIT_LIST_HEAD(&rc->dbg->rsvs);
+	spin_lock_init(&(rc->dbg)->list_lock);
 
 	uwb_pal_init(&rc->dbg->pal);
 	rc->dbg->pal.rc = rc;
diff --git a/drivers/uwb/uwb-internal.h b/drivers/uwb/uwb-internal.h
index f0f21f406bf0..d5bcfc1c227a 100644
--- a/drivers/uwb/uwb-internal.h
+++ b/drivers/uwb/uwb-internal.h
@@ -92,6 +92,12 @@ extern const char *uwb_rc_strerror(unsigned code);
 
 struct uwb_rc_neh;
 
+extern int uwb_rc_cmd_async(struct uwb_rc *rc, const char *cmd_name,
+			    struct uwb_rccb *cmd, size_t cmd_size,
+			    u8 expected_type, u16 expected_event,
+			    uwb_rc_cmd_cb_f cb, void *arg);
+
+
 void uwb_rc_neh_create(struct uwb_rc *rc);
 void uwb_rc_neh_destroy(struct uwb_rc *rc);
 
@@ -106,7 +112,69 @@ void uwb_rc_neh_put(struct uwb_rc_neh *neh);
 extern int uwb_est_create(void);
 extern void uwb_est_destroy(void);
 
+/*
+ * UWB conflicting alien reservations
+ */
+struct uwb_cnflt_alien {
+	struct uwb_rc *rc;
+	struct list_head rc_node;
+	struct uwb_mas_bm mas;
+	struct timer_list timer;
+	struct work_struct cnflt_update_work;
+};
+
+enum uwb_uwb_rsv_alloc_result {
+	UWB_RSV_ALLOC_FOUND = 0,
+	UWB_RSV_ALLOC_NOT_FOUND,
+};
+
+enum uwb_rsv_mas_status {
+	UWB_RSV_MAS_NOT_AVAIL = 1,
+	UWB_RSV_MAS_SAFE,
+	UWB_RSV_MAS_UNSAFE,
+};
+
+struct uwb_rsv_col_set_info {
+	unsigned char start_col;
+	unsigned char interval;
+	unsigned char safe_mas_per_col;
+	unsigned char unsafe_mas_per_col;
+};
+
+struct uwb_rsv_col_info {
+	unsigned char max_avail_safe;
+	unsigned char max_avail_unsafe;
+	unsigned char highest_mas[UWB_MAS_PER_ZONE];
+	struct uwb_rsv_col_set_info csi;
+};
+
+struct uwb_rsv_row_info {
+	unsigned char avail[UWB_MAS_PER_ZONE];
+	unsigned char free_rows;
+	unsigned char used_rows;
+};
+
+/*
+ * UWB find allocation
+ */
+struct uwb_rsv_alloc_info {
+	unsigned char bm[UWB_MAS_PER_ZONE * UWB_NUM_ZONES];
+	struct uwb_rsv_col_info ci[UWB_NUM_ZONES];
+	struct uwb_rsv_row_info ri;
+	struct uwb_mas_bm *not_available;
+	struct uwb_mas_bm *result;
+	int min_mas;
+	int max_mas;
+	int max_interval;
+	int total_allocated_mases;
+	int safe_allocated_mases;
+	int unsafe_allocated_mases;
+	int interval;
+};
 
+int uwb_rsv_find_best_allocation(struct uwb_rsv *rsv, struct uwb_mas_bm *available, 
+				 struct uwb_mas_bm *result);
+void uwb_rsv_handle_drp_avail_change(struct uwb_rc *rc);
 /*
  * UWB Events & management daemon
  */
@@ -254,18 +322,28 @@ void uwb_rsv_init(struct uwb_rc *rc);
 int uwb_rsv_setup(struct uwb_rc *rc);
 void uwb_rsv_cleanup(struct uwb_rc *rc);
 void uwb_rsv_remove_all(struct uwb_rc *rc);
+void uwb_rsv_get(struct uwb_rsv *rsv);
+void uwb_rsv_put(struct uwb_rsv *rsv);
+bool uwb_rsv_has_two_drp_ies(struct uwb_rsv *rsv);
+void uwb_rsv_dump(char *text, struct uwb_rsv *rsv);
+int uwb_rsv_try_move(struct uwb_rsv *rsv, struct uwb_mas_bm *available);
+void uwb_rsv_backoff_win_timer(unsigned long arg);
+void uwb_rsv_backoff_win_increment(struct uwb_rc *rc);
+int uwb_rsv_status(struct uwb_rsv *rsv);
+int uwb_rsv_companion_status(struct uwb_rsv *rsv);
 
 void uwb_rsv_set_state(struct uwb_rsv *rsv, enum uwb_rsv_state new_state);
 void uwb_rsv_remove(struct uwb_rsv *rsv);
 struct uwb_rsv *uwb_rsv_find(struct uwb_rc *rc, struct uwb_dev *src,
 			     struct uwb_ie_drp *drp_ie);
 void uwb_rsv_sched_update(struct uwb_rc *rc);
+void uwb_rsv_queue_update(struct uwb_rc *rc);
 
-void uwb_drp_handle_timeout(struct uwb_rsv *rsv);
 int uwb_drp_ie_update(struct uwb_rsv *rsv);
 void uwb_drp_ie_to_bm(struct uwb_mas_bm *bm, const struct uwb_ie_drp *drp_ie);
 
 void uwb_drp_avail_init(struct uwb_rc *rc);
+void uwb_drp_available(struct uwb_rc *rc, struct uwb_mas_bm *avail);
 int  uwb_drp_avail_reserve_pending(struct uwb_rc *rc, struct uwb_mas_bm *mas);
 void uwb_drp_avail_reserve(struct uwb_rc *rc, struct uwb_mas_bm *mas);
 void uwb_drp_avail_release(struct uwb_rc *rc, struct uwb_mas_bm *mas);
diff --git a/include/linux/uwb.h b/include/linux/uwb.h
index d7ed5201ade6..c02128991ff7 100644
--- a/include/linux/uwb.h
+++ b/include/linux/uwb.h
@@ -67,6 +67,7 @@ struct uwb_dev {
 	struct uwb_dev_addr dev_addr;
 	int beacon_slot;
 	DECLARE_BITMAP(streams, UWB_NUM_STREAMS);
+	DECLARE_BITMAP(last_availability_bm, UWB_NUM_MAS);
 };
 #define to_uwb_dev(d) container_of(d, struct uwb_dev, dev)
 
@@ -109,6 +110,9 @@ struct uwbd {
  */
 struct uwb_mas_bm {
 	DECLARE_BITMAP(bm, UWB_NUM_MAS);
+	DECLARE_BITMAP(unsafe_bm, UWB_NUM_MAS);
+	int safe;
+	int unsafe;
 };
 
 /**
@@ -134,14 +138,24 @@ struct uwb_mas_bm {
  * FIXME: further target states TBD.
  */
 enum uwb_rsv_state {
-	UWB_RSV_STATE_NONE,
+	UWB_RSV_STATE_NONE = 0,
 	UWB_RSV_STATE_O_INITIATED,
 	UWB_RSV_STATE_O_PENDING,
 	UWB_RSV_STATE_O_MODIFIED,
 	UWB_RSV_STATE_O_ESTABLISHED,
+	UWB_RSV_STATE_O_TO_BE_MOVED,
+	UWB_RSV_STATE_O_MOVE_EXPANDING,
+	UWB_RSV_STATE_O_MOVE_COMBINING,
+	UWB_RSV_STATE_O_MOVE_REDUCING,
 	UWB_RSV_STATE_T_ACCEPTED,
 	UWB_RSV_STATE_T_DENIED,
+	UWB_RSV_STATE_T_CONFLICT,
 	UWB_RSV_STATE_T_PENDING,
+	UWB_RSV_STATE_T_EXPANDING_ACCEPTED,
+	UWB_RSV_STATE_T_EXPANDING_CONFLICT,
+	UWB_RSV_STATE_T_EXPANDING_PENDING,
+	UWB_RSV_STATE_T_EXPANDING_DENIED,
+	UWB_RSV_STATE_T_RESIZED,
 
 	UWB_RSV_STATE_LAST,
 };
@@ -166,6 +180,12 @@ struct uwb_rsv_target {
 	};
 };
 
+struct uwb_rsv_move {
+	struct uwb_mas_bm final_mas;
+	struct uwb_ie_drp *companion_drp_ie;
+	struct uwb_mas_bm companion_mas;
+};
+
 /*
  * Number of streams reserved for reservations targeted at DevAddrs.
  */
@@ -203,6 +223,7 @@ typedef void (*uwb_rsv_cb_f)(struct uwb_rsv *rsv);
  *
  * @status:         negotiation status
  * @stream:         stream index allocated for this reservation
+ * @tiebreaker:     conflict tiebreaker for this reservation
  * @mas:            reserved MAS
  * @drp_ie:         the DRP IE
  * @ie_valid:       true iff the DRP IE matches the reservation parameters
@@ -225,19 +246,22 @@ struct uwb_rsv {
 	enum uwb_drp_type type;
 	int max_mas;
 	int min_mas;
-	int sparsity;
+	int max_interval;
 	bool is_multicast;
 
 	uwb_rsv_cb_f callback;
 	void *pal_priv;
 
 	enum uwb_rsv_state state;
+	bool needs_release_companion_mas;
 	u8 stream;
+	u8 tiebreaker;
 	struct uwb_mas_bm mas;
 	struct uwb_ie_drp *drp_ie;
+	struct uwb_rsv_move mv;
 	bool ie_valid;
 	struct timer_list timer;
-	bool expired;
+	struct work_struct handle_timeout_work;
 };
 
 static const
@@ -279,6 +303,13 @@ struct uwb_drp_avail {
 	bool ie_valid;
 };
 
+struct uwb_drp_backoff_win {
+	u8 window;
+	u8 n;
+	int total_expired;
+	struct timer_list timer;
+	bool can_reserve_extra_mases;
+};
 
 const char *uwb_rsv_state_str(enum uwb_rsv_state state);
 const char *uwb_rsv_type_str(enum uwb_drp_type type);
@@ -294,6 +325,8 @@ void uwb_rsv_terminate(struct uwb_rsv *rsv);
 
 void uwb_rsv_accept(struct uwb_rsv *rsv, uwb_rsv_cb_f cb, void *pal_priv);
 
+void uwb_rsv_get_usable_mas(struct uwb_rsv *orig_rsv, struct uwb_mas_bm *mas);
+
 /**
  * Radio Control Interface instance
  *
@@ -364,12 +397,18 @@ struct uwb_rc {
 
 	struct uwbd uwbd;
 
+	struct uwb_drp_backoff_win bow;
 	struct uwb_drp_avail drp_avail;
 	struct list_head reservations;
+	struct list_head cnflt_alien_list;
+	struct uwb_mas_bm cnflt_alien_bitmap;
 	struct mutex rsvs_mutex;
+	spinlock_t rsvs_lock;
 	struct workqueue_struct *rsv_workq;
-	struct work_struct rsv_update_work;
 
+	struct delayed_work rsv_update_work;
+	struct delayed_work rsv_alien_bp_work;
+	int set_drp_ie_pending;
 	struct mutex ies_mutex;
 	struct uwb_rc_cmd_set_ie *ies;
 	size_t ies_capacity;
diff --git a/include/linux/uwb/debug-cmd.h b/include/linux/uwb/debug-cmd.h
index 07efbe17db53..8da004e25628 100644
--- a/include/linux/uwb/debug-cmd.h
+++ b/include/linux/uwb/debug-cmd.h
@@ -43,7 +43,7 @@ struct uwb_dbg_cmd_rsv_establish {
 	__u8  type;
 	__u16 max_mas;
 	__u16 min_mas;
-	__u8  sparsity;
+	__u8  max_interval;
 };
 
 struct uwb_dbg_cmd_rsv_terminate {
diff --git a/include/linux/uwb/spec.h b/include/linux/uwb/spec.h
index a30436ea53aa..b52e44f1bd33 100644
--- a/include/linux/uwb/spec.h
+++ b/include/linux/uwb/spec.h
@@ -58,6 +58,11 @@ enum { UWB_NUM_ZONES = 16 };
  */
 #define UWB_MAS_PER_ZONE (UWB_NUM_MAS / UWB_NUM_ZONES)
 
+/*
+ * Number of MAS required before a row can be considered available.
+ */
+#define UWB_USABLE_MAS_PER_ROW (UWB_NUM_ZONES - 1)
+
 /*
  * Number of streams per DRP reservation between a pair of devices.
  *
@@ -93,6 +98,26 @@ enum { UWB_BEACON_SLOT_LENGTH_US = 85 };
  */
 enum { UWB_MAX_LOST_BEACONS = 3 };
 
+/*
+ * mDRPBackOffWinMin
+ *
+ * The minimum number of superframes to wait before trying to reserve
+ * extra MAS.
+ *
+ * [ECMA-368] section 17.16
+ */
+enum { UWB_DRP_BACKOFF_WIN_MIN = 2 };
+
+/*
+ * mDRPBackOffWinMax
+ *
+ * The maximum number of superframes to wait before trying to reserve
+ * extra MAS.
+ *
+ * [ECMA-368] section 17.16
+ */
+enum { UWB_DRP_BACKOFF_WIN_MAX = 16 };
+
 /*
  * Length of a superframe in microseconds.
  */
-- 
cgit v1.2.3


From 29c0177e6a4ac094302bed54a1d4bbb6b740a9ef Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Sat, 13 Dec 2008 21:20:25 +1030
Subject: cpumask: change cpumask_scnprintf, cpumask_parse_user, cpulist_parse,
 and cpulist_scnprintf to take pointers.

Impact: change calling convention of existing cpumask APIs

Most cpumask functions started with cpus_: these have been replaced by
cpumask_ ones which take struct cpumask pointers as expected.

These four functions don't have good replacement names; fortunately
they're rarely used, so we just change them over.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: paulus@samba.org
Cc: mingo@redhat.com
Cc: tony.luck@intel.com
Cc: ralf@linux-mips.org
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Cc: cl@linux-foundation.org
Cc: srostedt@redhat.com
---
 arch/ia64/kernel/topology.c           |  2 +-
 arch/mips/kernel/smp-cmp.c            |  4 +-
 arch/powerpc/platforms/pseries/xics.c |  2 +-
 arch/x86/kernel/cpu/intel_cacheinfo.c |  4 +-
 arch/x86/kernel/setup_percpu.c        |  2 +-
 drivers/base/cpu.c                    |  2 +-
 drivers/base/node.c                   |  4 +-
 drivers/base/topology.c               |  4 +-
 drivers/pci/pci-sysfs.c               |  4 +-
 drivers/pci/probe.c                   |  4 +-
 include/linux/cpumask.h               | 87 +++++++++++++++++++++++------------
 kernel/cpuset.c                       |  4 +-
 kernel/irq/proc.c                     |  4 +-
 kernel/profile.c                      |  4 +-
 kernel/sched.c                        |  4 +-
 kernel/sched_stats.h                  |  2 +-
 kernel/taskstats.c                    |  2 +-
 kernel/trace/trace.c                  |  4 +-
 mm/slub.c                             |  2 +-
 19 files changed, 86 insertions(+), 59 deletions(-)

(limited to 'include/linux')

diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c
index c75b914f2d6b..a8d61a3e9a94 100644
--- a/arch/ia64/kernel/topology.c
+++ b/arch/ia64/kernel/topology.c
@@ -219,7 +219,7 @@ static ssize_t show_shared_cpu_map(struct cache_info *this_leaf, char *buf)
 	cpumask_t shared_cpu_map;
 
 	cpus_and(shared_cpu_map, this_leaf->shared_cpu_map, cpu_online_map);
-	len = cpumask_scnprintf(buf, NR_CPUS+1, shared_cpu_map);
+	len = cpumask_scnprintf(buf, NR_CPUS+1, &shared_cpu_map);
 	len += sprintf(buf+len, "\n");
 	return len;
 }
diff --git a/arch/mips/kernel/smp-cmp.c b/arch/mips/kernel/smp-cmp.c
index 6789c1a12120..f27beca4b26d 100644
--- a/arch/mips/kernel/smp-cmp.c
+++ b/arch/mips/kernel/smp-cmp.c
@@ -51,10 +51,10 @@ static int __init allowcpus(char *str)
 	int len;
 
 	cpus_clear(cpu_allow_map);
-	if (cpulist_parse(str, cpu_allow_map) == 0) {
+	if (cpulist_parse(str, &cpu_allow_map) == 0) {
 		cpu_set(0, cpu_allow_map);
 		cpus_and(cpu_possible_map, cpu_possible_map, cpu_allow_map);
-		len = cpulist_scnprintf(buf, sizeof(buf)-1, cpu_possible_map);
+		len = cpulist_scnprintf(buf, sizeof(buf)-1, &cpu_possible_map);
 		buf[len] = '\0';
 		pr_debug("Allowable CPUs: %s\n", buf);
 		return 1;
diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c
index e1904774a70f..64d24310ce7e 100644
--- a/arch/powerpc/platforms/pseries/xics.c
+++ b/arch/powerpc/platforms/pseries/xics.c
@@ -358,7 +358,7 @@ static void xics_set_affinity(unsigned int virq, cpumask_t cpumask)
 	irq_server = get_irq_server(virq, 1);
 	if (irq_server == -1) {
 		char cpulist[128];
-		cpumask_scnprintf(cpulist, sizeof(cpulist), cpumask);
+		cpumask_scnprintf(cpulist, sizeof(cpulist), &cpumask);
 		printk(KERN_WARNING
 			"%s: No online cpus in the mask %s for irq %d\n",
 			__func__, cpulist, virq);
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 3f46afbb1cf1..43ea612d3e9d 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -626,8 +626,8 @@ static ssize_t show_shared_cpu_map_func(struct _cpuid4_info *this_leaf,
 		cpumask_t *mask = &this_leaf->shared_cpu_map;
 
 		n = type?
-			cpulist_scnprintf(buf, len-2, *mask):
-			cpumask_scnprintf(buf, len-2, *mask);
+			cpulist_scnprintf(buf, len-2, mask) :
+			cpumask_scnprintf(buf, len-2, mask);
 		buf[n++] = '\n';
 		buf[n] = '\0';
 	}
diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c
index ae0c0d3bb770..1c2084291f97 100644
--- a/arch/x86/kernel/setup_percpu.c
+++ b/arch/x86/kernel/setup_percpu.c
@@ -282,7 +282,7 @@ static void __cpuinit numa_set_cpumask(int cpu, int enable)
 	else
 		cpu_clear(cpu, *mask);
 
-	cpulist_scnprintf(buf, sizeof(buf), *mask);
+	cpulist_scnprintf(buf, sizeof(buf), mask);
 	printk(KERN_DEBUG "%s cpu %d node %d: mask now %s\n",
 		enable? "numa_add_cpu":"numa_remove_cpu", cpu, node, buf);
  }
diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c
index 64f5d54f7edc..4259072f5bd0 100644
--- a/drivers/base/cpu.c
+++ b/drivers/base/cpu.c
@@ -109,7 +109,7 @@ static SYSDEV_ATTR(crash_notes, 0400, show_crash_notes, NULL);
  */
 static ssize_t print_cpus_map(char *buf, cpumask_t *map)
 {
-	int n = cpulist_scnprintf(buf, PAGE_SIZE-2, *map);
+	int n = cpulist_scnprintf(buf, PAGE_SIZE-2, map);
 
 	buf[n++] = '\n';
 	buf[n] = '\0';
diff --git a/drivers/base/node.c b/drivers/base/node.c
index f5207090885a..91636cd8b6c9 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -30,8 +30,8 @@ static ssize_t node_read_cpumap(struct sys_device *dev, int type, char *buf)
 	BUILD_BUG_ON((NR_CPUS/32 * 9) > (PAGE_SIZE-1));
 
 	len = type?
-		cpulist_scnprintf(buf, PAGE_SIZE-2, *mask):
-		cpumask_scnprintf(buf, PAGE_SIZE-2, *mask);
+		cpulist_scnprintf(buf, PAGE_SIZE-2, mask) :
+		cpumask_scnprintf(buf, PAGE_SIZE-2, mask);
  	buf[len++] = '\n';
  	buf[len] = '\0';
 	return len;
diff --git a/drivers/base/topology.c b/drivers/base/topology.c
index 199cd97e32e6..a8bc1cbcfa7c 100644
--- a/drivers/base/topology.c
+++ b/drivers/base/topology.c
@@ -49,8 +49,8 @@ static ssize_t show_cpumap(int type, cpumask_t *mask, char *buf)
 
 	if (len > 1) {
 		n = type?
-			cpulist_scnprintf(buf, len-2, *mask):
-			cpumask_scnprintf(buf, len-2, *mask);
+			cpulist_scnprintf(buf, len-2, mask) :
+			cpumask_scnprintf(buf, len-2, mask);
 		buf[n++] = '\n';
 		buf[n] = '\0';
 	}
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index 5d72866897a8..c88485860a0a 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -74,7 +74,7 @@ static ssize_t local_cpus_show(struct device *dev,
 	int len;
 
 	mask = pcibus_to_cpumask(to_pci_dev(dev)->bus);
-	len = cpumask_scnprintf(buf, PAGE_SIZE-2, mask);
+	len = cpumask_scnprintf(buf, PAGE_SIZE-2, &mask);
 	buf[len++] = '\n';
 	buf[len] = '\0';
 	return len;
@@ -88,7 +88,7 @@ static ssize_t local_cpulist_show(struct device *dev,
 	int len;
 
 	mask = pcibus_to_cpumask(to_pci_dev(dev)->bus);
-	len = cpulist_scnprintf(buf, PAGE_SIZE-2, mask);
+	len = cpulist_scnprintf(buf, PAGE_SIZE-2, &mask);
 	buf[len++] = '\n';
 	buf[len] = '\0';
 	return len;
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index 003a9b3c293f..5b3f5937ecf5 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -55,8 +55,8 @@ static ssize_t pci_bus_show_cpuaffinity(struct device *dev,
 
 	cpumask = pcibus_to_cpumask(to_pci_bus(dev));
 	ret = type?
-		cpulist_scnprintf(buf, PAGE_SIZE-2, cpumask):
-		cpumask_scnprintf(buf, PAGE_SIZE-2, cpumask);
+		cpulist_scnprintf(buf, PAGE_SIZE-2, &cpumask) :
+		cpumask_scnprintf(buf, PAGE_SIZE-2, &cpumask);
 	buf[ret++] = '\n';
 	buf[ret] = '\0';
 	return ret;
diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 21e1dd43e52a..94a2ab88ae85 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -339,36 +339,6 @@ extern cpumask_t cpu_mask_all;
 #endif
 #define	CPUMASK_PTR(v, m) 	cpumask_t *v = &(m->v)
 
-#define cpumask_scnprintf(buf, len, src) \
-			__cpumask_scnprintf((buf), (len), &(src), NR_CPUS)
-static inline int __cpumask_scnprintf(char *buf, int len,
-					const cpumask_t *srcp, int nbits)
-{
-	return bitmap_scnprintf(buf, len, srcp->bits, nbits);
-}
-
-#define cpumask_parse_user(ubuf, ulen, dst) \
-			__cpumask_parse_user((ubuf), (ulen), &(dst), NR_CPUS)
-static inline int __cpumask_parse_user(const char __user *buf, int len,
-					cpumask_t *dstp, int nbits)
-{
-	return bitmap_parse_user(buf, len, dstp->bits, nbits);
-}
-
-#define cpulist_scnprintf(buf, len, src) \
-			__cpulist_scnprintf((buf), (len), &(src), NR_CPUS)
-static inline int __cpulist_scnprintf(char *buf, int len,
-					const cpumask_t *srcp, int nbits)
-{
-	return bitmap_scnlistprintf(buf, len, srcp->bits, nbits);
-}
-
-#define cpulist_parse(buf, dst) __cpulist_parse((buf), &(dst), NR_CPUS)
-static inline int __cpulist_parse(const char *buf, cpumask_t *dstp, int nbits)
-{
-	return bitmap_parselist(buf, dstp->bits, nbits);
-}
-
 #define cpu_remap(oldbit, old, new) \
 		__cpu_remap((oldbit), &(old), &(new), NR_CPUS)
 static inline int __cpu_remap(int oldbit,
@@ -945,6 +915,63 @@ static inline void cpumask_copy(struct cpumask *dstp,
  */
 #define cpumask_of(cpu) (get_cpu_mask(cpu))
 
+/**
+ * cpumask_scnprintf - print a cpumask into a string as comma-separated hex
+ * @buf: the buffer to sprintf into
+ * @len: the length of the buffer
+ * @srcp: the cpumask to print
+ *
+ * If len is zero, returns zero.  Otherwise returns the length of the
+ * (nul-terminated) @buf string.
+ */
+static inline int cpumask_scnprintf(char *buf, int len,
+				    const struct cpumask *srcp)
+{
+	return bitmap_scnprintf(buf, len, srcp->bits, nr_cpumask_bits);
+}
+
+/**
+ * cpumask_parse_user - extract a cpumask from a user string
+ * @buf: the buffer to extract from
+ * @len: the length of the buffer
+ * @dstp: the cpumask to set.
+ *
+ * Returns -errno, or 0 for success.
+ */
+static inline int cpumask_parse_user(const char __user *buf, int len,
+				     struct cpumask *dstp)
+{
+	return bitmap_parse_user(buf, len, dstp->bits, nr_cpumask_bits);
+}
+
+/**
+ * cpulist_scnprintf - print a cpumask into a string as comma-separated list
+ * @buf: the buffer to sprintf into
+ * @len: the length of the buffer
+ * @srcp: the cpumask to print
+ *
+ * If len is zero, returns zero.  Otherwise returns the length of the
+ * (nul-terminated) @buf string.
+ */
+static inline int cpulist_scnprintf(char *buf, int len,
+				    const struct cpumask *srcp)
+{
+	return bitmap_scnlistprintf(buf, len, srcp->bits, nr_cpumask_bits);
+}
+
+/**
+ * cpulist_parse_user - extract a cpumask from a user string of ranges
+ * @buf: the buffer to extract from
+ * @len: the length of the buffer
+ * @dstp: the cpumask to set.
+ *
+ * Returns -errno, or 0 for success.
+ */
+static inline int cpulist_parse(const char *buf, struct cpumask *dstp)
+{
+	return bitmap_parselist(buf, dstp->bits, nr_cpumask_bits);
+}
+
 /**
  * to_cpumask - convert an NR_CPUS bitmap to a struct cpumask *
  * @bitmap: the bitmap
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 96c0ba13b8cd..39c1a4c1c5a9 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -896,7 +896,7 @@ static int update_cpumask(struct cpuset *cs, const char *buf)
 	if (!*buf) {
 		cpus_clear(trialcs.cpus_allowed);
 	} else {
-		retval = cpulist_parse(buf, trialcs.cpus_allowed);
+		retval = cpulist_parse(buf, &trialcs.cpus_allowed);
 		if (retval < 0)
 			return retval;
 
@@ -1482,7 +1482,7 @@ static int cpuset_sprintf_cpulist(char *page, struct cpuset *cs)
 	mask = cs->cpus_allowed;
 	mutex_unlock(&callback_mutex);
 
-	return cpulist_scnprintf(page, PAGE_SIZE, mask);
+	return cpulist_scnprintf(page, PAGE_SIZE, &mask);
 }
 
 static int cpuset_sprintf_memlist(char *page, struct cpuset *cs)
diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c
index d257e7d6a8a4..f293349d49d0 100644
--- a/kernel/irq/proc.c
+++ b/kernel/irq/proc.c
@@ -47,7 +47,7 @@ static ssize_t irq_affinity_proc_write(struct file *file,
 	    irq_balancing_disabled(irq))
 		return -EIO;
 
-	err = cpumask_parse_user(buffer, count, new_value);
+	err = cpumask_parse_user(buffer, count, &new_value);
 	if (err)
 		return err;
 
@@ -95,7 +95,7 @@ static ssize_t default_affinity_write(struct file *file,
 	cpumask_t new_value;
 	int err;
 
-	err = cpumask_parse_user(buffer, count, new_value);
+	err = cpumask_parse_user(buffer, count, &new_value);
 	if (err)
 		return err;
 
diff --git a/kernel/profile.c b/kernel/profile.c
index dc41827fbfee..7d620dfdde59 100644
--- a/kernel/profile.c
+++ b/kernel/profile.c
@@ -442,7 +442,7 @@ void profile_tick(int type)
 static int prof_cpu_mask_read_proc(char *page, char **start, off_t off,
 			int count, int *eof, void *data)
 {
-	int len = cpumask_scnprintf(page, count, *(cpumask_t *)data);
+	int len = cpumask_scnprintf(page, count, (cpumask_t *)data);
 	if (count - len < 2)
 		return -EINVAL;
 	len += sprintf(page + len, "\n");
@@ -456,7 +456,7 @@ static int prof_cpu_mask_write_proc(struct file *file,
 	unsigned long full_count = count, err;
 	cpumask_t new_value;
 
-	err = cpumask_parse_user(buffer, count, new_value);
+	err = cpumask_parse_user(buffer, count, &new_value);
 	if (err)
 		return err;
 
diff --git a/kernel/sched.c b/kernel/sched.c
index e4bb1dd7b308..d2d16d1273b1 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6666,7 +6666,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
 	struct sched_group *group = sd->groups;
 	char str[256];
 
-	cpulist_scnprintf(str, sizeof(str), sd->span);
+	cpulist_scnprintf(str, sizeof(str), &sd->span);
 	cpus_clear(*groupmask);
 
 	printk(KERN_DEBUG "%*s domain %d: ", level, "", level);
@@ -6720,7 +6720,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
 
 		cpus_or(*groupmask, *groupmask, group->cpumask);
 
-		cpulist_scnprintf(str, sizeof(str), group->cpumask);
+		cpulist_scnprintf(str, sizeof(str), &group->cpumask);
 		printk(KERN_CONT " %s", str);
 
 		group = group->next;
diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h
index 7dbf72a2b02c..6beff1e4eeae 100644
--- a/kernel/sched_stats.h
+++ b/kernel/sched_stats.h
@@ -42,7 +42,7 @@ static int show_schedstat(struct seq_file *seq, void *v)
 		for_each_domain(cpu, sd) {
 			enum cpu_idle_type itype;
 
-			cpumask_scnprintf(mask_str, mask_len, sd->span);
+			cpumask_scnprintf(mask_str, mask_len, &sd->span);
 			seq_printf(seq, "domain%d %s", dcount++, mask_str);
 			for (itype = CPU_IDLE; itype < CPU_MAX_IDLE_TYPES;
 					itype++) {
diff --git a/kernel/taskstats.c b/kernel/taskstats.c
index bd6be76303cf..6d7dc4ec4aa5 100644
--- a/kernel/taskstats.c
+++ b/kernel/taskstats.c
@@ -352,7 +352,7 @@ static int parse(struct nlattr *na, cpumask_t *mask)
 	if (!data)
 		return -ENOMEM;
 	nla_strlcpy(data, na, len);
-	ret = cpulist_parse(data, *mask);
+	ret = cpulist_parse(data, mask);
 	kfree(data);
 	return ret;
 }
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index d86e3252f300..d2e75479dc50 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -2126,7 +2126,7 @@ tracing_cpumask_read(struct file *filp, char __user *ubuf,
 
 	mutex_lock(&tracing_cpumask_update_lock);
 
-	len = cpumask_scnprintf(mask_str, count, tracing_cpumask);
+	len = cpumask_scnprintf(mask_str, count, &tracing_cpumask);
 	if (count - len < 2) {
 		count = -EINVAL;
 		goto out_err;
@@ -2147,7 +2147,7 @@ tracing_cpumask_write(struct file *filp, const char __user *ubuf,
 	int err, cpu;
 
 	mutex_lock(&tracing_cpumask_update_lock);
-	err = cpumask_parse_user(ubuf, count, tracing_cpumask_new);
+	err = cpumask_parse_user(ubuf, count, &tracing_cpumask_new);
 	if (err)
 		goto err_unlock;
 
diff --git a/mm/slub.c b/mm/slub.c
index a2cd47d89e0a..8e516e29f989 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -3626,7 +3626,7 @@ static int list_locations(struct kmem_cache *s, char *buf,
 				len < PAGE_SIZE - 60) {
 			len += sprintf(buf + len, " cpus=");
 			len += cpulist_scnprintf(buf + len, PAGE_SIZE - len - 50,
-					l->cpus);
+					&l->cpus);
 		}
 
 		if (num_online_nodes() > 1 && !nodes_empty(l->nodes) &&
-- 
cgit v1.2.3


From 0de26520c7cabf36e1de090ea8092f011a6106ce Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Sat, 13 Dec 2008 21:20:26 +1030
Subject: cpumask: make irq_set_affinity() take a const struct cpumask

Impact: change existing irq_chip API

Not much point with gentle transition here: the struct irq_chip's
setaffinity method signature needs to change.

Fortunately, not widely used code, but hits a few architectures.

Note: In irq_select_affinity() I save a temporary in by mangling
irq_desc[irq].affinity directly.  Ingo, does this break anything?

(Folded in fix from KOSAKI Motohiro)

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Reviewed-by: Grant Grundler <grundler@parisc-linux.org>
Acked-by: Ingo Molnar <mingo@redhat.com>
Cc: ralf@linux-mips.org
Cc: grundler@parisc-linux.org
Cc: jeremy@xensource.com
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
---
 arch/alpha/kernel/irq.c               |  2 +-
 arch/alpha/kernel/sys_dp264.c         |  8 ++--
 arch/alpha/kernel/sys_titan.c         |  4 +-
 arch/arm/common/gic.c                 |  4 +-
 arch/arm/kernel/irq.c                 |  2 +-
 arch/arm/oprofile/op_model_mpcore.c   |  4 +-
 arch/cris/arch-v32/kernel/irq.c       |  4 +-
 arch/ia64/hp/sim/hpsim_irq.c          |  2 +-
 arch/ia64/kernel/iosapic.c            | 12 +++---
 arch/ia64/kernel/irq.c                |  9 ++--
 arch/ia64/kernel/msi_ia64.c           | 12 +++---
 arch/ia64/kernel/smpboot.c            |  4 +-
 arch/ia64/sn/kernel/irq.c             |  6 +--
 arch/ia64/sn/kernel/msi_sn.c          |  7 +--
 arch/mips/include/asm/irq.h           |  3 +-
 arch/mips/kernel/cevt-bcm1480.c       |  2 +-
 arch/mips/kernel/cevt-sb1250.c        |  2 +-
 arch/mips/kernel/irq-gic.c            |  6 +--
 arch/mips/mti-malta/malta-smtc.c      |  6 +--
 arch/mips/sibyte/bcm1480/irq.c        |  8 ++--
 arch/mips/sibyte/sb1250/irq.c         |  8 ++--
 arch/parisc/kernel/irq.c              |  6 +--
 arch/powerpc/kernel/irq.c             |  2 +-
 arch/powerpc/platforms/pseries/xics.c |  6 +--
 arch/powerpc/sysdev/mpic.c            |  4 +-
 arch/powerpc/sysdev/mpic.h            |  2 +-
 arch/sparc64/kernel/irq.c             | 11 +++--
 arch/sparc64/kernel/of_device.c       |  2 +-
 arch/sparc64/kernel/pci_msi.c         |  2 +-
 arch/x86/kernel/hpet.c                |  4 +-
 arch/x86/kernel/io_apic.c             | 81 +++++++++++++++++------------------
 arch/x86/kernel/irq_32.c              |  2 +-
 arch/x86/kernel/irq_64.c              |  2 +-
 drivers/parisc/iosapic.c              |  7 +--
 drivers/xen/events.c                  |  6 +--
 include/linux/interrupt.h             |  4 +-
 include/linux/irq.h                   |  3 +-
 kernel/irq/chip.c                     |  2 +-
 kernel/irq/manage.c                   | 22 +++++-----
 kernel/irq/migration.c                | 14 +++---
 kernel/irq/proc.c                     | 29 ++++++++-----
 kernel/time/tick-common.c             |  6 +--
 42 files changed, 171 insertions(+), 161 deletions(-)

(limited to 'include/linux')

diff --git a/arch/alpha/kernel/irq.c b/arch/alpha/kernel/irq.c
index c626a821cdcb..d0f1620007f7 100644
--- a/arch/alpha/kernel/irq.c
+++ b/arch/alpha/kernel/irq.c
@@ -55,7 +55,7 @@ int irq_select_affinity(unsigned int irq)
 	last_cpu = cpu;
 
 	irq_desc[irq].affinity = cpumask_of_cpu(cpu);
-	irq_desc[irq].chip->set_affinity(irq, cpumask_of_cpu(cpu));
+	irq_desc[irq].chip->set_affinity(irq, cpumask_of(cpu));
 	return 0;
 }
 #endif /* CONFIG_SMP */
diff --git a/arch/alpha/kernel/sys_dp264.c b/arch/alpha/kernel/sys_dp264.c
index c71b0fd7a61f..ab44c164d9d4 100644
--- a/arch/alpha/kernel/sys_dp264.c
+++ b/arch/alpha/kernel/sys_dp264.c
@@ -177,19 +177,19 @@ cpu_set_irq_affinity(unsigned int irq, cpumask_t affinity)
 }
 
 static void
-dp264_set_affinity(unsigned int irq, cpumask_t affinity)
+dp264_set_affinity(unsigned int irq, const struct cpumask *affinity)
 { 
 	spin_lock(&dp264_irq_lock);
-	cpu_set_irq_affinity(irq, affinity);
+	cpu_set_irq_affinity(irq, *affinity);
 	tsunami_update_irq_hw(cached_irq_mask);
 	spin_unlock(&dp264_irq_lock);
 }
 
 static void
-clipper_set_affinity(unsigned int irq, cpumask_t affinity)
+clipper_set_affinity(unsigned int irq, const struct cpumask *affinity)
 { 
 	spin_lock(&dp264_irq_lock);
-	cpu_set_irq_affinity(irq - 16, affinity);
+	cpu_set_irq_affinity(irq - 16, *affinity);
 	tsunami_update_irq_hw(cached_irq_mask);
 	spin_unlock(&dp264_irq_lock);
 }
diff --git a/arch/alpha/kernel/sys_titan.c b/arch/alpha/kernel/sys_titan.c
index 52c91ccc1648..27f840a4ad3d 100644
--- a/arch/alpha/kernel/sys_titan.c
+++ b/arch/alpha/kernel/sys_titan.c
@@ -158,10 +158,10 @@ titan_cpu_set_irq_affinity(unsigned int irq, cpumask_t affinity)
 }
 
 static void
-titan_set_irq_affinity(unsigned int irq, cpumask_t affinity)
+titan_set_irq_affinity(unsigned int irq, const struct cpumask *affinity)
 { 
 	spin_lock(&titan_irq_lock);
-	titan_cpu_set_irq_affinity(irq - 16, affinity);
+	titan_cpu_set_irq_affinity(irq - 16, *affinity);
 	titan_update_irq_hw(titan_cached_irq_mask);
 	spin_unlock(&titan_irq_lock);
 }
diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c
index 7fc9860a97d7..c6884ba1d5ed 100644
--- a/arch/arm/common/gic.c
+++ b/arch/arm/common/gic.c
@@ -109,11 +109,11 @@ static void gic_unmask_irq(unsigned int irq)
 }
 
 #ifdef CONFIG_SMP
-static void gic_set_cpu(unsigned int irq, cpumask_t mask_val)
+static void gic_set_cpu(unsigned int irq, const struct cpumask *mask_val)
 {
 	void __iomem *reg = gic_dist_base(irq) + GIC_DIST_TARGET + (gic_irq(irq) & ~3);
 	unsigned int shift = (irq % 4) * 8;
-	unsigned int cpu = first_cpu(mask_val);
+	unsigned int cpu = cpumask_first(mask_val);
 	u32 val;
 
 	spin_lock(&irq_controller_lock);
diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c
index 2f3eb795fa6e..7141cee1fab7 100644
--- a/arch/arm/kernel/irq.c
+++ b/arch/arm/kernel/irq.c
@@ -174,7 +174,7 @@ static void route_irq(struct irq_desc *desc, unsigned int irq, unsigned int cpu)
 	pr_debug("IRQ%u: moving from cpu%u to cpu%u\n", irq, desc->cpu, cpu);
 
 	spin_lock_irq(&desc->lock);
-	desc->chip->set_affinity(irq, cpumask_of_cpu(cpu));
+	desc->chip->set_affinity(irq, cpumask_of(cpu));
 	spin_unlock_irq(&desc->lock);
 }
 
diff --git a/arch/arm/oprofile/op_model_mpcore.c b/arch/arm/oprofile/op_model_mpcore.c
index 4de366e8b4c5..6d6bd5899240 100644
--- a/arch/arm/oprofile/op_model_mpcore.c
+++ b/arch/arm/oprofile/op_model_mpcore.c
@@ -260,10 +260,10 @@ static void em_stop(void)
 static void em_route_irq(int irq, unsigned int cpu)
 {
 	struct irq_desc *desc = irq_desc + irq;
-	cpumask_t mask = cpumask_of_cpu(cpu);
+	const struct cpumask *mask = cpumask_of(cpu);
 
 	spin_lock_irq(&desc->lock);
-	desc->affinity = mask;
+	desc->affinity = *mask;
 	desc->chip->set_affinity(irq, mask);
 	spin_unlock_irq(&desc->lock);
 }
diff --git a/arch/cris/arch-v32/kernel/irq.c b/arch/cris/arch-v32/kernel/irq.c
index 173c141ac9ba..295131fee710 100644
--- a/arch/cris/arch-v32/kernel/irq.c
+++ b/arch/cris/arch-v32/kernel/irq.c
@@ -325,11 +325,11 @@ static void end_crisv32_irq(unsigned int irq)
 {
 }
 
-void set_affinity_crisv32_irq(unsigned int irq, cpumask_t dest)
+void set_affinity_crisv32_irq(unsigned int irq, const struct cpumask *dest)
 {
 	unsigned long flags;
 	spin_lock_irqsave(&irq_lock, flags);
-	irq_allocations[irq - FIRST_IRQ].mask = dest;
+	irq_allocations[irq - FIRST_IRQ].mask = *dest;
 	spin_unlock_irqrestore(&irq_lock, flags);
 }
 
diff --git a/arch/ia64/hp/sim/hpsim_irq.c b/arch/ia64/hp/sim/hpsim_irq.c
index c2f58ff364e7..cc0a3182db3c 100644
--- a/arch/ia64/hp/sim/hpsim_irq.c
+++ b/arch/ia64/hp/sim/hpsim_irq.c
@@ -22,7 +22,7 @@ hpsim_irq_noop (unsigned int irq)
 }
 
 static void
-hpsim_set_affinity_noop (unsigned int a, cpumask_t b)
+hpsim_set_affinity_noop(unsigned int a, const struct cpumask *b)
 {
 }
 
diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c
index 5c4674ae8aea..c8adecd5b416 100644
--- a/arch/ia64/kernel/iosapic.c
+++ b/arch/ia64/kernel/iosapic.c
@@ -330,25 +330,25 @@ unmask_irq (unsigned int irq)
 
 
 static void
-iosapic_set_affinity (unsigned int irq, cpumask_t mask)
+iosapic_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 #ifdef CONFIG_SMP
 	u32 high32, low32;
-	int dest, rte_index;
+	int cpu, dest, rte_index;
 	int redir = (irq & IA64_IRQ_REDIRECTED) ? 1 : 0;
 	struct iosapic_rte_info *rte;
 	struct iosapic *iosapic;
 
 	irq &= (~IA64_IRQ_REDIRECTED);
 
-	cpus_and(mask, mask, cpu_online_map);
-	if (cpus_empty(mask))
+	cpu = cpumask_first_and(cpu_online_mask, mask);
+	if (cpu >= nr_cpu_ids)
 		return;
 
-	if (irq_prepare_move(irq, first_cpu(mask)))
+	if (irq_prepare_move(irq, cpu))
 		return;
 
-	dest = cpu_physical_id(first_cpu(mask));
+	dest = cpu_physical_id(cpu);
 
 	if (!iosapic_intr_info[irq].count)
 		return;			/* not an IOSAPIC interrupt */
diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c
index 7fd18f54c056..0b6db53fedcf 100644
--- a/arch/ia64/kernel/irq.c
+++ b/arch/ia64/kernel/irq.c
@@ -133,7 +133,6 @@ unsigned int vectors_in_migration[NR_IRQS];
  */
 static void migrate_irqs(void)
 {
-	cpumask_t	mask;
 	irq_desc_t *desc;
 	int 		irq, new_cpu;
 
@@ -152,15 +151,14 @@ static void migrate_irqs(void)
 		if (desc->status == IRQ_PER_CPU)
 			continue;
 
-		cpus_and(mask, irq_desc[irq].affinity, cpu_online_map);
-		if (any_online_cpu(mask) == NR_CPUS) {
+		if (cpumask_any_and(&irq_desc[irq].affinity, cpu_online_mask)
+		    >= nr_cpu_ids) {
 			/*
 			 * Save it for phase 2 processing
 			 */
 			vectors_in_migration[irq] = irq;
 
 			new_cpu = any_online_cpu(cpu_online_map);
-			mask = cpumask_of_cpu(new_cpu);
 
 			/*
 			 * Al three are essential, currently WARN_ON.. maybe panic?
@@ -168,7 +166,8 @@ static void migrate_irqs(void)
 			if (desc->chip && desc->chip->disable &&
 				desc->chip->enable && desc->chip->set_affinity) {
 				desc->chip->disable(irq);
-				desc->chip->set_affinity(irq, mask);
+				desc->chip->set_affinity(irq,
+							 cpumask_of(new_cpu));
 				desc->chip->enable(irq);
 			} else {
 				WARN_ON((!(desc->chip) || !(desc->chip->disable) ||
diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c
index 702a09c13238..890339339035 100644
--- a/arch/ia64/kernel/msi_ia64.c
+++ b/arch/ia64/kernel/msi_ia64.c
@@ -49,11 +49,12 @@
 static struct irq_chip	ia64_msi_chip;
 
 #ifdef CONFIG_SMP
-static void ia64_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask)
+static void ia64_set_msi_irq_affinity(unsigned int irq,
+				      const cpumask_t *cpu_mask)
 {
 	struct msi_msg msg;
 	u32 addr, data;
-	int cpu = first_cpu(cpu_mask);
+	int cpu = first_cpu(*cpu_mask);
 
 	if (!cpu_online(cpu))
 		return;
@@ -166,12 +167,11 @@ void arch_teardown_msi_irq(unsigned int irq)
 
 #ifdef CONFIG_DMAR
 #ifdef CONFIG_SMP
-static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask)
+static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg = irq_cfg + irq;
 	struct msi_msg msg;
-	int cpu = first_cpu(mask);
-
+	int cpu = cpumask_first(mask);
 
 	if (!cpu_online(cpu))
 		return;
@@ -187,7 +187,7 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask)
 	msg.address_lo |= MSI_ADDR_DESTID_CPU(cpu_physical_id(cpu));
 
 	dmar_msi_write(irq, &msg);
-	irq_desc[irq].affinity = mask;
+	irq_desc[irq].affinity = *mask;
 }
 #endif /* CONFIG_SMP */
 
diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c
index 4ede6e571c38..11463994a7d5 100644
--- a/arch/ia64/kernel/smpboot.c
+++ b/arch/ia64/kernel/smpboot.c
@@ -682,7 +682,7 @@ int migrate_platform_irqs(unsigned int cpu)
 {
 	int new_cpei_cpu;
 	irq_desc_t *desc = NULL;
-	cpumask_t 	mask;
+	const struct cpumask *mask;
 	int 		retval = 0;
 
 	/*
@@ -695,7 +695,7 @@ int migrate_platform_irqs(unsigned int cpu)
 			 * Now re-target the CPEI to a different processor
 			 */
 			new_cpei_cpu = any_online_cpu(cpu_online_map);
-			mask = cpumask_of_cpu(new_cpei_cpu);
+			mask = cpumask_of(new_cpei_cpu);
 			set_cpei_target_cpu(new_cpei_cpu);
 			desc = irq_desc + ia64_cpe_irq;
 			/*
diff --git a/arch/ia64/sn/kernel/irq.c b/arch/ia64/sn/kernel/irq.c
index 0c66dbdd1d72..66fd705e82c0 100644
--- a/arch/ia64/sn/kernel/irq.c
+++ b/arch/ia64/sn/kernel/irq.c
@@ -227,14 +227,14 @@ finish_up:
 	return new_irq_info;
 }
 
-static void sn_set_affinity_irq(unsigned int irq, cpumask_t mask)
+static void sn_set_affinity_irq(unsigned int irq, const struct cpumask *mask)
 {
 	struct sn_irq_info *sn_irq_info, *sn_irq_info_safe;
 	nasid_t nasid;
 	int slice;
 
-	nasid = cpuid_to_nasid(first_cpu(mask));
-	slice = cpuid_to_slice(first_cpu(mask));
+	nasid = cpuid_to_nasid(cpumask_first(mask));
+	slice = cpuid_to_slice(cpumask_first(mask));
 
 	list_for_each_entry_safe(sn_irq_info, sn_irq_info_safe,
 				 sn_irq_lh[irq], list)
diff --git a/arch/ia64/sn/kernel/msi_sn.c b/arch/ia64/sn/kernel/msi_sn.c
index 83f190ffe350..ca553b0429ce 100644
--- a/arch/ia64/sn/kernel/msi_sn.c
+++ b/arch/ia64/sn/kernel/msi_sn.c
@@ -151,7 +151,8 @@ int sn_setup_msi_irq(struct pci_dev *pdev, struct msi_desc *entry)
 }
 
 #ifdef CONFIG_SMP
-static void sn_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask)
+static void sn_set_msi_irq_affinity(unsigned int irq,
+				    const struct cpumask *cpu_mask)
 {
 	struct msi_msg msg;
 	int slice;
@@ -164,7 +165,7 @@ static void sn_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask)
 	struct sn_pcibus_provider *provider;
 	unsigned int cpu;
 
-	cpu = first_cpu(cpu_mask);
+	cpu = cpumask_first(cpu_mask);
 	sn_irq_info = sn_msi_info[irq].sn_irq_info;
 	if (sn_irq_info == NULL || sn_irq_info->irq_int_bit >= 0)
 		return;
@@ -204,7 +205,7 @@ static void sn_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask)
 	msg.address_lo = (u32)(bus_addr & 0x00000000ffffffff);
 
 	write_msi_msg(irq, &msg);
-	irq_desc[irq].affinity = cpu_mask;
+	irq_desc[irq].affinity = *cpu_mask;
 }
 #endif /* CONFIG_SMP */
 
diff --git a/arch/mips/include/asm/irq.h b/arch/mips/include/asm/irq.h
index a58f0eecc68f..abc62aa744ac 100644
--- a/arch/mips/include/asm/irq.h
+++ b/arch/mips/include/asm/irq.h
@@ -49,7 +49,8 @@ static inline void smtc_im_ack_irq(unsigned int irq)
 #ifdef CONFIG_MIPS_MT_SMTC_IRQAFF
 #include <linux/cpumask.h>
 
-extern void plat_set_irq_affinity(unsigned int irq, cpumask_t affinity);
+extern void plat_set_irq_affinity(unsigned int irq,
+				  const struct cpumask *affinity);
 extern void smtc_forward_irq(unsigned int irq);
 
 /*
diff --git a/arch/mips/kernel/cevt-bcm1480.c b/arch/mips/kernel/cevt-bcm1480.c
index 0a57f86945f1..d7e21bc8cd21 100644
--- a/arch/mips/kernel/cevt-bcm1480.c
+++ b/arch/mips/kernel/cevt-bcm1480.c
@@ -148,6 +148,6 @@ void __cpuinit sb1480_clockevent_init(void)
 	action->name	= name;
 	action->dev_id	= cd;
 
-	irq_set_affinity(irq, cpumask_of_cpu(cpu));
+	irq_set_affinity(irq, cpumask_of(cpu));
 	setup_irq(irq, action);
 }
diff --git a/arch/mips/kernel/cevt-sb1250.c b/arch/mips/kernel/cevt-sb1250.c
index 63ac3ad462bc..0f188cd46e03 100644
--- a/arch/mips/kernel/cevt-sb1250.c
+++ b/arch/mips/kernel/cevt-sb1250.c
@@ -147,6 +147,6 @@ void __cpuinit sb1250_clockevent_init(void)
 	action->name	= name;
 	action->dev_id	= cd;
 
-	irq_set_affinity(irq, cpumask_of_cpu(cpu));
+	irq_set_affinity(irq, cpumask_of(cpu));
 	setup_irq(irq, action);
 }
diff --git a/arch/mips/kernel/irq-gic.c b/arch/mips/kernel/irq-gic.c
index f0a4bb19e096..494a49a317e9 100644
--- a/arch/mips/kernel/irq-gic.c
+++ b/arch/mips/kernel/irq-gic.c
@@ -155,7 +155,7 @@ static void gic_unmask_irq(unsigned int irq)
 
 static DEFINE_SPINLOCK(gic_lock);
 
-static void gic_set_affinity(unsigned int irq, cpumask_t cpumask)
+static void gic_set_affinity(unsigned int irq, const struct cpumask *cpumask)
 {
 	cpumask_t	tmp = CPU_MASK_NONE;
 	unsigned long	flags;
@@ -164,7 +164,7 @@ static void gic_set_affinity(unsigned int irq, cpumask_t cpumask)
 	pr_debug(KERN_DEBUG "%s called\n", __func__);
 	irq -= _irqbase;
 
-	cpus_and(tmp, cpumask, cpu_online_map);
+	cpumask_and(&tmp, cpumask, cpu_online_mask);
 	if (cpus_empty(tmp))
 		return;
 
@@ -187,7 +187,7 @@ static void gic_set_affinity(unsigned int irq, cpumask_t cpumask)
 		set_bit(irq, pcpu_masks[first_cpu(tmp)].pcpu_mask);
 
 	}
-	irq_desc[irq].affinity = cpumask;
+	irq_desc[irq].affinity = *cpumask;
 	spin_unlock_irqrestore(&gic_lock, flags);
 
 }
diff --git a/arch/mips/mti-malta/malta-smtc.c b/arch/mips/mti-malta/malta-smtc.c
index f84a46a8ae6e..aabd7274507b 100644
--- a/arch/mips/mti-malta/malta-smtc.c
+++ b/arch/mips/mti-malta/malta-smtc.c
@@ -114,9 +114,9 @@ struct plat_smp_ops msmtc_smp_ops = {
  */
 
 
-void plat_set_irq_affinity(unsigned int irq, cpumask_t affinity)
+void plat_set_irq_affinity(unsigned int irq, const struct cpumask *affinity)
 {
-	cpumask_t tmask = affinity;
+	cpumask_t tmask = *affinity;
 	int cpu = 0;
 	void smtc_set_irq_affinity(unsigned int irq, cpumask_t aff);
 
@@ -139,7 +139,7 @@ void plat_set_irq_affinity(unsigned int irq, cpumask_t affinity)
 	 * be made to forward to an offline "CPU".
 	 */
 
-	for_each_cpu_mask(cpu, affinity) {
+	for_each_cpu(cpu, affinity) {
 		if ((cpu_data[cpu].vpe_id != 0) || !cpu_online(cpu))
 			cpu_clear(cpu, tmask);
 	}
diff --git a/arch/mips/sibyte/bcm1480/irq.c b/arch/mips/sibyte/bcm1480/irq.c
index a35818ed4263..12b465d404df 100644
--- a/arch/mips/sibyte/bcm1480/irq.c
+++ b/arch/mips/sibyte/bcm1480/irq.c
@@ -50,7 +50,7 @@ static void enable_bcm1480_irq(unsigned int irq);
 static void disable_bcm1480_irq(unsigned int irq);
 static void ack_bcm1480_irq(unsigned int irq);
 #ifdef CONFIG_SMP
-static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask);
+static void bcm1480_set_affinity(unsigned int irq, const struct cpumask *mask);
 #endif
 
 #ifdef CONFIG_PCI
@@ -109,7 +109,7 @@ void bcm1480_unmask_irq(int cpu, int irq)
 }
 
 #ifdef CONFIG_SMP
-static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask)
+static void bcm1480_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	int i = 0, old_cpu, cpu, int_on, k;
 	u64 cur_ints;
@@ -117,11 +117,11 @@ static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask)
 	unsigned long flags;
 	unsigned int irq_dirty;
 
-	if (cpus_weight(mask) != 1) {
+	if (cpumask_weight(mask) != 1) {
 		printk("attempted to set irq affinity for irq %d to multiple CPUs\n", irq);
 		return;
 	}
-	i = first_cpu(mask);
+	i = cpumask_first(mask);
 
 	/* Convert logical CPU to physical CPU */
 	cpu = cpu_logical_map(i);
diff --git a/arch/mips/sibyte/sb1250/irq.c b/arch/mips/sibyte/sb1250/irq.c
index a5158483986e..808ac2959b8c 100644
--- a/arch/mips/sibyte/sb1250/irq.c
+++ b/arch/mips/sibyte/sb1250/irq.c
@@ -50,7 +50,7 @@ static void enable_sb1250_irq(unsigned int irq);
 static void disable_sb1250_irq(unsigned int irq);
 static void ack_sb1250_irq(unsigned int irq);
 #ifdef CONFIG_SMP
-static void sb1250_set_affinity(unsigned int irq, cpumask_t mask);
+static void sb1250_set_affinity(unsigned int irq, const struct cpumask *mask);
 #endif
 
 #ifdef CONFIG_SIBYTE_HAS_LDT
@@ -103,16 +103,16 @@ void sb1250_unmask_irq(int cpu, int irq)
 }
 
 #ifdef CONFIG_SMP
-static void sb1250_set_affinity(unsigned int irq, cpumask_t mask)
+static void sb1250_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	int i = 0, old_cpu, cpu, int_on;
 	u64 cur_ints;
 	struct irq_desc *desc = irq_desc + irq;
 	unsigned long flags;
 
-	i = first_cpu(mask);
+	i = cpumask_first(mask);
 
-	if (cpus_weight(mask) > 1) {
+	if (cpumask_weight(mask) > 1) {
 		printk("attempted to set irq affinity for irq %d to multiple CPUs\n", irq);
 		return;
 	}
diff --git a/arch/parisc/kernel/irq.c b/arch/parisc/kernel/irq.c
index 23ef950df008..4cea935e2f99 100644
--- a/arch/parisc/kernel/irq.c
+++ b/arch/parisc/kernel/irq.c
@@ -131,12 +131,12 @@ int cpu_check_affinity(unsigned int irq, cpumask_t *dest)
 	return 0;
 }
 
-static void cpu_set_affinity_irq(unsigned int irq, cpumask_t dest)
+static void cpu_set_affinity_irq(unsigned int irq, const struct cpumask *dest)
 {
-	if (cpu_check_affinity(irq, &dest))
+	if (cpu_check_affinity(irq, dest))
 		return;
 
-	irq_desc[irq].affinity = dest;
+	irq_desc[irq].affinity = *dest;
 }
 #endif
 
diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c
index ac222d0ab12e..23b8b5e36f98 100644
--- a/arch/powerpc/kernel/irq.c
+++ b/arch/powerpc/kernel/irq.c
@@ -237,7 +237,7 @@ void fixup_irqs(cpumask_t map)
 			mask = map;
 		}
 		if (irq_desc[irq].chip->set_affinity)
-			irq_desc[irq].chip->set_affinity(irq, mask);
+			irq_desc[irq].chip->set_affinity(irq, &mask);
 		else if (irq_desc[irq].action && !(warned++))
 			printk("Cannot set affinity for irq %i\n", irq);
 	}
diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c
index 64d24310ce7e..424b335a71c8 100644
--- a/arch/powerpc/platforms/pseries/xics.c
+++ b/arch/powerpc/platforms/pseries/xics.c
@@ -332,7 +332,7 @@ static void xics_eoi_lpar(unsigned int virq)
 	lpar_xirr_info_set((0xff << 24) | irq);
 }
 
-static void xics_set_affinity(unsigned int virq, cpumask_t cpumask)
+static void xics_set_affinity(unsigned int virq, const struct cpumask *cpumask)
 {
 	unsigned int irq;
 	int status;
@@ -358,7 +358,7 @@ static void xics_set_affinity(unsigned int virq, cpumask_t cpumask)
 	irq_server = get_irq_server(virq, 1);
 	if (irq_server == -1) {
 		char cpulist[128];
-		cpumask_scnprintf(cpulist, sizeof(cpulist), &cpumask);
+		cpumask_scnprintf(cpulist, sizeof(cpulist), cpumask);
 		printk(KERN_WARNING
 			"%s: No online cpus in the mask %s for irq %d\n",
 			__func__, cpulist, virq);
@@ -845,7 +845,7 @@ void xics_migrate_irqs_away(void)
 
 		/* Reset affinity to all cpus */
 		irq_desc[virq].affinity = CPU_MASK_ALL;
-		desc->chip->set_affinity(virq, CPU_MASK_ALL);
+		desc->chip->set_affinity(virq, cpu_all_mask);
 unlock:
 		spin_unlock_irqrestore(&desc->lock, flags);
 	}
diff --git a/arch/powerpc/sysdev/mpic.c b/arch/powerpc/sysdev/mpic.c
index 1890fb085cde..5d7f9f0c93c3 100644
--- a/arch/powerpc/sysdev/mpic.c
+++ b/arch/powerpc/sysdev/mpic.c
@@ -817,7 +817,7 @@ static void mpic_end_ipi(unsigned int irq)
 
 #endif /* CONFIG_SMP */
 
-void mpic_set_affinity(unsigned int irq, cpumask_t cpumask)
+void mpic_set_affinity(unsigned int irq, const struct cpumask *cpumask)
 {
 	struct mpic *mpic = mpic_from_irq(irq);
 	unsigned int src = mpic_irq_to_hw(irq);
@@ -829,7 +829,7 @@ void mpic_set_affinity(unsigned int irq, cpumask_t cpumask)
 	} else {
 		cpumask_t tmp;
 
-		cpus_and(tmp, cpumask, cpu_online_map);
+		cpumask_and(&tmp, cpumask, cpu_online_mask);
 
 		mpic_irq_write(src, MPIC_INFO(IRQ_DESTINATION),
 			       mpic_physmask(cpus_addr(tmp)[0]));
diff --git a/arch/powerpc/sysdev/mpic.h b/arch/powerpc/sysdev/mpic.h
index 6209c62a426d..3cef2af10f42 100644
--- a/arch/powerpc/sysdev/mpic.h
+++ b/arch/powerpc/sysdev/mpic.h
@@ -36,6 +36,6 @@ static inline int mpic_pasemi_msi_init(struct mpic *mpic)
 
 extern int mpic_set_irq_type(unsigned int virq, unsigned int flow_type);
 extern void mpic_set_vector(unsigned int virq, unsigned int vector);
-extern void mpic_set_affinity(unsigned int irq, cpumask_t cpumask);
+extern void mpic_set_affinity(unsigned int irq, const struct cpumask *cpumask);
 
 #endif /* _POWERPC_SYSDEV_MPIC_H */
diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c
index 52fc836f464d..4aaf18e83c8c 100644
--- a/arch/sparc64/kernel/irq.c
+++ b/arch/sparc64/kernel/irq.c
@@ -312,7 +312,8 @@ static void sun4u_irq_enable(unsigned int virt_irq)
 	}
 }
 
-static void sun4u_set_affinity(unsigned int virt_irq, cpumask_t mask)
+static void sun4u_set_affinity(unsigned int virt_irq,
+			       const struct cpumask *mask)
 {
 	sun4u_irq_enable(virt_irq);
 }
@@ -362,7 +363,8 @@ static void sun4v_irq_enable(unsigned int virt_irq)
 		       ino, err);
 }
 
-static void sun4v_set_affinity(unsigned int virt_irq, cpumask_t mask)
+static void sun4v_set_affinity(unsigned int virt_irq,
+			       const struct cpumask *mask)
 {
 	unsigned int ino = virt_irq_table[virt_irq].dev_ino;
 	unsigned long cpuid = irq_choose_cpu(virt_irq);
@@ -429,7 +431,8 @@ static void sun4v_virq_enable(unsigned int virt_irq)
 		       dev_handle, dev_ino, err);
 }
 
-static void sun4v_virt_set_affinity(unsigned int virt_irq, cpumask_t mask)
+static void sun4v_virt_set_affinity(unsigned int virt_irq,
+				    const struct cpumask *mask)
 {
 	unsigned long cpuid, dev_handle, dev_ino;
 	int err;
@@ -788,7 +791,7 @@ void fixup_irqs(void)
 		    !(irq_desc[irq].status & IRQ_PER_CPU)) {
 			if (irq_desc[irq].chip->set_affinity)
 				irq_desc[irq].chip->set_affinity(irq,
-					irq_desc[irq].affinity);
+					&irq_desc[irq].affinity);
 		}
 		spin_unlock_irqrestore(&irq_desc[irq].lock, flags);
 	}
diff --git a/arch/sparc64/kernel/of_device.c b/arch/sparc64/kernel/of_device.c
index 0f616ae3246c..df2efb7fc14c 100644
--- a/arch/sparc64/kernel/of_device.c
+++ b/arch/sparc64/kernel/of_device.c
@@ -780,7 +780,7 @@ out:
 	if (nid != -1) {
 		cpumask_t numa_mask = node_to_cpumask(nid);
 
-		irq_set_affinity(irq, numa_mask);
+		irq_set_affinity(irq, &numa_mask);
 	}
 
 	return irq;
diff --git a/arch/sparc64/kernel/pci_msi.c b/arch/sparc64/kernel/pci_msi.c
index 2e680f34f727..0d0cd815e83e 100644
--- a/arch/sparc64/kernel/pci_msi.c
+++ b/arch/sparc64/kernel/pci_msi.c
@@ -288,7 +288,7 @@ static int bringup_one_msi_queue(struct pci_pbm_info *pbm,
 	if (nid != -1) {
 		cpumask_t numa_mask = node_to_cpumask(nid);
 
-		irq_set_affinity(irq, numa_mask);
+		irq_set_affinity(irq, &numa_mask);
 	}
 	err = request_irq(irq, sparc64_msiq_interrupt, 0,
 			  "MSIQ",
diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 067d8de913f6..940f25851e1e 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -301,7 +301,7 @@ static void hpet_set_mode(enum clock_event_mode mode,
 			struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt);
 			hpet_setup_msi_irq(hdev->irq);
 			disable_irq(hdev->irq);
-			irq_set_affinity(hdev->irq, cpumask_of_cpu(hdev->cpu));
+			irq_set_affinity(hdev->irq, cpumask_of(hdev->cpu));
 			enable_irq(hdev->irq);
 		}
 		break;
@@ -449,7 +449,7 @@ static int hpet_setup_irq(struct hpet_dev *dev)
 		return -1;
 
 	disable_irq(dev->irq);
-	irq_set_affinity(dev->irq, cpumask_of_cpu(dev->cpu));
+	irq_set_affinity(dev->irq, cpumask_of(dev->cpu));
 	enable_irq(dev->irq);
 
 	printk(KERN_DEBUG "hpet: %s irq %d for MSI\n",
diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c
index 9043251210fb..1184210e6d0c 100644
--- a/arch/x86/kernel/io_apic.c
+++ b/arch/x86/kernel/io_apic.c
@@ -361,7 +361,8 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, u8 vector)
 
 static int assign_irq_vector(int irq, cpumask_t mask);
 
-static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
+static void set_ioapic_affinity_irq(unsigned int irq,
+				    const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
 	unsigned long flags;
@@ -369,15 +370,14 @@ static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
 	cpumask_t tmp;
 	struct irq_desc *desc;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	if (!cpumask_intersects(mask, cpu_online_mask))
 		return;
 
 	cfg = irq_cfg(irq);
-	if (assign_irq_vector(irq, mask))
+	if (assign_irq_vector(irq, *mask))
 		return;
 
-	cpus_and(tmp, cfg->domain, mask);
+	cpumask_and(&tmp, &cfg->domain, mask);
 	dest = cpu_mask_to_apicid(tmp);
 	/*
 	 * Only the high 8 bits are valid.
@@ -387,7 +387,7 @@ static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
 	desc = irq_to_desc(irq);
 	spin_lock_irqsave(&ioapic_lock, flags);
 	__target_IO_APIC_irq(irq, dest, cfg->vector);
-	desc->affinity = mask;
+	cpumask_copy(&desc->affinity, mask);
 	spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 #endif /* CONFIG_SMP */
@@ -2189,7 +2189,7 @@ static void ir_irq_migration(struct work_struct *work)
 				continue;
 			}
 
-			desc->chip->set_affinity(irq, desc->pending_mask);
+			desc->chip->set_affinity(irq, &desc->pending_mask);
 			spin_unlock_irqrestore(&desc->lock, flags);
 		}
 	}
@@ -2198,18 +2198,19 @@ static void ir_irq_migration(struct work_struct *work)
 /*
  * Migrates the IRQ destination in the process context.
  */
-static void set_ir_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
+static void set_ir_ioapic_affinity_irq(unsigned int irq,
+				       const struct cpumask *mask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 
 	if (desc->status & IRQ_LEVEL) {
 		desc->status |= IRQ_MOVE_PENDING;
-		desc->pending_mask = mask;
+		cpumask_copy(&desc->pending_mask, mask);
 		migrate_irq_remapped_level(irq);
 		return;
 	}
 
-	migrate_ioapic_irq(irq, mask);
+	migrate_ioapic_irq(irq, *mask);
 }
 #endif
 
@@ -3027,7 +3028,7 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
 }
 
 #ifdef CONFIG_SMP
-static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
+static void set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
 	struct msi_msg msg;
@@ -3035,15 +3036,14 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
 	cpumask_t tmp;
 	struct irq_desc *desc;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	if (!cpumask_intersects(mask, cpu_online_mask))
 		return;
 
-	if (assign_irq_vector(irq, mask))
+	if (assign_irq_vector(irq, *mask))
 		return;
 
 	cfg = irq_cfg(irq);
-	cpus_and(tmp, cfg->domain, mask);
+	cpumask_and(&tmp, &cfg->domain, mask);
 	dest = cpu_mask_to_apicid(tmp);
 
 	read_msi_msg(irq, &msg);
@@ -3055,7 +3055,7 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
 
 	write_msi_msg(irq, &msg);
 	desc = irq_to_desc(irq);
-	desc->affinity = mask;
+	cpumask_copy(&desc->affinity, mask);
 }
 
 #ifdef CONFIG_INTR_REMAP
@@ -3063,7 +3063,8 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
  * Migrate the MSI irq to another cpumask. This migration is
  * done in the process context using interrupt-remapping hardware.
  */
-static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
+static void ir_set_msi_irq_affinity(unsigned int irq,
+				    const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
 	unsigned int dest;
@@ -3071,18 +3072,17 @@ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
 	struct irte irte;
 	struct irq_desc *desc;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	if (!cpumask_intersects(mask, cpu_online_mask))
 		return;
 
 	if (get_irte(irq, &irte))
 		return;
 
-	if (assign_irq_vector(irq, mask))
+	if (assign_irq_vector(irq, *mask))
 		return;
 
 	cfg = irq_cfg(irq);
-	cpus_and(tmp, cfg->domain, mask);
+	cpumask_and(&tmp, &cfg->domain, mask);
 	dest = cpu_mask_to_apicid(tmp);
 
 	irte.vector = cfg->vector;
@@ -3106,7 +3106,7 @@ static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
 	}
 
 	desc = irq_to_desc(irq);
-	desc->affinity = mask;
+	cpumask_copy(&desc->affinity, mask);
 }
 #endif
 #endif /* CONFIG_SMP */
@@ -3308,7 +3308,7 @@ void arch_teardown_msi_irq(unsigned int irq)
 
 #ifdef CONFIG_DMAR
 #ifdef CONFIG_SMP
-static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask)
+static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
 	struct msi_msg msg;
@@ -3316,15 +3316,14 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask)
 	cpumask_t tmp;
 	struct irq_desc *desc;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	if (!cpumask_intersects(mask, cpu_online_mask))
 		return;
 
-	if (assign_irq_vector(irq, mask))
+	if (assign_irq_vector(irq, *mask))
 		return;
 
 	cfg = irq_cfg(irq);
-	cpus_and(tmp, cfg->domain, mask);
+	cpumask_and(&tmp, &cfg->domain, mask);
 	dest = cpu_mask_to_apicid(tmp);
 
 	dmar_msi_read(irq, &msg);
@@ -3336,7 +3335,7 @@ static void dmar_msi_set_affinity(unsigned int irq, cpumask_t mask)
 
 	dmar_msi_write(irq, &msg);
 	desc = irq_to_desc(irq);
-	desc->affinity = mask;
+	cpumask_copy(&desc->affinity, mask);
 }
 #endif /* CONFIG_SMP */
 
@@ -3369,7 +3368,7 @@ int arch_setup_dmar_msi(unsigned int irq)
 #ifdef CONFIG_HPET_TIMER
 
 #ifdef CONFIG_SMP
-static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask)
+static void hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
 	struct irq_desc *desc;
@@ -3377,15 +3376,14 @@ static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask)
 	unsigned int dest;
 	cpumask_t tmp;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	if (!cpumask_intersects(mask, cpu_online_mask))
 		return;
 
-	if (assign_irq_vector(irq, mask))
+	if (assign_irq_vector(irq, *mask))
 		return;
 
 	cfg = irq_cfg(irq);
-	cpus_and(tmp, cfg->domain, mask);
+	cpumask_and(&tmp, &cfg->domain, mask);
 	dest = cpu_mask_to_apicid(tmp);
 
 	hpet_msi_read(irq, &msg);
@@ -3397,7 +3395,7 @@ static void hpet_msi_set_affinity(unsigned int irq, cpumask_t mask)
 
 	hpet_msi_write(irq, &msg);
 	desc = irq_to_desc(irq);
-	desc->affinity = mask;
+	cpumask_copy(&desc->affinity, mask);
 }
 #endif /* CONFIG_SMP */
 
@@ -3451,27 +3449,26 @@ static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector)
 	write_ht_irq_msg(irq, &msg);
 }
 
-static void set_ht_irq_affinity(unsigned int irq, cpumask_t mask)
+static void set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
 	struct irq_cfg *cfg;
 	unsigned int dest;
 	cpumask_t tmp;
 	struct irq_desc *desc;
 
-	cpus_and(tmp, mask, cpu_online_map);
-	if (cpus_empty(tmp))
+	if (!cpumask_intersects(mask, cpu_online_mask))
 		return;
 
-	if (assign_irq_vector(irq, mask))
+	if (assign_irq_vector(irq, *mask))
 		return;
 
 	cfg = irq_cfg(irq);
-	cpus_and(tmp, cfg->domain, mask);
+	cpumask_and(&tmp, &cfg->domain, mask);
 	dest = cpu_mask_to_apicid(tmp);
 
 	target_ht_irq(irq, dest, cfg->vector);
 	desc = irq_to_desc(irq);
-	desc->affinity = mask;
+	cpumask_copy(&desc->affinity, mask);
 }
 #endif
 
@@ -3794,10 +3791,10 @@ void __init setup_ioapic_dest(void)
 
 #ifdef CONFIG_INTR_REMAP
 			if (intr_remapping_enabled)
-				set_ir_ioapic_affinity_irq(irq, mask);
+				set_ir_ioapic_affinity_irq(irq, &mask);
 			else
 #endif
-				set_ioapic_affinity_irq(irq, mask);
+				set_ioapic_affinity_irq(irq, &mask);
 		}
 
 	}
diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c
index a51382672de0..87870a49be4e 100644
--- a/arch/x86/kernel/irq_32.c
+++ b/arch/x86/kernel/irq_32.c
@@ -251,7 +251,7 @@ void fixup_irqs(cpumask_t map)
 			mask = map;
 		}
 		if (desc->chip->set_affinity)
-			desc->chip->set_affinity(irq, mask);
+			desc->chip->set_affinity(irq, &mask);
 		else if (desc->action && !(warned++))
 			printk("Cannot set affinity for irq %i\n", irq);
 	}
diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c
index 60eb84eb77a0..7d37f847544d 100644
--- a/arch/x86/kernel/irq_64.c
+++ b/arch/x86/kernel/irq_64.c
@@ -116,7 +116,7 @@ void fixup_irqs(cpumask_t map)
 			desc->chip->mask(irq);
 
 		if (desc->chip->set_affinity)
-			desc->chip->set_affinity(irq, mask);
+			desc->chip->set_affinity(irq, &mask);
 		else if (!(warned++))
 			set_affinity = 0;
 
diff --git a/drivers/parisc/iosapic.c b/drivers/parisc/iosapic.c
index 7beffcab2745..9dedbbd218c3 100644
--- a/drivers/parisc/iosapic.c
+++ b/drivers/parisc/iosapic.c
@@ -704,16 +704,17 @@ static unsigned int iosapic_startup_irq(unsigned int irq)
 }
 
 #ifdef CONFIG_SMP
-static void iosapic_set_affinity_irq(unsigned int irq, cpumask_t dest)
+static void iosapic_set_affinity_irq(unsigned int irq,
+				     const struct cpumask *dest)
 {
 	struct vector_info *vi = iosapic_get_vector(irq);
 	u32 d0, d1, dummy_d0;
 	unsigned long flags;
 
-	if (cpu_check_affinity(irq, &dest))
+	if (cpu_check_affinity(irq, dest))
 		return;
 
-	vi->txn_addr = txn_affinity_addr(irq, first_cpu(dest));
+	vi->txn_addr = txn_affinity_addr(irq, cpumask_first(dest));
 
 	spin_lock_irqsave(&iosapic_lock, flags);
 	/* d1 contains the destination CPU, so only want to set that
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 1e3b934a4cf7..eba5ec5b020e 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -579,7 +579,7 @@ void rebind_evtchn_irq(int evtchn, int irq)
 	spin_unlock(&irq_mapping_update_lock);
 
 	/* new event channels are always bound to cpu 0 */
-	irq_set_affinity(irq, cpumask_of_cpu(0));
+	irq_set_affinity(irq, cpumask_of(0));
 
 	/* Unmask the event channel. */
 	enable_irq(irq);
@@ -608,9 +608,9 @@ static void rebind_irq_to_cpu(unsigned irq, unsigned tcpu)
 }
 
 
-static void set_affinity_irq(unsigned irq, cpumask_t dest)
+static void set_affinity_irq(unsigned irq, const struct cpumask *dest)
 {
-	unsigned tcpu = first_cpu(dest);
+	unsigned tcpu = cpumask_first(dest);
 	rebind_irq_to_cpu(irq, tcpu);
 }
 
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index f58a0cf8929a..48e63934fabe 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -109,13 +109,13 @@ extern void enable_irq(unsigned int irq);
 
 extern cpumask_t irq_default_affinity;
 
-extern int irq_set_affinity(unsigned int irq, cpumask_t cpumask);
+extern int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask);
 extern int irq_can_set_affinity(unsigned int irq);
 extern int irq_select_affinity(unsigned int irq);
 
 #else /* CONFIG_SMP */
 
-static inline int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
+static inline int irq_set_affinity(unsigned int irq, const struct cpumask *m)
 {
 	return -EINVAL;
 }
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 3dddfa703ebd..ab70fd604d3a 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -113,7 +113,8 @@ struct irq_chip {
 	void		(*eoi)(unsigned int irq);
 
 	void		(*end)(unsigned int irq);
-	void		(*set_affinity)(unsigned int irq, cpumask_t dest);
+	void		(*set_affinity)(unsigned int irq,
+					const struct cpumask *dest);
 	int		(*retrigger)(unsigned int irq);
 	int		(*set_type)(unsigned int irq, unsigned int flow_type);
 	int		(*set_wake)(unsigned int irq, unsigned int on);
diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index 10b5092e9bfe..58d8e31daa49 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -45,7 +45,7 @@ void dynamic_irq_init(unsigned int irq)
 	desc->irq_count = 0;
 	desc->irqs_unhandled = 0;
 #ifdef CONFIG_SMP
-	cpus_setall(desc->affinity);
+	cpumask_setall(&desc->affinity);
 #endif
 	spin_unlock_irqrestore(&desc->lock, flags);
 }
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 801addda3c43..10ad2f87ed9a 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -79,7 +79,7 @@ int irq_can_set_affinity(unsigned int irq)
  *	@cpumask:	cpumask
  *
  */
-int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
+int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
 	unsigned long flags;
@@ -91,14 +91,14 @@ int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
 
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	if (desc->status & IRQ_MOVE_PCNTXT || desc->status & IRQ_DISABLED) {
-		desc->affinity = cpumask;
+		cpumask_copy(&desc->affinity, cpumask);
 		desc->chip->set_affinity(irq, cpumask);
 	} else {
 		desc->status |= IRQ_MOVE_PENDING;
-		desc->pending_mask = cpumask;
+		cpumask_copy(&desc->pending_mask, cpumask);
 	}
 #else
-	desc->affinity = cpumask;
+	cpumask_copy(&desc->affinity, cpumask);
 	desc->chip->set_affinity(irq, cpumask);
 #endif
 	desc->status |= IRQ_AFFINITY_SET;
@@ -112,26 +112,24 @@ int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
  */
 int do_irq_select_affinity(unsigned int irq, struct irq_desc *desc)
 {
-	cpumask_t mask;
-
 	if (!irq_can_set_affinity(irq))
 		return 0;
 
-	cpus_and(mask, cpu_online_map, irq_default_affinity);
-
 	/*
 	 * Preserve an userspace affinity setup, but make sure that
 	 * one of the targets is online.
 	 */
 	if (desc->status & (IRQ_AFFINITY_SET | IRQ_NO_BALANCING)) {
-		if (cpus_intersects(desc->affinity, cpu_online_map))
-			mask = desc->affinity;
+		if (cpumask_any_and(&desc->affinity, cpu_online_mask)
+		    < nr_cpu_ids)
+			goto set_affinity;
 		else
 			desc->status &= ~IRQ_AFFINITY_SET;
 	}
 
-	desc->affinity = mask;
-	desc->chip->set_affinity(irq, mask);
+	cpumask_and(&desc->affinity, cpu_online_mask, &irq_default_affinity);
+set_affinity:
+	desc->chip->set_affinity(irq, &desc->affinity);
 
 	return 0;
 }
diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c
index 9db681d95814..bd72329e630c 100644
--- a/kernel/irq/migration.c
+++ b/kernel/irq/migration.c
@@ -4,7 +4,6 @@
 void move_masked_irq(int irq)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
-	cpumask_t tmp;
 
 	if (likely(!(desc->status & IRQ_MOVE_PENDING)))
 		return;
@@ -19,7 +18,7 @@ void move_masked_irq(int irq)
 
 	desc->status &= ~IRQ_MOVE_PENDING;
 
-	if (unlikely(cpus_empty(desc->pending_mask)))
+	if (unlikely(cpumask_empty(&desc->pending_mask)))
 		return;
 
 	if (!desc->chip->set_affinity)
@@ -27,8 +26,6 @@ void move_masked_irq(int irq)
 
 	assert_spin_locked(&desc->lock);
 
-	cpus_and(tmp, desc->pending_mask, cpu_online_map);
-
 	/*
 	 * If there was a valid mask to work with, please
 	 * do the disable, re-program, enable sequence.
@@ -41,10 +38,13 @@ void move_masked_irq(int irq)
 	 * For correct operation this depends on the caller
 	 * masking the irqs.
 	 */
-	if (likely(!cpus_empty(tmp))) {
-		desc->chip->set_affinity(irq,tmp);
+	if (likely(cpumask_any_and(&desc->pending_mask, cpu_online_mask)
+		   < nr_cpu_ids)) {
+		cpumask_and(&desc->affinity,
+			    &desc->pending_mask, cpu_online_mask);
+		desc->chip->set_affinity(irq, &desc->affinity);
 	}
-	cpus_clear(desc->pending_mask);
+	cpumask_clear(&desc->pending_mask);
 }
 
 void move_native_irq(int irq)
diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c
index f293349d49d0..8e91c9762520 100644
--- a/kernel/irq/proc.c
+++ b/kernel/irq/proc.c
@@ -40,33 +40,42 @@ static ssize_t irq_affinity_proc_write(struct file *file,
 		const char __user *buffer, size_t count, loff_t *pos)
 {
 	unsigned int irq = (int)(long)PDE(file->f_path.dentry->d_inode)->data;
-	cpumask_t new_value;
+	cpumask_var_t new_value;
 	int err;
 
 	if (!irq_to_desc(irq)->chip->set_affinity || no_irq_affinity ||
 	    irq_balancing_disabled(irq))
 		return -EIO;
 
-	err = cpumask_parse_user(buffer, count, &new_value);
+	if (!alloc_cpumask_var(&new_value, GFP_KERNEL))
+		return -ENOMEM;
+
+	err = cpumask_parse_user(buffer, count, new_value);
 	if (err)
-		return err;
+		goto free_cpumask;
 
-	if (!is_affinity_mask_valid(new_value))
-		return -EINVAL;
+	if (!is_affinity_mask_valid(*new_value)) {
+		err = -EINVAL;
+		goto free_cpumask;
+	}
 
 	/*
 	 * Do not allow disabling IRQs completely - it's a too easy
 	 * way to make the system unusable accidentally :-) At least
 	 * one online CPU still has to be targeted.
 	 */
-	if (!cpus_intersects(new_value, cpu_online_map))
+	if (!cpumask_intersects(new_value, cpu_online_mask)) {
 		/* Special case for empty set - allow the architecture
 		   code to set default SMP affinity. */
-		return irq_select_affinity_usr(irq) ? -EINVAL : count;
-
-	irq_set_affinity(irq, new_value);
+		err = irq_select_affinity_usr(irq) ? -EINVAL : count;
+	} else {
+		irq_set_affinity(irq, new_value);
+		err = count;
+	}
 
-	return count;
+free_cpumask:
+	free_cpumask_var(new_value);
+	return err;
 }
 
 static int irq_affinity_proc_open(struct inode *inode, struct file *file)
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index df12434b43ca..ab65d217583f 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -136,7 +136,7 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast)
  */
 static void tick_setup_device(struct tick_device *td,
 			      struct clock_event_device *newdev, int cpu,
-			      const cpumask_t *cpumask)
+			      const struct cpumask *cpumask)
 {
 	ktime_t next_event;
 	void (*handler)(struct clock_event_device *) = NULL;
@@ -171,8 +171,8 @@ static void tick_setup_device(struct tick_device *td,
 	 * When the device is not per cpu, pin the interrupt to the
 	 * current cpu:
 	 */
-	if (!cpus_equal(newdev->cpumask, *cpumask))
-		irq_set_affinity(newdev->irq, *cpumask);
+	if (!cpumask_equal(&newdev->cpumask, cpumask))
+		irq_set_affinity(newdev->irq, cpumask);
 
 	/*
 	 * When global broadcasting is active, check if the current
-- 
cgit v1.2.3


From 320ab2b0b1e08e3805a3e1084a2f0eb1938d5d67 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Sat, 13 Dec 2008 21:20:26 +1030
Subject: cpumask: convert struct clock_event_device to cpumask pointers.

Impact: change calling convention of existing clock_event APIs

struct clock_event_timer's cpumask field gets changed to take pointer,
as does the ->broadcast function.

Another single-patch change.  For safety, we BUG_ON() in
clockevents_register_device() if it's not set.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Ingo Molnar <mingo@elte.hu>
---
 arch/arm/mach-at91/at91rm9200_time.c    | 3 +--
 arch/arm/mach-at91/at91sam926x_time.c   | 2 +-
 arch/arm/mach-davinci/time.c            | 2 +-
 arch/arm/mach-imx/time.c                | 2 +-
 arch/arm/mach-ixp4xx/common.c           | 2 +-
 arch/arm/mach-msm/timer.c               | 2 +-
 arch/arm/mach-ns9xxx/time-ns9360.c      | 2 +-
 arch/arm/mach-omap1/time.c              | 2 +-
 arch/arm/mach-omap1/timer32k.c          | 2 +-
 arch/arm/mach-omap2/timer-gp.c          | 2 +-
 arch/arm/mach-pxa/time.c                | 2 +-
 arch/arm/mach-realview/core.c           | 2 +-
 arch/arm/mach-realview/localtimer.c     | 4 ++--
 arch/arm/mach-sa1100/time.c             | 2 +-
 arch/arm/mach-versatile/core.c          | 2 +-
 arch/arm/plat-mxc/time.c                | 2 +-
 arch/arm/plat-orion/time.c              | 2 +-
 arch/avr32/kernel/time.c                | 2 +-
 arch/blackfin/kernel/time-ts.c          | 2 +-
 arch/m68knommu/platform/coldfire/pit.c  | 2 +-
 arch/mips/jazz/irq.c                    | 2 +-
 arch/mips/kernel/cevt-bcm1480.c         | 2 +-
 arch/mips/kernel/cevt-ds1287.c          | 2 +-
 arch/mips/kernel/cevt-gt641xx.c         | 2 +-
 arch/mips/kernel/cevt-r4k.c             | 2 +-
 arch/mips/kernel/cevt-sb1250.c          | 2 +-
 arch/mips/kernel/cevt-smtc.c            | 2 +-
 arch/mips/kernel/cevt-txx9.c            | 2 +-
 arch/mips/kernel/i8253.c                | 2 +-
 arch/mips/nxp/pnx8550/common/time.c     | 1 +
 arch/mips/sgi-ip27/ip27-timer.c         | 2 +-
 arch/mips/sni/time.c                    | 2 +-
 arch/powerpc/kernel/time.c              | 2 +-
 arch/s390/kernel/time.c                 | 2 +-
 arch/sh/include/asm/smp.h               | 2 +-
 arch/sh/kernel/smp.c                    | 4 ++--
 arch/sh/kernel/timers/timer-broadcast.c | 2 +-
 arch/sh/kernel/timers/timer-tmu.c       | 2 +-
 arch/sparc64/kernel/time.c              | 2 +-
 arch/um/kernel/time.c                   | 2 +-
 arch/x86/kernel/apic.c                  | 8 ++++----
 arch/x86/kernel/hpet.c                  | 4 ++--
 arch/x86/kernel/i8253.c                 | 2 +-
 arch/x86/kernel/mfgpt_32.c              | 2 +-
 arch/x86/kernel/vmiclock_32.c           | 2 +-
 arch/x86/lguest/boot.c                  | 2 +-
 arch/x86/xen/time.c                     | 2 +-
 drivers/clocksource/tcb_clksrc.c        | 2 +-
 include/linux/clockchips.h              | 4 ++--
 kernel/time/clockevents.c               | 2 ++
 kernel/time/tick-broadcast.c            | 2 +-
 kernel/time/tick-common.c               | 8 ++++----
 52 files changed, 63 insertions(+), 61 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/mach-at91/at91rm9200_time.c b/arch/arm/mach-at91/at91rm9200_time.c
index a72e798a2a40..72f51d39202c 100644
--- a/arch/arm/mach-at91/at91rm9200_time.c
+++ b/arch/arm/mach-at91/at91rm9200_time.c
@@ -169,7 +169,6 @@ static struct clock_event_device clkevt = {
 	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
 	.shift		= 32,
 	.rating		= 150,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event	= clkevt32k_next_event,
 	.set_mode	= clkevt32k_mode,
 };
@@ -197,7 +196,7 @@ void __init at91rm9200_timer_init(void)
 	clkevt.mult = div_sc(AT91_SLOW_CLOCK, NSEC_PER_SEC, clkevt.shift);
 	clkevt.max_delta_ns = clockevent_delta2ns(AT91_ST_ALMV, &clkevt);
 	clkevt.min_delta_ns = clockevent_delta2ns(2, &clkevt) + 1;
-	clkevt.cpumask = cpumask_of_cpu(0);
+	clkevt.cpumask = cpumask_of(0);
 	clockevents_register_device(&clkevt);
 
 	/* register clocksource */
diff --git a/arch/arm/mach-at91/at91sam926x_time.c b/arch/arm/mach-at91/at91sam926x_time.c
index 122fd77ed580..b63e1d5f1bad 100644
--- a/arch/arm/mach-at91/at91sam926x_time.c
+++ b/arch/arm/mach-at91/at91sam926x_time.c
@@ -91,7 +91,6 @@ static struct clock_event_device pit_clkevt = {
 	.features	= CLOCK_EVT_FEAT_PERIODIC,
 	.shift		= 32,
 	.rating		= 100,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_mode	= pit_clkevt_mode,
 };
 
@@ -173,6 +172,7 @@ static void __init at91sam926x_pit_init(void)
 
 	/* Set up and register clockevents */
 	pit_clkevt.mult = div_sc(pit_rate, NSEC_PER_SEC, pit_clkevt.shift);
+	pit_clkevt.cpumask = cpumask_of(0);
 	clockevents_register_device(&pit_clkevt);
 }
 
diff --git a/arch/arm/mach-davinci/time.c b/arch/arm/mach-davinci/time.c
index 3b9a296b5c4b..f8bcd29d17a6 100644
--- a/arch/arm/mach-davinci/time.c
+++ b/arch/arm/mach-davinci/time.c
@@ -322,7 +322,7 @@ static void __init davinci_timer_init(void)
 	clockevent_davinci.min_delta_ns =
 		clockevent_delta2ns(1, &clockevent_davinci);
 
-	clockevent_davinci.cpumask = cpumask_of_cpu(0);
+	clockevent_davinci.cpumask = cpumask_of(0);
 	clockevents_register_device(&clockevent_davinci);
 }
 
diff --git a/arch/arm/mach-imx/time.c b/arch/arm/mach-imx/time.c
index a11765f5f23b..aff0ebcfa847 100644
--- a/arch/arm/mach-imx/time.c
+++ b/arch/arm/mach-imx/time.c
@@ -184,7 +184,7 @@ static int __init imx_clockevent_init(unsigned long rate)
 	clockevent_imx.min_delta_ns =
 		clockevent_delta2ns(0xf, &clockevent_imx);
 
-	clockevent_imx.cpumask = cpumask_of_cpu(0);
+	clockevent_imx.cpumask = cpumask_of(0);
 
 	clockevents_register_device(&clockevent_imx);
 
diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c
index 7766f469456b..f4656d2ac8a8 100644
--- a/arch/arm/mach-ixp4xx/common.c
+++ b/arch/arm/mach-ixp4xx/common.c
@@ -487,7 +487,7 @@ static int __init ixp4xx_clockevent_init(void)
 		clockevent_delta2ns(0xfffffffe, &clockevent_ixp4xx);
 	clockevent_ixp4xx.min_delta_ns =
 		clockevent_delta2ns(0xf, &clockevent_ixp4xx);
-	clockevent_ixp4xx.cpumask = cpumask_of_cpu(0);
+	clockevent_ixp4xx.cpumask = cpumask_of(0);
 
 	clockevents_register_device(&clockevent_ixp4xx);
 	return 0;
diff --git a/arch/arm/mach-msm/timer.c b/arch/arm/mach-msm/timer.c
index 345a14cb73c3..444d9c0f5ca6 100644
--- a/arch/arm/mach-msm/timer.c
+++ b/arch/arm/mach-msm/timer.c
@@ -182,7 +182,7 @@ static void __init msm_timer_init(void)
 			clockevent_delta2ns(0xf0000000 >> clock->shift, ce);
 		/* 4 gets rounded down to 3 */
 		ce->min_delta_ns = clockevent_delta2ns(4, ce);
-		ce->cpumask = cpumask_of_cpu(0);
+		ce->cpumask = cpumask_of(0);
 
 		cs->mult = clocksource_hz2mult(clock->freq, cs->shift);
 		res = clocksource_register(cs);
diff --git a/arch/arm/mach-ns9xxx/time-ns9360.c b/arch/arm/mach-ns9xxx/time-ns9360.c
index a63424d083d9..41df69721769 100644
--- a/arch/arm/mach-ns9xxx/time-ns9360.c
+++ b/arch/arm/mach-ns9xxx/time-ns9360.c
@@ -173,7 +173,7 @@ static void __init ns9360_timer_init(void)
 	ns9360_clockevent_device.min_delta_ns =
 		clockevent_delta2ns(1, &ns9360_clockevent_device);
 
-	ns9360_clockevent_device.cpumask = cpumask_of_cpu(0);
+	ns9360_clockevent_device.cpumask = cpumask_of(0);
 	clockevents_register_device(&ns9360_clockevent_device);
 
 	setup_irq(IRQ_NS9360_TIMER0 + TIMER_CLOCKEVENT,
diff --git a/arch/arm/mach-omap1/time.c b/arch/arm/mach-omap1/time.c
index 2cf7e32bd293..495a32c287b4 100644
--- a/arch/arm/mach-omap1/time.c
+++ b/arch/arm/mach-omap1/time.c
@@ -173,7 +173,7 @@ static __init void omap_init_mpu_timer(unsigned long rate)
 	clockevent_mpu_timer1.min_delta_ns =
 		clockevent_delta2ns(1, &clockevent_mpu_timer1);
 
-	clockevent_mpu_timer1.cpumask = cpumask_of_cpu(0);
+	clockevent_mpu_timer1.cpumask = cpumask_of(0);
 	clockevents_register_device(&clockevent_mpu_timer1);
 }
 
diff --git a/arch/arm/mach-omap1/timer32k.c b/arch/arm/mach-omap1/timer32k.c
index 705367ece174..fd3f7396e162 100644
--- a/arch/arm/mach-omap1/timer32k.c
+++ b/arch/arm/mach-omap1/timer32k.c
@@ -187,7 +187,7 @@ static __init void omap_init_32k_timer(void)
 	clockevent_32k_timer.min_delta_ns =
 		clockevent_delta2ns(1, &clockevent_32k_timer);
 
-	clockevent_32k_timer.cpumask = cpumask_of_cpu(0);
+	clockevent_32k_timer.cpumask = cpumask_of(0);
 	clockevents_register_device(&clockevent_32k_timer);
 }
 
diff --git a/arch/arm/mach-omap2/timer-gp.c b/arch/arm/mach-omap2/timer-gp.c
index 589393bedade..ae6036300f60 100644
--- a/arch/arm/mach-omap2/timer-gp.c
+++ b/arch/arm/mach-omap2/timer-gp.c
@@ -120,7 +120,7 @@ static void __init omap2_gp_clockevent_init(void)
 	clockevent_gpt.min_delta_ns =
 		clockevent_delta2ns(1, &clockevent_gpt);
 
-	clockevent_gpt.cpumask = cpumask_of_cpu(0);
+	clockevent_gpt.cpumask = cpumask_of(0);
 	clockevents_register_device(&clockevent_gpt);
 }
 
diff --git a/arch/arm/mach-pxa/time.c b/arch/arm/mach-pxa/time.c
index f8a9a62959e5..bf3c9a4aad50 100644
--- a/arch/arm/mach-pxa/time.c
+++ b/arch/arm/mach-pxa/time.c
@@ -122,7 +122,6 @@ static struct clock_event_device ckevt_pxa_osmr0 = {
 	.features	= CLOCK_EVT_FEAT_ONESHOT,
 	.shift		= 32,
 	.rating		= 200,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event	= pxa_osmr0_set_next_event,
 	.set_mode	= pxa_osmr0_set_mode,
 };
@@ -170,6 +169,7 @@ static void __init pxa_timer_init(void)
 		clockevent_delta2ns(0x7fffffff, &ckevt_pxa_osmr0);
 	ckevt_pxa_osmr0.min_delta_ns =
 		clockevent_delta2ns(MIN_OSCR_DELTA * 2, &ckevt_pxa_osmr0) + 1;
+	ckevt_pxa_osmr0.cpumask = cpumask_of(0);
 
 	cksrc_pxa_oscr0.mult =
 		clocksource_hz2mult(clock_tick_rate, cksrc_pxa_oscr0.shift);
diff --git a/arch/arm/mach-realview/core.c b/arch/arm/mach-realview/core.c
index 2f04d54711e7..b07cb9b7adb1 100644
--- a/arch/arm/mach-realview/core.c
+++ b/arch/arm/mach-realview/core.c
@@ -511,7 +511,7 @@ static struct clock_event_device timer0_clockevent =	 {
 	.set_mode	= timer_set_mode,
 	.set_next_event	= timer_set_next_event,
 	.rating		= 300,
-	.cpumask	= CPU_MASK_ALL,
+	.cpumask	= cpu_all_mask,
 };
 
 static void __init realview_clockevents_init(unsigned int timer_irq)
diff --git a/arch/arm/mach-realview/localtimer.c b/arch/arm/mach-realview/localtimer.c
index 44d178cd5733..504961ef343c 100644
--- a/arch/arm/mach-realview/localtimer.c
+++ b/arch/arm/mach-realview/localtimer.c
@@ -161,7 +161,7 @@ void __cpuinit local_timer_setup(unsigned int cpu)
 	clk->set_mode		= local_timer_set_mode;
 	clk->set_next_event	= local_timer_set_next_event;
 	clk->irq		= IRQ_LOCALTIMER;
-	clk->cpumask		= cpumask_of_cpu(cpu);
+	clk->cpumask		= cpumask_of(cpu);
 	clk->shift		= 20;
 	clk->mult		= div_sc(mpcore_timer_rate, NSEC_PER_SEC, clk->shift);
 	clk->max_delta_ns	= clockevent_delta2ns(0xffffffff, clk);
@@ -199,7 +199,7 @@ void __cpuinit local_timer_setup(unsigned int cpu)
 	clk->rating		= 200;
 	clk->set_mode		= dummy_timer_set_mode;
 	clk->broadcast		= smp_timer_broadcast;
-	clk->cpumask		= cpumask_of_cpu(cpu);
+	clk->cpumask		= cpumask_of(cpu);
 
 	clockevents_register_device(clk);
 }
diff --git a/arch/arm/mach-sa1100/time.c b/arch/arm/mach-sa1100/time.c
index 24c0a4bae850..1cac4ac0b4b8 100644
--- a/arch/arm/mach-sa1100/time.c
+++ b/arch/arm/mach-sa1100/time.c
@@ -73,7 +73,6 @@ static struct clock_event_device ckevt_sa1100_osmr0 = {
 	.features	= CLOCK_EVT_FEAT_ONESHOT,
 	.shift		= 32,
 	.rating		= 200,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event	= sa1100_osmr0_set_next_event,
 	.set_mode	= sa1100_osmr0_set_mode,
 };
@@ -110,6 +109,7 @@ static void __init sa1100_timer_init(void)
 		clockevent_delta2ns(0x7fffffff, &ckevt_sa1100_osmr0);
 	ckevt_sa1100_osmr0.min_delta_ns =
 		clockevent_delta2ns(MIN_OSCR_DELTA * 2, &ckevt_sa1100_osmr0) + 1;
+	ckevt_sa1100_osmr0.cpumask = cpumask_of(0);
 
 	cksrc_sa1100_oscr.mult =
 		clocksource_hz2mult(CLOCK_TICK_RATE, cksrc_sa1100_oscr.shift);
diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c
index 565e0ba0d67e..a3f1933434e2 100644
--- a/arch/arm/mach-versatile/core.c
+++ b/arch/arm/mach-versatile/core.c
@@ -965,7 +965,7 @@ static void __init versatile_timer_init(void)
 	timer0_clockevent.min_delta_ns =
 		clockevent_delta2ns(0xf, &timer0_clockevent);
 
-	timer0_clockevent.cpumask = cpumask_of_cpu(0);
+	timer0_clockevent.cpumask = cpumask_of(0);
 	clockevents_register_device(&timer0_clockevent);
 }
 
diff --git a/arch/arm/plat-mxc/time.c b/arch/arm/plat-mxc/time.c
index fd28f5194f71..758a1293bcfa 100644
--- a/arch/arm/plat-mxc/time.c
+++ b/arch/arm/plat-mxc/time.c
@@ -190,7 +190,7 @@ static int __init mxc_clockevent_init(void)
 	clockevent_mxc.min_delta_ns =
 			clockevent_delta2ns(0xff, &clockevent_mxc);
 
-	clockevent_mxc.cpumask = cpumask_of_cpu(0);
+	clockevent_mxc.cpumask = cpumask_of(0);
 
 	clockevents_register_device(&clockevent_mxc);
 
diff --git a/arch/arm/plat-orion/time.c b/arch/arm/plat-orion/time.c
index 544d6b327f3a..6fa2923e6dca 100644
--- a/arch/arm/plat-orion/time.c
+++ b/arch/arm/plat-orion/time.c
@@ -149,7 +149,6 @@ static struct clock_event_device orion_clkevt = {
 	.features	= CLOCK_EVT_FEAT_ONESHOT | CLOCK_EVT_FEAT_PERIODIC,
 	.shift		= 32,
 	.rating		= 300,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event	= orion_clkevt_next_event,
 	.set_mode	= orion_clkevt_mode,
 };
@@ -199,5 +198,6 @@ void __init orion_time_init(unsigned int irq, unsigned int tclk)
 	orion_clkevt.mult = div_sc(tclk, NSEC_PER_SEC, orion_clkevt.shift);
 	orion_clkevt.max_delta_ns = clockevent_delta2ns(0xfffffffe, &orion_clkevt);
 	orion_clkevt.min_delta_ns = clockevent_delta2ns(1, &orion_clkevt);
+	orion_clkevt.cpumask = cpumask_of(0);
 	clockevents_register_device(&orion_clkevt);
 }
diff --git a/arch/avr32/kernel/time.c b/arch/avr32/kernel/time.c
index 283481d74a5b..0ff46bf873b0 100644
--- a/arch/avr32/kernel/time.c
+++ b/arch/avr32/kernel/time.c
@@ -106,7 +106,6 @@ static struct clock_event_device comparator = {
 	.features	= CLOCK_EVT_FEAT_ONESHOT,
 	.shift		= 16,
 	.rating		= 50,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event	= comparator_next_event,
 	.set_mode	= comparator_mode,
 };
@@ -134,6 +133,7 @@ void __init time_init(void)
 	comparator.mult = div_sc(counter_hz, NSEC_PER_SEC, comparator.shift);
 	comparator.max_delta_ns = clockevent_delta2ns((u32)~0, &comparator);
 	comparator.min_delta_ns = clockevent_delta2ns(50, &comparator) + 1;
+	comparator.cpumask = cpumask_of(0);
 
 	sysreg_write(COMPARE, 0);
 	timer_irqaction.dev_id = &comparator;
diff --git a/arch/blackfin/kernel/time-ts.c b/arch/blackfin/kernel/time-ts.c
index e887efc86c29..0ed2badfd746 100644
--- a/arch/blackfin/kernel/time-ts.c
+++ b/arch/blackfin/kernel/time-ts.c
@@ -162,7 +162,6 @@ static struct clock_event_device clockevent_bfin = {
 	.name		= "bfin_core_timer",
 	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
 	.shift		= 32,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event = bfin_timer_set_next_event,
 	.set_mode	= bfin_timer_set_mode,
 };
@@ -193,6 +192,7 @@ static int __init bfin_clockevent_init(void)
 	clockevent_bfin.mult = div_sc(timer_clk, NSEC_PER_SEC, clockevent_bfin.shift);
 	clockevent_bfin.max_delta_ns = clockevent_delta2ns(-1, &clockevent_bfin);
 	clockevent_bfin.min_delta_ns = clockevent_delta2ns(100, &clockevent_bfin);
+	clockevent_bfin.cpumask = cpumask_of(0);
 	clockevents_register_device(&clockevent_bfin);
 
 	return 0;
diff --git a/arch/m68knommu/platform/coldfire/pit.c b/arch/m68knommu/platform/coldfire/pit.c
index c5b916700b22..2a12e7fa9748 100644
--- a/arch/m68knommu/platform/coldfire/pit.c
+++ b/arch/m68knommu/platform/coldfire/pit.c
@@ -156,7 +156,7 @@ void hw_timer_init(void)
 {
 	u32 imr;
 
-	cf_pit_clockevent.cpumask = cpumask_of_cpu(smp_processor_id());
+	cf_pit_clockevent.cpumask = cpumask_of(smp_processor_id());
 	cf_pit_clockevent.mult = div_sc(FREQ, NSEC_PER_SEC, 32);
 	cf_pit_clockevent.max_delta_ns =
 		clockevent_delta2ns(0xFFFF, &cf_pit_clockevent);
diff --git a/arch/mips/jazz/irq.c b/arch/mips/jazz/irq.c
index d7f8a782aae4..03965cb1b252 100644
--- a/arch/mips/jazz/irq.c
+++ b/arch/mips/jazz/irq.c
@@ -146,7 +146,7 @@ void __init plat_time_init(void)
 
 	BUG_ON(HZ != 100);
 
-	cd->cpumask             = cpumask_of_cpu(cpu);
+	cd->cpumask             = cpumask_of(cpu);
 	clockevents_register_device(cd);
 	action->dev_id = cd;
 	setup_irq(JAZZ_TIMER_IRQ, action);
diff --git a/arch/mips/kernel/cevt-bcm1480.c b/arch/mips/kernel/cevt-bcm1480.c
index d7e21bc8cd21..b820661678b0 100644
--- a/arch/mips/kernel/cevt-bcm1480.c
+++ b/arch/mips/kernel/cevt-bcm1480.c
@@ -126,7 +126,7 @@ void __cpuinit sb1480_clockevent_init(void)
 	cd->min_delta_ns	= clockevent_delta2ns(2, cd);
 	cd->rating		= 200;
 	cd->irq			= irq;
-	cd->cpumask		= cpumask_of_cpu(cpu);
+	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= sibyte_next_event;
 	cd->set_mode		= sibyte_set_mode;
 	clockevents_register_device(cd);
diff --git a/arch/mips/kernel/cevt-ds1287.c b/arch/mips/kernel/cevt-ds1287.c
index df4acb68bfb5..1ada45ea0700 100644
--- a/arch/mips/kernel/cevt-ds1287.c
+++ b/arch/mips/kernel/cevt-ds1287.c
@@ -88,7 +88,6 @@ static void ds1287_event_handler(struct clock_event_device *dev)
 static struct clock_event_device ds1287_clockevent = {
 	.name		= "ds1287",
 	.features	= CLOCK_EVT_FEAT_PERIODIC,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_next_event	= ds1287_set_next_event,
 	.set_mode	= ds1287_set_mode,
 	.event_handler	= ds1287_event_handler,
@@ -122,6 +121,7 @@ int __init ds1287_clockevent_init(int irq)
 	clockevent_set_clock(cd, 32768);
 	cd->max_delta_ns = clockevent_delta2ns(0x7fffffff, cd);
 	cd->min_delta_ns = clockevent_delta2ns(0x300, cd);
+	cd->cpumask = cpumask_of(0);
 
 	clockevents_register_device(&ds1287_clockevent);
 
diff --git a/arch/mips/kernel/cevt-gt641xx.c b/arch/mips/kernel/cevt-gt641xx.c
index 6e2f58520afb..e9b787feedcb 100644
--- a/arch/mips/kernel/cevt-gt641xx.c
+++ b/arch/mips/kernel/cevt-gt641xx.c
@@ -96,7 +96,6 @@ static void gt641xx_timer0_event_handler(struct clock_event_device *dev)
 static struct clock_event_device gt641xx_timer0_clockevent = {
 	.name		= "gt641xx-timer0",
 	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
-	.cpumask	= CPU_MASK_CPU0,
 	.irq		= GT641XX_TIMER0_IRQ,
 	.set_next_event	= gt641xx_timer0_set_next_event,
 	.set_mode	= gt641xx_timer0_set_mode,
@@ -132,6 +131,7 @@ static int __init gt641xx_timer0_clockevent_init(void)
 	clockevent_set_clock(cd, gt641xx_base_clock);
 	cd->max_delta_ns = clockevent_delta2ns(0x7fffffff, cd);
 	cd->min_delta_ns = clockevent_delta2ns(0x300, cd);
+	cd->cpumask = cpumask_of(0);
 
 	clockevents_register_device(&gt641xx_timer0_clockevent);
 
diff --git a/arch/mips/kernel/cevt-r4k.c b/arch/mips/kernel/cevt-r4k.c
index 4a4c59f2737a..e1ec83b68031 100644
--- a/arch/mips/kernel/cevt-r4k.c
+++ b/arch/mips/kernel/cevt-r4k.c
@@ -195,7 +195,7 @@ int __cpuinit mips_clockevent_init(void)
 
 	cd->rating		= 300;
 	cd->irq			= irq;
-	cd->cpumask		= cpumask_of_cpu(cpu);
+	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= mips_next_event;
 	cd->set_mode		= mips_set_clock_mode;
 	cd->event_handler	= mips_event_handler;
diff --git a/arch/mips/kernel/cevt-sb1250.c b/arch/mips/kernel/cevt-sb1250.c
index 0f188cd46e03..a2eebaafda52 100644
--- a/arch/mips/kernel/cevt-sb1250.c
+++ b/arch/mips/kernel/cevt-sb1250.c
@@ -125,7 +125,7 @@ void __cpuinit sb1250_clockevent_init(void)
 	cd->min_delta_ns	= clockevent_delta2ns(2, cd);
 	cd->rating		= 200;
 	cd->irq			= irq;
-	cd->cpumask		= cpumask_of_cpu(cpu);
+	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= sibyte_next_event;
 	cd->set_mode		= sibyte_set_mode;
 	clockevents_register_device(cd);
diff --git a/arch/mips/kernel/cevt-smtc.c b/arch/mips/kernel/cevt-smtc.c
index 5162fe4b5952..6d45e24db5bf 100644
--- a/arch/mips/kernel/cevt-smtc.c
+++ b/arch/mips/kernel/cevt-smtc.c
@@ -292,7 +292,7 @@ int __cpuinit mips_clockevent_init(void)
 
 	cd->rating		= 300;
 	cd->irq			= irq;
-	cd->cpumask		= cpumask_of_cpu(cpu);
+	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= mips_next_event;
 	cd->set_mode		= mips_set_clock_mode;
 	cd->event_handler	= mips_event_handler;
diff --git a/arch/mips/kernel/cevt-txx9.c b/arch/mips/kernel/cevt-txx9.c
index b5fc4eb412d2..eccf7d6096bd 100644
--- a/arch/mips/kernel/cevt-txx9.c
+++ b/arch/mips/kernel/cevt-txx9.c
@@ -112,7 +112,6 @@ static struct clock_event_device txx9tmr_clock_event_device = {
 	.name		= "TXx9",
 	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
 	.rating		= 200,
-	.cpumask	= CPU_MASK_CPU0,
 	.set_mode	= txx9tmr_set_mode,
 	.set_next_event	= txx9tmr_set_next_event,
 };
@@ -150,6 +149,7 @@ void __init txx9_clockevent_init(unsigned long baseaddr, int irq,
 		clockevent_delta2ns(0xffffffff >> (32 - TXX9_TIMER_BITS), cd);
 	cd->min_delta_ns = clockevent_delta2ns(0xf, cd);
 	cd->irq = irq;
+	cd->cpumask = cpumask_of(0),
 	clockevents_register_device(cd);
 	setup_irq(irq, &txx9tmr_irq);
 	printk(KERN_INFO "TXx9: clockevent device at 0x%lx, irq %d\n",
diff --git a/arch/mips/kernel/i8253.c b/arch/mips/kernel/i8253.c
index b6ac55162b9a..f4d187825f96 100644
--- a/arch/mips/kernel/i8253.c
+++ b/arch/mips/kernel/i8253.c
@@ -115,7 +115,7 @@ void __init setup_pit_timer(void)
 	 * Start pit with the boot cpu mask and make it global after the
 	 * IO_APIC has been initialized.
 	 */
-	cd->cpumask = cpumask_of_cpu(cpu);
+	cd->cpumask = cpumask_of(cpu);
 	clockevent_set_clock(cd, CLOCK_TICK_RATE);
 	cd->max_delta_ns = clockevent_delta2ns(0x7FFF, cd);
 	cd->min_delta_ns = clockevent_delta2ns(0xF, cd);
diff --git a/arch/mips/nxp/pnx8550/common/time.c b/arch/mips/nxp/pnx8550/common/time.c
index 62f495b57f93..cf293b279098 100644
--- a/arch/mips/nxp/pnx8550/common/time.c
+++ b/arch/mips/nxp/pnx8550/common/time.c
@@ -102,6 +102,7 @@ __init void plat_time_init(void)
 	unsigned int p;
 	unsigned int pow2p;
 
+	pnx8xxx_clockevent.cpumask = cpu_none_mask;
 	clockevents_register_device(&pnx8xxx_clockevent);
 	clocksource_register(&pnx_clocksource);
 
diff --git a/arch/mips/sgi-ip27/ip27-timer.c b/arch/mips/sgi-ip27/ip27-timer.c
index 1327c2746fb7..f024057a35f8 100644
--- a/arch/mips/sgi-ip27/ip27-timer.c
+++ b/arch/mips/sgi-ip27/ip27-timer.c
@@ -134,7 +134,7 @@ void __cpuinit hub_rt_clock_event_init(void)
 	cd->min_delta_ns        = clockevent_delta2ns(0x300, cd);
 	cd->rating		= 200;
 	cd->irq			= irq;
-	cd->cpumask		= cpumask_of_cpu(cpu);
+	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= rt_next_event;
 	cd->set_mode		= rt_set_mode;
 	clockevents_register_device(cd);
diff --git a/arch/mips/sni/time.c b/arch/mips/sni/time.c
index 796e3ce28720..69f5f88711cc 100644
--- a/arch/mips/sni/time.c
+++ b/arch/mips/sni/time.c
@@ -80,7 +80,7 @@ static void __init sni_a20r_timer_setup(void)
 	struct irqaction *action = &a20r_irqaction;
 	unsigned int cpu = smp_processor_id();
 
-	cd->cpumask             = cpumask_of_cpu(cpu);
+	cd->cpumask             = cpumask_of(cpu);
 	clockevents_register_device(cd);
 	action->dev_id = cd;
 	setup_irq(SNI_A20R_IRQ_TIMER, &a20r_irqaction);
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index e2ee66b5831d..6f39d35d6f55 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -869,7 +869,7 @@ static void register_decrementer_clockevent(int cpu)
 	struct clock_event_device *dec = &per_cpu(decrementers, cpu).event;
 
 	*dec = decrementer_clockevent;
-	dec->cpumask = cpumask_of_cpu(cpu);
+	dec->cpumask = cpumask_of(cpu);
 
 	printk(KERN_DEBUG "clockevent: %s mult[%lx] shift[%d] cpu[%d]\n",
 	       dec->name, dec->mult, dec->shift, cpu);
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index eccefbbff887..f5bd141c8443 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -154,7 +154,7 @@ void init_cpu_timer(void)
 	cd->min_delta_ns	= 1;
 	cd->max_delta_ns	= LONG_MAX;
 	cd->rating		= 400;
-	cd->cpumask		= cpumask_of_cpu(cpu);
+	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= s390_next_event;
 	cd->set_mode		= s390_set_mode;
 
diff --git a/arch/sh/include/asm/smp.h b/arch/sh/include/asm/smp.h
index 85b660c17eb0..c24e9c6a1736 100644
--- a/arch/sh/include/asm/smp.h
+++ b/arch/sh/include/asm/smp.h
@@ -31,7 +31,7 @@ enum {
 };
 
 void smp_message_recv(unsigned int msg);
-void smp_timer_broadcast(cpumask_t mask);
+void smp_timer_broadcast(const struct cpumask *mask);
 
 void local_timer_interrupt(void);
 void local_timer_setup(unsigned int cpu);
diff --git a/arch/sh/kernel/smp.c b/arch/sh/kernel/smp.c
index 593937d0c495..8f4027412614 100644
--- a/arch/sh/kernel/smp.c
+++ b/arch/sh/kernel/smp.c
@@ -184,11 +184,11 @@ void arch_send_call_function_single_ipi(int cpu)
 	plat_send_ipi(cpu, SMP_MSG_FUNCTION_SINGLE);
 }
 
-void smp_timer_broadcast(cpumask_t mask)
+void smp_timer_broadcast(const struct cpumask *mask)
 {
 	int cpu;
 
-	for_each_cpu_mask(cpu, mask)
+	for_each_cpu(cpu, mask)
 		plat_send_ipi(cpu, SMP_MSG_TIMER);
 }
 
diff --git a/arch/sh/kernel/timers/timer-broadcast.c b/arch/sh/kernel/timers/timer-broadcast.c
index c2317635230f..96e8eaea1e62 100644
--- a/arch/sh/kernel/timers/timer-broadcast.c
+++ b/arch/sh/kernel/timers/timer-broadcast.c
@@ -51,7 +51,7 @@ void __cpuinit local_timer_setup(unsigned int cpu)
 	clk->mult		= 1;
 	clk->set_mode		= dummy_timer_set_mode;
 	clk->broadcast		= smp_timer_broadcast;
-	clk->cpumask		= cpumask_of_cpu(cpu);
+	clk->cpumask		= cpumask_of(cpu);
 
 	clockevents_register_device(clk);
 }
diff --git a/arch/sh/kernel/timers/timer-tmu.c b/arch/sh/kernel/timers/timer-tmu.c
index 3c61ddd4d43e..0db3f9510336 100644
--- a/arch/sh/kernel/timers/timer-tmu.c
+++ b/arch/sh/kernel/timers/timer-tmu.c
@@ -263,7 +263,7 @@ static int tmu_timer_init(void)
 	tmu0_clockevent.min_delta_ns =
 			clockevent_delta2ns(1, &tmu0_clockevent);
 
-	tmu0_clockevent.cpumask = cpumask_of_cpu(0);
+	tmu0_clockevent.cpumask = cpumask_of(0);
 
 	clockevents_register_device(&tmu0_clockevent);
 
diff --git a/arch/sparc64/kernel/time.c b/arch/sparc64/kernel/time.c
index 141da3759091..9df8f095a8b1 100644
--- a/arch/sparc64/kernel/time.c
+++ b/arch/sparc64/kernel/time.c
@@ -763,7 +763,7 @@ void __devinit setup_sparc64_timer(void)
 	sevt = &__get_cpu_var(sparc64_events);
 
 	memcpy(sevt, &sparc64_clockevent, sizeof(*sevt));
-	sevt->cpumask = cpumask_of_cpu(smp_processor_id());
+	sevt->cpumask = cpumask_of(smp_processor_id());
 
 	clockevents_register_device(sevt);
 }
diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c
index 47f04f4a3464..b13a87a3ec95 100644
--- a/arch/um/kernel/time.c
+++ b/arch/um/kernel/time.c
@@ -50,7 +50,7 @@ static int itimer_next_event(unsigned long delta,
 static struct clock_event_device itimer_clockevent = {
 	.name		= "itimer",
 	.rating		= 250,
-	.cpumask	= CPU_MASK_ALL,
+	.cpumask	= cpu_all_mask,
 	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
 	.set_mode	= itimer_set_mode,
 	.set_next_event = itimer_next_event,
diff --git a/arch/x86/kernel/apic.c b/arch/x86/kernel/apic.c
index 16f94879b525..b2cef49f3085 100644
--- a/arch/x86/kernel/apic.c
+++ b/arch/x86/kernel/apic.c
@@ -141,7 +141,7 @@ static int lapic_next_event(unsigned long delta,
 			    struct clock_event_device *evt);
 static void lapic_timer_setup(enum clock_event_mode mode,
 			      struct clock_event_device *evt);
-static void lapic_timer_broadcast(cpumask_t mask);
+static void lapic_timer_broadcast(const struct cpumask *mask);
 static void apic_pm_activate(void);
 
 /*
@@ -453,10 +453,10 @@ static void lapic_timer_setup(enum clock_event_mode mode,
 /*
  * Local APIC timer broadcast function
  */
-static void lapic_timer_broadcast(cpumask_t mask)
+static void lapic_timer_broadcast(const struct cpumask *mask)
 {
 #ifdef CONFIG_SMP
-	send_IPI_mask(mask, LOCAL_TIMER_VECTOR);
+	send_IPI_mask(*mask, LOCAL_TIMER_VECTOR);
 #endif
 }
 
@@ -469,7 +469,7 @@ static void __cpuinit setup_APIC_timer(void)
 	struct clock_event_device *levt = &__get_cpu_var(lapic_events);
 
 	memcpy(levt, &lapic_clockevent, sizeof(*levt));
-	levt->cpumask = cpumask_of_cpu(smp_processor_id());
+	levt->cpumask = cpumask_of(smp_processor_id());
 
 	clockevents_register_device(levt);
 }
diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 940f25851e1e..e76d7e272974 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -246,7 +246,7 @@ static void hpet_legacy_clockevent_register(void)
 	 * Start hpet with the boot cpu mask and make it
 	 * global after the IO_APIC has been initialized.
 	 */
-	hpet_clockevent.cpumask = cpumask_of_cpu(smp_processor_id());
+	hpet_clockevent.cpumask = cpumask_of(smp_processor_id());
 	clockevents_register_device(&hpet_clockevent);
 	global_clock_event = &hpet_clockevent;
 	printk(KERN_DEBUG "hpet clockevent registered\n");
@@ -500,7 +500,7 @@ static void init_one_hpet_msi_clockevent(struct hpet_dev *hdev, int cpu)
 	/* 5 usec minimum reprogramming delta. */
 	evt->min_delta_ns = 5000;
 
-	evt->cpumask = cpumask_of_cpu(hdev->cpu);
+	evt->cpumask = cpumask_of(hdev->cpu);
 	clockevents_register_device(evt);
 }
 
diff --git a/arch/x86/kernel/i8253.c b/arch/x86/kernel/i8253.c
index c1b5e3ece1f2..10f92fb532f3 100644
--- a/arch/x86/kernel/i8253.c
+++ b/arch/x86/kernel/i8253.c
@@ -114,7 +114,7 @@ void __init setup_pit_timer(void)
 	 * Start pit with the boot cpu mask and make it global after the
 	 * IO_APIC has been initialized.
 	 */
-	pit_clockevent.cpumask = cpumask_of_cpu(smp_processor_id());
+	pit_clockevent.cpumask = cpumask_of(smp_processor_id());
 	pit_clockevent.mult = div_sc(CLOCK_TICK_RATE, NSEC_PER_SEC,
 				     pit_clockevent.shift);
 	pit_clockevent.max_delta_ns =
diff --git a/arch/x86/kernel/mfgpt_32.c b/arch/x86/kernel/mfgpt_32.c
index 3b599518c322..c12314c9e86f 100644
--- a/arch/x86/kernel/mfgpt_32.c
+++ b/arch/x86/kernel/mfgpt_32.c
@@ -287,7 +287,7 @@ static struct clock_event_device mfgpt_clockevent = {
 	.set_mode = mfgpt_set_mode,
 	.set_next_event = mfgpt_next_event,
 	.rating = 250,
-	.cpumask = CPU_MASK_ALL,
+	.cpumask = cpu_all_mask,
 	.shift = 32
 };
 
diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c
index 254ee07f8635..c4c1f9e09402 100644
--- a/arch/x86/kernel/vmiclock_32.c
+++ b/arch/x86/kernel/vmiclock_32.c
@@ -226,7 +226,7 @@ static void __devinit vmi_time_init_clockevent(void)
 	/* Upper bound is clockevent's use of ulong for cycle deltas. */
 	evt->max_delta_ns = clockevent_delta2ns(ULONG_MAX, evt);
 	evt->min_delta_ns = clockevent_delta2ns(1, evt);
-	evt->cpumask = cpumask_of_cpu(cpu);
+	evt->cpumask = cpumask_of(cpu);
 
 	printk(KERN_WARNING "vmi: registering clock event %s. mult=%lu shift=%u\n",
 	       evt->name, evt->mult, evt->shift);
diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c
index a5d8e1ace1cf..104c8220a383 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -737,7 +737,7 @@ static void lguest_time_init(void)
 
 	/* We can't set cpumask in the initializer: damn C limitations!  Set it
 	 * here and register our timer device. */
-	lguest_clockevent.cpumask = cpumask_of_cpu(0);
+	lguest_clockevent.cpumask = cpumask_of(0);
 	clockevents_register_device(&lguest_clockevent);
 
 	/* Finally, we unblock the timer interrupt. */
diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c
index c9f7cda48ed7..65d75a6be0ba 100644
--- a/arch/x86/xen/time.c
+++ b/arch/x86/xen/time.c
@@ -437,7 +437,7 @@ void xen_setup_timer(int cpu)
 	evt = &per_cpu(xen_clock_events, cpu);
 	memcpy(evt, xen_clockevent, sizeof(*evt));
 
-	evt->cpumask = cpumask_of_cpu(cpu);
+	evt->cpumask = cpumask_of(cpu);
 	evt->irq = irq;
 
 	setup_runstate_info(cpu);
diff --git a/drivers/clocksource/tcb_clksrc.c b/drivers/clocksource/tcb_clksrc.c
index f450588e5858..254f1064d973 100644
--- a/drivers/clocksource/tcb_clksrc.c
+++ b/drivers/clocksource/tcb_clksrc.c
@@ -154,7 +154,6 @@ static struct tc_clkevt_device clkevt = {
 		.shift		= 32,
 		/* Should be lower than at91rm9200's system timer */
 		.rating		= 125,
-		.cpumask	= CPU_MASK_CPU0,
 		.set_next_event	= tc_next_event,
 		.set_mode	= tc_mode,
 	},
@@ -195,6 +194,7 @@ static void __init setup_clkevents(struct atmel_tc *tc, int clk32k_divisor_idx)
 	clkevt.clkevt.max_delta_ns
 		= clockevent_delta2ns(0xffff, &clkevt.clkevt);
 	clkevt.clkevt.min_delta_ns = clockevent_delta2ns(1, &clkevt.clkevt) + 1;
+	clkevt.clkevt.cpumask = cpumask_of(0);
 
 	setup_irq(irq, &tc_irqaction);
 
diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h
index ed3a5d473e52..cea153697ec7 100644
--- a/include/linux/clockchips.h
+++ b/include/linux/clockchips.h
@@ -82,13 +82,13 @@ struct clock_event_device {
 	int			shift;
 	int			rating;
 	int			irq;
-	cpumask_t		cpumask;
+	const struct cpumask	*cpumask;
 	int			(*set_next_event)(unsigned long evt,
 						  struct clock_event_device *);
 	void			(*set_mode)(enum clock_event_mode mode,
 					    struct clock_event_device *);
 	void			(*event_handler)(struct clock_event_device *);
-	void			(*broadcast)(cpumask_t mask);
+	void			(*broadcast)(const struct cpumask *mask);
 	struct list_head	list;
 	enum clock_event_mode	mode;
 	ktime_t			next_event;
diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index f8d968063cea..ea2f48af83cf 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -166,6 +166,8 @@ static void clockevents_notify_released(void)
 void clockevents_register_device(struct clock_event_device *dev)
 {
 	BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED);
+	BUG_ON(!dev->cpumask);
+
 	/*
 	 * A nsec2cyc multiplicator of 0 is invalid and we'd crash
 	 * on it, so fix it up and emit a warning:
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index f98a1b7b16e9..9590af2327be 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -150,7 +150,7 @@ static void tick_do_broadcast(cpumask_t mask)
 		 */
 		cpu = first_cpu(mask);
 		td = &per_cpu(tick_cpu_device, cpu);
-		td->evtdev->broadcast(mask);
+		td->evtdev->broadcast(&mask);
 	}
 }
 
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index ab65d217583f..f8372be74122 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -171,7 +171,7 @@ static void tick_setup_device(struct tick_device *td,
 	 * When the device is not per cpu, pin the interrupt to the
 	 * current cpu:
 	 */
-	if (!cpumask_equal(&newdev->cpumask, cpumask))
+	if (!cpumask_equal(newdev->cpumask, cpumask))
 		irq_set_affinity(newdev->irq, cpumask);
 
 	/*
@@ -202,14 +202,14 @@ static int tick_check_new_device(struct clock_event_device *newdev)
 	spin_lock_irqsave(&tick_device_lock, flags);
 
 	cpu = smp_processor_id();
-	if (!cpu_isset(cpu, newdev->cpumask))
+	if (!cpumask_test_cpu(cpu, newdev->cpumask))
 		goto out_bc;
 
 	td = &per_cpu(tick_cpu_device, cpu);
 	curdev = td->evtdev;
 
 	/* cpu local device ? */
-	if (!cpus_equal(newdev->cpumask, cpumask_of_cpu(cpu))) {
+	if (!cpumask_equal(newdev->cpumask, cpumask_of(cpu))) {
 
 		/*
 		 * If the cpu affinity of the device interrupt can not
@@ -222,7 +222,7 @@ static int tick_check_new_device(struct clock_event_device *newdev)
 		 * If we have a cpu local device already, do not replace it
 		 * by a non cpu local device
 		 */
-		if (curdev && cpus_equal(curdev->cpumask, cpumask_of_cpu(cpu)))
+		if (curdev && cpumask_equal(curdev->cpumask, cpumask_of(cpu)))
 			goto out_bc;
 	}
 
-- 
cgit v1.2.3


From 7be7585393d311866653564fbcd10a3232773c0b Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Sat, 13 Dec 2008 21:20:28 +1030
Subject: cpumask: Use all NR_CPUS bits unless CONFIG_CPUMASK_OFFSTACK

Impact: futureproof as we convert more code to new APIs

The old cpumask operators treat all NR_CPUS bits as relevent, the new
ones use nr_cpumask_bits.  For large NR_CPUS and small nr_cpu_ids, this
makes a difference.

However, mixing the two can cause problems with undefined bits.  An
arch which sets CONFIG_CPUMASK_OFFSTACK should have converted across
to the new operators, so it's safe in that case.

(Thanks to Stephen Rothwell for bisecting the initial unused-bits bug,
and Mike Travis for this solution).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Mike Travis <travis@sgi.com>
---
 include/linux/cpumask.h | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 94a2ab88ae85..d4bf52603e6b 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -510,9 +510,6 @@ extern cpumask_t cpu_active_map;
 	[BITS_TO_LONGS(NR_CPUS)-1] = CPU_MASK_LAST_WORD	\
 }
 
-/* This produces more efficient code. */
-#define nr_cpumask_bits	NR_CPUS
-
 #else /* NR_CPUS > BITS_PER_LONG */
 
 #define CPU_BITS_ALL						\
@@ -520,9 +517,15 @@ extern cpumask_t cpu_active_map;
 	[0 ... BITS_TO_LONGS(NR_CPUS)-2] = ~0UL,		\
 	[BITS_TO_LONGS(NR_CPUS)-1] = CPU_MASK_LAST_WORD		\
 }
+#endif /* NR_CPUS > BITS_PER_LONG */
 
+#ifdef CONFIG_CPUMASK_OFFSTACK
+/* Assuming NR_CPUS is huge, a runtime limit is more efficient.  Also,
+ * not all bits may be allocated. */
 #define nr_cpumask_bits	nr_cpu_ids
-#endif /* NR_CPUS > BITS_PER_LONG */
+#else
+#define nr_cpumask_bits	NR_CPUS
+#endif
 
 /* verify cpu argument to cpumask_* operators */
 static inline unsigned int cpumask_check(unsigned int cpu)
-- 
cgit v1.2.3


From 7b4967c532045a1983d6d4af5c69cc7c5109f62b Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Fri, 19 Dec 2008 16:56:37 +1030
Subject: cpumask: Add alloc_cpumask_var_node()

Impact: New API

This will be needed in x86 code to allocate the domain and old_domain
cpumasks on the same node as where the containing irq_cfg struct is
allocated.

(Also fixes double-dump_stack on rare CONFIG_DEBUG_PER_CPU_MAPS case)

Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (re-impl alloc_cpumask_var)
---
 include/linux/cpumask.h |  7 +++++++
 lib/cpumask.c           | 11 ++++++++---
 2 files changed, 15 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index d4bf52603e6b..b5ad19a6f43f 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -1025,6 +1025,7 @@ static inline size_t cpumask_size(void)
 #ifdef CONFIG_CPUMASK_OFFSTACK
 typedef struct cpumask *cpumask_var_t;
 
+bool alloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags, int node);
 bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags);
 void alloc_bootmem_cpumask_var(cpumask_var_t *mask);
 void free_cpumask_var(cpumask_var_t mask);
@@ -1038,6 +1039,12 @@ static inline bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
 	return true;
 }
 
+static inline bool alloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags,
+					  int node)
+{
+	return true;
+}
+
 static inline void alloc_bootmem_cpumask_var(cpumask_var_t *mask)
 {
 }
diff --git a/lib/cpumask.c b/lib/cpumask.c
index 8d03f22c6ced..3f258f58c85b 100644
--- a/lib/cpumask.c
+++ b/lib/cpumask.c
@@ -76,15 +76,14 @@ int cpumask_any_but(const struct cpumask *mask, unsigned int cpu)
 
 /* These are not inline because of header tangles. */
 #ifdef CONFIG_CPUMASK_OFFSTACK
-bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
+bool alloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags, int node)
 {
 	if (likely(slab_is_available()))
-		*mask = kmalloc(cpumask_size(), flags);
+		*mask = kmalloc_node(cpumask_size(), flags, node);
 	else {
 #ifdef CONFIG_DEBUG_PER_CPU_MAPS
 		printk(KERN_ERR
 			"=> alloc_cpumask_var: kmalloc not available!\n");
-		dump_stack();
 #endif
 		*mask = NULL;
 	}
@@ -96,6 +95,12 @@ bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
 #endif
 	return *mask != NULL;
 }
+EXPORT_SYMBOL(alloc_cpumask_var_node);
+
+bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
+{
+	return alloc_cpumask_var_node(mask, flags, numa_node_id());
+}
 EXPORT_SYMBOL(alloc_cpumask_var);
 
 void __init alloc_bootmem_cpumask_var(cpumask_var_t *mask)
-- 
cgit v1.2.3


From e057d7aea9d8f2a46cd440d8bfb72245d4e72d79 Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Mon, 15 Dec 2008 20:26:48 -0800
Subject: cpumask: add sysfs displays for configured and disabled cpu maps

Impact: add new sysfs files.

Add sysfs files "kernel_max" and "offline" to display the max CPU index
allowed (NR_CPUS-1), and the map of cpus that are offline.

Cpus can be offlined via HOTPLUG, disabled by the BIOS ACPI tables, or
if they exceed the number of cpus allowed by the NR_CPUS config option,
or the "maxcpus=NUM" kernel start parameter.

The "possible_cpus=NUM" parameter can also extend the number of possible
cpus allowed, in which case the cpus not present at startup will be
in the offline state.  (These cpus can be HOTPLUGGED ON after system
startup [pending a follow-on patch to provide the capability via the
/sys/devices/sys/cpu/cpuN/online mechanism to bring them online.])

By design, the "offlined cpus > possible cpus" display will always
use the following formats:

  * all possible cpus online:   "x$"    or "x-y$"
  * some possible cpus offline: ".*,x$" or ".*,x-y$"

where:
  x == number of possible cpus (nr_cpu_ids); and
  y == number of cpus >= NR_CPUS or maxcpus (if y > x).

One use of this feature is for distros to select (or configure) the
appropriate kernel to install for the resident system.

Notes:
  * cpus offlined <= possible cpus will be printed for all architectures.
  * cpus offlined >  possible cpus will only be printed for arches that
  	set 'total_cpus' [X86 only in this patch].

Based on tip/cpus4096 + .../rusty/linux-2.6-for-ingo.git/master +
	 x86-only-patches sent 12/15.

Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/base/cpu.c  | 44 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/smp.h |  3 +++
 2 files changed, 47 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c
index 4259072f5bd0..2aef96f20b30 100644
--- a/drivers/base/cpu.c
+++ b/drivers/base/cpu.c
@@ -128,10 +128,54 @@ print_cpus_func(online);
 print_cpus_func(possible);
 print_cpus_func(present);
 
+/*
+ * Print values for NR_CPUS and offlined cpus
+ */
+static ssize_t print_cpus_kernel_max(struct sysdev_class *class, char *buf)
+{
+	int n = snprintf(buf, PAGE_SIZE-2, "%d\n", CONFIG_NR_CPUS - 1);
+	return n;
+}
+static SYSDEV_CLASS_ATTR(kernel_max, 0444, print_cpus_kernel_max, NULL);
+
+/* arch-optional setting to enable display of offline cpus >= nr_cpu_ids */
+unsigned int total_cpus;
+
+static ssize_t print_cpus_offline(struct sysdev_class *class, char *buf)
+{
+	int n = 0, len = PAGE_SIZE-2;
+	cpumask_var_t offline;
+
+	/* display offline cpus < nr_cpu_ids */
+	if (!alloc_cpumask_var(&offline, GFP_KERNEL))
+		return -ENOMEM;
+	cpumask_complement(offline, cpu_online_mask);
+	n = cpulist_scnprintf(buf, len, offline);
+	free_cpumask_var(offline);
+
+	/* display offline cpus >= nr_cpu_ids */
+	if (total_cpus && nr_cpu_ids < total_cpus) {
+		if (n && n < len)
+			buf[n++] = ',';
+
+		if (nr_cpu_ids == total_cpus-1)
+			n += snprintf(&buf[n], len - n, "%d", nr_cpu_ids);
+		else
+			n += snprintf(&buf[n], len - n, "%d-%d",
+						      nr_cpu_ids, total_cpus-1);
+	}
+
+	n += snprintf(&buf[n], len - n, "\n");
+	return n;
+}
+static SYSDEV_CLASS_ATTR(offline, 0444, print_cpus_offline, NULL);
+
 static struct sysdev_class_attribute *cpu_state_attr[] = {
 	&attr_online_map,
 	&attr_possible_map,
 	&attr_present_map,
+	&attr_kernel_max,
+	&attr_offline,
 };
 
 static int cpu_states_init(void)
diff --git a/include/linux/smp.h b/include/linux/smp.h
index 3f9a60043a97..0d5770c2e43a 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -21,6 +21,9 @@ struct call_single_data {
 	u16 priv;
 };
 
+/* total number of cpus in this system (may exceed NR_CPUS) */
+extern unsigned int total_cpus;
+
 #ifdef CONFIG_SMP
 
 #include <linux/preempt.h>
-- 
cgit v1.2.3


From 716707b29906e1d8d190defe3d646610b097a861 Mon Sep 17 00:00:00 2001
From: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Date: Thu, 18 Dec 2008 23:26:02 +0530
Subject: sched: convert BALANCE_FOR_xx_POWER to inline functions

Impact: cleanup

BALANCE_FOR_MC_POWER and similar macros defined in sched.h are
not constants and have various condition checks and significant
amount of code that is not suitable to be contain in a macro.
Also there could be side effects on the expressions passed to
some of them like test_sd_parent().

This patch converts all complex macros related to power savings
balance to inline functions.

Signed-off-by: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h    | 33 ++++++++++++++++++++++++---------
 include/linux/topology.h |  4 ++--
 2 files changed, 26 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4240f6bfa812..1210fb0e45ff 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -763,15 +763,23 @@ enum cpu_idle_type {
 #define SD_SERIALIZE		1024	/* Only a single load balancing instance */
 #define SD_WAKE_IDLE_FAR	2048	/* Gain latency sacrificing cache hit */
 
-#define BALANCE_FOR_MC_POWER	\
-	(sched_smt_power_savings ? SD_POWERSAVINGS_BALANCE : 0)
+extern int sched_mc_power_savings, sched_smt_power_savings;
+
+static inline int sd_balance_for_mc_power(void)
+{
+	if (sched_smt_power_savings)
+		return SD_POWERSAVINGS_BALANCE;
 
-#define BALANCE_FOR_PKG_POWER	\
-	((sched_mc_power_savings || sched_smt_power_savings) ?	\
-	 SD_POWERSAVINGS_BALANCE : 0)
+	return 0;
+}
 
-#define test_sd_parent(sd, flag)	((sd->parent &&		\
-					 (sd->parent->flags & flag)) ? 1 : 0)
+static inline int sd_balance_for_package_power(void)
+{
+	if (sched_mc_power_savings | sched_smt_power_savings)
+		return SD_POWERSAVINGS_BALANCE;
+
+	return 0;
+}
 
 
 struct sched_group {
@@ -1399,6 +1407,15 @@ struct task_struct {
 #endif
 };
 
+/* Test a flag in parent sched domain */
+static inline int test_sd_parent(struct sched_domain *sd, int flag)
+{
+	if (sd->parent && (sd->parent->flags & flag))
+		return 1;
+
+	return 0;
+}
+
 /*
  * Priority of a process goes from 0..MAX_PRIO-1, valid RT
  * priority is 0..MAX_RT_PRIO-1, and SCHED_NORMAL/SCHED_BATCH
@@ -2256,8 +2273,6 @@ __trace_special(void *__tr, void *__data,
 extern long sched_setaffinity(pid_t pid, const struct cpumask *new_mask);
 extern long sched_getaffinity(pid_t pid, struct cpumask *mask);
 
-extern int sched_mc_power_savings, sched_smt_power_savings;
-
 extern void normalize_rt_tasks(void);
 
 #ifdef CONFIG_GROUP_SCHED
diff --git a/include/linux/topology.h b/include/linux/topology.h
index 0c5b5ac36d8e..0ce7c0dac06c 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -125,7 +125,7 @@ int arch_update_cpu_topology(void);
 				| SD_WAKE_AFFINE	\
 				| SD_WAKE_BALANCE	\
 				| SD_SHARE_PKG_RESOURCES\
-				| BALANCE_FOR_MC_POWER,	\
+				| sd_balance_for_mc_power(),\
 	.last_balance		= jiffies,		\
 	.balance_interval	= 1,			\
 }
@@ -150,7 +150,7 @@ int arch_update_cpu_topology(void);
 				| SD_BALANCE_FORK	\
 				| SD_WAKE_AFFINE	\
 				| SD_WAKE_BALANCE	\
-				| BALANCE_FOR_PKG_POWER,\
+				| sd_balance_for_package_power(),\
 	.last_balance		= jiffies,		\
 	.balance_interval	= 1,			\
 }
-- 
cgit v1.2.3


From afb8a9b70b86866a60e08b2956ae4e1406390336 Mon Sep 17 00:00:00 2001
From: Gautham R Shenoy <ego@in.ibm.com>
Date: Thu, 18 Dec 2008 23:26:09 +0530
Subject: sched: framework for sched_mc/smt_power_savings=N

Impact: extend range of /sys/devices/system/cpu/sched_mc_power_savings

Currently the sched_mc/smt_power_savings variable is a boolean,
which either enables or disables topology based power savings.
This patch extends the behaviour of the variable from boolean to
multivalued, such that based on the value, we decide how
aggressively do we want to perform powersavings balance at
appropriate sched domain based on topology.

Variable levels of power saving tunable would benefit end user to
match the required level of power savings vs performance
trade-off depending on the system configuration and workloads.

This version makes the sched_mc_power_savings global variable to
take more values (0,1,2).  Later versions can have a single
tunable called sched_power_savings instead of
sched_{mc,smt}_power_savings.

Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Signed-off-by: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h | 11 +++++++++++
 kernel/sched.c        | 17 ++++++++++++++---
 2 files changed, 25 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 1210fb0e45ff..a96726658eca 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -763,6 +763,17 @@ enum cpu_idle_type {
 #define SD_SERIALIZE		1024	/* Only a single load balancing instance */
 #define SD_WAKE_IDLE_FAR	2048	/* Gain latency sacrificing cache hit */
 
+enum powersavings_balance_level {
+	POWERSAVINGS_BALANCE_NONE = 0,  /* No power saving load balance */
+	POWERSAVINGS_BALANCE_BASIC,	/* Fill one thread/core/package
+					 * first for long running threads
+					 */
+	POWERSAVINGS_BALANCE_WAKEUP,	/* Also bias task wakeups to semi-idle
+					 * cpu package for power savings
+					 */
+	MAX_POWERSAVINGS_BALANCE_LEVELS
+};
+
 extern int sched_mc_power_savings, sched_smt_power_savings;
 
 static inline int sd_balance_for_mc_power(void)
diff --git a/kernel/sched.c b/kernel/sched.c
index b309027bf9e8..56b285cd5350 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -7906,14 +7906,25 @@ int arch_reinit_sched_domains(void)
 static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt)
 {
 	int ret;
+	unsigned int level = 0;
 
-	if (buf[0] != '0' && buf[0] != '1')
+	if (sscanf(buf, "%u", &level) != 1)
+		return -EINVAL;
+
+	/*
+	 * level is always be positive so don't check for
+	 * level < POWERSAVINGS_BALANCE_NONE which is 0
+	 * What happens on 0 or 1 byte write,
+	 * need to check for count as well?
+	 */
+
+	if (level >= MAX_POWERSAVINGS_BALANCE_LEVELS)
 		return -EINVAL;
 
 	if (smt)
-		sched_smt_power_savings = (buf[0] == '1');
+		sched_smt_power_savings = level;
 	else
-		sched_mc_power_savings = (buf[0] == '1');
+		sched_mc_power_savings = level;
 
 	ret = arch_reinit_sched_domains();
 
-- 
cgit v1.2.3


From 100fdaee70ebf5f31b9451fbc01300c627091328 Mon Sep 17 00:00:00 2001
From: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Date: Thu, 18 Dec 2008 23:26:47 +0530
Subject: sched: add SD_BALANCE_NEWIDLE at MC and CPU level for sched_mc>0

Impact: change task balancing to save power more agressively

Add SD_BALANCE_NEWIDLE flag at MC level and CPU level
if sched_mc is set.  This helps power savings and
will not affect performance when sched_mc=0

Ingo and Mike Galbraith have optimised the SD flags by
removing SD_BALANCE_NEWIDLE at MC and CPU level.  This
helps performance but hurts power savings since this
slows down task consolidation by reducing the number
of times load_balance is run.

    sched: fine-tune SD_MC_INIT
        commit 14800984706bf6936bbec5187f736e928be5c218
        Author: Mike Galbraith <efault@gmx.de>
        Date:   Fri Nov 7 15:26:50 2008 +0100

    sched: re-tune balancing -- revert
        commit 9fcd18c9e63e325dbd2b4c726623f760788d5aa8
        Author: Ingo Molnar <mingo@elte.hu>
        Date:   Wed Nov 5 16:52:08 2008 +0100

This patch selectively enables SD_BALANCE_NEWIDLE flag
only when sched_mc is set to 1 or 2.  This helps power savings
by task consolidation and also does not hurt performance at
sched_mc=0 where all power saving optimisations are turned off.

Signed-off-by: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h    | 13 +++++++++++++
 include/linux/topology.h |  6 ++++--
 2 files changed, 17 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index a96726658eca..5a933d925473 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -792,6 +792,19 @@ static inline int sd_balance_for_package_power(void)
 	return 0;
 }
 
+/*
+ * Optimise SD flags for power savings:
+ * SD_BALANCE_NEWIDLE helps agressive task consolidation and power savings.
+ * Keep default SD flags if sched_{smt,mc}_power_saving=0
+ */
+
+static inline int sd_power_saving_flags(void)
+{
+	if (sched_mc_power_savings | sched_smt_power_savings)
+		return SD_BALANCE_NEWIDLE;
+
+	return 0;
+}
 
 struct sched_group {
 	struct sched_group *next;	/* Must be a circular list */
diff --git a/include/linux/topology.h b/include/linux/topology.h
index 0ce7c0dac06c..e632d29f0544 100644
--- a/include/linux/topology.h
+++ b/include/linux/topology.h
@@ -125,7 +125,8 @@ int arch_update_cpu_topology(void);
 				| SD_WAKE_AFFINE	\
 				| SD_WAKE_BALANCE	\
 				| SD_SHARE_PKG_RESOURCES\
-				| sd_balance_for_mc_power(),\
+				| sd_balance_for_mc_power()\
+				| sd_power_saving_flags(),\
 	.last_balance		= jiffies,		\
 	.balance_interval	= 1,			\
 }
@@ -150,7 +151,8 @@ int arch_update_cpu_topology(void);
 				| SD_BALANCE_FORK	\
 				| SD_WAKE_AFFINE	\
 				| SD_WAKE_BALANCE	\
-				| sd_balance_for_package_power(),\
+				| sd_balance_for_package_power()\
+				| sd_power_saving_flags(),\
 	.last_balance		= jiffies,		\
 	.balance_interval	= 1,			\
 }
-- 
cgit v1.2.3


From 06aaf76a7e2e4cc57eabcb8f43ec99c961fe55fe Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 18 Dec 2008 21:30:23 +0100
Subject: sched: move test_sd_parent() to an SMP section of sched.h

Impact: build fix

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 5a933d925473..e5f928a079e8 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -920,6 +920,15 @@ extern void partition_sched_domains(int ndoms_new, struct cpumask *doms_new,
 				    struct sched_domain_attr *dattr_new);
 extern int arch_reinit_sched_domains(void);
 
+/* Test a flag in parent sched domain */
+static inline int test_sd_parent(struct sched_domain *sd, int flag)
+{
+	if (sd->parent && (sd->parent->flags & flag))
+		return 1;
+
+	return 0;
+}
+
 #else /* CONFIG_SMP */
 
 struct sched_domain_attr;
@@ -1431,15 +1440,6 @@ struct task_struct {
 #endif
 };
 
-/* Test a flag in parent sched domain */
-static inline int test_sd_parent(struct sched_domain *sd, int flag)
-{
-	if (sd->parent && (sd->parent->flags & flag))
-		return 1;
-
-	return 0;
-}
-
 /*
  * Priority of a process goes from 0..MAX_PRIO-1, valid RT
  * priority is 0..MAX_RT_PRIO-1, and SCHED_NORMAL/SCHED_BATCH
-- 
cgit v1.2.3


From 3f4b0ef7f2899c91b1d6958779f084b44dd59d32 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Sun, 26 Oct 2008 20:52:15 +0100
Subject: ACPI hibernate: Add a mechanism to save/restore ACPI NVS memory

According to the ACPI Specification 3.0b, Section 15.3.2,
"OSPM will call the _PTS control method some time before entering a
sleeping state, to allow the platform's AML code to update this
memory image before entering the sleeping state. After the system
awakes from an S4 state, OSPM will restore this memory area and call
the _WAK control method to enable the BIOS to reclaim its memory
image."  For this reason, implement a mechanism allowing us to save
the NVS memory during hibernation and to restore it during the
subsequent resume.

Based on a patch by Zhang Rui.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Nigel Cunningham <nigel@tuxonice.net>
Cc: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/sleep/main.c |  53 +++++++++++++++++---
 include/linux/suspend.h   |  13 +++++
 kernel/power/swsusp.c     | 122 ++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 180 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/acpi/sleep/main.c b/drivers/acpi/sleep/main.c
index 28a691cc625e..45a8015e4217 100644
--- a/drivers/acpi/sleep/main.c
+++ b/drivers/acpi/sleep/main.c
@@ -394,9 +394,25 @@ void __init acpi_no_s4_hw_signature(void)
 
 static int acpi_hibernation_begin(void)
 {
-	acpi_target_sleep_state = ACPI_STATE_S4;
-	acpi_sleep_tts_switch(acpi_target_sleep_state);
-	return 0;
+	int error;
+
+	error = hibernate_nvs_alloc();
+	if (!error) {
+		acpi_target_sleep_state = ACPI_STATE_S4;
+		acpi_sleep_tts_switch(acpi_target_sleep_state);
+	}
+
+	return error;
+}
+
+static int acpi_hibernation_pre_snapshot(void)
+{
+	int error = acpi_pm_prepare();
+
+	if (!error)
+		hibernate_nvs_save();
+
+	return error;
 }
 
 static int acpi_hibernation_enter(void)
@@ -417,6 +433,12 @@ static int acpi_hibernation_enter(void)
 	return ACPI_SUCCESS(status) ? 0 : -EFAULT;
 }
 
+static void acpi_hibernation_finish(void)
+{
+	hibernate_nvs_free();
+	acpi_pm_finish();
+}
+
 static void acpi_hibernation_leave(void)
 {
 	/*
@@ -432,6 +454,8 @@ static void acpi_hibernation_leave(void)
 			"cannot resume!\n");
 		panic("ACPI S4 hardware signature mismatch");
 	}
+	/* Restore the NVS memory area */
+	hibernate_nvs_restore();
 }
 
 static void acpi_pm_enable_gpes(void)
@@ -442,8 +466,8 @@ static void acpi_pm_enable_gpes(void)
 static struct platform_hibernation_ops acpi_hibernation_ops = {
 	.begin = acpi_hibernation_begin,
 	.end = acpi_pm_end,
-	.pre_snapshot = acpi_pm_prepare,
-	.finish = acpi_pm_finish,
+	.pre_snapshot = acpi_hibernation_pre_snapshot,
+	.finish = acpi_hibernation_finish,
 	.prepare = acpi_pm_prepare,
 	.enter = acpi_hibernation_enter,
 	.leave = acpi_hibernation_leave,
@@ -469,8 +493,21 @@ static int acpi_hibernation_begin_old(void)
 
 	error = acpi_sleep_prepare(ACPI_STATE_S4);
 
+	if (!error) {
+		error = hibernate_nvs_alloc();
+		if (!error)
+			acpi_target_sleep_state = ACPI_STATE_S4;
+	}
+	return error;
+}
+
+static int acpi_hibernation_pre_snapshot_old(void)
+{
+	int error = acpi_pm_disable_gpes();
+
 	if (!error)
-		acpi_target_sleep_state = ACPI_STATE_S4;
+		hibernate_nvs_save();
+
 	return error;
 }
 
@@ -481,8 +518,8 @@ static int acpi_hibernation_begin_old(void)
 static struct platform_hibernation_ops acpi_hibernation_ops_old = {
 	.begin = acpi_hibernation_begin_old,
 	.end = acpi_pm_end,
-	.pre_snapshot = acpi_pm_disable_gpes,
-	.finish = acpi_pm_finish,
+	.pre_snapshot = acpi_hibernation_pre_snapshot_old,
+	.finish = acpi_hibernation_finish,
 	.prepare = acpi_pm_disable_gpes,
 	.enter = acpi_hibernation_enter,
 	.leave = acpi_hibernation_leave,
diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index 2ce8207686e2..2b409c44db83 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -232,6 +232,11 @@ extern unsigned long get_safe_page(gfp_t gfp_mask);
 
 extern void hibernation_set_ops(struct platform_hibernation_ops *ops);
 extern int hibernate(void);
+extern int hibernate_nvs_register(unsigned long start, unsigned long size);
+extern int hibernate_nvs_alloc(void);
+extern void hibernate_nvs_free(void);
+extern void hibernate_nvs_save(void);
+extern void hibernate_nvs_restore(void);
 #else /* CONFIG_HIBERNATION */
 static inline int swsusp_page_is_forbidden(struct page *p) { return 0; }
 static inline void swsusp_set_page_free(struct page *p) {}
@@ -239,6 +244,14 @@ static inline void swsusp_unset_page_free(struct page *p) {}
 
 static inline void hibernation_set_ops(struct platform_hibernation_ops *ops) {}
 static inline int hibernate(void) { return -ENOSYS; }
+static inline int hibernate_nvs_register(unsigned long a, unsigned long b)
+{
+	return 0;
+}
+static inline int hibernate_nvs_alloc(void) { return 0; }
+static inline void hibernate_nvs_free(void) {}
+static inline void hibernate_nvs_save(void) {}
+static inline void hibernate_nvs_restore(void) {}
 #endif /* CONFIG_HIBERNATION */
 
 #ifdef CONFIG_PM_SLEEP
diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c
index 023ff2a31d89..a92c91451559 100644
--- a/kernel/power/swsusp.c
+++ b/kernel/power/swsusp.c
@@ -262,3 +262,125 @@ int swsusp_shrink_memory(void)
 
 	return 0;
 }
+
+/*
+ * Platforms, like ACPI, may want us to save some memory used by them during
+ * hibernation and to restore the contents of this memory during the subsequent
+ * resume.  The code below implements a mechanism allowing us to do that.
+ */
+
+struct nvs_page {
+	unsigned long phys_start;
+	unsigned int size;
+	void *kaddr;
+	void *data;
+	struct list_head node;
+};
+
+static LIST_HEAD(nvs_list);
+
+/**
+ *	hibernate_nvs_register - register platform NVS memory region to save
+ *	@start - physical address of the region
+ *	@size - size of the region
+ *
+ *	The NVS region need not be page-aligned (both ends) and we arrange
+ *	things so that the data from page-aligned addresses in this region will
+ *	be copied into separate RAM pages.
+ */
+int hibernate_nvs_register(unsigned long start, unsigned long size)
+{
+	struct nvs_page *entry, *next;
+
+	while (size > 0) {
+		unsigned int nr_bytes;
+
+		entry = kzalloc(sizeof(struct nvs_page), GFP_KERNEL);
+		if (!entry)
+			goto Error;
+
+		list_add_tail(&entry->node, &nvs_list);
+		entry->phys_start = start;
+		nr_bytes = PAGE_SIZE - (start & ~PAGE_MASK);
+		entry->size = (size < nr_bytes) ? size : nr_bytes;
+
+		start += entry->size;
+		size -= entry->size;
+	}
+	return 0;
+
+ Error:
+	list_for_each_entry_safe(entry, next, &nvs_list, node) {
+		list_del(&entry->node);
+		kfree(entry);
+	}
+	return -ENOMEM;
+}
+
+/**
+ *	hibernate_nvs_free - free data pages allocated for saving NVS regions
+ */
+void hibernate_nvs_free(void)
+{
+	struct nvs_page *entry;
+
+	list_for_each_entry(entry, &nvs_list, node)
+		if (entry->data) {
+			free_page((unsigned long)entry->data);
+			entry->data = NULL;
+			if (entry->kaddr) {
+				iounmap(entry->kaddr);
+				entry->kaddr = NULL;
+			}
+		}
+}
+
+/**
+ *	hibernate_nvs_alloc - allocate memory necessary for saving NVS regions
+ */
+int hibernate_nvs_alloc(void)
+{
+	struct nvs_page *entry;
+
+	list_for_each_entry(entry, &nvs_list, node) {
+		entry->data = (void *)__get_free_page(GFP_KERNEL);
+		if (!entry->data) {
+			hibernate_nvs_free();
+			return -ENOMEM;
+		}
+	}
+	return 0;
+}
+
+/**
+ *	hibernate_nvs_save - save NVS memory regions
+ */
+void hibernate_nvs_save(void)
+{
+	struct nvs_page *entry;
+
+	printk(KERN_INFO "PM: Saving platform NVS memory\n");
+
+	list_for_each_entry(entry, &nvs_list, node)
+		if (entry->data) {
+			entry->kaddr = ioremap(entry->phys_start, entry->size);
+			memcpy(entry->data, entry->kaddr, entry->size);
+		}
+}
+
+/**
+ *	hibernate_nvs_restore - restore NVS memory regions
+ *
+ *	This function is going to be called with interrupts disabled, so it
+ *	cannot iounmap the virtual addresses used to access the NVS region.
+ */
+void hibernate_nvs_restore(void)
+{
+	struct nvs_page *entry;
+
+	printk(KERN_INFO "PM: Restoring platform NVS memory\n");
+
+	list_for_each_entry(entry, &nvs_list, node)
+		if (entry->data)
+			memcpy(entry->kaddr, entry->data, entry->size);
+}
-- 
cgit v1.2.3


From ba84ed9546e91348fdf3ff2bff859b0ee53b407a Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Sun, 26 Oct 2008 20:56:30 +0100
Subject: ACPI hibernate: Introduce new kernel parameter acpi_sleep=s4_nonvs

On some machines it may be necessary to disable the saving/restoring
of the ACPI NVS memory region during hibernation/resume.  For this
purpose, introduce new ACPI kernel command line option
acpi_sleep=s4_nonvs.

Based on a patch by Zhang Rui.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Nigel Cunningham <nigel@tuxonice.net>
Acked-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 Documentation/kernel-parameters.txt |  5 ++++-
 arch/x86/kernel/acpi/sleep.c        |  2 ++
 drivers/acpi/sleep/main.c           | 18 ++++++++++++++++--
 include/linux/acpi.h                |  1 +
 4 files changed, 23 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index e0f346d201ed..1d089eeff3cf 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -149,7 +149,8 @@ and is between 256 and 4096 characters. It is defined in the file
 			default: 0
 
 	acpi_sleep=	[HW,ACPI] Sleep options
-			Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig, old_ordering }
+			Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig,
+				  old_ordering, s4_nonvs }
 			See Documentation/power/video.txt for s3_bios and s3_mode.
 			s3_beep is for debugging; it makes the PC's speaker beep
 			as soon as the kernel's real-mode entry point is called.
@@ -159,6 +160,8 @@ and is between 256 and 4096 characters. It is defined in the file
 			control method, wrt putting devices into low power
 			states, to be enforced (the ACPI 2.0 ordering of _PTS is
 			used by default).
+			s4_nonvs prevents the kernel from saving/restoring the
+			ACPI NVS memory during hibernation.
 
 	acpi_sci=	[HW,ACPI] ACPI System Control Interrupt trigger mode
 			Format: { level | edge | high | low }
diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c
index 806b4e9051b4..707c1f6f95fa 100644
--- a/arch/x86/kernel/acpi/sleep.c
+++ b/arch/x86/kernel/acpi/sleep.c
@@ -159,6 +159,8 @@ static int __init acpi_sleep_setup(char *str)
 #endif
 		if (strncmp(str, "old_ordering", 12) == 0)
 			acpi_old_suspend_ordering();
+		if (strncmp(str, "s4_nonvs", 8) == 0)
+			acpi_s4_no_nvs();
 		str = strchr(str, ',');
 		if (str != NULL)
 			str += strspn(str, ", \t");
diff --git a/drivers/acpi/sleep/main.c b/drivers/acpi/sleep/main.c
index 45a8015e4217..bef41fd4c877 100644
--- a/drivers/acpi/sleep/main.c
+++ b/drivers/acpi/sleep/main.c
@@ -101,6 +101,19 @@ void __init acpi_old_suspend_ordering(void)
  * cases.
  */
 static bool set_sci_en_on_resume;
+/*
+ * The ACPI specification wants us to save NVS memory regions during hibernation
+ * and to restore them during the subsequent resume.  However, it is not certain
+ * if this mechanism is going to work on all machines, so we allow the user to
+ * disable this mechanism using the 'acpi_sleep=s4_nonvs' kernel command line
+ * option.
+ */
+static bool s4_no_nvs;
+
+void __init acpi_s4_no_nvs(void)
+{
+	s4_no_nvs = true;
+}
 
 /**
  *	acpi_pm_disable_gpes - Disable the GPEs.
@@ -396,7 +409,7 @@ static int acpi_hibernation_begin(void)
 {
 	int error;
 
-	error = hibernate_nvs_alloc();
+	error = s4_no_nvs ? 0 : hibernate_nvs_alloc();
 	if (!error) {
 		acpi_target_sleep_state = ACPI_STATE_S4;
 		acpi_sleep_tts_switch(acpi_target_sleep_state);
@@ -494,7 +507,8 @@ static int acpi_hibernation_begin_old(void)
 	error = acpi_sleep_prepare(ACPI_STATE_S4);
 
 	if (!error) {
-		error = hibernate_nvs_alloc();
+		if (!s4_no_nvs)
+			error = hibernate_nvs_alloc();
 		if (!error)
 			acpi_target_sleep_state = ACPI_STATE_S4;
 	}
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index fba8051fb297..dfa0a5356c53 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -270,6 +270,7 @@ int acpi_check_mem_region(resource_size_t start, resource_size_t n,
 #ifdef CONFIG_PM_SLEEP
 void __init acpi_no_s4_hw_signature(void);
 void __init acpi_old_suspend_ordering(void);
+void __init acpi_s4_no_nvs(void);
 #endif /* CONFIG_PM_SLEEP */
 #else	/* CONFIG_ACPI */
 
-- 
cgit v1.2.3


From 50b6f1f4a430608f7345f66ecd68a129bff11649 Mon Sep 17 00:00:00 2001
From: Kwangwoo Lee <kwangwoo.lee@gmail.com>
Date: Sat, 20 Dec 2008 04:26:01 -0500
Subject: Input: add tsc2007 based touchscreen driver

This drive has been tested on ARM9 based SoC - MV86XX.

Signed-off-by: Kwangwoo Lee <kwangwoo.lee@gmail.com>
Acked-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/touchscreen/Kconfig   |  11 ++
 drivers/input/touchscreen/Makefile  |   1 +
 drivers/input/touchscreen/tsc2007.c | 381 ++++++++++++++++++++++++++++++++++++
 include/linux/i2c/tsc2007.h         |  17 ++
 4 files changed, 410 insertions(+)
 create mode 100644 drivers/input/touchscreen/tsc2007.c
 create mode 100644 include/linux/i2c/tsc2007.h

(limited to 'include/linux')

diff --git a/drivers/input/touchscreen/Kconfig b/drivers/input/touchscreen/Kconfig
index 20eb52ed176d..83747dc9ba05 100644
--- a/drivers/input/touchscreen/Kconfig
+++ b/drivers/input/touchscreen/Kconfig
@@ -389,4 +389,15 @@ config TOUCHSCREEN_TOUCHIT213
 	  To compile this driver as a module, choose M here: the
 	  module will be called touchit213.
 
+config TOUCHSCREEN_TSC2007
+	tristate "TSC2007 based touchscreens"
+	depends on I2C
+	help
+	  Say Y here if you have a TSC2007 based touchscreen.
+
+	  If unsure, say N.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called tsc2007.
+
 endif
diff --git a/drivers/input/touchscreen/Makefile b/drivers/input/touchscreen/Makefile
index 3dc84d3846c1..127f87cc2e2e 100644
--- a/drivers/input/touchscreen/Makefile
+++ b/drivers/input/touchscreen/Makefile
@@ -25,6 +25,7 @@ obj-$(CONFIG_TOUCHSCREEN_PENMOUNT)	+= penmount.o
 obj-$(CONFIG_TOUCHSCREEN_TOUCHIT213)	+= touchit213.o
 obj-$(CONFIG_TOUCHSCREEN_TOUCHRIGHT)	+= touchright.o
 obj-$(CONFIG_TOUCHSCREEN_TOUCHWIN)	+= touchwin.o
+obj-$(CONFIG_TOUCHSCREEN_TSC2007)	+= tsc2007.o
 obj-$(CONFIG_TOUCHSCREEN_UCB1400)	+= ucb1400_ts.o
 obj-$(CONFIG_TOUCHSCREEN_WACOM_W8001)	+= wacom_w8001.o
 obj-$(CONFIG_TOUCHSCREEN_WM97XX)	+= wm97xx-ts.o
diff --git a/drivers/input/touchscreen/tsc2007.c b/drivers/input/touchscreen/tsc2007.c
new file mode 100644
index 000000000000..b75dc2990574
--- /dev/null
+++ b/drivers/input/touchscreen/tsc2007.c
@@ -0,0 +1,381 @@
+/*
+ * drivers/input/touchscreen/tsc2007.c
+ *
+ * Copyright (c) 2008 MtekVision Co., Ltd.
+ *	Kwangwoo Lee <kwlee@mtekvision.com>
+ *
+ * Using code from:
+ *  - ads7846.c
+ *	Copyright (c) 2005 David Brownell
+ *	Copyright (c) 2006 Nokia Corporation
+ *  - corgi_ts.c
+ *	Copyright (C) 2004-2005 Richard Purdie
+ *  - omap_ts.[hc], ads7846.h, ts_osk.c
+ *	Copyright (C) 2002 MontaVista Software
+ *	Copyright (C) 2004 Texas Instruments
+ *	Copyright (C) 2005 Dirk Behme
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/hrtimer.h>
+#include <linux/slab.h>
+#include <linux/input.h>
+#include <linux/interrupt.h>
+#include <linux/i2c.h>
+#include <linux/i2c/tsc2007.h>
+
+#define TS_POLL_DELAY	(10 * 1000)	/* ns delay before the first sample */
+#define TS_POLL_PERIOD	(5 * 1000)	/* ns delay between samples */
+
+#define TSC2007_MEASURE_TEMP0		(0x0 << 4)
+#define TSC2007_MEASURE_AUX		(0x2 << 4)
+#define TSC2007_MEASURE_TEMP1		(0x4 << 4)
+#define TSC2007_ACTIVATE_XN		(0x8 << 4)
+#define TSC2007_ACTIVATE_YN		(0x9 << 4)
+#define TSC2007_ACTIVATE_YP_XN		(0xa << 4)
+#define TSC2007_SETUP			(0xb << 4)
+#define TSC2007_MEASURE_X		(0xc << 4)
+#define TSC2007_MEASURE_Y		(0xd << 4)
+#define TSC2007_MEASURE_Z1		(0xe << 4)
+#define TSC2007_MEASURE_Z2		(0xf << 4)
+
+#define TSC2007_POWER_OFF_IRQ_EN	(0x0 << 2)
+#define TSC2007_ADC_ON_IRQ_DIS0		(0x1 << 2)
+#define TSC2007_ADC_OFF_IRQ_EN		(0x2 << 2)
+#define TSC2007_ADC_ON_IRQ_DIS1		(0x3 << 2)
+
+#define TSC2007_12BIT			(0x0 << 1)
+#define TSC2007_8BIT			(0x1 << 1)
+
+#define	MAX_12BIT			((1 << 12) - 1)
+
+#define ADC_ON_12BIT	(TSC2007_12BIT | TSC2007_ADC_ON_IRQ_DIS0)
+
+#define READ_Y		(ADC_ON_12BIT | TSC2007_MEASURE_Y)
+#define READ_Z1		(ADC_ON_12BIT | TSC2007_MEASURE_Z1)
+#define READ_Z2		(ADC_ON_12BIT | TSC2007_MEASURE_Z2)
+#define READ_X		(ADC_ON_12BIT | TSC2007_MEASURE_X)
+#define PWRDOWN		(TSC2007_12BIT | TSC2007_POWER_OFF_IRQ_EN)
+
+struct ts_event {
+	u16	x;
+	u16	y;
+	u16	z1, z2;
+};
+
+struct tsc2007 {
+	struct input_dev	*input;
+	char			phys[32];
+	struct hrtimer		timer;
+	struct ts_event		tc;
+
+	struct i2c_client	*client;
+
+	spinlock_t		lock;
+
+	u16			model;
+	u16			x_plate_ohms;
+
+	unsigned		pendown;
+	int			irq;
+
+	int			(*get_pendown_state)(void);
+	void			(*clear_penirq)(void);
+};
+
+static inline int tsc2007_xfer(struct tsc2007 *tsc, u8 cmd)
+{
+	s32 data;
+	u16 val;
+
+	data = i2c_smbus_read_word_data(tsc->client, cmd);
+	if (data < 0) {
+		dev_err(&tsc->client->dev, "i2c io error: %d\n", data);
+		return data;
+	}
+
+	/* The protocol and raw data format from i2c interface:
+	 * S Addr Wr [A] Comm [A] S Addr Rd [A] [DataLow] A [DataHigh] NA P
+	 * Where DataLow has [D11-D4], DataHigh has [D3-D0 << 4 | Dummy 4bit].
+	 */
+	val = swab16(data) >> 4;
+
+	dev_dbg(&tsc->client->dev, "data: 0x%x, val: 0x%x\n", data, val);
+
+	return val;
+}
+
+static void tsc2007_send_event(void *tsc)
+{
+	struct tsc2007	*ts = tsc;
+	u32		rt;
+	u16		x, y, z1, z2;
+
+	x = ts->tc.x;
+	y = ts->tc.y;
+	z1 = ts->tc.z1;
+	z2 = ts->tc.z2;
+
+	/* range filtering */
+	if (x == MAX_12BIT)
+		x = 0;
+
+	if (likely(x && z1)) {
+		/* compute touch pressure resistance using equation #1 */
+		rt = z2;
+		rt -= z1;
+		rt *= x;
+		rt *= ts->x_plate_ohms;
+		rt /= z1;
+		rt = (rt + 2047) >> 12;
+	} else
+		rt = 0;
+
+	/* Sample found inconsistent by debouncing or pressure is beyond
+	 * the maximum. Don't report it to user space, repeat at least
+	 * once more the measurement
+	 */
+	if (rt > MAX_12BIT) {
+		dev_dbg(&ts->client->dev, "ignored pressure %d\n", rt);
+
+		hrtimer_start(&ts->timer, ktime_set(0, TS_POLL_PERIOD),
+			      HRTIMER_MODE_REL);
+		return;
+	}
+
+	/* NOTE: We can't rely on the pressure to determine the pen down
+	 * state, even this controller has a pressure sensor.  The pressure
+	 * value can fluctuate for quite a while after lifting the pen and
+	 * in some cases may not even settle at the expected value.
+	 *
+	 * The only safe way to check for the pen up condition is in the
+	 * timer by reading the pen signal state (it's a GPIO _and_ IRQ).
+	 */
+	if (rt) {
+		struct input_dev *input = ts->input;
+
+		if (!ts->pendown) {
+			dev_dbg(&ts->client->dev, "DOWN\n");
+
+			input_report_key(input, BTN_TOUCH, 1);
+			ts->pendown = 1;
+		}
+
+		input_report_abs(input, ABS_X, x);
+		input_report_abs(input, ABS_Y, y);
+		input_report_abs(input, ABS_PRESSURE, rt);
+
+		input_sync(input);
+
+		dev_dbg(&ts->client->dev, "point(%4d,%4d), pressure (%4u)\n",
+			x, y, rt);
+	}
+
+	hrtimer_start(&ts->timer, ktime_set(0, TS_POLL_PERIOD),
+			HRTIMER_MODE_REL);
+}
+
+static int tsc2007_read_values(struct tsc2007 *tsc)
+{
+	/* y- still on; turn on only y+ (and ADC) */
+	tsc->tc.y = tsc2007_xfer(tsc, READ_Y);
+
+	/* turn y- off, x+ on, then leave in lowpower */
+	tsc->tc.x = tsc2007_xfer(tsc, READ_X);
+
+	/* turn y+ off, x- on; we'll use formula #1 */
+	tsc->tc.z1 = tsc2007_xfer(tsc, READ_Z1);
+	tsc->tc.z2 = tsc2007_xfer(tsc, READ_Z2);
+
+	/* power down */
+	tsc2007_xfer(tsc, PWRDOWN);
+
+	return 0;
+}
+
+static enum hrtimer_restart tsc2007_timer(struct hrtimer *handle)
+{
+	struct tsc2007 *ts = container_of(handle, struct tsc2007, timer);
+
+	spin_lock_irq(&ts->lock);
+
+	if (unlikely(!ts->get_pendown_state() && ts->pendown)) {
+		struct input_dev *input = ts->input;
+
+		dev_dbg(&ts->client->dev, "UP\n");
+
+		input_report_key(input, BTN_TOUCH, 0);
+		input_report_abs(input, ABS_PRESSURE, 0);
+		input_sync(input);
+
+		ts->pendown = 0;
+		enable_irq(ts->irq);
+	} else {
+		/* pen is still down, continue with the measurement */
+		dev_dbg(&ts->client->dev, "pen is still down\n");
+
+		tsc2007_read_values(ts);
+		tsc2007_send_event(ts);
+	}
+
+	spin_unlock_irq(&ts->lock);
+
+	return HRTIMER_NORESTART;
+}
+
+static irqreturn_t tsc2007_irq(int irq, void *handle)
+{
+	struct tsc2007 *ts = handle;
+	unsigned long flags;
+
+	spin_lock_irqsave(&ts->lock, flags);
+
+	if (likely(ts->get_pendown_state())) {
+		disable_irq(ts->irq);
+		hrtimer_start(&ts->timer, ktime_set(0, TS_POLL_DELAY),
+					HRTIMER_MODE_REL);
+	}
+
+	if (ts->clear_penirq)
+		ts->clear_penirq();
+
+	spin_unlock_irqrestore(&ts->lock, flags);
+
+	return IRQ_HANDLED;
+}
+
+static int tsc2007_probe(struct i2c_client *client,
+			const struct i2c_device_id *id)
+{
+	struct tsc2007 *ts;
+	struct tsc2007_platform_data *pdata = pdata = client->dev.platform_data;
+	struct input_dev *input_dev;
+	int err;
+
+	if (!pdata) {
+		dev_err(&client->dev, "platform data is required!\n");
+		return -EINVAL;
+	}
+
+	if (!i2c_check_functionality(client->adapter,
+				     I2C_FUNC_SMBUS_READ_WORD_DATA))
+		return -EIO;
+
+	ts = kzalloc(sizeof(struct tsc2007), GFP_KERNEL);
+	input_dev = input_allocate_device();
+	if (!ts || !input_dev) {
+		err = -ENOMEM;
+		goto err_free_mem;
+	}
+
+	ts->client = client;
+	i2c_set_clientdata(client, ts);
+
+	ts->input = input_dev;
+
+	hrtimer_init(&ts->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	ts->timer.function = tsc2007_timer;
+
+	spin_lock_init(&ts->lock);
+
+	ts->model             = pdata->model;
+	ts->x_plate_ohms      = pdata->x_plate_ohms;
+	ts->get_pendown_state = pdata->get_pendown_state;
+	ts->clear_penirq      = pdata->clear_penirq;
+
+	pdata->init_platform_hw();
+
+	snprintf(ts->phys, sizeof(ts->phys), "%s/input0", client->dev.bus_id);
+
+	input_dev->name = "TSC2007 Touchscreen";
+	input_dev->phys = ts->phys;
+	input_dev->id.bustype = BUS_I2C;
+
+	input_dev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_ABS);
+	input_dev->keybit[BIT_WORD(BTN_TOUCH)] = BIT_MASK(BTN_TOUCH);
+
+	input_set_abs_params(input_dev, ABS_X, 0, MAX_12BIT, 0, 0);
+	input_set_abs_params(input_dev, ABS_Y, 0, MAX_12BIT, 0, 0);
+	input_set_abs_params(input_dev, ABS_PRESSURE, 0, MAX_12BIT, 0, 0);
+
+	tsc2007_read_values(ts);
+
+	ts->irq = client->irq;
+
+	err = request_irq(ts->irq, tsc2007_irq, 0,
+			client->dev.driver->name, ts);
+	if (err < 0) {
+		dev_err(&client->dev, "irq %d busy?\n", ts->irq);
+		goto err_free_mem;
+	}
+
+	err = input_register_device(input_dev);
+	if (err)
+		goto err_free_irq;
+
+	dev_info(&client->dev, "registered with irq (%d)\n", ts->irq);
+
+	return 0;
+
+ err_free_irq:
+	free_irq(ts->irq, ts);
+	hrtimer_cancel(&ts->timer);
+ err_free_mem:
+	input_free_device(input_dev);
+	kfree(ts);
+	return err;
+}
+
+static int tsc2007_remove(struct i2c_client *client)
+{
+	struct tsc2007	*ts = i2c_get_clientdata(client);
+	struct tsc2007_platform_data *pdata;
+
+	pdata = client->dev.platform_data;
+	pdata->exit_platform_hw();
+
+	free_irq(ts->irq, ts);
+	hrtimer_cancel(&ts->timer);
+	input_unregister_device(ts->input);
+	kfree(ts);
+
+	return 0;
+}
+
+static struct i2c_device_id tsc2007_idtable[] = {
+	{ "tsc2007", 0 },
+	{ }
+};
+
+MODULE_DEVICE_TABLE(i2c, tsc2007_idtable);
+
+static struct i2c_driver tsc2007_driver = {
+	.driver = {
+		.owner	= THIS_MODULE,
+		.name	= "tsc2007"
+	},
+	.id_table	= tsc2007_idtable,
+	.probe		= tsc2007_probe,
+	.remove		= tsc2007_remove,
+};
+
+static int __init tsc2007_init(void)
+{
+	return i2c_add_driver(&tsc2007_driver);
+}
+
+static void __exit tsc2007_exit(void)
+{
+	i2c_del_driver(&tsc2007_driver);
+}
+
+module_init(tsc2007_init);
+module_exit(tsc2007_exit);
+
+MODULE_AUTHOR("Kwangwoo Lee <kwlee@mtekvision.com>");
+MODULE_DESCRIPTION("TSC2007 TouchScreen Driver");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/i2c/tsc2007.h b/include/linux/i2c/tsc2007.h
new file mode 100644
index 000000000000..c6361fbb7bf9
--- /dev/null
+++ b/include/linux/i2c/tsc2007.h
@@ -0,0 +1,17 @@
+#ifndef __LINUX_I2C_TSC2007_H
+#define __LINUX_I2C_TSC2007_H
+
+/* linux/i2c/tsc2007.h */
+
+struct tsc2007_platform_data {
+	u16	model;				/* 2007. */
+	u16	x_plate_ohms;
+
+	int	(*get_pendown_state)(void);
+	void	(*clear_penirq)(void);		/* If needed, clear 2nd level
+						   interrupt source */
+	int	(*init_platform_hw)(void);
+	void	(*exit_platform_hw)(void);
+};
+
+#endif
-- 
cgit v1.2.3


From a01777ecf227de735d7e525ecda48fe74b838a17 Mon Sep 17 00:00:00 2001
From: David Vrabel <david.vrabel@csr.com>
Date: Mon, 22 Dec 2008 18:30:29 +0000
Subject: uwb: remove unused include/linux/uwb/debug.h

Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/uwb/address.c                 |  2 +-
 drivers/uwb/driver.c                  |  2 +-
 drivers/uwb/i1480/i1480u-wlp/lc.c     |  2 +-
 drivers/uwb/i1480/i1480u-wlp/netdev.c |  2 +-
 drivers/uwb/i1480/i1480u-wlp/sysfs.c  |  2 +-
 include/linux/uwb/debug.h             | 82 -----------------------------------
 6 files changed, 5 insertions(+), 87 deletions(-)
 delete mode 100644 include/linux/uwb/debug.h

(limited to 'include/linux')

diff --git a/drivers/uwb/address.c b/drivers/uwb/address.c
index 1664ae5f1706..ad21b1d7218c 100644
--- a/drivers/uwb/address.c
+++ b/drivers/uwb/address.c
@@ -28,7 +28,7 @@
 #include <linux/device.h>
 #include <linux/random.h>
 #include <linux/etherdevice.h>
-#include <linux/uwb/debug.h>
+
 #include "uwb-internal.h"
 
 
diff --git a/drivers/uwb/driver.c b/drivers/uwb/driver.c
index f57c26580de2..da77e41de990 100644
--- a/drivers/uwb/driver.c
+++ b/drivers/uwb/driver.c
@@ -53,7 +53,7 @@
 #include <linux/err.h>
 #include <linux/kdev_t.h>
 #include <linux/random.h>
-#include <linux/uwb/debug.h>
+
 #include "uwb-internal.h"
 
 
diff --git a/drivers/uwb/i1480/i1480u-wlp/lc.c b/drivers/uwb/i1480/i1480u-wlp/lc.c
index 488b2e30a0a8..049c05d4cc6a 100644
--- a/drivers/uwb/i1480/i1480u-wlp/lc.c
+++ b/drivers/uwb/i1480/i1480u-wlp/lc.c
@@ -57,7 +57,7 @@
  */
 #include <linux/if_arp.h>
 #include <linux/etherdevice.h>
-#include <linux/uwb/debug.h>
+
 #include "i1480u-wlp.h"
 
 
diff --git a/drivers/uwb/i1480/i1480u-wlp/netdev.c b/drivers/uwb/i1480/i1480u-wlp/netdev.c
index 2eafb973cd05..e3873ffb942c 100644
--- a/drivers/uwb/i1480/i1480u-wlp/netdev.c
+++ b/drivers/uwb/i1480/i1480u-wlp/netdev.c
@@ -41,7 +41,7 @@
 
 #include <linux/if_arp.h>
 #include <linux/etherdevice.h>
-#include <linux/uwb/debug.h>
+
 #include "i1480u-wlp.h"
 
 struct i1480u_cmd_set_ip_mas {
diff --git a/drivers/uwb/i1480/i1480u-wlp/sysfs.c b/drivers/uwb/i1480/i1480u-wlp/sysfs.c
index a92a787725fc..4ffaf546cc6c 100644
--- a/drivers/uwb/i1480/i1480u-wlp/sysfs.c
+++ b/drivers/uwb/i1480/i1480u-wlp/sysfs.c
@@ -25,8 +25,8 @@
 
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
-#include <linux/uwb/debug.h>
 #include <linux/device.h>
+
 #include "i1480u-wlp.h"
 
 
diff --git a/include/linux/uwb/debug.h b/include/linux/uwb/debug.h
deleted file mode 100644
index 67a240527145..000000000000
--- a/include/linux/uwb/debug.h
+++ /dev/null
@@ -1,82 +0,0 @@
-/*
- * Ultra Wide Band
- * Debug Support
- *
- * Copyright (C) 2005-2006 Intel Corporation
- * Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License version
- * 2 as published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
- * 02110-1301, USA.
- *
- *
- * FIXME: doc
- * Invoke like:
- *
- * #define D_LOCAL 4
- * #include <linux/uwb/debug.h>
- *
- * At the end of your include files.
- */
-#include <linux/types.h>
-
-struct device;
-extern void dump_bytes(struct device *dev, const void *_buf, size_t rsize);
-
-/* Master debug switch; !0 enables, 0 disables */
-#define D_MASTER (!0)
-
-/* Local (per-file) debug switch; #define before #including */
-#ifndef D_LOCAL
-#define D_LOCAL 0
-#endif
-
-#undef __d_printf
-#undef d_fnstart
-#undef d_fnend
-#undef d_printf
-#undef d_dump
-
-#define __d_printf(l, _tag, _dev, f, a...)				\
-do {									\
-	struct device *__dev = (_dev);					\
-	if (D_MASTER && D_LOCAL >= (l)) {				\
-		char __head[64] = "";					\
-		if (_dev != NULL) {					\
-			if ((unsigned long)__dev < 4096)		\
-				printk(KERN_ERR "E: Corrupt dev %p\n",	\
-					__dev);				\
-			else						\
-				snprintf(__head, sizeof(__head),	\
-					 "%s %s: ",			\
-					 dev_driver_string(__dev),	\
-					 dev_name(__dev));		\
-		}							\
-		printk(KERN_ERR "%s%s" _tag ": " f, __head,		\
-			__func__, ## a);				\
-	}								\
-} while (0 && _dev)
-
-#define d_fnstart(l, _dev, f, a...)	\
-	__d_printf(l, " FNSTART", _dev, f, ## a)
-#define d_fnend(l, _dev, f, a...)	\
-	__d_printf(l, " FNEND", _dev, f, ## a)
-#define d_printf(l, _dev, f, a...)	\
-	__d_printf(l, "", _dev, f, ## a)
-#define d_dump(l, _dev, ptr, size)		\
-do {						\
-	struct device *__dev = _dev;		\
-	if (D_MASTER && D_LOCAL >= (l))		\
-		dump_bytes(__dev, ptr, size);	\
-} while (0 && _dev)
-#define d_test(l) (D_MASTER && D_LOCAL >= (l))
-- 
cgit v1.2.3


From 160bbab3000dafccbe43688e48208cecf4deb879 Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Tue, 23 Dec 2008 10:00:14 +0000
Subject: [MTD] struct device - replace bus_id with dev_name(), dev_set_name()

Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/devices/m25p80.c        | 20 ++++++++++----------
 drivers/mtd/devices/mtd_dataflash.c | 32 ++++++++++++++++----------------
 drivers/mtd/maps/integrator-flash.c |  2 +-
 drivers/mtd/maps/ixp2000.c          |  4 ++--
 drivers/mtd/maps/ixp4xx.c           |  2 +-
 drivers/mtd/maps/omap_nor.c         |  2 +-
 drivers/mtd/maps/physmap.c          |  6 +++---
 drivers/mtd/maps/physmap_of.c       |  4 ++--
 drivers/mtd/mtdconcat.c             |  2 +-
 drivers/mtd/nand/fsl_upm.c          |  2 +-
 drivers/mtd/nand/plat_nand.c        |  2 +-
 drivers/mtd/nand/tmio_nand.c        |  2 +-
 drivers/mtd/onenand/generic.c       |  2 +-
 drivers/mtd/onenand/omap2.c         |  2 +-
 drivers/mtd/ubi/build.c             |  2 +-
 drivers/mtd/ubi/vmt.c               |  4 ++--
 include/linux/mtd/concat.h          |  2 +-
 17 files changed, 46 insertions(+), 46 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mtd/devices/m25p80.c b/drivers/mtd/devices/m25p80.c
index 9be0229c3d30..7c3fc766dcf1 100644
--- a/drivers/mtd/devices/m25p80.c
+++ b/drivers/mtd/devices/m25p80.c
@@ -171,8 +171,8 @@ static int wait_till_ready(struct m25p *flash)
 static int erase_chip(struct m25p *flash)
 {
 	DEBUG(MTD_DEBUG_LEVEL3, "%s: %s %lldKiB\n",
-			flash->spi->dev.bus_id, __func__,
-			(long long)(flash->mtd.size >> 10));
+	      dev_name(&flash->spi->dev), __func__,
+	      (long long)(flash->mtd.size >> 10));
 
 	/* Wait until finished previous write command. */
 	if (wait_till_ready(flash))
@@ -198,7 +198,7 @@ static int erase_chip(struct m25p *flash)
 static int erase_sector(struct m25p *flash, u32 offset)
 {
 	DEBUG(MTD_DEBUG_LEVEL3, "%s: %s %dKiB at 0x%08x\n",
-			flash->spi->dev.bus_id, __func__,
+			dev_name(&flash->spi->dev), __func__,
 			flash->mtd.erasesize / 1024, offset);
 
 	/* Wait until finished previous write command. */
@@ -236,8 +236,8 @@ static int m25p80_erase(struct mtd_info *mtd, struct erase_info *instr)
 	uint32_t rem;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: %s %s 0x%llx, len %lld\n",
-			flash->spi->dev.bus_id, __func__, "at",
-			(long long)instr->addr, (long long)instr->len);
+	      dev_name(&flash->spi->dev), __func__, "at",
+	      (long long)instr->addr, (long long)instr->len);
 
 	/* sanity checks */
 	if (instr->addr + instr->len > flash->mtd.size)
@@ -296,7 +296,7 @@ static int m25p80_read(struct mtd_info *mtd, loff_t from, size_t len,
 	struct spi_message m;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: %s %s 0x%08x, len %zd\n",
-			flash->spi->dev.bus_id, __func__, "from",
+			dev_name(&flash->spi->dev), __func__, "from",
 			(u32)from, len);
 
 	/* sanity checks */
@@ -368,7 +368,7 @@ static int m25p80_write(struct mtd_info *mtd, loff_t to, size_t len,
 	struct spi_message m;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: %s %s 0x%08x, len %zd\n",
-			flash->spi->dev.bus_id, __func__, "to",
+			dev_name(&flash->spi->dev), __func__, "to",
 			(u32)to, len);
 
 	if (retlen)
@@ -564,7 +564,7 @@ static struct flash_info *__devinit jedec_probe(struct spi_device *spi)
 	tmp = spi_write_then_read(spi, &code, 1, id, 5);
 	if (tmp < 0) {
 		DEBUG(MTD_DEBUG_LEVEL0, "%s: error %d reading JEDEC ID\n",
-			spi->dev.bus_id, tmp);
+			dev_name(&spi->dev), tmp);
 		return NULL;
 	}
 	jedec = id[0];
@@ -618,7 +618,7 @@ static int __devinit m25p_probe(struct spi_device *spi)
 		/* unrecognized chip? */
 		if (i == ARRAY_SIZE(m25p_data)) {
 			DEBUG(MTD_DEBUG_LEVEL0, "%s: unrecognized id %s\n",
-					spi->dev.bus_id, data->type);
+					dev_name(&spi->dev), data->type);
 			info = NULL;
 
 		/* recognized; is that chip really what's there? */
@@ -659,7 +659,7 @@ static int __devinit m25p_probe(struct spi_device *spi)
 	if (data && data->name)
 		flash->mtd.name = data->name;
 	else
-		flash->mtd.name = spi->dev.bus_id;
+		flash->mtd.name = dev_name(&spi->dev);
 
 	flash->mtd.type = MTD_NORFLASH;
 	flash->mtd.writesize = 1;
diff --git a/drivers/mtd/devices/mtd_dataflash.c b/drivers/mtd/devices/mtd_dataflash.c
index 68068975940b..d44f741ae229 100644
--- a/drivers/mtd/devices/mtd_dataflash.c
+++ b/drivers/mtd/devices/mtd_dataflash.c
@@ -129,7 +129,7 @@ static int dataflash_waitready(struct spi_device *spi)
 		status = dataflash_status(spi);
 		if (status < 0) {
 			DEBUG(MTD_DEBUG_LEVEL1, "%s: status %d?\n",
-					spi->dev.bus_id, status);
+					dev_name(&spi->dev), status);
 			status = 0;
 		}
 
@@ -156,8 +156,8 @@ static int dataflash_erase(struct mtd_info *mtd, struct erase_info *instr)
 	uint32_t		rem;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: erase addr=0x%llx len 0x%llx\n",
-			spi->dev.bus_id, (long long)instr->addr,
-			(long long)instr->len);
+	      dev_name(&spi->dev), (long long)instr->addr,
+	      (long long)instr->len);
 
 	/* Sanity checks */
 	if (instr->addr + instr->len > mtd->size)
@@ -203,7 +203,7 @@ static int dataflash_erase(struct mtd_info *mtd, struct erase_info *instr)
 
 		if (status < 0) {
 			printk(KERN_ERR "%s: erase %x, err %d\n",
-				spi->dev.bus_id, pageaddr, status);
+				dev_name(&spi->dev), pageaddr, status);
 			/* REVISIT:  can retry instr->retries times; or
 			 * giveup and instr->fail_addr = instr->addr;
 			 */
@@ -245,7 +245,7 @@ static int dataflash_read(struct mtd_info *mtd, loff_t from, size_t len,
 	int			status;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: read 0x%x..0x%x\n",
-		priv->spi->dev.bus_id, (unsigned)from, (unsigned)(from + len));
+		dev_name(&priv->spi->dev), (unsigned)from, (unsigned)(from + len));
 
 	*retlen = 0;
 
@@ -294,7 +294,7 @@ static int dataflash_read(struct mtd_info *mtd, loff_t from, size_t len,
 		status = 0;
 	} else
 		DEBUG(MTD_DEBUG_LEVEL1, "%s: read %x..%x --> %d\n",
-			priv->spi->dev.bus_id,
+			dev_name(&priv->spi->dev),
 			(unsigned)from, (unsigned)(from + len),
 			status);
 	return status;
@@ -321,7 +321,7 @@ static int dataflash_write(struct mtd_info *mtd, loff_t to, size_t len,
 	uint8_t			*command;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: write 0x%x..0x%x\n",
-		spi->dev.bus_id, (unsigned)to, (unsigned)(to + len));
+		dev_name(&spi->dev), (unsigned)to, (unsigned)(to + len));
 
 	*retlen = 0;
 
@@ -380,7 +380,7 @@ static int dataflash_write(struct mtd_info *mtd, loff_t to, size_t len,
 			status = spi_sync(spi, &msg);
 			if (status < 0)
 				DEBUG(MTD_DEBUG_LEVEL1, "%s: xfer %u -> %d \n",
-					spi->dev.bus_id, addr, status);
+					dev_name(&spi->dev), addr, status);
 
 			(void) dataflash_waitready(priv->spi);
 		}
@@ -402,7 +402,7 @@ static int dataflash_write(struct mtd_info *mtd, loff_t to, size_t len,
 		spi_transfer_del(x + 1);
 		if (status < 0)
 			DEBUG(MTD_DEBUG_LEVEL1, "%s: pgm %u/%u -> %d \n",
-				spi->dev.bus_id, addr, writelen, status);
+				dev_name(&spi->dev), addr, writelen, status);
 
 		(void) dataflash_waitready(priv->spi);
 
@@ -422,14 +422,14 @@ static int dataflash_write(struct mtd_info *mtd, loff_t to, size_t len,
 		status = spi_sync(spi, &msg);
 		if (status < 0)
 			DEBUG(MTD_DEBUG_LEVEL1, "%s: compare %u -> %d \n",
-				spi->dev.bus_id, addr, status);
+				dev_name(&spi->dev), addr, status);
 
 		status = dataflash_waitready(priv->spi);
 
 		/* Check result of the compare operation */
 		if (status & (1 << 6)) {
 			printk(KERN_ERR "%s: compare page %u, err %d\n",
-				spi->dev.bus_id, pageaddr, status);
+				dev_name(&spi->dev), pageaddr, status);
 			remaining = 0;
 			status = -EIO;
 			break;
@@ -785,7 +785,7 @@ static struct flash_info *__devinit jedec_probe(struct spi_device *spi)
 	tmp = spi_write_then_read(spi, &code, 1, id, 3);
 	if (tmp < 0) {
 		DEBUG(MTD_DEBUG_LEVEL0, "%s: error %d reading JEDEC ID\n",
-			spi->dev.bus_id, tmp);
+			dev_name(&spi->dev), tmp);
 		return ERR_PTR(tmp);
 	}
 	if (id[0] != 0x1f)
@@ -875,7 +875,7 @@ static int __devinit dataflash_probe(struct spi_device *spi)
 	status = dataflash_status(spi);
 	if (status <= 0 || status == 0xff) {
 		DEBUG(MTD_DEBUG_LEVEL1, "%s: status error %d\n",
-				spi->dev.bus_id, status);
+				dev_name(&spi->dev), status);
 		if (status == 0 || status == 0xff)
 			status = -ENODEV;
 		return status;
@@ -911,13 +911,13 @@ static int __devinit dataflash_probe(struct spi_device *spi)
 	/* obsolete AT45DB1282 not (yet?) supported */
 	default:
 		DEBUG(MTD_DEBUG_LEVEL1, "%s: unsupported device (%x)\n",
-				spi->dev.bus_id, status & 0x3c);
+				dev_name(&spi->dev), status & 0x3c);
 		status = -ENODEV;
 	}
 
 	if (status < 0)
 		DEBUG(MTD_DEBUG_LEVEL1, "%s: add_dataflash --> %d\n",
-				spi->dev.bus_id, status);
+				dev_name(&spi->dev), status);
 
 	return status;
 }
@@ -927,7 +927,7 @@ static int __devexit dataflash_remove(struct spi_device *spi)
 	struct dataflash	*flash = dev_get_drvdata(&spi->dev);
 	int			status;
 
-	DEBUG(MTD_DEBUG_LEVEL1, "%s: remove\n", spi->dev.bus_id);
+	DEBUG(MTD_DEBUG_LEVEL1, "%s: remove\n", dev_name(&spi->dev));
 
 	if (mtd_has_partitions() && flash->partitioned)
 		status = del_mtd_partitions(&flash->mtd);
diff --git a/drivers/mtd/maps/integrator-flash.c b/drivers/mtd/maps/integrator-flash.c
index 7100ee3c7b01..d2ec262666c7 100644
--- a/drivers/mtd/maps/integrator-flash.c
+++ b/drivers/mtd/maps/integrator-flash.c
@@ -105,7 +105,7 @@ static int armflash_probe(struct platform_device *dev)
 	info->map.bankwidth	= plat->width;
 	info->map.phys		= res->start;
 	info->map.virt		= base;
-	info->map.name		= dev->dev.bus_id;
+	info->map.name		= dev_name(&dev->dev);
 	info->map.set_vpp	= armflash_set_vpp;
 
 	simple_map_init(&info->map);
diff --git a/drivers/mtd/maps/ixp2000.c b/drivers/mtd/maps/ixp2000.c
index dcdb1f17577d..d76880d91bdb 100644
--- a/drivers/mtd/maps/ixp2000.c
+++ b/drivers/mtd/maps/ixp2000.c
@@ -188,7 +188,7 @@ static int ixp2000_flash_probe(struct platform_device *dev)
  	 */
 	info->map.map_priv_2 = (unsigned long) ixp_data->bank_setup;
 
-	info->map.name = dev->dev.bus_id;
+	info->map.name = dev_name(&dev->dev);
 	info->map.read = ixp2000_flash_read8;
 	info->map.write = ixp2000_flash_write8;
 	info->map.copy_from = ixp2000_flash_copy_from;
@@ -196,7 +196,7 @@ static int ixp2000_flash_probe(struct platform_device *dev)
 
 	info->res = request_mem_region(dev->resource->start,
 			dev->resource->end - dev->resource->start + 1,
-			dev->dev.bus_id);
+			dev_name(&dev->dev));
 	if (!info->res) {
 		dev_err(&dev->dev, "Could not reserve memory region\n");
 		err = -ENOMEM;
diff --git a/drivers/mtd/maps/ixp4xx.c b/drivers/mtd/maps/ixp4xx.c
index 9c7a5fbd4e51..4d0be2f1503f 100644
--- a/drivers/mtd/maps/ixp4xx.c
+++ b/drivers/mtd/maps/ixp4xx.c
@@ -218,7 +218,7 @@ static int ixp4xx_flash_probe(struct platform_device *dev)
 	 * handle that.
 	 */
 	info->map.bankwidth = 2;
-	info->map.name = dev->dev.bus_id;
+	info->map.name = dev_name(&dev->dev);
 	info->map.read = ixp4xx_read16,
 	info->map.write = ixp4xx_probe_write16,
 	info->map.copy_from = ixp4xx_copy_from,
diff --git a/drivers/mtd/maps/omap_nor.c b/drivers/mtd/maps/omap_nor.c
index 05f276af15da..7e50e9b1b781 100644
--- a/drivers/mtd/maps/omap_nor.c
+++ b/drivers/mtd/maps/omap_nor.c
@@ -101,7 +101,7 @@ static int __init omapflash_probe(struct platform_device *pdev)
 		err = -ENOMEM;
 		goto out_release_mem_region;
 	}
-	info->map.name		= pdev->dev.bus_id;
+	info->map.name		= dev_name(&pdev->dev);
 	info->map.phys		= res->start;
 	info->map.size		= size;
 	info->map.bankwidth	= pdata->width;
diff --git a/drivers/mtd/maps/physmap.c b/drivers/mtd/maps/physmap.c
index 58207b3b9411..d3a2acc7e9be 100644
--- a/drivers/mtd/maps/physmap.c
+++ b/drivers/mtd/maps/physmap.c
@@ -106,13 +106,13 @@ static int physmap_flash_probe(struct platform_device *dev)
 		if (!devm_request_mem_region(&dev->dev,
 			dev->resource[i].start,
 			dev->resource[i].end - dev->resource[i].start + 1,
-			dev->dev.bus_id)) {
+			dev_name(&dev->dev))) {
 			dev_err(&dev->dev, "Could not reserve memory region\n");
 			err = -ENOMEM;
 			goto err_out;
 		}
 
-		info->map[i].name = dev->dev.bus_id;
+		info->map[i].name = dev_name(&dev->dev);
 		info->map[i].phys = dev->resource[i].start;
 		info->map[i].size = dev->resource[i].end - dev->resource[i].start + 1;
 		info->map[i].bankwidth = physmap_data->width;
@@ -148,7 +148,7 @@ static int physmap_flash_probe(struct platform_device *dev)
 		 * We detected multiple devices. Concatenate them together.
 		 */
 #ifdef CONFIG_MTD_CONCAT
-		info->cmtd = mtd_concat_create(info->mtd, devices_found, dev->dev.bus_id);
+		info->cmtd = mtd_concat_create(info->mtd, devices_found, dev_name(&dev->dev));
 		if (info->cmtd == NULL)
 			err = -ENXIO;
 #else
diff --git a/drivers/mtd/maps/physmap_of.c b/drivers/mtd/maps/physmap_of.c
index 5fcfec034a94..fbf0ca939d72 100644
--- a/drivers/mtd/maps/physmap_of.c
+++ b/drivers/mtd/maps/physmap_of.c
@@ -183,7 +183,7 @@ static int __devinit of_flash_probe(struct of_device *dev,
 
 	err = -EBUSY;
 	info->res = request_mem_region(res.start, res.end - res.start + 1,
-				       dev->dev.bus_id);
+				       dev_name(&dev->dev));
 	if (!info->res)
 		goto err_out;
 
@@ -194,7 +194,7 @@ static int __devinit of_flash_probe(struct of_device *dev,
 		goto err_out;
 	}
 
-	info->map.name = dev->dev.bus_id;
+	info->map.name = dev_name(&dev->dev);
 	info->map.phys = res.start;
 	info->map.size = res.end - res.start + 1;
 	info->map.bankwidth = *width;
diff --git a/drivers/mtd/mtdconcat.c b/drivers/mtd/mtdconcat.c
index c26dd528d094..3dbb1b38db66 100644
--- a/drivers/mtd/mtdconcat.c
+++ b/drivers/mtd/mtdconcat.c
@@ -691,7 +691,7 @@ static int concat_block_markbad(struct mtd_info *mtd, loff_t ofs)
  */
 struct mtd_info *mtd_concat_create(struct mtd_info *subdev[],	/* subdevices to concatenate */
 				   int num_devs,	/* number of subdevices      */
-				   char *name)
+				   const char *name)
 {				/* name for the new device   */
 	int i;
 	size_t size;
diff --git a/drivers/mtd/nand/fsl_upm.c b/drivers/mtd/nand/fsl_upm.c
index a83192f80eba..7815a404a632 100644
--- a/drivers/mtd/nand/fsl_upm.c
+++ b/drivers/mtd/nand/fsl_upm.c
@@ -222,7 +222,7 @@ static int __devinit fun_probe(struct of_device *ofdev,
 
 	fun->rnb_gpio = of_get_gpio(ofdev->node, 0);
 	if (fun->rnb_gpio >= 0) {
-		ret = gpio_request(fun->rnb_gpio, ofdev->dev.bus_id);
+		ret = gpio_request(fun->rnb_gpio, dev_name(&ofdev->dev));
 		if (ret) {
 			dev_err(&ofdev->dev, "can't request RNB gpio\n");
 			goto err2;
diff --git a/drivers/mtd/nand/plat_nand.c b/drivers/mtd/nand/plat_nand.c
index f674c5427b17..75f9f4874ecf 100644
--- a/drivers/mtd/nand/plat_nand.c
+++ b/drivers/mtd/nand/plat_nand.c
@@ -54,7 +54,7 @@ static int __init plat_nand_probe(struct platform_device *pdev)
 	data->chip.priv = &data;
 	data->mtd.priv = &data->chip;
 	data->mtd.owner = THIS_MODULE;
-	data->mtd.name = pdev->dev.bus_id;
+	data->mtd.name = dev_name(&pdev->dev);
 
 	data->chip.IO_ADDR_R = data->io_base;
 	data->chip.IO_ADDR_W = data->io_base;
diff --git a/drivers/mtd/nand/tmio_nand.c b/drivers/mtd/nand/tmio_nand.c
index edb1e322113d..daa6a4c3b8ce 100644
--- a/drivers/mtd/nand/tmio_nand.c
+++ b/drivers/mtd/nand/tmio_nand.c
@@ -433,7 +433,7 @@ static int tmio_probe(struct platform_device *dev)
 	nand_chip->chip_delay = 15;
 
 	retval = request_irq(irq, &tmio_irq,
-				IRQF_DISABLED, dev->dev.bus_id, tmio);
+				IRQF_DISABLED, dev_name(&dev->dev), tmio);
 	if (retval) {
 		dev_err(&dev->dev, "request_irq error %d\n", retval);
 		goto err_irq;
diff --git a/drivers/mtd/onenand/generic.c b/drivers/mtd/onenand/generic.c
index ad81ab8e95e2..5b69e7773c6c 100644
--- a/drivers/mtd/onenand/generic.c
+++ b/drivers/mtd/onenand/generic.c
@@ -63,7 +63,7 @@ static int __devinit generic_onenand_probe(struct device *dev)
 	info->onenand.mmcontrol = pdata->mmcontrol;
 	info->onenand.irq = platform_get_irq(pdev, 0);
 
-	info->mtd.name = pdev->dev.bus_id;
+	info->mtd.name = dev_name(&pdev->dev);
 	info->mtd.priv = &info->onenand;
 	info->mtd.owner = THIS_MODULE;
 
diff --git a/drivers/mtd/onenand/omap2.c b/drivers/mtd/onenand/omap2.c
index a7e4d985f5ef..710edee790b5 100644
--- a/drivers/mtd/onenand/omap2.c
+++ b/drivers/mtd/onenand/omap2.c
@@ -668,7 +668,7 @@ static int __devinit omap2_onenand_probe(struct platform_device *pdev)
 		 c->onenand.base);
 
 	c->pdev = pdev;
-	c->mtd.name = pdev->dev.bus_id;
+	c->mtd.name = dev_name(&pdev->dev);
 	c->mtd.priv = &c->onenand;
 	c->mtd.owner = THIS_MODULE;
 
diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index 634e2e86525f..84a134ead7cc 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -280,7 +280,7 @@ static int ubi_sysfs_init(struct ubi_device *ubi)
 	ubi->dev.release = dev_release;
 	ubi->dev.devt = ubi->cdev.dev;
 	ubi->dev.class = ubi_class;
-	sprintf(&ubi->dev.bus_id[0], UBI_NAME_STR"%d", ubi->ubi_num);
+	dev_set_name(&ubi->dev, UBI_NAME_STR"%d", ubi->ubi_num);
 	err = device_register(&ubi->dev);
 	if (err)
 		return err;
diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c
index 3531ca9a1e24..22e1d7398fce 100644
--- a/drivers/mtd/ubi/vmt.c
+++ b/drivers/mtd/ubi/vmt.c
@@ -329,7 +329,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
 	vol->dev.devt = dev;
 	vol->dev.class = ubi_class;
 
-	sprintf(&vol->dev.bus_id[0], "%s_%d", ubi->ubi_name, vol->vol_id);
+	dev_set_name(&vol->dev, "%s_%d", ubi->ubi_name, vol->vol_id);
 	err = device_register(&vol->dev);
 	if (err) {
 		ubi_err("cannot register device");
@@ -678,7 +678,7 @@ int ubi_add_volume(struct ubi_device *ubi, struct ubi_volume *vol)
 	vol->dev.parent = &ubi->dev;
 	vol->dev.devt = dev;
 	vol->dev.class = ubi_class;
-	sprintf(&vol->dev.bus_id[0], "%s_%d", ubi->ubi_name, vol->vol_id);
+	dev_set_name(&vol->dev, "%s_%d", ubi->ubi_name, vol->vol_id);
 	err = device_register(&vol->dev);
 	if (err)
 		goto out_gluebi;
diff --git a/include/linux/mtd/concat.h b/include/linux/mtd/concat.h
index c02f3d264ecf..e80c674daeb3 100644
--- a/include/linux/mtd/concat.h
+++ b/include/linux/mtd/concat.h
@@ -13,7 +13,7 @@
 struct mtd_info *mtd_concat_create(
     struct mtd_info *subdev[],  /* subdevices to concatenate */
     int num_devs,               /* number of subdevices      */
-    char *name);                /* name for the new device   */
+    const char *name);          /* name for the new device   */
 
 void mtd_concat_destroy(struct mtd_info *mtd);
 
-- 
cgit v1.2.3


From f9af0e70911e9d6cc9a68f784dca86415486084d Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Fri, 26 Dec 2008 12:24:24 +0900
Subject: irq: for_each_irq_desc() move to irqnr.h

Impact: cleanup

before CONFIG_SPARSE_IRQ age, for_each_irq_desc() sat in irqnr.h and
could be called from generic code.

CONFIG_SPARSE_IRQ breaks this assumption, but SPARSE_IRQ version
for_each_irq_desc() also can move into irqnr.h easily.

Also, this patch unifies CONFIG_SPARSE_IRQ and !CONFIG_SPARSE_IRQ
for_each_irq_desc().

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/irq.h   | 24 ++++--------------------
 include/linux/irqnr.h | 19 +++++++++----------
 kernel/irq/handle.c   | 13 +++++++++++--
 3 files changed, 24 insertions(+), 32 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/irq.h b/include/linux/irq.h
index 98564dc64476..69da275c0ebd 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -202,33 +202,17 @@ extern void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc
 
 #ifndef CONFIG_SPARSE_IRQ
 extern struct irq_desc irq_desc[NR_IRQS];
-
-static inline struct irq_desc *irq_to_desc(unsigned int irq)
-{
-	return (irq < NR_IRQS) ? irq_desc + irq : NULL;
-}
-static inline struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu)
-{
-	return irq_to_desc(irq);
-}
-
-#else
-
-extern struct irq_desc *irq_to_desc(unsigned int irq);
-extern struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu);
+#else /* CONFIG_SPARSE_IRQ */
 extern struct irq_desc *move_irq_desc(struct irq_desc *old_desc, int cpu);
 
-# define for_each_irq_desc(irq, desc)		\
-	for (irq = 0, desc = irq_to_desc(irq); irq < nr_irqs; irq++, desc = irq_to_desc(irq))
-# define for_each_irq_desc_reverse(irq, desc)                          \
-	for (irq = nr_irqs - 1, desc = irq_to_desc(irq); irq >= 0; irq--, desc = irq_to_desc(irq))
-
 #define kstat_irqs_this_cpu(DESC) \
 	((DESC)->kstat_irqs[smp_processor_id()])
 #define kstat_incr_irqs_this_cpu(irqno, DESC) \
 	((DESC)->kstat_irqs[smp_processor_id()]++)
 
-#endif
+#endif /* CONFIG_SPARSE_IRQ */
+
+extern struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu);
 
 static inline struct irq_desc *
 irq_remap_to_desc(unsigned int irq, struct irq_desc *desc)
diff --git a/include/linux/irqnr.h b/include/linux/irqnr.h
index 95d2b74641f5..c4a59c7a478b 100644
--- a/include/linux/irqnr.h
+++ b/include/linux/irqnr.h
@@ -15,20 +15,19 @@
 
 # define for_each_irq_desc_reverse(irq, desc)                          \
 	for (irq = nr_irqs - 1; irq >= 0; irq--)
-#else
+#else /* CONFIG_GENERIC_HARDIRQS */
 
 extern int nr_irqs;
+extern struct irq_desc *irq_to_desc(unsigned int irq);
 
-#ifndef CONFIG_SPARSE_IRQ
+# define for_each_irq_desc(irq, desc)					\
+	for (irq = 0, desc = irq_to_desc(irq); irq < nr_irqs;		\
+	     irq++, desc = irq_to_desc(irq))
+# define for_each_irq_desc_reverse(irq, desc)				\
+	for (irq = nr_irqs - 1, desc = irq_to_desc(irq); irq >= 0;	\
+	     irq--, desc = irq_to_desc(irq))
 
-struct irq_desc;
-# define for_each_irq_desc(irq, desc)		\
-	for (irq = 0, desc = irq_desc; irq < nr_irqs; irq++, desc++)
-# define for_each_irq_desc_reverse(irq, desc)                          \
-	for (irq = nr_irqs - 1, desc = irq_desc + (nr_irqs - 1);        \
-	    irq >= 0; irq--, desc--)
-#endif
-#endif
+#endif /* CONFIG_GENERIC_HARDIRQS */
 
 #define for_each_irq_nr(irq)                   \
        for (irq = 0; irq < nr_irqs; irq++)
diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index 6492400cb50d..4db7d2df86b6 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -203,7 +203,7 @@ out_unlock:
 	return desc;
 }
 
-#else
+#else /* !CONFIG_SPARSE_IRQ */
 
 struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = {
 	[0 ... NR_IRQS-1] = {
@@ -218,7 +218,16 @@ struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = {
 	}
 };
 
-#endif
+struct irq_desc *irq_to_desc(unsigned int irq)
+{
+	return (irq < NR_IRQS) ? irq_desc + irq : NULL;
+}
+
+struct irq_desc *irq_to_desc_alloc_cpu(unsigned int irq, int cpu)
+{
+	return irq_to_desc(irq);
+}
+#endif /* !CONFIG_SPARSE_IRQ */
 
 /*
  * What should we do if we get a hw irq event on an illegal vector?
-- 
cgit v1.2.3


From 18eefedfe8ad33e8fc7614c13359e29a9fab4644 Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Fri, 26 Dec 2008 12:29:48 +0900
Subject: irq: simplify for_each_irq_desc() usage

Impact: cleanup

all for_each_irq_desc() usage point have !desc check.
then its check can move into for_each_irq_desc() macro.

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/io_apic.c | 10 ----------
 drivers/xen/events.c      |  3 ---
 include/linux/irqnr.h     |  8 ++++++--
 kernel/irq/autoprobe.c    | 15 ---------------
 kernel/irq/handle.c       |  3 ---
 kernel/irq/spurious.c     |  5 -----
 6 files changed, 6 insertions(+), 38 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c
index a74887b416cc..2fe543f58ac8 100644
--- a/arch/x86/kernel/io_apic.c
+++ b/arch/x86/kernel/io_apic.c
@@ -1345,8 +1345,6 @@ void __setup_vector_irq(int cpu)
 
 	/* Mark the inuse vectors */
 	for_each_irq_desc(irq, desc) {
-		if (!desc)
-			continue;
 		cfg = desc->chip_data;
 		if (!cpu_isset(cpu, cfg->domain))
 			continue;
@@ -1730,8 +1728,6 @@ __apicdebuginit(void) print_IO_APIC(void)
 	for_each_irq_desc(irq, desc) {
 		struct irq_pin_list *entry;
 
-		if (!desc)
-			continue;
 		cfg = desc->chip_data;
 		entry = cfg->irq_2_pin;
 		if (!entry)
@@ -2378,9 +2374,6 @@ static void ir_irq_migration(struct work_struct *work)
 	struct irq_desc *desc;
 
 	for_each_irq_desc(irq, desc) {
-		if (!desc)
-			continue;
-
 		if (desc->status & IRQ_MOVE_PENDING) {
 			unsigned long flags;
 
@@ -2671,9 +2664,6 @@ static inline void init_IO_APIC_traps(void)
 	 * 0x80, because int 0x80 is hm, kind of importantish. ;)
 	 */
 	for_each_irq_desc(irq, desc) {
-		if (!desc)
-			continue;
-
 		cfg = desc->chip_data;
 		if (IO_APIC_IRQ(irq) && cfg && !cfg->vector) {
 			/*
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 46625cd38743..e26733a9df21 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -142,9 +142,6 @@ static void init_evtchn_cpu_bindings(void)
 
 	/* By default all event channels notify CPU#0. */
 	for_each_irq_desc(i, desc) {
-		if (!desc)
-			continue;
-
 		desc->affinity = cpumask_of_cpu(0);
 	}
 #endif
diff --git a/include/linux/irqnr.h b/include/linux/irqnr.h
index c4a59c7a478b..5504a5c97836 100644
--- a/include/linux/irqnr.h
+++ b/include/linux/irqnr.h
@@ -22,10 +22,14 @@ extern struct irq_desc *irq_to_desc(unsigned int irq);
 
 # define for_each_irq_desc(irq, desc)					\
 	for (irq = 0, desc = irq_to_desc(irq); irq < nr_irqs;		\
-	     irq++, desc = irq_to_desc(irq))
+	     irq++, desc = irq_to_desc(irq))				\
+		if (desc)
+
+
 # define for_each_irq_desc_reverse(irq, desc)				\
 	for (irq = nr_irqs - 1, desc = irq_to_desc(irq); irq >= 0;	\
-	     irq--, desc = irq_to_desc(irq))
+	     irq--, desc = irq_to_desc(irq))				\
+		if (desc)
 
 #endif /* CONFIG_GENERIC_HARDIRQS */
 
diff --git a/kernel/irq/autoprobe.c b/kernel/irq/autoprobe.c
index 650ce4102a63..cc0f7321b8ce 100644
--- a/kernel/irq/autoprobe.c
+++ b/kernel/irq/autoprobe.c
@@ -40,9 +40,6 @@ unsigned long probe_irq_on(void)
 	 * flush such a longstanding irq before considering it as spurious.
 	 */
 	for_each_irq_desc_reverse(i, desc) {
-		if (!desc)
-			continue;
-
 		spin_lock_irq(&desc->lock);
 		if (!desc->action && !(desc->status & IRQ_NOPROBE)) {
 			/*
@@ -71,9 +68,6 @@ unsigned long probe_irq_on(void)
 	 * happened in the previous stage, it may have masked itself)
 	 */
 	for_each_irq_desc_reverse(i, desc) {
-		if (!desc)
-			continue;
-
 		spin_lock_irq(&desc->lock);
 		if (!desc->action && !(desc->status & IRQ_NOPROBE)) {
 			desc->status |= IRQ_AUTODETECT | IRQ_WAITING;
@@ -92,9 +86,6 @@ unsigned long probe_irq_on(void)
 	 * Now filter out any obviously spurious interrupts
 	 */
 	for_each_irq_desc(i, desc) {
-		if (!desc)
-			continue;
-
 		spin_lock_irq(&desc->lock);
 		status = desc->status;
 
@@ -133,9 +124,6 @@ unsigned int probe_irq_mask(unsigned long val)
 	int i;
 
 	for_each_irq_desc(i, desc) {
-		if (!desc)
-			continue;
-
 		spin_lock_irq(&desc->lock);
 		status = desc->status;
 
@@ -178,9 +166,6 @@ int probe_irq_off(unsigned long val)
 	unsigned int status;
 
 	for_each_irq_desc(i, desc) {
-		if (!desc)
-			continue;
-
 		spin_lock_irq(&desc->lock);
 		status = desc->status;
 
diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index 03479dfdebb8..7dbdfe524693 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -437,9 +437,6 @@ void early_init_irq_lock_class(void)
 	int i;
 
 	for_each_irq_desc(i, desc) {
-		if (!desc)
-			continue;
-
 		lockdep_set_class(&desc->lock, &irq_desc_lock_class);
 	}
 }
diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c
index 3738107531fd..dd364c11e56e 100644
--- a/kernel/irq/spurious.c
+++ b/kernel/irq/spurious.c
@@ -91,9 +91,6 @@ static int misrouted_irq(int irq)
 	int i, ok = 0;
 
 	for_each_irq_desc(i, desc) {
-		if (!desc)
-			continue;
-
 		if (!i)
 			 continue;
 
@@ -115,8 +112,6 @@ static void poll_spurious_irqs(unsigned long dummy)
 	for_each_irq_desc(i, desc) {
 		unsigned int status;
 
-		if (!desc)
-			continue;
 		if (!i)
 			 continue;
 
-- 
cgit v1.2.3


From 13a0c3c269b223f60abfac8a9811d77111a8b4ba Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Fri, 26 Dec 2008 02:05:47 -0800
Subject: sparseirq: work around compiler optimizing away __weak functions

Impact: fix panic on null pointer with sparseirq

Some GCC versions seem to inline the weak global function,
when that function is empty.

Work it around, by making the functions return a (dummy) integer.

Signed-off-by: Yinghai <yinghai@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/io_apic.c | 8 ++++++--
 include/linux/irq.h       | 6 +++---
 init/main.c               | 7 ++++---
 kernel/irq/handle.c       | 7 ++++---
 4 files changed, 17 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c
index 2fe543f58ac8..976039377846 100644
--- a/arch/x86/kernel/io_apic.c
+++ b/arch/x86/kernel/io_apic.c
@@ -170,7 +170,7 @@ static struct irq_cfg irq_cfgx[NR_IRQS] = {
 	[15] = { .domain = CPU_MASK_ALL, .vector = IRQ15_VECTOR, },
 };
 
-void __init arch_early_irq_init(void)
+int __init arch_early_irq_init(void)
 {
 	struct irq_cfg *cfg;
 	struct irq_desc *desc;
@@ -184,6 +184,8 @@ void __init arch_early_irq_init(void)
 		desc = irq_to_desc(i);
 		desc->chip_data = &cfg[i];
 	}
+
+	return 0;
 }
 
 #ifdef CONFIG_SPARSE_IRQ
@@ -212,7 +214,7 @@ static struct irq_cfg *get_one_free_irq_cfg(int cpu)
 	return cfg;
 }
 
-void arch_init_chip_data(struct irq_desc *desc, int cpu)
+int arch_init_chip_data(struct irq_desc *desc, int cpu)
 {
 	struct irq_cfg *cfg;
 
@@ -224,6 +226,8 @@ void arch_init_chip_data(struct irq_desc *desc, int cpu)
 			BUG_ON(1);
 		}
 	}
+
+	return 0;
 }
 
 #ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 69da275c0ebd..0e40af4bac40 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -193,9 +193,9 @@ struct irq_desc {
 	const char		*name;
 } ____cacheline_internodealigned_in_smp;
 
-extern void early_irq_init(void);
-extern void arch_early_irq_init(void);
-extern void arch_init_chip_data(struct irq_desc *desc, int cpu);
+extern int early_irq_init(void);
+extern int arch_early_irq_init(void);
+extern int arch_init_chip_data(struct irq_desc *desc, int cpu);
 extern void arch_init_copy_chip_data(struct irq_desc *old_desc,
 					struct irq_desc *desc, int cpu);
 extern void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc);
diff --git a/init/main.c b/init/main.c
index c1f999a3cf31..c314aa15370e 100644
--- a/init/main.c
+++ b/init/main.c
@@ -539,13 +539,14 @@ void __init __weak thread_info_cache_init(void)
 {
 }
 
-void __init __weak arch_early_irq_init(void)
+int __init __weak arch_early_irq_init(void)
 {
+	return 0;
 }
 
-void __init __weak early_irq_init(void)
+int __init __weak early_irq_init(void)
 {
-	arch_early_irq_init();
+	return arch_early_irq_init();
 }
 
 asmlinkage void __init start_kernel(void)
diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index 893da67b7781..0bef3ecb7a0e 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -86,8 +86,9 @@ void init_kstat_irqs(struct irq_desc *desc, int cpu, int nr)
 		desc->kstat_irqs = (unsigned int *)ptr;
 }
 
-void __attribute__((weak)) arch_init_chip_data(struct irq_desc *desc, int cpu)
+int __weak arch_init_chip_data(struct irq_desc *desc, int cpu)
 {
+	return 0;
 }
 
 static void init_one_irq_desc(int irq, struct irq_desc *desc, int cpu)
@@ -132,7 +133,7 @@ static struct irq_desc irq_desc_legacy[NR_IRQS_LEGACY] __cacheline_aligned_in_sm
 /* FIXME: use bootmem alloc ...*/
 static unsigned int kstat_irqs_legacy[NR_IRQS_LEGACY][NR_CPUS];
 
-void __init early_irq_init(void)
+int __init early_irq_init(void)
 {
 	struct irq_desc *desc;
 	int legacy_count;
@@ -151,7 +152,7 @@ void __init early_irq_init(void)
 	for (i = legacy_count; i < NR_IRQS; i++)
 		irq_desc_ptrs[i] = NULL;
 
-	arch_early_irq_init();
+	return arch_early_irq_init();
 }
 
 struct irq_desc *irq_to_desc(unsigned int irq)
-- 
cgit v1.2.3


From 70a7d3cc1308a55104fbe505d76f2aca8a4cf53e Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Mon, 22 Dec 2008 10:26:05 -0800
Subject: swiotlb: add hwdev to swiotlb_phys_to_bus() / swiotlb_sg_to_bus()

Impact: extend functions with a (yet unused) parameter, update callsites

Some architectures need it - in preparation for highmem swiotlb.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/pci-swiotlb_64.c |  2 +-
 include/linux/swiotlb.h          |  3 ++-
 lib/swiotlb.c                    | 53 +++++++++++++++++-----------------------
 3 files changed, 25 insertions(+), 33 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/pci-swiotlb_64.c b/arch/x86/kernel/pci-swiotlb_64.c
index 242c3440687f..6cf8a816dc29 100644
--- a/arch/x86/kernel/pci-swiotlb_64.c
+++ b/arch/x86/kernel/pci-swiotlb_64.c
@@ -23,7 +23,7 @@ void *swiotlb_alloc(unsigned order, unsigned long nslabs)
 	return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order);
 }
 
-dma_addr_t swiotlb_phys_to_bus(phys_addr_t paddr)
+dma_addr_t swiotlb_phys_to_bus(struct device *hwdev, phys_addr_t paddr)
 {
 	return paddr;
 }
diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h
index 325af1de0351..dedd3c0cfe30 100644
--- a/include/linux/swiotlb.h
+++ b/include/linux/swiotlb.h
@@ -27,7 +27,8 @@ swiotlb_init(void);
 extern void *swiotlb_alloc_boot(size_t bytes, unsigned long nslabs);
 extern void *swiotlb_alloc(unsigned order, unsigned long nslabs);
 
-extern dma_addr_t swiotlb_phys_to_bus(phys_addr_t address);
+extern dma_addr_t swiotlb_phys_to_bus(struct device *hwdev,
+				      phys_addr_t address);
 extern phys_addr_t swiotlb_bus_to_phys(dma_addr_t address);
 
 extern int swiotlb_arch_range_needs_mapping(void *ptr, size_t size);
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index fa2dc4e5f9ba..3657da8ebbc3 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -126,7 +126,7 @@ void * __weak swiotlb_alloc(unsigned order, unsigned long nslabs)
 	return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order);
 }
 
-dma_addr_t __weak swiotlb_phys_to_bus(phys_addr_t paddr)
+dma_addr_t __weak swiotlb_phys_to_bus(struct device *hwdev, phys_addr_t paddr)
 {
 	return paddr;
 }
@@ -136,9 +136,10 @@ phys_addr_t __weak swiotlb_bus_to_phys(dma_addr_t baddr)
 	return baddr;
 }
 
-static dma_addr_t swiotlb_virt_to_bus(volatile void *address)
+static dma_addr_t swiotlb_virt_to_bus(struct device *hwdev,
+				      volatile void *address)
 {
-	return swiotlb_phys_to_bus(virt_to_phys(address));
+	return swiotlb_phys_to_bus(hwdev, virt_to_phys(address));
 }
 
 static void *swiotlb_bus_to_virt(dma_addr_t address)
@@ -151,35 +152,23 @@ int __weak swiotlb_arch_range_needs_mapping(void *ptr, size_t size)
 	return 0;
 }
 
-static dma_addr_t swiotlb_sg_to_bus(struct scatterlist *sg)
+static dma_addr_t swiotlb_sg_to_bus(struct device *hwdev, struct scatterlist *sg)
 {
-	return swiotlb_phys_to_bus(page_to_phys(sg_page(sg)) + sg->offset);
+	return swiotlb_phys_to_bus(hwdev, page_to_phys(sg_page(sg)) + sg->offset);
 }
 
 static void swiotlb_print_info(unsigned long bytes)
 {
 	phys_addr_t pstart, pend;
-	dma_addr_t bstart, bend;
 
 	pstart = virt_to_phys(io_tlb_start);
 	pend = virt_to_phys(io_tlb_end);
 
-	bstart = swiotlb_phys_to_bus(pstart);
-	bend = swiotlb_phys_to_bus(pend);
-
 	printk(KERN_INFO "Placing %luMB software IO TLB between %p - %p\n",
 	       bytes >> 20, io_tlb_start, io_tlb_end);
-	if (pstart != bstart || pend != bend)
-		printk(KERN_INFO "software IO TLB at phys %#llx - %#llx"
-		       " bus %#llx - %#llx\n",
-		       (unsigned long long)pstart,
-		       (unsigned long long)pend,
-		       (unsigned long long)bstart,
-		       (unsigned long long)bend);
-	else
-		printk(KERN_INFO "software IO TLB at phys %#llx - %#llx\n",
-		       (unsigned long long)pstart,
-		       (unsigned long long)pend);
+	printk(KERN_INFO "software IO TLB at phys %#llx - %#llx\n",
+	       (unsigned long long)pstart,
+	       (unsigned long long)pend);
 }
 
 /*
@@ -406,7 +395,7 @@ map_single(struct device *hwdev, struct swiotlb_phys_addr buffer, size_t size, i
 	struct swiotlb_phys_addr slot_buf;
 
 	mask = dma_get_seg_boundary(hwdev);
-	start_dma_addr = swiotlb_virt_to_bus(io_tlb_start) & mask;
+	start_dma_addr = swiotlb_virt_to_bus(hwdev, io_tlb_start) & mask;
 
 	offset_slots = ALIGN(start_dma_addr, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT;
 
@@ -585,7 +574,9 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 		dma_mask = hwdev->coherent_dma_mask;
 
 	ret = (void *)__get_free_pages(flags, order);
-	if (ret && !is_buffer_dma_capable(dma_mask, swiotlb_virt_to_bus(ret), size)) {
+	if (ret &&
+	    !is_buffer_dma_capable(dma_mask, swiotlb_virt_to_bus(hwdev, ret),
+				   size)) {
 		/*
 		 * The allocated memory isn't reachable by the device.
 		 * Fall back on swiotlb_map_single().
@@ -609,7 +600,7 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 	}
 
 	memset(ret, 0, size);
-	dev_addr = swiotlb_virt_to_bus(ret);
+	dev_addr = swiotlb_virt_to_bus(hwdev, ret);
 
 	/* Confirm address can be DMA'd by device */
 	if (!is_buffer_dma_capable(dma_mask, dev_addr, size)) {
@@ -669,7 +660,7 @@ dma_addr_t
 swiotlb_map_single_attrs(struct device *hwdev, void *ptr, size_t size,
 			 int dir, struct dma_attrs *attrs)
 {
-	dma_addr_t dev_addr = swiotlb_virt_to_bus(ptr);
+	dma_addr_t dev_addr = swiotlb_virt_to_bus(hwdev, ptr);
 	void *map;
 	struct swiotlb_phys_addr buffer;
 
@@ -694,7 +685,7 @@ swiotlb_map_single_attrs(struct device *hwdev, void *ptr, size_t size,
 		map = io_tlb_overflow_buffer;
 	}
 
-	dev_addr = swiotlb_virt_to_bus(map);
+	dev_addr = swiotlb_virt_to_bus(hwdev, map);
 
 	/*
 	 * Ensure that the address returned is DMA'ble
@@ -840,7 +831,7 @@ swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl, int nelems,
 	BUG_ON(dir == DMA_NONE);
 
 	for_each_sg(sgl, sg, nelems, i) {
-		dev_addr = swiotlb_sg_to_bus(sg);
+		dev_addr = swiotlb_sg_to_bus(hwdev, sg);
 		if (range_needs_mapping(sg_virt(sg), sg->length) ||
 		    address_needs_mapping(hwdev, dev_addr, sg->length)) {
 			void *map;
@@ -856,7 +847,7 @@ swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl, int nelems,
 				sgl[0].dma_length = 0;
 				return 0;
 			}
-			sg->dma_address = swiotlb_virt_to_bus(map);
+			sg->dma_address = swiotlb_virt_to_bus(hwdev, map);
 		} else
 			sg->dma_address = dev_addr;
 		sg->dma_length = sg->length;
@@ -886,7 +877,7 @@ swiotlb_unmap_sg_attrs(struct device *hwdev, struct scatterlist *sgl,
 	BUG_ON(dir == DMA_NONE);
 
 	for_each_sg(sgl, sg, nelems, i) {
-		if (sg->dma_address != swiotlb_sg_to_bus(sg))
+		if (sg->dma_address != swiotlb_sg_to_bus(hwdev, sg))
 			unmap_single(hwdev, swiotlb_bus_to_virt(sg->dma_address),
 				     sg->dma_length, dir);
 		else if (dir == DMA_FROM_DEVICE)
@@ -919,7 +910,7 @@ swiotlb_sync_sg(struct device *hwdev, struct scatterlist *sgl,
 	BUG_ON(dir == DMA_NONE);
 
 	for_each_sg(sgl, sg, nelems, i) {
-		if (sg->dma_address != swiotlb_sg_to_bus(sg))
+		if (sg->dma_address != swiotlb_sg_to_bus(hwdev, sg))
 			sync_single(hwdev, swiotlb_bus_to_virt(sg->dma_address),
 				    sg->dma_length, dir, target);
 		else if (dir == DMA_FROM_DEVICE)
@@ -944,7 +935,7 @@ swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
 int
 swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t dma_addr)
 {
-	return (dma_addr == swiotlb_virt_to_bus(io_tlb_overflow_buffer));
+	return (dma_addr == swiotlb_virt_to_bus(hwdev, io_tlb_overflow_buffer));
 }
 
 /*
@@ -956,7 +947,7 @@ swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t dma_addr)
 int
 swiotlb_dma_supported(struct device *hwdev, u64 mask)
 {
-	return swiotlb_virt_to_bus(io_tlb_end - 1) <= mask;
+	return swiotlb_virt_to_bus(hwdev, io_tlb_end - 1) <= mask;
 }
 
 EXPORT_SYMBOL(swiotlb_map_single);
-- 
cgit v1.2.3


From 43a256322ac1fc105c181b3cade3b9bfc0b63ca1 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yhlu.kernel@gmail.com>
Date: Sun, 28 Dec 2008 16:01:13 -0800
Subject: sparseirq: move __weak symbols into separate compilation unit

GCC has a bug with __weak alias functions: if the functions are in
the same compilation unit as their call site, GCC can decide to
inline them - and thus rob the linker of the opportunity to override
the weak alias with the real thing.

So move all the IRQ handling related __weak symbols to kernel/irq/chip.c.

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/interrupt.h |  6 ++++++
 include/linux/irq.h       |  3 ---
 init/main.c               | 10 ----------
 kernel/irq/manage.c       |  9 ---------
 kernel/softirq.c          | 20 ++++++++++++++++++++
 5 files changed, 26 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index 777f89e00b4a..d9a370325ae2 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -467,4 +467,10 @@ static inline void init_irq_proc(void)
 
 int show_interrupts(struct seq_file *p, void *v);
 
+struct irq_desc;
+
+extern int early_irq_init(void);
+extern int arch_early_irq_init(void);
+extern int arch_init_chip_data(struct irq_desc *desc, int cpu);
+
 #endif
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 0e40af4bac40..d64a6d49bdef 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -193,9 +193,6 @@ struct irq_desc {
 	const char		*name;
 } ____cacheline_internodealigned_in_smp;
 
-extern int early_irq_init(void);
-extern int arch_early_irq_init(void);
-extern int arch_init_chip_data(struct irq_desc *desc, int cpu);
 extern void arch_init_copy_chip_data(struct irq_desc *old_desc,
 					struct irq_desc *desc, int cpu);
 extern void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc);
diff --git a/init/main.c b/init/main.c
index c314aa15370e..2c183abbf61c 100644
--- a/init/main.c
+++ b/init/main.c
@@ -539,16 +539,6 @@ void __init __weak thread_info_cache_init(void)
 {
 }
 
-int __init __weak arch_early_irq_init(void)
-{
-	return 0;
-}
-
-int __init __weak early_irq_init(void)
-{
-	return arch_early_irq_init();
-}
-
 asmlinkage void __init start_kernel(void)
 {
 	char * command_line;
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index c2741b02ad38..46953a06f4a8 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -261,15 +261,6 @@ void enable_irq(unsigned int irq)
 }
 EXPORT_SYMBOL(enable_irq);
 
-/*
- * [ Not in kernel/irq/handle.c, so that GCC does not
- *   inline the __weak alias: ]
- */
-int __weak arch_init_chip_data(struct irq_desc *desc, int cpu)
-{
-	return 0;
-}
-
 static int set_irq_wake_real(unsigned int irq, unsigned int on)
 {
 	struct irq_desc *desc = irq_to_desc(irq);
diff --git a/kernel/softirq.c b/kernel/softirq.c
index e7c69a720d69..daf46358d2dd 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -797,3 +797,23 @@ int on_each_cpu(void (*func) (void *info), void *info, int wait)
 }
 EXPORT_SYMBOL(on_each_cpu);
 #endif
+
+/*
+ * [ These __weak aliases are kept in a separate compilation unit, so that
+ *   GCC does not inline them incorrectly. ]
+ */
+
+int __init __weak early_irq_init(void)
+{
+	return 0;
+}
+
+int __init __weak arch_early_irq_init(void)
+{
+	return 0;
+}
+
+int __weak arch_init_chip_data(struct irq_desc *desc, int cpu)
+{
+	return 0;
+}
-- 
cgit v1.2.3


From ea319518ba3de282c13ae1cf4bf2215c5e03e67e Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 26 Dec 2008 15:08:55 +0100
Subject: locking, percpu counters: introduce separate lock classes

Impact: fix lockdep false positives

Classify percpu_counter instances similar to regular lock objects --
that is, per instantiation site.

The networking code has increased its use of percpu_counters, which
leads to false positives if they are treated as a single class.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/percpu_counter.h | 14 ++++++++++----
 lib/percpu_counter.c           | 18 ++++--------------
 lib/proportions.c              |  6 +++---
 mm/backing-dev.c               |  2 +-
 4 files changed, 18 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/percpu_counter.h b/include/linux/percpu_counter.h
index 9007ccdfc112..96bdde36599f 100644
--- a/include/linux/percpu_counter.h
+++ b/include/linux/percpu_counter.h
@@ -30,8 +30,16 @@ struct percpu_counter {
 #define FBC_BATCH	(NR_CPUS*4)
 #endif
 
-int percpu_counter_init(struct percpu_counter *fbc, s64 amount);
-int percpu_counter_init_irq(struct percpu_counter *fbc, s64 amount);
+int __percpu_counter_init(struct percpu_counter *fbc, s64 amount,
+			  struct lock_class_key *key);
+
+#define percpu_counter_init(fbc, value)					\
+	({								\
+		static struct lock_class_key __key;			\
+									\
+		__percpu_counter_init(fbc, value, &__key);		\
+	})
+
 void percpu_counter_destroy(struct percpu_counter *fbc);
 void percpu_counter_set(struct percpu_counter *fbc, s64 amount);
 void __percpu_counter_add(struct percpu_counter *fbc, s64 amount, s32 batch);
@@ -85,8 +93,6 @@ static inline int percpu_counter_init(struct percpu_counter *fbc, s64 amount)
 	return 0;
 }
 
-#define percpu_counter_init_irq percpu_counter_init
-
 static inline void percpu_counter_destroy(struct percpu_counter *fbc)
 {
 }
diff --git a/lib/percpu_counter.c b/lib/percpu_counter.c
index a8663890a88c..c7fe2e4e8ed1 100644
--- a/lib/percpu_counter.c
+++ b/lib/percpu_counter.c
@@ -71,11 +71,11 @@ s64 __percpu_counter_sum(struct percpu_counter *fbc)
 }
 EXPORT_SYMBOL(__percpu_counter_sum);
 
-static struct lock_class_key percpu_counter_irqsafe;
-
-int percpu_counter_init(struct percpu_counter *fbc, s64 amount)
+int __percpu_counter_init(struct percpu_counter *fbc, s64 amount,
+			  struct lock_class_key *key)
 {
 	spin_lock_init(&fbc->lock);
+	lockdep_set_class(&fbc->lock, key);
 	fbc->count = amount;
 	fbc->counters = alloc_percpu(s32);
 	if (!fbc->counters)
@@ -87,17 +87,7 @@ int percpu_counter_init(struct percpu_counter *fbc, s64 amount)
 #endif
 	return 0;
 }
-EXPORT_SYMBOL(percpu_counter_init);
-
-int percpu_counter_init_irq(struct percpu_counter *fbc, s64 amount)
-{
-	int err;
-
-	err = percpu_counter_init(fbc, amount);
-	if (!err)
-		lockdep_set_class(&fbc->lock, &percpu_counter_irqsafe);
-	return err;
-}
+EXPORT_SYMBOL(__percpu_counter_init);
 
 void percpu_counter_destroy(struct percpu_counter *fbc)
 {
diff --git a/lib/proportions.c b/lib/proportions.c
index 4f387a643d72..7367f2b727d0 100644
--- a/lib/proportions.c
+++ b/lib/proportions.c
@@ -83,11 +83,11 @@ int prop_descriptor_init(struct prop_descriptor *pd, int shift)
 	pd->index = 0;
 	pd->pg[0].shift = shift;
 	mutex_init(&pd->mutex);
-	err = percpu_counter_init_irq(&pd->pg[0].events, 0);
+	err = percpu_counter_init(&pd->pg[0].events, 0);
 	if (err)
 		goto out;
 
-	err = percpu_counter_init_irq(&pd->pg[1].events, 0);
+	err = percpu_counter_init(&pd->pg[1].events, 0);
 	if (err)
 		percpu_counter_destroy(&pd->pg[0].events);
 
@@ -191,7 +191,7 @@ int prop_local_init_percpu(struct prop_local_percpu *pl)
 	spin_lock_init(&pl->lock);
 	pl->shift = 0;
 	pl->period = 0;
-	return percpu_counter_init_irq(&pl->events, 0);
+	return percpu_counter_init(&pl->events, 0);
 }
 
 void prop_local_destroy_percpu(struct prop_local_percpu *pl)
diff --git a/mm/backing-dev.c b/mm/backing-dev.c
index f2e574dbc300..f3b125857827 100644
--- a/mm/backing-dev.c
+++ b/mm/backing-dev.c
@@ -220,7 +220,7 @@ int bdi_init(struct backing_dev_info *bdi)
 	bdi->max_prop_frac = PROP_FRAC_BASE;
 
 	for (i = 0; i < NR_BDI_STAT_ITEMS; i++) {
-		err = percpu_counter_init_irq(&bdi->bdi_stat[i], 0);
+		err = percpu_counter_init(&bdi->bdi_stat[i], 0);
 		if (err)
 			goto err;
 	}
-- 
cgit v1.2.3


From 34a4c5eb421dab6fe8381aa12c990f9d6f645b17 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Mon, 29 Dec 2008 04:00:23 -0800
Subject: Input: map_to_7segment.h - convert to __inline__ for userspace

Use __inline__ rather than inline for map_to_seg7() since it is exported
to userspace.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 include/linux/map_to_7segment.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/map_to_7segment.h b/include/linux/map_to_7segment.h
index 7df8432c4402..12d62a54d470 100644
--- a/include/linux/map_to_7segment.h
+++ b/include/linux/map_to_7segment.h
@@ -75,7 +75,7 @@ struct seg7_conversion_map {
 	unsigned char	table[128];
 };
 
-static inline int map_to_seg7(struct seg7_conversion_map *map, int c)
+static __inline__ int map_to_seg7(struct seg7_conversion_map *map, int c)
 {
 	return c >= 0 && c < sizeof(map->table) ? map->table[c] : -EINVAL;
 }
-- 
cgit v1.2.3


From 278d1ed65e25d80af7c3a112d707b3f70516ddb4 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 30 Dec 2008 09:05:12 +1030
Subject: cpumask: make CONFIG_NR_CPUS always valid.

Impact: cleanup

Currently we have NR_CPUS, which is 1 on UP, and CONFIG_NR_CPUS on
SMP.  If we make CONFIG_NR_CPUS always valid (and always 1 on !SMP),
we can skip the middleman.

This also allows us to find and check all the unaudited NR_CPUS usage
as we prepare for v. large NR_CPUS.

To avoid breaking every arch, we cheat and do this for the moment
in the header if the arch doesn't.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
---
 include/linux/threads.h | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/threads.h b/include/linux/threads.h
index 38d1a5d6568e..052b12bec8bd 100644
--- a/include/linux/threads.h
+++ b/include/linux/threads.h
@@ -8,17 +8,17 @@
  */
 
 /*
- * Maximum supported processors that can run under SMP.  This value is
- * set via configure setting.  The maximum is equal to the size of the
- * bitmasks used on that platform, i.e. 32 or 64.  Setting this smaller
- * saves quite a bit of memory.
+ * Maximum supported processors.  Setting this smaller saves quite a
+ * bit of memory.  Use nr_cpu_ids instead of this except for static bitmaps.
  */
-#ifdef CONFIG_SMP
-#define NR_CPUS		CONFIG_NR_CPUS
-#else
-#define NR_CPUS		1
+#ifndef CONFIG_NR_CPUS
+/* FIXME: This should be fixed in the arch's Kconfig */
+#define CONFIG_NR_CPUS	1
 #endif
 
+/* Places which use this should consider cpumask_var_t. */
+#define NR_CPUS		CONFIG_NR_CPUS
+
 #define MIN_THREADS_LEFT_FOR_ROOT 4
 
 /*
-- 
cgit v1.2.3


From 4b0bc0bca83f3fb7cf920e2ec80684c15d2269c0 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 30 Dec 2008 09:05:13 +1030
Subject: bitmap: test for constant as well as small size for inline versions

Impact: reduce text size

bitmap_zero et al have a fastpath for nbits <= BITS_PER_LONG, but this
should really only apply where the nbits is known at compile time.

This only saves about 1200 bytes on an allyesconfig kernel, but with
cpumasks going variable that number will increase.

   text		data	bss	dec		hex	filename
35327852        5035607 6782976 47146435        2cf65c3 vmlinux-before
35326640        5035607 6782976 47145223        2cf6107 vmlinux-after

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 include/linux/bitmap.h | 35 +++++++++++++++++++----------------
 1 file changed, 19 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/bitmap.h b/include/linux/bitmap.h
index a08c33a26ca9..2878811c6134 100644
--- a/include/linux/bitmap.h
+++ b/include/linux/bitmap.h
@@ -137,9 +137,12 @@ extern void bitmap_copy_le(void *dst, const unsigned long *src, int nbits);
 		(1UL<<((nbits) % BITS_PER_LONG))-1 : ~0UL		\
 )
 
+#define small_const_nbits(nbits) \
+	(__builtin_constant_p(nbits) && (nbits) <= BITS_PER_LONG)
+
 static inline void bitmap_zero(unsigned long *dst, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = 0UL;
 	else {
 		int len = BITS_TO_LONGS(nbits) * sizeof(unsigned long);
@@ -150,7 +153,7 @@ static inline void bitmap_zero(unsigned long *dst, int nbits)
 static inline void bitmap_fill(unsigned long *dst, int nbits)
 {
 	size_t nlongs = BITS_TO_LONGS(nbits);
-	if (nlongs > 1) {
+	if (!small_const_nbits(nbits)) {
 		int len = (nlongs - 1) * sizeof(unsigned long);
 		memset(dst, 0xff,  len);
 	}
@@ -160,7 +163,7 @@ static inline void bitmap_fill(unsigned long *dst, int nbits)
 static inline void bitmap_copy(unsigned long *dst, const unsigned long *src,
 			int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = *src;
 	else {
 		int len = BITS_TO_LONGS(nbits) * sizeof(unsigned long);
@@ -171,7 +174,7 @@ static inline void bitmap_copy(unsigned long *dst, const unsigned long *src,
 static inline void bitmap_and(unsigned long *dst, const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = *src1 & *src2;
 	else
 		__bitmap_and(dst, src1, src2, nbits);
@@ -180,7 +183,7 @@ static inline void bitmap_and(unsigned long *dst, const unsigned long *src1,
 static inline void bitmap_or(unsigned long *dst, const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = *src1 | *src2;
 	else
 		__bitmap_or(dst, src1, src2, nbits);
@@ -189,7 +192,7 @@ static inline void bitmap_or(unsigned long *dst, const unsigned long *src1,
 static inline void bitmap_xor(unsigned long *dst, const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = *src1 ^ *src2;
 	else
 		__bitmap_xor(dst, src1, src2, nbits);
@@ -198,7 +201,7 @@ static inline void bitmap_xor(unsigned long *dst, const unsigned long *src1,
 static inline void bitmap_andnot(unsigned long *dst, const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = *src1 & ~(*src2);
 	else
 		__bitmap_andnot(dst, src1, src2, nbits);
@@ -207,7 +210,7 @@ static inline void bitmap_andnot(unsigned long *dst, const unsigned long *src1,
 static inline void bitmap_complement(unsigned long *dst, const unsigned long *src,
 			int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = ~(*src) & BITMAP_LAST_WORD_MASK(nbits);
 	else
 		__bitmap_complement(dst, src, nbits);
@@ -216,7 +219,7 @@ static inline void bitmap_complement(unsigned long *dst, const unsigned long *sr
 static inline int bitmap_equal(const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		return ! ((*src1 ^ *src2) & BITMAP_LAST_WORD_MASK(nbits));
 	else
 		return __bitmap_equal(src1, src2, nbits);
@@ -225,7 +228,7 @@ static inline int bitmap_equal(const unsigned long *src1,
 static inline int bitmap_intersects(const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		return ((*src1 & *src2) & BITMAP_LAST_WORD_MASK(nbits)) != 0;
 	else
 		return __bitmap_intersects(src1, src2, nbits);
@@ -234,7 +237,7 @@ static inline int bitmap_intersects(const unsigned long *src1,
 static inline int bitmap_subset(const unsigned long *src1,
 			const unsigned long *src2, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		return ! ((*src1 & ~(*src2)) & BITMAP_LAST_WORD_MASK(nbits));
 	else
 		return __bitmap_subset(src1, src2, nbits);
@@ -242,7 +245,7 @@ static inline int bitmap_subset(const unsigned long *src1,
 
 static inline int bitmap_empty(const unsigned long *src, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		return ! (*src & BITMAP_LAST_WORD_MASK(nbits));
 	else
 		return __bitmap_empty(src, nbits);
@@ -250,7 +253,7 @@ static inline int bitmap_empty(const unsigned long *src, int nbits)
 
 static inline int bitmap_full(const unsigned long *src, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		return ! (~(*src) & BITMAP_LAST_WORD_MASK(nbits));
 	else
 		return __bitmap_full(src, nbits);
@@ -258,7 +261,7 @@ static inline int bitmap_full(const unsigned long *src, int nbits)
 
 static inline int bitmap_weight(const unsigned long *src, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		return hweight_long(*src & BITMAP_LAST_WORD_MASK(nbits));
 	return __bitmap_weight(src, nbits);
 }
@@ -266,7 +269,7 @@ static inline int bitmap_weight(const unsigned long *src, int nbits)
 static inline void bitmap_shift_right(unsigned long *dst,
 			const unsigned long *src, int n, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = *src >> n;
 	else
 		__bitmap_shift_right(dst, src, n, nbits);
@@ -275,7 +278,7 @@ static inline void bitmap_shift_right(unsigned long *dst,
 static inline void bitmap_shift_left(unsigned long *dst,
 			const unsigned long *src, int n, int nbits)
 {
-	if (nbits <= BITS_PER_LONG)
+	if (small_const_nbits(nbits))
 		*dst = (*src << n) & BITMAP_LAST_WORD_MASK(nbits);
 	else
 		__bitmap_shift_left(dst, src, n, nbits);
-- 
cgit v1.2.3


From cb78a0ce69fad2026825f957e24e2d9cda1ec9f1 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 30 Dec 2008 09:05:14 +1030
Subject: bitmap: fix seq_bitmap and seq_cpumask to take const pointer

Impact: cleanup

seq_bitmap just calls bitmap_scnprintf on the bits: that arg can be const.
Similarly, seq_cpumask just calls seq_bitmap.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 fs/seq_file.c            | 3 ++-
 include/linux/seq_file.h | 5 +++--
 2 files changed, 5 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/fs/seq_file.c b/fs/seq_file.c
index 16c211558c22..c99358a52176 100644
--- a/fs/seq_file.c
+++ b/fs/seq_file.c
@@ -462,7 +462,8 @@ int seq_dentry(struct seq_file *m, struct dentry *dentry, char *esc)
 	return -1;
 }
 
-int seq_bitmap(struct seq_file *m, unsigned long *bits, unsigned int nr_bits)
+int seq_bitmap(struct seq_file *m, const unsigned long *bits,
+				   unsigned int nr_bits)
 {
 	if (m->count < m->size) {
 		int len = bitmap_scnprintf(m->buf + m->count,
diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
index b3dfa72f13b9..952e0187ba16 100644
--- a/include/linux/seq_file.h
+++ b/include/linux/seq_file.h
@@ -50,8 +50,9 @@ int seq_path(struct seq_file *, struct path *, char *);
 int seq_dentry(struct seq_file *, struct dentry *, char *);
 int seq_path_root(struct seq_file *m, struct path *path, struct path *root,
 		  char *esc);
-int seq_bitmap(struct seq_file *m, unsigned long *bits, unsigned int nr_bits);
-static inline int seq_cpumask(struct seq_file *m, cpumask_t *mask)
+int seq_bitmap(struct seq_file *m, const unsigned long *bits,
+				   unsigned int nr_bits);
+static inline int seq_cpumask(struct seq_file *m, const struct cpumask *mask)
 {
 	return seq_bitmap(m, mask->bits, NR_CPUS);
 }
-- 
cgit v1.2.3


From b3199c025d1646e25e7d1d640dd605db251dccf8 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 30 Dec 2008 09:05:14 +1030
Subject: cpumask: switch over to cpu_online/possible/active/present_mask: core

Impact: cleanup

This implements the obsolescent cpu_online_map in terms of
cpu_online_mask, rather than the other way around.  Same for the other
maps.

The documentation comments are also updated to refer to _mask rather
than _map.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
---
 include/linux/cpumask.h | 75 +++++++++++++++++++------------------------------
 kernel/cpu.c            | 49 +++++++++++++++-----------------
 2 files changed, 52 insertions(+), 72 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index b5ad19a6f43f..db2341beca45 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -416,65 +416,54 @@ int __next_cpu_nr(int n, const cpumask_t *srcp);
 
 /*
  * The following particular system cpumasks and operations manage
- * possible, present, active and online cpus.  Each of them is a fixed size
- * bitmap of size NR_CPUS.
+ * possible, present, active and online cpus.
  *
- *  #ifdef CONFIG_HOTPLUG_CPU
- *     cpu_possible_map - has bit 'cpu' set iff cpu is populatable
- *     cpu_present_map  - has bit 'cpu' set iff cpu is populated
- *     cpu_online_map   - has bit 'cpu' set iff cpu available to scheduler
- *     cpu_active_map   - has bit 'cpu' set iff cpu available to migration
- *  #else
- *     cpu_possible_map - has bit 'cpu' set iff cpu is populated
- *     cpu_present_map  - copy of cpu_possible_map
- *     cpu_online_map   - has bit 'cpu' set iff cpu available to scheduler
- *  #endif
+ *     cpu_possible_mask- has bit 'cpu' set iff cpu is populatable
+ *     cpu_present_mask - has bit 'cpu' set iff cpu is populated
+ *     cpu_online_mask  - has bit 'cpu' set iff cpu available to scheduler
+ *     cpu_active_mask  - has bit 'cpu' set iff cpu available to migration
  *
- *  In either case, NR_CPUS is fixed at compile time, as the static
- *  size of these bitmaps.  The cpu_possible_map is fixed at boot
- *  time, as the set of CPU id's that it is possible might ever
- *  be plugged in at anytime during the life of that system boot.
- *  The cpu_present_map is dynamic(*), representing which CPUs
- *  are currently plugged in.  And cpu_online_map is the dynamic
- *  subset of cpu_present_map, indicating those CPUs available
- *  for scheduling.
+ *  If !CONFIG_HOTPLUG_CPU, present == possible, and active == online.
  *
- *  If HOTPLUG is enabled, then cpu_possible_map is forced to have
+ *  The cpu_possible_mask is fixed at boot time, as the set of CPU id's
+ *  that it is possible might ever be plugged in at anytime during the
+ *  life of that system boot.  The cpu_present_mask is dynamic(*),
+ *  representing which CPUs are currently plugged in.  And
+ *  cpu_online_mask is the dynamic subset of cpu_present_mask,
+ *  indicating those CPUs available for scheduling.
+ *
+ *  If HOTPLUG is enabled, then cpu_possible_mask is forced to have
  *  all NR_CPUS bits set, otherwise it is just the set of CPUs that
  *  ACPI reports present at boot.
  *
- *  If HOTPLUG is enabled, then cpu_present_map varies dynamically,
+ *  If HOTPLUG is enabled, then cpu_present_mask varies dynamically,
  *  depending on what ACPI reports as currently plugged in, otherwise
- *  cpu_present_map is just a copy of cpu_possible_map.
+ *  cpu_present_mask is just a copy of cpu_possible_mask.
  *
- *  (*) Well, cpu_present_map is dynamic in the hotplug case.  If not
- *      hotplug, it's a copy of cpu_possible_map, hence fixed at boot.
+ *  (*) Well, cpu_present_mask is dynamic in the hotplug case.  If not
+ *      hotplug, it's a copy of cpu_possible_mask, hence fixed at boot.
  *
  * Subtleties:
  * 1) UP arch's (NR_CPUS == 1, CONFIG_SMP not defined) hardcode
  *    assumption that their single CPU is online.  The UP
- *    cpu_{online,possible,present}_maps are placebos.  Changing them
+ *    cpu_{online,possible,present}_masks are placebos.  Changing them
  *    will have no useful affect on the following num_*_cpus()
  *    and cpu_*() macros in the UP case.  This ugliness is a UP
  *    optimization - don't waste any instructions or memory references
  *    asking if you're online or how many CPUs there are if there is
  *    only one CPU.
- * 2) Most SMP arch's #define some of these maps to be some
- *    other map specific to that arch.  Therefore, the following
- *    must be #define macros, not inlines.  To see why, examine
- *    the assembly code produced by the following.  Note that
- *    set1() writes phys_x_map, but set2() writes x_map:
- *        int x_map, phys_x_map;
- *        #define set1(a) x_map = a
- *        inline void set2(int a) { x_map = a; }
- *        #define x_map phys_x_map
- *        main(){ set1(3); set2(5); }
  */
 
-extern cpumask_t cpu_possible_map;
-extern cpumask_t cpu_online_map;
-extern cpumask_t cpu_present_map;
-extern cpumask_t cpu_active_map;
+extern const struct cpumask *const cpu_possible_mask;
+extern const struct cpumask *const cpu_online_mask;
+extern const struct cpumask *const cpu_present_mask;
+extern const struct cpumask *const cpu_active_mask;
+
+/* These strip const, as traditionally they weren't const. */
+#define cpu_possible_map	(*(cpumask_t *)cpu_possible_mask)
+#define cpu_online_map		(*(cpumask_t *)cpu_online_mask)
+#define cpu_present_map		(*(cpumask_t *)cpu_present_mask)
+#define cpu_active_map		(*(cpumask_t *)cpu_active_mask)
 
 #if NR_CPUS > 1
 #define num_online_cpus()	cpus_weight_nr(cpu_online_map)
@@ -1058,12 +1047,6 @@ static inline void free_bootmem_cpumask_var(cpumask_var_t mask)
 }
 #endif /* CONFIG_CPUMASK_OFFSTACK */
 
-/* The pointer versions of the maps, these will become the primary versions. */
-#define cpu_possible_mask ((const struct cpumask *)&cpu_possible_map)
-#define cpu_online_mask ((const struct cpumask *)&cpu_online_map)
-#define cpu_present_mask ((const struct cpumask *)&cpu_present_map)
-#define cpu_active_mask ((const struct cpumask *)&cpu_active_map)
-
 /* It's common to want to use cpu_all_mask in struct member initializers,
  * so it has to refer to an address rather than a pointer. */
 extern const DECLARE_BITMAP(cpu_all_bits, NR_CPUS);
diff --git a/kernel/cpu.c b/kernel/cpu.c
index bae131a1211b..3ddc509b19c5 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -15,30 +15,8 @@
 #include <linux/stop_machine.h>
 #include <linux/mutex.h>
 
-/*
- * Represents all cpu's present in the system
- * In systems capable of hotplug, this map could dynamically grow
- * as new cpu's are detected in the system via any platform specific
- * method, such as ACPI for e.g.
- */
-cpumask_t cpu_present_map __read_mostly;
-EXPORT_SYMBOL(cpu_present_map);
-
-/*
- * Represents all cpu's that are currently online.
- */
-cpumask_t cpu_online_map __read_mostly;
-EXPORT_SYMBOL(cpu_online_map);
-
-#ifdef CONFIG_INIT_ALL_POSSIBLE
-cpumask_t cpu_possible_map __read_mostly = CPU_MASK_ALL;
-#else
-cpumask_t cpu_possible_map __read_mostly;
-#endif
-EXPORT_SYMBOL(cpu_possible_map);
-
 #ifdef CONFIG_SMP
-/* Serializes the updates to cpu_online_map, cpu_present_map */
+/* Serializes the updates to cpu_online_mask, cpu_present_mask */
 static DEFINE_MUTEX(cpu_add_remove_lock);
 
 static __cpuinitdata RAW_NOTIFIER_HEAD(cpu_chain);
@@ -65,8 +43,6 @@ void __init cpu_hotplug_init(void)
 	cpu_hotplug.refcount = 0;
 }
 
-cpumask_t cpu_active_map;
-
 #ifdef CONFIG_HOTPLUG_CPU
 
 void get_online_cpus(void)
@@ -97,7 +73,7 @@ EXPORT_SYMBOL_GPL(put_online_cpus);
 
 /*
  * The following two API's must be used when attempting
- * to serialize the updates to cpu_online_map, cpu_present_map.
+ * to serialize the updates to cpu_online_mask, cpu_present_mask.
  */
 void cpu_maps_update_begin(void)
 {
@@ -503,3 +479,24 @@ EXPORT_SYMBOL_GPL(cpu_bit_bitmap);
 
 const DECLARE_BITMAP(cpu_all_bits, NR_CPUS) = CPU_BITS_ALL;
 EXPORT_SYMBOL(cpu_all_bits);
+
+#ifdef CONFIG_INIT_ALL_POSSIBLE
+static DECLARE_BITMAP(cpu_possible_bits, CONFIG_NR_CPUS) __read_mostly
+	= CPU_BITS_ALL;
+#else
+static DECLARE_BITMAP(cpu_possible_bits, CONFIG_NR_CPUS) __read_mostly;
+#endif
+const struct cpumask *const cpu_possible_mask = to_cpumask(cpu_possible_bits);
+EXPORT_SYMBOL(cpu_possible_mask);
+
+static DECLARE_BITMAP(cpu_online_bits, CONFIG_NR_CPUS) __read_mostly;
+const struct cpumask *const cpu_online_mask = to_cpumask(cpu_online_bits);
+EXPORT_SYMBOL(cpu_online_mask);
+
+static DECLARE_BITMAP(cpu_present_bits, CONFIG_NR_CPUS) __read_mostly;
+const struct cpumask *const cpu_present_mask = to_cpumask(cpu_present_bits);
+EXPORT_SYMBOL(cpu_present_mask);
+
+static DECLARE_BITMAP(cpu_active_bits, CONFIG_NR_CPUS) __read_mostly;
+const struct cpumask *const cpu_active_mask = to_cpumask(cpu_active_bits);
+EXPORT_SYMBOL(cpu_active_mask);
-- 
cgit v1.2.3


From ae7a47e72e1a0b5e2b46d1596bc2c22942a73023 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 30 Dec 2008 09:05:15 +1030
Subject: cpumask: make cpumask.h eat its own dogfood.

Changes:
1) cpumask_t to struct cpumask,
2) cpus_weight_nr to cpumask_weight,
3) cpu_isset to cpumask_test_cpu,
4) ->bits to cpumask_bits()
5) cpu_*_map to cpu_*_mask.
6) for_each_cpu_mask_nr to for_each_cpu

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 include/linux/cpumask.h | 75 +++++++++++++++++++++++++------------------------
 1 file changed, 38 insertions(+), 37 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index db2341beca45..e62a67156c53 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -268,6 +268,25 @@ static inline void __cpus_shift_left(cpumask_t *dstp,
 	bitmap_shift_left(dstp->bits, srcp->bits, n, nbits);
 }
 
+/**
+ * to_cpumask - convert an NR_CPUS bitmap to a struct cpumask *
+ * @bitmap: the bitmap
+ *
+ * There are a few places where cpumask_var_t isn't appropriate and
+ * static cpumasks must be used (eg. very early boot), yet we don't
+ * expose the definition of 'struct cpumask'.
+ *
+ * This does the conversion, and can be used as a constant initializer.
+ */
+#define to_cpumask(bitmap)						\
+	((struct cpumask *)(1 ? (bitmap)				\
+			    : (void *)sizeof(__check_is_bitmap(bitmap))))
+
+static inline int __check_is_bitmap(const unsigned long *bitmap)
+{
+	return 1;
+}
+
 /*
  * Special-case data structure for "single bit set only" constant CPU masks.
  *
@@ -278,11 +297,11 @@ static inline void __cpus_shift_left(cpumask_t *dstp,
 extern const unsigned long
 	cpu_bit_bitmap[BITS_PER_LONG+1][BITS_TO_LONGS(NR_CPUS)];
 
-static inline const cpumask_t *get_cpu_mask(unsigned int cpu)
+static inline const struct cpumask *get_cpu_mask(unsigned int cpu)
 {
 	const unsigned long *p = cpu_bit_bitmap[1 + cpu % BITS_PER_LONG];
 	p -= cpu / BITS_PER_LONG;
-	return (const cpumask_t *)p;
+	return to_cpumask(p);
 }
 
 /*
@@ -466,13 +485,13 @@ extern const struct cpumask *const cpu_active_mask;
 #define cpu_active_map		(*(cpumask_t *)cpu_active_mask)
 
 #if NR_CPUS > 1
-#define num_online_cpus()	cpus_weight_nr(cpu_online_map)
-#define num_possible_cpus()	cpus_weight_nr(cpu_possible_map)
-#define num_present_cpus()	cpus_weight_nr(cpu_present_map)
-#define cpu_online(cpu)		cpu_isset((cpu), cpu_online_map)
-#define cpu_possible(cpu)	cpu_isset((cpu), cpu_possible_map)
-#define cpu_present(cpu)	cpu_isset((cpu), cpu_present_map)
-#define cpu_active(cpu)		cpu_isset((cpu), cpu_active_map)
+#define num_online_cpus()	cpumask_weight(cpu_online_mask)
+#define num_possible_cpus()	cpumask_weight(cpu_possible_mask)
+#define num_present_cpus()	cpumask_weight(cpu_present_mask)
+#define cpu_online(cpu)		cpumask_test_cpu((cpu), cpu_online_mask)
+#define cpu_possible(cpu)	cpumask_test_cpu((cpu), cpu_possible_mask)
+#define cpu_present(cpu)	cpumask_test_cpu((cpu), cpu_present_mask)
+#define cpu_active(cpu)		cpumask_test_cpu((cpu), cpu_active_mask)
 #else
 #define num_online_cpus()	1
 #define num_possible_cpus()	1
@@ -485,10 +504,6 @@ extern const struct cpumask *const cpu_active_mask;
 
 #define cpu_is_offline(cpu)	unlikely(!cpu_online(cpu))
 
-#define for_each_possible_cpu(cpu) for_each_cpu_mask_nr((cpu), cpu_possible_map)
-#define for_each_online_cpu(cpu)   for_each_cpu_mask_nr((cpu), cpu_online_map)
-#define for_each_present_cpu(cpu)  for_each_cpu_mask_nr((cpu), cpu_present_map)
-
 /* These are the new versions of the cpumask operators: passed by pointer.
  * The older versions will be implemented in terms of these, then deleted. */
 #define cpumask_bits(maskp) ((maskp)->bits)
@@ -676,7 +691,7 @@ static inline void cpumask_clear_cpu(int cpu, struct cpumask *dstp)
  * No static inline type checking - see Subtlety (1) above.
  */
 #define cpumask_test_cpu(cpu, cpumask) \
-	test_bit(cpumask_check(cpu), (cpumask)->bits)
+	test_bit(cpumask_check(cpu), cpumask_bits((cpumask)))
 
 /**
  * cpumask_test_and_set_cpu - atomically test and set a cpu in a cpumask
@@ -919,7 +934,7 @@ static inline void cpumask_copy(struct cpumask *dstp,
 static inline int cpumask_scnprintf(char *buf, int len,
 				    const struct cpumask *srcp)
 {
-	return bitmap_scnprintf(buf, len, srcp->bits, nr_cpumask_bits);
+	return bitmap_scnprintf(buf, len, cpumask_bits(srcp), nr_cpumask_bits);
 }
 
 /**
@@ -933,7 +948,7 @@ static inline int cpumask_scnprintf(char *buf, int len,
 static inline int cpumask_parse_user(const char __user *buf, int len,
 				     struct cpumask *dstp)
 {
-	return bitmap_parse_user(buf, len, dstp->bits, nr_cpumask_bits);
+	return bitmap_parse_user(buf, len, cpumask_bits(dstp), nr_cpumask_bits);
 }
 
 /**
@@ -948,7 +963,8 @@ static inline int cpumask_parse_user(const char __user *buf, int len,
 static inline int cpulist_scnprintf(char *buf, int len,
 				    const struct cpumask *srcp)
 {
-	return bitmap_scnlistprintf(buf, len, srcp->bits, nr_cpumask_bits);
+	return bitmap_scnlistprintf(buf, len, cpumask_bits(srcp),
+				    nr_cpumask_bits);
 }
 
 /**
@@ -961,26 +977,7 @@ static inline int cpulist_scnprintf(char *buf, int len,
  */
 static inline int cpulist_parse(const char *buf, struct cpumask *dstp)
 {
-	return bitmap_parselist(buf, dstp->bits, nr_cpumask_bits);
-}
-
-/**
- * to_cpumask - convert an NR_CPUS bitmap to a struct cpumask *
- * @bitmap: the bitmap
- *
- * There are a few places where cpumask_var_t isn't appropriate and
- * static cpumasks must be used (eg. very early boot), yet we don't
- * expose the definition of 'struct cpumask'.
- *
- * This does the conversion, and can be used as a constant initializer.
- */
-#define to_cpumask(bitmap)						\
-	((struct cpumask *)(1 ? (bitmap)				\
-			    : (void *)sizeof(__check_is_bitmap(bitmap))))
-
-static inline int __check_is_bitmap(const unsigned long *bitmap)
-{
-	return 1;
+	return bitmap_parselist(buf, cpumask_bits(dstp), nr_cpumask_bits);
 }
 
 /**
@@ -1055,6 +1052,10 @@ extern const DECLARE_BITMAP(cpu_all_bits, NR_CPUS);
 /* First bits of cpu_bit_bitmap are in fact unset. */
 #define cpu_none_mask to_cpumask(cpu_bit_bitmap[0])
 
+#define for_each_possible_cpu(cpu) for_each_cpu((cpu), cpu_possible_mask)
+#define for_each_online_cpu(cpu)   for_each_cpu((cpu), cpu_online_mask)
+#define for_each_present_cpu(cpu)  for_each_cpu((cpu), cpu_present_mask)
+
 /* Wrappers for arch boot code to manipulate normally-constant masks */
 static inline void set_cpu_possible(unsigned int cpu, bool possible)
 {
-- 
cgit v1.2.3


From 3fa41520696fec2815e2d88fbcccdda77ba4d693 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 30 Dec 2008 09:05:16 +1030
Subject: cpumask: make set_cpu_*/init_cpu_* out-of-line

They're only for use in boot/cpu hotplug code anyway, and this avoids
the use of deprecated cpu_*_map.

Stephen Rothwell points out that gcc 4.2.4 (on powerpc at least)
didn't like the cast away of const anyway:

  include/linux/cpumask.h: In function 'set_cpu_possible':
  include/linux/cpumask.h:1052: warning: passing argument 2 of 'cpumask_set_cpu' discards qualifiers from pointer target type

So this kills two birds with one stone.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 include/linux/cpumask.h | 53 +++++++------------------------------------------
 kernel/cpu.c            | 47 +++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 54 insertions(+), 46 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index e62a67156c53..7c178a6baae3 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -1057,50 +1057,11 @@ extern const DECLARE_BITMAP(cpu_all_bits, NR_CPUS);
 #define for_each_present_cpu(cpu)  for_each_cpu((cpu), cpu_present_mask)
 
 /* Wrappers for arch boot code to manipulate normally-constant masks */
-static inline void set_cpu_possible(unsigned int cpu, bool possible)
-{
-	if (possible)
-		cpumask_set_cpu(cpu, &cpu_possible_map);
-	else
-		cpumask_clear_cpu(cpu, &cpu_possible_map);
-}
-
-static inline void set_cpu_present(unsigned int cpu, bool present)
-{
-	if (present)
-		cpumask_set_cpu(cpu, &cpu_present_map);
-	else
-		cpumask_clear_cpu(cpu, &cpu_present_map);
-}
-
-static inline void set_cpu_online(unsigned int cpu, bool online)
-{
-	if (online)
-		cpumask_set_cpu(cpu, &cpu_online_map);
-	else
-		cpumask_clear_cpu(cpu, &cpu_online_map);
-}
-
-static inline void set_cpu_active(unsigned int cpu, bool active)
-{
-	if (active)
-		cpumask_set_cpu(cpu, &cpu_active_map);
-	else
-		cpumask_clear_cpu(cpu, &cpu_active_map);
-}
-
-static inline void init_cpu_present(const struct cpumask *src)
-{
-	cpumask_copy(&cpu_present_map, src);
-}
-
-static inline void init_cpu_possible(const struct cpumask *src)
-{
-	cpumask_copy(&cpu_possible_map, src);
-}
-
-static inline void init_cpu_online(const struct cpumask *src)
-{
-	cpumask_copy(&cpu_online_map, src);
-}
+void set_cpu_possible(unsigned int cpu, bool possible);
+void set_cpu_present(unsigned int cpu, bool present);
+void set_cpu_online(unsigned int cpu, bool online);
+void set_cpu_active(unsigned int cpu, bool active);
+void init_cpu_present(const struct cpumask *src);
+void init_cpu_possible(const struct cpumask *src);
+void init_cpu_online(const struct cpumask *src);
 #endif /* __LINUX_CPUMASK_H */
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 3ddc509b19c5..2c9f78f3a2fc 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -500,3 +500,50 @@ EXPORT_SYMBOL(cpu_present_mask);
 static DECLARE_BITMAP(cpu_active_bits, CONFIG_NR_CPUS) __read_mostly;
 const struct cpumask *const cpu_active_mask = to_cpumask(cpu_active_bits);
 EXPORT_SYMBOL(cpu_active_mask);
+
+void set_cpu_possible(unsigned int cpu, bool possible)
+{
+	if (possible)
+		cpumask_set_cpu(cpu, to_cpumask(cpu_possible_bits));
+	else
+		cpumask_clear_cpu(cpu, to_cpumask(cpu_possible_bits));
+}
+
+void set_cpu_present(unsigned int cpu, bool present)
+{
+	if (present)
+		cpumask_set_cpu(cpu, to_cpumask(cpu_present_bits));
+	else
+		cpumask_clear_cpu(cpu, to_cpumask(cpu_present_bits));
+}
+
+void set_cpu_online(unsigned int cpu, bool online)
+{
+	if (online)
+		cpumask_set_cpu(cpu, to_cpumask(cpu_online_bits));
+	else
+		cpumask_clear_cpu(cpu, to_cpumask(cpu_online_bits));
+}
+
+void set_cpu_active(unsigned int cpu, bool active)
+{
+	if (active)
+		cpumask_set_cpu(cpu, to_cpumask(cpu_active_bits));
+	else
+		cpumask_clear_cpu(cpu, to_cpumask(cpu_active_bits));
+}
+
+void init_cpu_present(const struct cpumask *src)
+{
+	cpumask_copy(to_cpumask(cpu_present_bits), src);
+}
+
+void init_cpu_possible(const struct cpumask *src)
+{
+	cpumask_copy(to_cpumask(cpu_possible_bits), src);
+}
+
+void init_cpu_online(const struct cpumask *src)
+{
+	cpumask_copy(to_cpumask(cpu_online_bits), src);
+}
-- 
cgit v1.2.3


From 54b11e6d57a10aa9d0009efd93873e17bffd5d30 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 30 Dec 2008 09:05:16 +1030
Subject: cpumask: smp_call_function_many()

Impact: Implementation change to remove cpumask_t from stack.

Actually change smp_call_function_mask() to smp_call_function_many().
We avoid cpumasks on the stack in this version.

(S390 has its own version, but that's going away apparently).

We have to do some dancing to figure out if 0 or 1 other cpus are in
the mask supplied and the online mask without allocating a tmp
cpumask.  It's still fairly cheap.

We allocate the cpumask at the end of the call_function_data
structure: if allocation fails we fallback to smp_call_function_single
rather than using the baroque quiescing code (which needs a cpumask on
stack).

(Thanks to Hiroshi Shimamoto for spotting several bugs in previous versions!)

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Cc: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Cc: npiggin@suse.de
Cc: axboe@kernel.dk
---
 include/linux/smp.h |  15 +++---
 kernel/smp.c        | 139 ++++++++++++++++++----------------------------------
 2 files changed, 57 insertions(+), 97 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/smp.h b/include/linux/smp.h
index 2f85f3b04bc4..b82466968101 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -67,15 +67,16 @@ extern void smp_cpus_done(unsigned int max_cpus);
  * Call a function on all other processors
  */
 int smp_call_function(void(*func)(void *info), void *info, int wait);
-/* Deprecated: use smp_call_function_many() which uses a cpumask ptr. */
-int smp_call_function_mask(cpumask_t mask, void(*func)(void *info), void *info,
-				int wait);
+void smp_call_function_many(const struct cpumask *mask,
+			    void (*func)(void *info), void *info, bool wait);
 
-static inline void smp_call_function_many(const struct cpumask *mask,
-					  void (*func)(void *info), void *info,
-					  int wait)
+/* Deprecated: Use smp_call_function_many which takes a pointer to the mask. */
+static inline int
+smp_call_function_mask(cpumask_t mask, void(*func)(void *info), void *info,
+		       int wait)
 {
-	smp_call_function_mask(*mask, func, info, wait);
+	smp_call_function_many(&mask, func, info, wait);
+	return 0;
 }
 
 int smp_call_function_single(int cpuid, void (*func) (void *info), void *info,
diff --git a/kernel/smp.c b/kernel/smp.c
index 75c8dde58c55..9f0eafed1399 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -24,8 +24,8 @@ struct call_function_data {
 	struct call_single_data csd;
 	spinlock_t lock;
 	unsigned int refs;
-	cpumask_t cpumask;
 	struct rcu_head rcu_head;
+	unsigned long cpumask_bits[];
 };
 
 struct call_single_queue {
@@ -110,13 +110,13 @@ void generic_smp_call_function_interrupt(void)
 	list_for_each_entry_rcu(data, &call_function_queue, csd.list) {
 		int refs;
 
-		if (!cpu_isset(cpu, data->cpumask))
+		if (!cpumask_test_cpu(cpu, to_cpumask(data->cpumask_bits)))
 			continue;
 
 		data->csd.func(data->csd.info);
 
 		spin_lock(&data->lock);
-		cpu_clear(cpu, data->cpumask);
+		cpumask_clear_cpu(cpu, to_cpumask(data->cpumask_bits));
 		WARN_ON(data->refs == 0);
 		data->refs--;
 		refs = data->refs;
@@ -266,51 +266,13 @@ void __smp_call_function_single(int cpu, struct call_single_data *data)
 	generic_exec_single(cpu, data);
 }
 
-/* Dummy function */
-static void quiesce_dummy(void *unused)
-{
-}
-
-/*
- * Ensure stack based data used in call function mask is safe to free.
- *
- * This is needed by smp_call_function_mask when using on-stack data, because
- * a single call function queue is shared by all CPUs, and any CPU may pick up
- * the data item on the queue at any time before it is deleted. So we need to
- * ensure that all CPUs have transitioned through a quiescent state after
- * this call.
- *
- * This is a very slow function, implemented by sending synchronous IPIs to
- * all possible CPUs. For this reason, we have to alloc data rather than use
- * stack based data even in the case of synchronous calls. The stack based
- * data is then just used for deadlock/oom fallback which will be very rare.
- *
- * If a faster scheme can be made, we could go back to preferring stack based
- * data -- the data allocation/free is non-zero cost.
- */
-static void smp_call_function_mask_quiesce_stack(cpumask_t mask)
-{
-	struct call_single_data data;
-	int cpu;
-
-	data.func = quiesce_dummy;
-	data.info = NULL;
-
-	for_each_cpu_mask(cpu, mask) {
-		data.flags = CSD_FLAG_WAIT;
-		generic_exec_single(cpu, &data);
-	}
-}
-
 /**
- * smp_call_function_mask(): Run a function on a set of other CPUs.
- * @mask: The set of cpus to run on.
+ * smp_call_function_many(): Run a function on a set of other CPUs.
+ * @mask: The set of cpus to run on (only runs on online subset).
  * @func: The function to run. This must be fast and non-blocking.
  * @info: An arbitrary pointer to pass to the function.
  * @wait: If true, wait (atomically) until function has completed on other CPUs.
  *
- * Returns 0 on success, else a negative status code.
- *
  * If @wait is true, then returns once @func has returned. Note that @wait
  * will be implicitly turned on in case of allocation failures, since
  * we fall back to on-stack allocation.
@@ -319,53 +281,57 @@ static void smp_call_function_mask_quiesce_stack(cpumask_t mask)
  * hardware interrupt handler or from a bottom half handler. Preemption
  * must be disabled when calling this function.
  */
-int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info,
-			   int wait)
+void smp_call_function_many(const struct cpumask *mask,
+			    void (*func)(void *), void *info,
+			    bool wait)
 {
-	struct call_function_data d;
-	struct call_function_data *data = NULL;
-	cpumask_t allbutself;
+	struct call_function_data *data;
 	unsigned long flags;
-	int cpu, num_cpus;
-	int slowpath = 0;
+	int cpu, next_cpu;
 
 	/* Can deadlock when called with interrupts disabled */
 	WARN_ON(irqs_disabled());
 
-	cpu = smp_processor_id();
-	allbutself = cpu_online_map;
-	cpu_clear(cpu, allbutself);
-	cpus_and(mask, mask, allbutself);
-	num_cpus = cpus_weight(mask);
-
-	/*
-	 * If zero CPUs, return. If just a single CPU, turn this request
-	 * into a targetted single call instead since it's faster.
-	 */
-	if (!num_cpus)
-		return 0;
-	else if (num_cpus == 1) {
-		cpu = first_cpu(mask);
-		return smp_call_function_single(cpu, func, info, wait);
+	/* So, what's a CPU they want?  Ignoring this one. */
+	cpu = cpumask_first_and(mask, cpu_online_mask);
+	if (cpu == smp_processor_id())
+		cpu = cpumask_next_and(cpu, mask, cpu_online_mask);
+	/* No online cpus?  We're done. */
+	if (cpu >= nr_cpu_ids)
+		return;
+
+	/* Do we have another CPU which isn't us? */
+	next_cpu = cpumask_next_and(cpu, mask, cpu_online_mask);
+	if (next_cpu == smp_processor_id())
+		next_cpu = cpumask_next_and(next_cpu, mask, cpu_online_mask);
+
+	/* Fastpath: do that cpu by itself. */
+	if (next_cpu >= nr_cpu_ids) {
+		smp_call_function_single(cpu, func, info, wait);
+		return;
 	}
 
-	data = kmalloc(sizeof(*data), GFP_ATOMIC);
-	if (data) {
-		data->csd.flags = CSD_FLAG_ALLOC;
-		if (wait)
-			data->csd.flags |= CSD_FLAG_WAIT;
-	} else {
-		data = &d;
-		data->csd.flags = CSD_FLAG_WAIT;
-		wait = 1;
-		slowpath = 1;
+	data = kmalloc(sizeof(*data) + cpumask_size(), GFP_ATOMIC);
+	if (unlikely(!data)) {
+		/* Slow path. */
+		for_each_online_cpu(cpu) {
+			if (cpu == smp_processor_id())
+				continue;
+			if (cpumask_test_cpu(cpu, mask))
+				smp_call_function_single(cpu, func, info, wait);
+		}
+		return;
 	}
 
 	spin_lock_init(&data->lock);
+	data->csd.flags = CSD_FLAG_ALLOC;
+	if (wait)
+		data->csd.flags |= CSD_FLAG_WAIT;
 	data->csd.func = func;
 	data->csd.info = info;
-	data->refs = num_cpus;
-	data->cpumask = mask;
+	cpumask_and(to_cpumask(data->cpumask_bits), mask, cpu_online_mask);
+	cpumask_clear_cpu(smp_processor_id(), to_cpumask(data->cpumask_bits));
+	data->refs = cpumask_weight(to_cpumask(data->cpumask_bits));
 
 	spin_lock_irqsave(&call_function_lock, flags);
 	list_add_tail_rcu(&data->csd.list, &call_function_queue);
@@ -377,18 +343,13 @@ int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info,
 	smp_mb();
 
 	/* Send a message to all CPUs in the map */
-	arch_send_call_function_ipi(mask);
+	arch_send_call_function_ipi(*to_cpumask(data->cpumask_bits));
 
 	/* optionally wait for the CPUs to complete */
-	if (wait) {
+	if (wait)
 		csd_flag_wait(&data->csd);
-		if (unlikely(slowpath))
-			smp_call_function_mask_quiesce_stack(mask);
-	}
-
-	return 0;
 }
-EXPORT_SYMBOL(smp_call_function_mask);
+EXPORT_SYMBOL(smp_call_function_many);
 
 /**
  * smp_call_function(): Run a function on all other CPUs.
@@ -396,7 +357,7 @@ EXPORT_SYMBOL(smp_call_function_mask);
  * @info: An arbitrary pointer to pass to the function.
  * @wait: If true, wait (atomically) until function has completed on other CPUs.
  *
- * Returns 0 on success, else a negative status code.
+ * Returns 0.
  *
  * If @wait is true, then returns once @func has returned; otherwise
  * it returns just before the target cpu calls @func. In case of allocation
@@ -407,12 +368,10 @@ EXPORT_SYMBOL(smp_call_function_mask);
  */
 int smp_call_function(void (*func)(void *), void *info, int wait)
 {
-	int ret;
-
 	preempt_disable();
-	ret = smp_call_function_mask(cpu_online_map, func, info, wait);
+	smp_call_function_many(cpu_online_mask, func, info, wait);
 	preempt_enable();
-	return ret;
+	return 0;
 }
 EXPORT_SYMBOL(smp_call_function);
 
-- 
cgit v1.2.3


From e12f0102ac81d660c9f801d0a0e10ccf4537a9de Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 30 Dec 2008 09:05:19 +1030
Subject: cpumask: Use nr_cpu_ids in seq_cpumask

Impact: cleanup, futureproof

nr_cpu_ids is the (badly named) runtime limit on possible CPU numbers;
ie. the variable version of NR_CPUS.

With the new cpumask operators, only bits less than this are defined.
So we should use it everywhere, rather than NR_CPUS.  Eventually this
will make it possible to allocate cpumasks of the minimal length at runtime.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/seq_file.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
index 952e0187ba16..40ea5058c2ec 100644
--- a/include/linux/seq_file.h
+++ b/include/linux/seq_file.h
@@ -54,7 +54,7 @@ int seq_bitmap(struct seq_file *m, const unsigned long *bits,
 				   unsigned int nr_bits);
 static inline int seq_cpumask(struct seq_file *m, const struct cpumask *mask)
 {
-	return seq_bitmap(m, mask->bits, NR_CPUS);
+	return seq_bitmap(m, mask->bits, nr_cpu_ids);
 }
 
 static inline int seq_nodemask(struct seq_file *m, nodemask_t *mask)
-- 
cgit v1.2.3


From 47fea2adfc9e16846bc57c2f64ff233b354fef39 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@infradead.org>
Date: Mon, 29 Dec 2008 23:39:17 +0530
Subject: sched: sched.c declare variables before they get used

Impact: cleanup, avoid sparse warnings

In linux/sched.h moved out sysctl_sched_latency, sysctl_sched_min_granularity,
sysctl_sched_wakeup_granularity, sysctl_sched_shares_ratelimit and
sysctl_sched_shares_thresh from #ifdef CONFIG_SCHED_DEBUG as these variables
are common for both.

Fixes these sparse warnings:
  kernel/sched.c:825:14: warning: symbol 'sysctl_sched_shares_ratelimit' was not declared. Should it be static?
  kernel/sched.c:832:14: warning: symbol 'sysctl_sched_shares_thresh' was not declared. Should it be static?
  kernel/sched_fair.c:37:14: warning: symbol 'sysctl_sched_latency' was not declared. Should it be static?
  kernel/sched_fair.c:43:14: warning: symbol 'sysctl_sched_min_granularity' was not declared. Should it be static?
  kernel/sched_fair.c:72:14: warning: symbol 'sysctl_sched_wakeup_granularity' was not declared. Should it be static?

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 8395e715809d..01d9fd268eb0 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1651,16 +1651,16 @@ extern void wake_up_idle_cpu(int cpu);
 static inline void wake_up_idle_cpu(int cpu) { }
 #endif
 
-#ifdef CONFIG_SCHED_DEBUG
 extern unsigned int sysctl_sched_latency;
 extern unsigned int sysctl_sched_min_granularity;
 extern unsigned int sysctl_sched_wakeup_granularity;
+extern unsigned int sysctl_sched_shares_ratelimit;
+extern unsigned int sysctl_sched_shares_thresh;
+#ifdef CONFIG_SCHED_DEBUG
 extern unsigned int sysctl_sched_child_runs_first;
 extern unsigned int sysctl_sched_features;
 extern unsigned int sysctl_sched_migration_cost;
 extern unsigned int sysctl_sched_nr_migrate;
-extern unsigned int sysctl_sched_shares_ratelimit;
-extern unsigned int sysctl_sched_shares_thresh;
 
 int sched_nr_latency_handler(struct ctl_table *table, int write,
 		struct file *file, void __user *buffer, size_t *length,
-- 
cgit v1.2.3


From f748bafa3ca1fb056e63afdeecacc1c68d8104df Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Mon, 8 Dec 2008 21:30:31 -0700
Subject: ACPI: PCI: move struct acpi_prt_entry declaration out of public
 header file

The struct acpi_prt_entry is used only in pci_irq.c, so there's no
need for the declaration to be public.  This patch moves it into
pci_irq.c.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_irq.c | 16 ++++++++++++++++
 include/linux/acpi.h   | 16 ----------------
 2 files changed, 16 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/acpi/pci_irq.c b/drivers/acpi/pci_irq.c
index 9302f4bb89e2..ea003bab7ecd 100644
--- a/drivers/acpi/pci_irq.c
+++ b/drivers/acpi/pci_irq.c
@@ -41,6 +41,22 @@
 #define _COMPONENT		ACPI_PCI_COMPONENT
 ACPI_MODULE_NAME("pci_irq");
 
+struct acpi_prt_entry {
+	struct list_head	node;
+	struct acpi_pci_id	id;
+	u8			pin;
+	struct {
+		acpi_handle		handle;
+		u32			index;
+	}			link;
+	u32			irq;
+};
+
+struct acpi_prt_list {
+	int			count;
+	struct list_head	entries;
+};
+
 static struct acpi_prt_list acpi_prt;
 static DEFINE_SPINLOCK(acpi_prt_lock);
 
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index fba8051fb297..813f937b3ab4 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -131,22 +131,6 @@ extern int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity);
  */
 void acpi_unregister_gsi (u32 gsi);
 
-struct acpi_prt_entry {
-	struct list_head	node;
-	struct acpi_pci_id	id;
-	u8			pin;
-	struct {
-		acpi_handle		handle;
-		u32			index;
-	}			link;
-	u32			irq;
-};
-
-struct acpi_prt_list {
-	int			count;
-	struct list_head	entries;
-};
-
 struct pci_dev;
 
 int acpi_pci_irq_enable (struct pci_dev *dev);
-- 
cgit v1.2.3


From ea7e96e0f2277107d9ea14c3f16c86ba82b2e560 Mon Sep 17 00:00:00 2001
From: Lin Ming <ming.m.lin@intel.com>
Date: Tue, 16 Dec 2008 16:28:17 +0800
Subject: ACPI: remove private acpica headers from driver files

External driver files should not include any private acpica headers.

Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 arch/ia64/include/asm/acpi-ext.h   | 1 -
 arch/ia64/include/asm/sn/acpi.h    | 2 --
 arch/ia64/sn/kernel/io_acpi_init.c | 1 -
 arch/ia64/sn/kernel/io_common.c    | 1 -
 drivers/acpi/cm_sbs.c              | 3 ---
 drivers/acpi/debug.c               | 1 -
 drivers/acpi/ec.c                  | 1 -
 drivers/acpi/numa.c                | 1 -
 drivers/acpi/sbshc.c               | 1 -
 drivers/acpi/scan.c                | 1 -
 drivers/acpi/sleep/wakeup.c        | 1 -
 drivers/ata/libata-acpi.c          | 6 ------
 drivers/ata/pata_acpi.c            | 6 ------
 drivers/char/tpm/tpm_bios.c        | 2 --
 drivers/ide/ide-acpi.c             | 6 ------
 drivers/misc/tc1100-wmi.c          | 1 -
 drivers/misc/thinkpad_acpi.c       | 1 -
 drivers/pci/hotplug/acpi_pcihp.c   | 1 -
 drivers/pci/hotplug/pciehp.h       | 1 -
 drivers/pci/pci-acpi.c             | 2 --
 drivers/pnp/pnpacpi/core.c         | 1 -
 include/linux/pci_hotplug.h        | 1 -
 22 files changed, 42 deletions(-)

(limited to 'include/linux')

diff --git a/arch/ia64/include/asm/acpi-ext.h b/arch/ia64/include/asm/acpi-ext.h
index 734d137dda6e..7f8362b379eb 100644
--- a/arch/ia64/include/asm/acpi-ext.h
+++ b/arch/ia64/include/asm/acpi-ext.h
@@ -14,7 +14,6 @@
 #define _ASM_IA64_ACPI_EXT_H
 
 #include <linux/types.h>
-#include <acpi/actypes.h>
 
 extern acpi_status hp_acpi_csr_space (acpi_handle, u64 *base, u64 *length);
 
diff --git a/arch/ia64/include/asm/sn/acpi.h b/arch/ia64/include/asm/sn/acpi.h
index 9ce2801cbd57..fd480db25565 100644
--- a/arch/ia64/include/asm/sn/acpi.h
+++ b/arch/ia64/include/asm/sn/acpi.h
@@ -9,8 +9,6 @@
 #ifndef _ASM_IA64_SN_ACPI_H
 #define _ASM_IA64_SN_ACPI_H
 
-#include "acpi/acglobal.h"
-
 extern int sn_acpi_rev;
 #define SN_ACPI_BASE_SUPPORT()   (sn_acpi_rev >= 0x20101)
 
diff --git a/arch/ia64/sn/kernel/io_acpi_init.c b/arch/ia64/sn/kernel/io_acpi_init.c
index 4c8bc8eee5ad..c5a214026a77 100644
--- a/arch/ia64/sn/kernel/io_acpi_init.c
+++ b/arch/ia64/sn/kernel/io_acpi_init.c
@@ -13,7 +13,6 @@
 #include <asm/sn/sn_sal.h>
 #include "xtalk/hubdev.h"
 #include <linux/acpi.h>
-#include <acpi/acnamesp.h>
 
 
 /*
diff --git a/arch/ia64/sn/kernel/io_common.c b/arch/ia64/sn/kernel/io_common.c
index e1917ede0d60..0d4ffa4da1da 100644
--- a/arch/ia64/sn/kernel/io_common.c
+++ b/arch/ia64/sn/kernel/io_common.c
@@ -26,7 +26,6 @@
 #include <linux/acpi.h>
 #include <asm/sn/sn2/sn_hwperf.h>
 #include <asm/sn/acpi.h>
-#include "acpi/acglobal.h"
 
 extern void sn_init_cpei_timer(void);
 extern void register_sn_procfs(void);
diff --git a/drivers/acpi/cm_sbs.c b/drivers/acpi/cm_sbs.c
index 307963bd1043..332fe4b21708 100644
--- a/drivers/acpi/cm_sbs.c
+++ b/drivers/acpi/cm_sbs.c
@@ -27,9 +27,6 @@
 #include <linux/seq_file.h>
 #include <acpi/acpi_bus.h>
 #include <acpi/acpi_drivers.h>
-#include <acpi/acmacros.h>
-#include <acpi/actypes.h>
-#include <acpi/acutils.h>
 
 ACPI_MODULE_NAME("cm_sbs");
 #define ACPI_AC_CLASS		"ac_adapter"
diff --git a/drivers/acpi/debug.c b/drivers/acpi/debug.c
index c48396892008..20223cbd0d1c 100644
--- a/drivers/acpi/debug.c
+++ b/drivers/acpi/debug.c
@@ -9,7 +9,6 @@
 #include <linux/moduleparam.h>
 #include <asm/uaccess.h>
 #include <acpi/acpi_drivers.h>
-#include <acpi/acglobal.h>
 
 #define _COMPONENT		ACPI_SYSTEM_COMPONENT
 ACPI_MODULE_NAME("debug");
diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c
index 2cbc2c9c07ac..3105e0410e9b 100644
--- a/drivers/acpi/ec.c
+++ b/drivers/acpi/ec.c
@@ -42,7 +42,6 @@
 #include <asm/io.h>
 #include <acpi/acpi_bus.h>
 #include <acpi/acpi_drivers.h>
-#include <acpi/actypes.h>
 
 #define ACPI_EC_CLASS			"embedded_controller"
 #define ACPI_EC_DEVICE_NAME		"Embedded Controller"
diff --git a/drivers/acpi/numa.c b/drivers/acpi/numa.c
index 25ceae9191ef..c5e292aab0e3 100644
--- a/drivers/acpi/numa.c
+++ b/drivers/acpi/numa.c
@@ -29,7 +29,6 @@
 #include <linux/errno.h>
 #include <linux/acpi.h>
 #include <acpi/acpi_bus.h>
-#include <acpi/acmacros.h>
 
 #define ACPI_NUMA	0x80000000
 #define _COMPONENT	ACPI_NUMA
diff --git a/drivers/acpi/sbshc.c b/drivers/acpi/sbshc.c
index e53e590252c0..0619734895b2 100644
--- a/drivers/acpi/sbshc.c
+++ b/drivers/acpi/sbshc.c
@@ -10,7 +10,6 @@
 
 #include <acpi/acpi_bus.h>
 #include <acpi/acpi_drivers.h>
-#include <acpi/actypes.h>
 #include <linux/wait.h>
 #include <linux/delay.h>
 #include <linux/interrupt.h>
diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index 39b7233c3485..c54d7b6c4066 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -10,7 +10,6 @@
 #include <linux/kthread.h>
 
 #include <acpi/acpi_drivers.h>
-#include <acpi/acinterp.h>	/* for acpi_ex_eisa_id_to_string() */
 
 #define _COMPONENT		ACPI_BUS_COMPONENT
 ACPI_MODULE_NAME("scan");
diff --git a/drivers/acpi/sleep/wakeup.c b/drivers/acpi/sleep/wakeup.c
index db325c28a1fb..2d34806d45dd 100644
--- a/drivers/acpi/sleep/wakeup.c
+++ b/drivers/acpi/sleep/wakeup.c
@@ -8,7 +8,6 @@
 #include <acpi/acpi_drivers.h>
 #include <linux/kernel.h>
 #include <linux/types.h>
-#include <acpi/acevents.h>
 #include "sleep.h"
 
 #define _COMPONENT		ACPI_SYSTEM_COMPONENT
diff --git a/drivers/ata/libata-acpi.c b/drivers/ata/libata-acpi.c
index c012307d0ba6..246987f0b88c 100644
--- a/drivers/ata/libata-acpi.c
+++ b/drivers/ata/libata-acpi.c
@@ -19,12 +19,6 @@
 #include "libata.h"
 
 #include <acpi/acpi_bus.h>
-#include <acpi/acnames.h>
-#include <acpi/acnamesp.h>
-#include <acpi/acparser.h>
-#include <acpi/acexcep.h>
-#include <acpi/acmacros.h>
-#include <acpi/actypes.h>
 
 enum {
 	ATA_ACPI_FILTER_SETXFER	= 1 << 0,
diff --git a/drivers/ata/pata_acpi.c b/drivers/ata/pata_acpi.c
index e2e332d8ff95..8b77a9802df1 100644
--- a/drivers/ata/pata_acpi.c
+++ b/drivers/ata/pata_acpi.c
@@ -13,12 +13,6 @@
 #include <linux/device.h>
 #include <scsi/scsi_host.h>
 #include <acpi/acpi_bus.h>
-#include <acpi/acnames.h>
-#include <acpi/acnamesp.h>
-#include <acpi/acparser.h>
-#include <acpi/acexcep.h>
-#include <acpi/acmacros.h>
-#include <acpi/actypes.h>
 
 #include <linux/libata.h>
 #include <linux/ata.h>
diff --git a/drivers/char/tpm/tpm_bios.c b/drivers/char/tpm/tpm_bios.c
index 68f052b42ed7..ed306eb1057f 100644
--- a/drivers/char/tpm/tpm_bios.c
+++ b/drivers/char/tpm/tpm_bios.c
@@ -23,8 +23,6 @@
 #include <linux/security.h>
 #include <linux/module.h>
 #include <acpi/acpi.h>
-#include <acpi/actypes.h>
-#include <acpi/actbl.h>
 #include "tpm.h"
 
 #define TCG_EVENT_NAME_LEN_MAX	255
diff --git a/drivers/ide/ide-acpi.c b/drivers/ide/ide-acpi.c
index 244a8a052ce8..9e8d52a4f306 100644
--- a/drivers/ide/ide-acpi.c
+++ b/drivers/ide/ide-acpi.c
@@ -18,12 +18,6 @@
 #include <linux/dmi.h>
 
 #include <acpi/acpi_bus.h>
-#include <acpi/acnames.h>
-#include <acpi/acnamesp.h>
-#include <acpi/acparser.h>
-#include <acpi/acexcep.h>
-#include <acpi/acmacros.h>
-#include <acpi/actypes.h>
 
 #define REGS_PER_GTF		7
 struct taskfile_array {
diff --git a/drivers/misc/tc1100-wmi.c b/drivers/misc/tc1100-wmi.c
index f25e4c974dcf..b4a4aa9ee482 100644
--- a/drivers/misc/tc1100-wmi.c
+++ b/drivers/misc/tc1100-wmi.c
@@ -30,7 +30,6 @@
 #include <linux/init.h>
 #include <linux/types.h>
 #include <acpi/acpi.h>
-#include <acpi/actypes.h>
 #include <acpi/acpi_bus.h>
 #include <acpi/acpi_drivers.h>
 #include <linux/platform_device.h>
diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c
index 899766e16fa8..3478453eba7a 100644
--- a/drivers/misc/thinkpad_acpi.c
+++ b/drivers/misc/thinkpad_acpi.c
@@ -76,7 +76,6 @@
 #include <linux/workqueue.h>
 
 #include <acpi/acpi_drivers.h>
-#include <acpi/acnamesp.h>
 
 #include <linux/pci_ids.h>
 
diff --git a/drivers/pci/hotplug/acpi_pcihp.c b/drivers/pci/hotplug/acpi_pcihp.c
index e17ef54f0efc..2c981cbb0719 100644
--- a/drivers/pci/hotplug/acpi_pcihp.c
+++ b/drivers/pci/hotplug/acpi_pcihp.c
@@ -33,7 +33,6 @@
 #include <linux/pci-acpi.h>
 #include <acpi/acpi.h>
 #include <acpi/acpi_bus.h>
-#include <acpi/actypes.h>
 
 #define MY_NAME	"acpi_pcihp"
 
diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h
index b2801a7ee37f..7072952ea1d2 100644
--- a/drivers/pci/hotplug/pciehp.h
+++ b/drivers/pci/hotplug/pciehp.h
@@ -217,7 +217,6 @@ struct hpc_ops {
 #ifdef CONFIG_ACPI
 #include <acpi/acpi.h>
 #include <acpi/acpi_bus.h>
-#include <acpi/actypes.h>
 #include <linux/pci-acpi.h>
 
 static inline int pciehp_get_hp_hw_control_from_firmware(struct pci_dev *dev)
diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c
index ae5ec76dca77..9d976d51d406 100644
--- a/drivers/pci/pci-acpi.c
+++ b/drivers/pci/pci-acpi.c
@@ -13,8 +13,6 @@
 #include <linux/module.h>
 #include <linux/pci-aspm.h>
 #include <acpi/acpi.h>
-#include <acpi/acnamesp.h>
-#include <acpi/acresrc.h>
 #include <acpi/acpi_bus.h>
 
 #include <linux/pci-acpi.h>
diff --git a/drivers/pnp/pnpacpi/core.c b/drivers/pnp/pnpacpi/core.c
index 383e47c392a4..2834846a185d 100644
--- a/drivers/pnp/pnpacpi/core.c
+++ b/drivers/pnp/pnpacpi/core.c
@@ -23,7 +23,6 @@
 #include <linux/pnp.h>
 #include <linux/mod_devicetable.h>
 #include <acpi/acpi_bus.h>
-#include <acpi/actypes.h>
 
 #include "../base.h"
 #include "pnpacpi.h"
diff --git a/include/linux/pci_hotplug.h b/include/linux/pci_hotplug.h
index a00bd1a0f156..c2d1a7d1886a 100644
--- a/include/linux/pci_hotplug.h
+++ b/include/linux/pci_hotplug.h
@@ -223,7 +223,6 @@ struct hotplug_params {
 #ifdef CONFIG_ACPI
 #include <acpi/acpi.h>
 #include <acpi/acpi_bus.h>
-#include <acpi/actypes.h>
 extern acpi_status acpi_get_hp_params_from_firmware(struct pci_bus *bus,
 				struct hotplug_params *hpp);
 int acpi_get_hp_hw_control_from_firmware(struct pci_dev *dev, u32 flags);
-- 
cgit v1.2.3


From 1c5745aa380efb6417b5681104b007c8612fb496 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 22 Dec 2008 23:05:28 +0100
Subject: sched_clock: prevent scd->clock from moving backwards, take #2

Redo:

  5b7dba4: sched_clock: prevent scd->clock from moving backwards

which had to be reverted due to s2ram hangs:

  ca7e716: Revert "sched_clock: prevent scd->clock from moving backwards"

... this time with resume restoring GTOD later in the sequence
taken into account as well.

The "timekeeping_suspended" flag is not very nice but we cannot call into
GTOD before it has been properly resumed and the scheduler will run very
early in the resume sequence.

Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/time.h      | 1 +
 kernel/sched_clock.c      | 5 ++++-
 kernel/time/timekeeping.c | 7 +++++--
 3 files changed, 10 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/time.h b/include/linux/time.h
index ce321ac5c8f8..fbbd2a1c92ba 100644
--- a/include/linux/time.h
+++ b/include/linux/time.h
@@ -105,6 +105,7 @@ extern unsigned long read_persistent_clock(void);
 extern int update_persistent_clock(struct timespec now);
 extern int no_sync_cmos_clock __read_mostly;
 void timekeeping_init(void);
+extern int timekeeping_suspended;
 
 unsigned long get_seconds(void);
 struct timespec current_kernel_time(void);
diff --git a/kernel/sched_clock.c b/kernel/sched_clock.c
index e8ab096ddfe3..a0b0852414cc 100644
--- a/kernel/sched_clock.c
+++ b/kernel/sched_clock.c
@@ -124,7 +124,7 @@ static u64 __update_sched_clock(struct sched_clock_data *scd, u64 now)
 
 	clock = scd->tick_gtod + delta;
 	min_clock = wrap_max(scd->tick_gtod, scd->clock);
-	max_clock = scd->tick_gtod + TICK_NSEC;
+	max_clock = wrap_max(scd->clock, scd->tick_gtod + TICK_NSEC);
 
 	clock = wrap_max(clock, min_clock);
 	clock = wrap_min(clock, max_clock);
@@ -227,6 +227,9 @@ EXPORT_SYMBOL_GPL(sched_clock_idle_sleep_event);
  */
 void sched_clock_idle_wakeup_event(u64 delta_ns)
 {
+	if (timekeeping_suspended)
+		return;
+
 	sched_clock_tick();
 	touch_softlockup_watchdog();
 }
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index fa05e88aa76f..900f1b6598d1 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -46,6 +46,9 @@ struct timespec xtime __attribute__ ((aligned (16)));
 struct timespec wall_to_monotonic __attribute__ ((aligned (16)));
 static unsigned long total_sleep_time;		/* seconds */
 
+/* flag for if timekeeping is suspended */
+int __read_mostly timekeeping_suspended;
+
 static struct timespec xtime_cache __attribute__ ((aligned (16)));
 void update_xtime_cache(u64 nsec)
 {
@@ -92,6 +95,8 @@ void getnstimeofday(struct timespec *ts)
 	unsigned long seq;
 	s64 nsecs;
 
+	WARN_ON(timekeeping_suspended);
+
 	do {
 		seq = read_seqbegin(&xtime_lock);
 
@@ -299,8 +304,6 @@ void __init timekeeping_init(void)
 	write_sequnlock_irqrestore(&xtime_lock, flags);
 }
 
-/* flag for if timekeeping is suspended */
-static int timekeeping_suspended;
 /* time in seconds when suspend began */
 static unsigned long timekeeping_suspend_time;
 
-- 
cgit v1.2.3


From 457533a7d3402d1d91fbc125c8bd1bd16dcd3cd4 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Wed, 31 Dec 2008 15:11:37 +0100
Subject: [PATCH] fix scaled & unscaled cputime accounting

The utimescaled / stimescaled fields in the task structure and the
global cpustat should be set on all architectures. On s390 the calls
to account_user_time_scaled and account_system_time_scaled never have
been added. In addition system time that is accounted as guest time
to the user time of a process is accounted to the scaled system time
instead of the scaled user time.
To fix the bugs and to prevent future forgetfulness this patch merges
account_system_time_scaled into account_system_time and
account_user_time_scaled into account_user_time.

Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Chris Wright <chrisw@sous-sol.org>
Cc: Michael Neuling <mikey@neuling.org>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/ia64/kernel/time.c     | 12 ++++--------
 arch/powerpc/kernel/time.c  |  7 ++-----
 arch/s390/kernel/vtime.c    | 10 +++++-----
 include/linux/kernel_stat.h |  6 ++----
 kernel/sched.c              | 41 ++++++++++++++++-------------------------
 kernel/time/tick-sched.c    |  5 +++--
 kernel/timer.c              | 12 +++++-------
 7 files changed, 37 insertions(+), 56 deletions(-)

(limited to 'include/linux')

diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c
index 65c10a42c88f..4ee367817049 100644
--- a/arch/ia64/kernel/time.c
+++ b/arch/ia64/kernel/time.c
@@ -93,13 +93,11 @@ void ia64_account_on_switch(struct task_struct *prev, struct task_struct *next)
 	now = ia64_get_itc();
 
 	delta_stime = cycle_to_cputime(pi->ac_stime + (now - pi->ac_stamp));
-	account_system_time(prev, 0, delta_stime);
-	account_system_time_scaled(prev, delta_stime);
+	account_system_time(prev, 0, delta_stime, delta_stime);
 
 	if (pi->ac_utime) {
 		delta_utime = cycle_to_cputime(pi->ac_utime);
-		account_user_time(prev, delta_utime);
-		account_user_time_scaled(prev, delta_utime);
+		account_user_time(prev, delta_utime, delta_utime);
 	}
 
 	pi->ac_stamp = ni->ac_stamp = now;
@@ -122,8 +120,7 @@ void account_system_vtime(struct task_struct *tsk)
 	now = ia64_get_itc();
 
 	delta_stime = cycle_to_cputime(ti->ac_stime + (now - ti->ac_stamp));
-	account_system_time(tsk, 0, delta_stime);
-	account_system_time_scaled(tsk, delta_stime);
+	account_system_time(tsk, 0, delta_stime, delta_stime);
 	ti->ac_stime = 0;
 
 	ti->ac_stamp = now;
@@ -143,8 +140,7 @@ void account_process_tick(struct task_struct *p, int user_tick)
 
 	if (ti->ac_utime) {
 		delta_utime = cycle_to_cputime(ti->ac_utime);
-		account_user_time(p, delta_utime);
-		account_user_time_scaled(p, delta_utime);
+		account_user_time(p, delta_utime, delta_utime);
 		ti->ac_utime = 0;
 	}
 }
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index e1f3a5140429..92650ccad2e1 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -256,8 +256,7 @@ void account_system_vtime(struct task_struct *tsk)
 		delta += sys_time;
 		get_paca()->system_time = 0;
 	}
-	account_system_time(tsk, 0, delta);
-	account_system_time_scaled(tsk, deltascaled);
+	account_system_time(tsk, 0, delta, deltascaled);
 	per_cpu(cputime_last_delta, smp_processor_id()) = delta;
 	per_cpu(cputime_scaled_last_delta, smp_processor_id()) = deltascaled;
 	local_irq_restore(flags);
@@ -275,10 +274,8 @@ void account_process_tick(struct task_struct *tsk, int user_tick)
 
 	utime = get_paca()->user_time;
 	get_paca()->user_time = 0;
-	account_user_time(tsk, utime);
-
 	utimescaled = cputime_to_scaled(utime);
-	account_user_time_scaled(tsk, utimescaled);
+	account_user_time(tsk, utime, utimescaled);
 }
 
 /*
diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c
index 75a6e62ea973..07283aea2e56 100644
--- a/arch/s390/kernel/vtime.c
+++ b/arch/s390/kernel/vtime.c
@@ -50,12 +50,12 @@ void account_process_tick(struct task_struct *tsk, int user_tick)
 	rcu_user_flag = cputime != 0;
 	S390_lowcore.user_timer -= cputime << 12;
 	S390_lowcore.steal_clock -= cputime << 12;
-	account_user_time(tsk, cputime);
+	account_user_time(tsk, cputime, cputime);
 
 	cputime =  S390_lowcore.system_timer >> 12;
 	S390_lowcore.system_timer -= cputime << 12;
 	S390_lowcore.steal_clock -= cputime << 12;
-	account_system_time(tsk, HARDIRQ_OFFSET, cputime);
+	account_system_time(tsk, HARDIRQ_OFFSET, cputime, cputime);
 
 	cputime = S390_lowcore.steal_clock;
 	if ((__s64) cputime > 0) {
@@ -82,12 +82,12 @@ void account_vtime(struct task_struct *tsk)
 	cputime = S390_lowcore.user_timer >> 12;
 	S390_lowcore.user_timer -= cputime << 12;
 	S390_lowcore.steal_clock -= cputime << 12;
-	account_user_time(tsk, cputime);
+	account_user_time(tsk, cputime, cputime);
 
 	cputime =  S390_lowcore.system_timer >> 12;
 	S390_lowcore.system_timer -= cputime << 12;
 	S390_lowcore.steal_clock -= cputime << 12;
-	account_system_time(tsk, 0, cputime);
+	account_system_time(tsk, 0, cputime, cputime);
 }
 
 /*
@@ -107,7 +107,7 @@ void account_system_vtime(struct task_struct *tsk)
 	cputime =  S390_lowcore.system_timer >> 12;
 	S390_lowcore.system_timer -= cputime << 12;
 	S390_lowcore.steal_clock -= cputime << 12;
-	account_system_time(tsk, 0, cputime);
+	account_system_time(tsk, 0, cputime, cputime);
 }
 EXPORT_SYMBOL_GPL(account_system_vtime);
 
diff --git a/include/linux/kernel_stat.h b/include/linux/kernel_stat.h
index 4ee4b3d2316f..c78a459662a6 100644
--- a/include/linux/kernel_stat.h
+++ b/include/linux/kernel_stat.h
@@ -79,10 +79,8 @@ static inline unsigned int kstat_irqs(unsigned int irq)
 }
 
 extern unsigned long long task_delta_exec(struct task_struct *);
-extern void account_user_time(struct task_struct *, cputime_t);
-extern void account_user_time_scaled(struct task_struct *, cputime_t);
-extern void account_system_time(struct task_struct *, int, cputime_t);
-extern void account_system_time_scaled(struct task_struct *, cputime_t);
+extern void account_user_time(struct task_struct *, cputime_t, cputime_t);
+extern void account_system_time(struct task_struct *, int, cputime_t, cputime_t);
 extern void account_steal_time(struct task_struct *, cputime_t);
 
 #endif /* _LINUX_KERNEL_STAT_H */
diff --git a/kernel/sched.c b/kernel/sched.c
index fff1c4a20b65..5b03679ff712 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4080,13 +4080,17 @@ unsigned long long task_delta_exec(struct task_struct *p)
  * Account user cpu time to a process.
  * @p: the process that the cpu time gets accounted to
  * @cputime: the cpu time spent in user space since the last update
+ * @cputime_scaled: cputime scaled by cpu frequency
  */
-void account_user_time(struct task_struct *p, cputime_t cputime)
+void account_user_time(struct task_struct *p, cputime_t cputime,
+		       cputime_t cputime_scaled)
 {
 	struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
 	cputime64_t tmp;
 
+	/* Add user time to process. */
 	p->utime = cputime_add(p->utime, cputime);
+	p->utimescaled = cputime_add(p->utimescaled, cputime_scaled);
 	account_group_user_time(p, cputime);
 
 	/* Add user time to cpustat. */
@@ -4103,51 +4107,49 @@ void account_user_time(struct task_struct *p, cputime_t cputime)
  * Account guest cpu time to a process.
  * @p: the process that the cpu time gets accounted to
  * @cputime: the cpu time spent in virtual machine since the last update
+ * @cputime_scaled: cputime scaled by cpu frequency
  */
-static void account_guest_time(struct task_struct *p, cputime_t cputime)
+static void account_guest_time(struct task_struct *p, cputime_t cputime,
+			       cputime_t cputime_scaled)
 {
 	cputime64_t tmp;
 	struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
 
 	tmp = cputime_to_cputime64(cputime);
 
+	/* Add guest time to process. */
 	p->utime = cputime_add(p->utime, cputime);
+	p->utimescaled = cputime_add(p->utimescaled, cputime_scaled);
 	account_group_user_time(p, cputime);
 	p->gtime = cputime_add(p->gtime, cputime);
 
+	/* Add guest time to cpustat. */
 	cpustat->user = cputime64_add(cpustat->user, tmp);
 	cpustat->guest = cputime64_add(cpustat->guest, tmp);
 }
 
-/*
- * Account scaled user cpu time to a process.
- * @p: the process that the cpu time gets accounted to
- * @cputime: the cpu time spent in user space since the last update
- */
-void account_user_time_scaled(struct task_struct *p, cputime_t cputime)
-{
-	p->utimescaled = cputime_add(p->utimescaled, cputime);
-}
-
 /*
  * Account system cpu time to a process.
  * @p: the process that the cpu time gets accounted to
  * @hardirq_offset: the offset to subtract from hardirq_count()
  * @cputime: the cpu time spent in kernel space since the last update
+ * @cputime_scaled: cputime scaled by cpu frequency
  */
 void account_system_time(struct task_struct *p, int hardirq_offset,
-			 cputime_t cputime)
+			 cputime_t cputime, cputime_t cputime_scaled)
 {
 	struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
 	struct rq *rq = this_rq();
 	cputime64_t tmp;
 
 	if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) {
-		account_guest_time(p, cputime);
+		account_guest_time(p, cputime, cputime_scaled);
 		return;
 	}
 
+	/* Add system time to process. */
 	p->stime = cputime_add(p->stime, cputime);
+	p->stimescaled = cputime_add(p->stimescaled, cputime_scaled);
 	account_group_system_time(p, cputime);
 
 	/* Add system time to cpustat. */
@@ -4166,17 +4168,6 @@ void account_system_time(struct task_struct *p, int hardirq_offset,
 	acct_update_integrals(p);
 }
 
-/*
- * Account scaled system cpu time to a process.
- * @p: the process that the cpu time gets accounted to
- * @hardirq_offset: the offset to subtract from hardirq_count()
- * @cputime: the cpu time spent in kernel space since the last update
- */
-void account_system_time_scaled(struct task_struct *p, cputime_t cputime)
-{
-	p->stimescaled = cputime_add(p->stimescaled, cputime);
-}
-
 /*
  * Account for involuntary wait time.
  * @p: the process from which the cpu time has been stolen
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 8f3fc2582d38..1f2fce2479fe 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -420,6 +420,7 @@ void tick_nohz_restart_sched_tick(void)
 	int cpu = smp_processor_id();
 	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
 	unsigned long ticks;
+	cputime_t cputime;
 	ktime_t now;
 
 	local_irq_disable();
@@ -452,8 +453,8 @@ void tick_nohz_restart_sched_tick(void)
 	 */
 	if (ticks && ticks < LONG_MAX) {
 		add_preempt_count(HARDIRQ_OFFSET);
-		account_system_time(current, HARDIRQ_OFFSET,
-				    jiffies_to_cputime(ticks));
+		cputime = jiffies_to_cputime(ticks);
+		account_system_time(current, HARDIRQ_OFFSET, cputime, cputime);
 		sub_preempt_count(HARDIRQ_OFFSET);
 	}
 
diff --git a/kernel/timer.c b/kernel/timer.c
index 566257d1dc10..b5efb528aa1d 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -1023,13 +1023,11 @@ void account_process_tick(struct task_struct *p, int user_tick)
 {
 	cputime_t one_jiffy = jiffies_to_cputime(1);
 
-	if (user_tick) {
-		account_user_time(p, one_jiffy);
-		account_user_time_scaled(p, cputime_to_scaled(one_jiffy));
-	} else {
-		account_system_time(p, HARDIRQ_OFFSET, one_jiffy);
-		account_system_time_scaled(p, cputime_to_scaled(one_jiffy));
-	}
+	if (user_tick)
+		account_user_time(p, one_jiffy, cputime_to_scaled(one_jiffy));
+	else
+		account_system_time(p, HARDIRQ_OFFSET, one_jiffy,
+				    cputime_to_scaled(one_jiffy));
 }
 #endif
 
-- 
cgit v1.2.3


From 79741dd35713ff4f6fd0eafd59fa94e8a4ba922d Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Wed, 31 Dec 2008 15:11:38 +0100
Subject: [PATCH] idle cputime accounting

The cpu time spent by the idle process actually doing something is
currently accounted as idle time. This is plain wrong, the architectures
that support VIRT_CPU_ACCOUNTING=y can do better: distinguish between the
time spent doing nothing and the time spent by idle doing work. The first
is accounted with account_idle_time and the second with account_system_time.
The architectures that use the account_xxx_time interface directly and not
the account_xxx_ticks interface now need to do the check for the idle
process in their arch code. In particular to improve the system vs true
idle time accounting the arch code needs to measure the true idle time
instead of just testing for the idle process.
To improve the tick based accounting as well we would need an architecture
primitive that can tell us if the pt_regs of the interrupted context
points to the magic instruction that halts the cpu.

In addition idle time is no more added to the stime of the idle process.
This field now contains the system time of the idle process as it should
be. On systems without VIRT_CPU_ACCOUNTING this will always be zero as
every tick that occurs while idle is running will be accounted as idle
time.

This patch contains the necessary common code changes to be able to
distinguish idle system time and true idle time. The architectures with
support for VIRT_CPU_ACCOUNTING need some changes to exploit this.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/ia64/kernel/time.c       | 10 ++++--
 arch/powerpc/kernel/process.c |  1 +
 arch/powerpc/kernel/time.c    | 13 +++++--
 arch/s390/kernel/vtime.c      | 20 ++++++++---
 arch/x86/xen/time.c           | 10 +++---
 include/linux/kernel_stat.h   |  7 +++-
 include/linux/sched.h         |  1 -
 kernel/sched.c                | 80 ++++++++++++++++++++++++++++++++++---------
 kernel/time/tick-sched.c      | 13 ++++---
 kernel/timer.c                | 13 -------
 10 files changed, 114 insertions(+), 54 deletions(-)

(limited to 'include/linux')

diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c
index 4ee367817049..f0ebb342409d 100644
--- a/arch/ia64/kernel/time.c
+++ b/arch/ia64/kernel/time.c
@@ -93,7 +93,10 @@ void ia64_account_on_switch(struct task_struct *prev, struct task_struct *next)
 	now = ia64_get_itc();
 
 	delta_stime = cycle_to_cputime(pi->ac_stime + (now - pi->ac_stamp));
-	account_system_time(prev, 0, delta_stime, delta_stime);
+	if (idle_task(smp_processor_id()) != prev)
+		account_system_time(prev, 0, delta_stime, delta_stime);
+	else
+		account_idle_time(delta_stime);
 
 	if (pi->ac_utime) {
 		delta_utime = cycle_to_cputime(pi->ac_utime);
@@ -120,7 +123,10 @@ void account_system_vtime(struct task_struct *tsk)
 	now = ia64_get_itc();
 
 	delta_stime = cycle_to_cputime(ti->ac_stime + (now - ti->ac_stamp));
-	account_system_time(tsk, 0, delta_stime, delta_stime);
+	if (irq_count() || idle_task(smp_processor_id()) != tsk)
+		account_system_time(tsk, 0, delta_stime, delta_stime);
+	else
+		account_idle_time(delta_stime);
 	ti->ac_stime = 0;
 
 	ti->ac_stamp = now;
diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index 51b201ddf9a1..fb7049c054c0 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -33,6 +33,7 @@
 #include <linux/mqueue.h>
 #include <linux/hardirq.h>
 #include <linux/utsname.h>
+#include <linux/kernel_stat.h>
 
 #include <asm/pgtable.h>
 #include <asm/uaccess.h>
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index 92650ccad2e1..3be355c1cfa7 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -256,7 +256,10 @@ void account_system_vtime(struct task_struct *tsk)
 		delta += sys_time;
 		get_paca()->system_time = 0;
 	}
-	account_system_time(tsk, 0, delta, deltascaled);
+	if (in_irq() || idle_task(smp_processor_id()) != tsk)
+		account_system_time(tsk, 0, delta, deltascaled);
+	else
+		account_idle_time(delta);
 	per_cpu(cputime_last_delta, smp_processor_id()) = delta;
 	per_cpu(cputime_scaled_last_delta, smp_processor_id()) = deltascaled;
 	local_irq_restore(flags);
@@ -335,8 +338,12 @@ void calculate_steal_time(void)
 	tb = mftb();
 	purr = mfspr(SPRN_PURR);
 	stolen = (tb - pme->tb) - (purr - pme->purr);
-	if (stolen > 0)
-		account_steal_time(current, stolen);
+	if (stolen > 0) {
+		if (idle_task(smp_processor_id()) != current)
+			account_steal_time(stolen);
+		else
+			account_idle_time(stolen);
+	}
 	pme->tb = tb;
 	pme->purr = purr;
 }
diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c
index 07283aea2e56..4a4a34caec55 100644
--- a/arch/s390/kernel/vtime.c
+++ b/arch/s390/kernel/vtime.c
@@ -55,13 +55,19 @@ void account_process_tick(struct task_struct *tsk, int user_tick)
 	cputime =  S390_lowcore.system_timer >> 12;
 	S390_lowcore.system_timer -= cputime << 12;
 	S390_lowcore.steal_clock -= cputime << 12;
-	account_system_time(tsk, HARDIRQ_OFFSET, cputime, cputime);
+	if (idle_task(smp_processor_id()) != current)
+		account_system_time(tsk, HARDIRQ_OFFSET, cputime, cputime);
+	else
+		account_idle_time(cputime);
 
 	cputime = S390_lowcore.steal_clock;
 	if ((__s64) cputime > 0) {
 		cputime >>= 12;
 		S390_lowcore.steal_clock -= cputime << 12;
-		account_steal_time(tsk, cputime);
+		if (idle_task(smp_processor_id()) != current)
+			account_steal_time(cputime);
+		else
+			account_idle_time(cputime);
 	}
 }
 
@@ -87,7 +93,10 @@ void account_vtime(struct task_struct *tsk)
 	cputime =  S390_lowcore.system_timer >> 12;
 	S390_lowcore.system_timer -= cputime << 12;
 	S390_lowcore.steal_clock -= cputime << 12;
-	account_system_time(tsk, 0, cputime, cputime);
+	if (idle_task(smp_processor_id()) != current)
+		account_system_time(tsk, 0, cputime, cputime);
+	else
+		account_idle_time(cputime);
 }
 
 /*
@@ -107,7 +116,10 @@ void account_system_vtime(struct task_struct *tsk)
 	cputime =  S390_lowcore.system_timer >> 12;
 	S390_lowcore.system_timer -= cputime << 12;
 	S390_lowcore.steal_clock -= cputime << 12;
-	account_system_time(tsk, 0, cputime, cputime);
+	if (in_irq() || idle_task(smp_processor_id()) != current)
+		account_system_time(tsk, 0, cputime, cputime);
+	else
+		account_idle_time(cputime);
 }
 EXPORT_SYMBOL_GPL(account_system_vtime);
 
diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c
index c9f7cda48ed7..732e52dc991a 100644
--- a/arch/x86/xen/time.c
+++ b/arch/x86/xen/time.c
@@ -132,8 +132,7 @@ static void do_stolen_accounting(void)
 	*snap = state;
 
 	/* Add the appropriate number of ticks of stolen time,
-	   including any left-overs from last time.  Passing NULL to
-	   account_steal_time accounts the time as stolen. */
+	   including any left-overs from last time. */
 	stolen = runnable + offline + __get_cpu_var(residual_stolen);
 
 	if (stolen < 0)
@@ -141,11 +140,10 @@ static void do_stolen_accounting(void)
 
 	ticks = iter_div_u64_rem(stolen, NS_PER_TICK, &stolen);
 	__get_cpu_var(residual_stolen) = stolen;
-	account_steal_time(NULL, ticks);
+	account_steal_ticks(ticks);
 
 	/* Add the appropriate number of ticks of blocked time,
-	   including any left-overs from last time.  Passing idle to
-	   account_steal_time accounts the time as idle/wait. */
+	   including any left-overs from last time. */
 	blocked += __get_cpu_var(residual_blocked);
 
 	if (blocked < 0)
@@ -153,7 +151,7 @@ static void do_stolen_accounting(void)
 
 	ticks = iter_div_u64_rem(blocked, NS_PER_TICK, &blocked);
 	__get_cpu_var(residual_blocked) = blocked;
-	account_steal_time(idle_task(smp_processor_id()), ticks);
+	account_idle_ticks(ticks);
 }
 
 /*
diff --git a/include/linux/kernel_stat.h b/include/linux/kernel_stat.h
index c78a459662a6..570d20413119 100644
--- a/include/linux/kernel_stat.h
+++ b/include/linux/kernel_stat.h
@@ -81,6 +81,11 @@ static inline unsigned int kstat_irqs(unsigned int irq)
 extern unsigned long long task_delta_exec(struct task_struct *);
 extern void account_user_time(struct task_struct *, cputime_t, cputime_t);
 extern void account_system_time(struct task_struct *, int, cputime_t, cputime_t);
-extern void account_steal_time(struct task_struct *, cputime_t);
+extern void account_steal_time(cputime_t);
+extern void account_idle_time(cputime_t);
+
+extern void account_process_tick(struct task_struct *, int user);
+extern void account_steal_ticks(unsigned long ticks);
+extern void account_idle_ticks(unsigned long ticks);
 
 #endif /* _LINUX_KERNEL_STAT_H */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 8395e715809d..b475d4db8053 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -284,7 +284,6 @@ long io_schedule_timeout(long timeout);
 
 extern void cpu_init (void);
 extern void trap_init(void);
-extern void account_process_tick(struct task_struct *task, int user);
 extern void update_process_times(int user);
 extern void scheduler_tick(void);
 
diff --git a/kernel/sched.c b/kernel/sched.c
index 5b03679ff712..635eaffe1e4c 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4139,7 +4139,6 @@ void account_system_time(struct task_struct *p, int hardirq_offset,
 			 cputime_t cputime, cputime_t cputime_scaled)
 {
 	struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
-	struct rq *rq = this_rq();
 	cputime64_t tmp;
 
 	if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) {
@@ -4158,37 +4157,84 @@ void account_system_time(struct task_struct *p, int hardirq_offset,
 		cpustat->irq = cputime64_add(cpustat->irq, tmp);
 	else if (softirq_count())
 		cpustat->softirq = cputime64_add(cpustat->softirq, tmp);
-	else if (p != rq->idle)
-		cpustat->system = cputime64_add(cpustat->system, tmp);
-	else if (atomic_read(&rq->nr_iowait) > 0)
-		cpustat->iowait = cputime64_add(cpustat->iowait, tmp);
 	else
-		cpustat->idle = cputime64_add(cpustat->idle, tmp);
+		cpustat->system = cputime64_add(cpustat->system, tmp);
+
 	/* Account for system time used */
 	acct_update_integrals(p);
 }
 
 /*
  * Account for involuntary wait time.
- * @p: the process from which the cpu time has been stolen
  * @steal: the cpu time spent in involuntary wait
  */
-void account_steal_time(struct task_struct *p, cputime_t steal)
+void account_steal_time(cputime_t cputime)
+{
+	struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
+	cputime64_t cputime64 = cputime_to_cputime64(cputime);
+
+	cpustat->steal = cputime64_add(cpustat->steal, cputime64);
+}
+
+/*
+ * Account for idle time.
+ * @cputime: the cpu time spent in idle wait
+ */
+void account_idle_time(cputime_t cputime)
 {
 	struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
-	cputime64_t tmp = cputime_to_cputime64(steal);
+	cputime64_t cputime64 = cputime_to_cputime64(cputime);
 	struct rq *rq = this_rq();
 
-	if (p == rq->idle) {
-		p->stime = cputime_add(p->stime, steal);
-		if (atomic_read(&rq->nr_iowait) > 0)
-			cpustat->iowait = cputime64_add(cpustat->iowait, tmp);
-		else
-			cpustat->idle = cputime64_add(cpustat->idle, tmp);
-	} else
-		cpustat->steal = cputime64_add(cpustat->steal, tmp);
+	if (atomic_read(&rq->nr_iowait) > 0)
+		cpustat->iowait = cputime64_add(cpustat->iowait, cputime64);
+	else
+		cpustat->idle = cputime64_add(cpustat->idle, cputime64);
+}
+
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING
+
+/*
+ * Account a single tick of cpu time.
+ * @p: the process that the cpu time gets accounted to
+ * @user_tick: indicates if the tick is a user or a system tick
+ */
+void account_process_tick(struct task_struct *p, int user_tick)
+{
+	cputime_t one_jiffy = jiffies_to_cputime(1);
+	cputime_t one_jiffy_scaled = cputime_to_scaled(one_jiffy);
+	struct rq *rq = this_rq();
+
+	if (user_tick)
+		account_user_time(p, one_jiffy, one_jiffy_scaled);
+	else if (p != rq->idle)
+		account_system_time(p, HARDIRQ_OFFSET, one_jiffy,
+				    one_jiffy_scaled);
+	else
+		account_idle_time(one_jiffy);
+}
+
+/*
+ * Account multiple ticks of steal time.
+ * @p: the process from which the cpu time has been stolen
+ * @ticks: number of stolen ticks
+ */
+void account_steal_ticks(unsigned long ticks)
+{
+	account_steal_time(jiffies_to_cputime(ticks));
+}
+
+/*
+ * Account multiple ticks of idle time.
+ * @ticks: number of stolen ticks
+ */
+void account_idle_ticks(unsigned long ticks)
+{
+	account_idle_time(jiffies_to_cputime(ticks));
 }
 
+#endif
+
 /*
  * Use precise platform statistics if available:
  */
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 1f2fce2479fe..611fa4c0baab 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -419,8 +419,9 @@ void tick_nohz_restart_sched_tick(void)
 {
 	int cpu = smp_processor_id();
 	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING
 	unsigned long ticks;
-	cputime_t cputime;
+#endif
 	ktime_t now;
 
 	local_irq_disable();
@@ -442,6 +443,7 @@ void tick_nohz_restart_sched_tick(void)
 	tick_do_update_jiffies64(now);
 	cpu_clear(cpu, nohz_cpu_mask);
 
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING
 	/*
 	 * We stopped the tick in idle. Update process times would miss the
 	 * time we slept as update_process_times does only a 1 tick
@@ -451,12 +453,9 @@ void tick_nohz_restart_sched_tick(void)
 	/*
 	 * We might be one off. Do not randomly account a huge number of ticks!
 	 */
-	if (ticks && ticks < LONG_MAX) {
-		add_preempt_count(HARDIRQ_OFFSET);
-		cputime = jiffies_to_cputime(ticks);
-		account_system_time(current, HARDIRQ_OFFSET, cputime, cputime);
-		sub_preempt_count(HARDIRQ_OFFSET);
-	}
+	if (ticks && ticks < LONG_MAX)
+		account_idle_ticks(ticks);
+#endif
 
 	touch_softlockup_watchdog();
 	/*
diff --git a/kernel/timer.c b/kernel/timer.c
index b5efb528aa1d..dee3f641a7a7 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -1018,19 +1018,6 @@ unsigned long get_next_timer_interrupt(unsigned long now)
 }
 #endif
 
-#ifndef CONFIG_VIRT_CPU_ACCOUNTING
-void account_process_tick(struct task_struct *p, int user_tick)
-{
-	cputime_t one_jiffy = jiffies_to_cputime(1);
-
-	if (user_tick)
-		account_user_time(p, one_jiffy, cputime_to_scaled(one_jiffy));
-	else
-		account_system_time(p, HARDIRQ_OFFSET, one_jiffy,
-				    cputime_to_scaled(one_jiffy));
-}
-#endif
-
 /*
  * Called from the timer interrupt handler to charge one tick to the current
  * process.  user_tick is 1 if the tick is user time, 0 for system.
-- 
cgit v1.2.3


From c4abb7c9cde24b7351a47328ef866e6a2bbb1ad0 Mon Sep 17 00:00:00 2001
From: Jan Kiszka <jan.kiszka@siemens.com>
Date: Fri, 26 Sep 2008 09:30:55 +0200
Subject: KVM: x86: Support for user space injected NMIs

Introduces the KVM_NMI IOCTL to the generic x86 part of KVM for
injecting NMIs from user space and also extends the statistic report
accordingly.

Based on the original patch by Sheng Yang.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Sheng Yang <sheng.yang@intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/x86/include/asm/kvm_host.h |  2 ++
 arch/x86/kvm/x86.c              | 46 +++++++++++++++++++++++++++++++++++++++--
 include/linux/kvm.h             | 11 ++++++++--
 3 files changed, 55 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index bfbbdea869bf..a40fa8478920 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -398,6 +398,7 @@ struct kvm_vcpu_stat {
 	u32 halt_exits;
 	u32 halt_wakeup;
 	u32 request_irq_exits;
+	u32 request_nmi_exits;
 	u32 irq_exits;
 	u32 host_state_reload;
 	u32 efer_reload;
@@ -406,6 +407,7 @@ struct kvm_vcpu_stat {
 	u32 insn_emulation_fail;
 	u32 hypercalls;
 	u32 irq_injections;
+	u32 nmi_injections;
 };
 
 struct descriptor_table {
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 1fa9a6db633d..07971451b947 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -86,6 +86,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
 	{ "hypercalls", VCPU_STAT(hypercalls) },
 	{ "request_irq", VCPU_STAT(request_irq_exits) },
+	{ "request_nmi", VCPU_STAT(request_nmi_exits) },
 	{ "irq_exits", VCPU_STAT(irq_exits) },
 	{ "host_state_reload", VCPU_STAT(host_state_reload) },
 	{ "efer_reload", VCPU_STAT(efer_reload) },
@@ -93,6 +94,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ "insn_emulation", VCPU_STAT(insn_emulation) },
 	{ "insn_emulation_fail", VCPU_STAT(insn_emulation_fail) },
 	{ "irq_injections", VCPU_STAT(irq_injections) },
+	{ "nmi_injections", VCPU_STAT(nmi_injections) },
 	{ "mmu_shadow_zapped", VM_STAT(mmu_shadow_zapped) },
 	{ "mmu_pte_write", VM_STAT(mmu_pte_write) },
 	{ "mmu_pte_updated", VM_STAT(mmu_pte_updated) },
@@ -1318,6 +1320,15 @@ static int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu,
 	return 0;
 }
 
+static int kvm_vcpu_ioctl_nmi(struct kvm_vcpu *vcpu)
+{
+	vcpu_load(vcpu);
+	kvm_inject_nmi(vcpu);
+	vcpu_put(vcpu);
+
+	return 0;
+}
+
 static int vcpu_ioctl_tpr_access_reporting(struct kvm_vcpu *vcpu,
 					   struct kvm_tpr_access_ctl *tac)
 {
@@ -1377,6 +1388,13 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 		r = 0;
 		break;
 	}
+	case KVM_NMI: {
+		r = kvm_vcpu_ioctl_nmi(vcpu);
+		if (r)
+			goto out;
+		r = 0;
+		break;
+	}
 	case KVM_SET_CPUID: {
 		struct kvm_cpuid __user *cpuid_arg = argp;
 		struct kvm_cpuid cpuid;
@@ -2812,18 +2830,37 @@ static int dm_request_for_irq_injection(struct kvm_vcpu *vcpu,
 		(kvm_x86_ops->get_rflags(vcpu) & X86_EFLAGS_IF));
 }
 
+/*
+ * Check if userspace requested a NMI window, and that the NMI window
+ * is open.
+ *
+ * No need to exit to userspace if we already have a NMI queued.
+ */
+static int dm_request_for_nmi_injection(struct kvm_vcpu *vcpu,
+					struct kvm_run *kvm_run)
+{
+	return (!vcpu->arch.nmi_pending &&
+		kvm_run->request_nmi_window &&
+		vcpu->arch.nmi_window_open);
+}
+
 static void post_kvm_run_save(struct kvm_vcpu *vcpu,
 			      struct kvm_run *kvm_run)
 {
 	kvm_run->if_flag = (kvm_x86_ops->get_rflags(vcpu) & X86_EFLAGS_IF) != 0;
 	kvm_run->cr8 = kvm_get_cr8(vcpu);
 	kvm_run->apic_base = kvm_get_apic_base(vcpu);
-	if (irqchip_in_kernel(vcpu->kvm))
+	if (irqchip_in_kernel(vcpu->kvm)) {
 		kvm_run->ready_for_interrupt_injection = 1;
-	else
+		kvm_run->ready_for_nmi_injection = 1;
+	} else {
 		kvm_run->ready_for_interrupt_injection =
 					(vcpu->arch.interrupt_window_open &&
 					 vcpu->arch.irq_summary == 0);
+		kvm_run->ready_for_nmi_injection =
+					(vcpu->arch.nmi_window_open &&
+					 vcpu->arch.nmi_pending == 0);
+	}
 }
 
 static void vapic_enter(struct kvm_vcpu *vcpu)
@@ -2999,6 +3036,11 @@ static int __vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		}
 
 		if (r > 0) {
+			if (dm_request_for_nmi_injection(vcpu, kvm_run)) {
+				r = -EINTR;
+				kvm_run->exit_reason = KVM_EXIT_NMI;
+				++vcpu->stat.request_nmi_exits;
+			}
 			if (dm_request_for_irq_injection(vcpu, kvm_run)) {
 				r = -EINTR;
 				kvm_run->exit_reason = KVM_EXIT_INTR;
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index f18b86fa8655..44fd7fa0af2b 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -83,18 +83,22 @@ struct kvm_irqchip {
 #define KVM_EXIT_S390_SIEIC       13
 #define KVM_EXIT_S390_RESET       14
 #define KVM_EXIT_DCR              15
+#define KVM_EXIT_NMI              16
+#define KVM_EXIT_NMI_WINDOW_OPEN  17
 
 /* for KVM_RUN, returned by mmap(vcpu_fd, offset=0) */
 struct kvm_run {
 	/* in */
 	__u8 request_interrupt_window;
-	__u8 padding1[7];
+	__u8 request_nmi_window;
+	__u8 padding1[6];
 
 	/* out */
 	__u32 exit_reason;
 	__u8 ready_for_interrupt_injection;
 	__u8 if_flag;
-	__u8 padding2[2];
+	__u8 ready_for_nmi_injection;
+	__u8 padding2;
 
 	/* in (pre_kvm_run), out (post_kvm_run) */
 	__u64 cr8;
@@ -387,6 +391,7 @@ struct kvm_trace_rec {
 #define KVM_CAP_DEVICE_ASSIGNMENT 17
 #endif
 #define KVM_CAP_IOMMU 18
+#define KVM_CAP_NMI 19
 
 /*
  * ioctls for VM fds
@@ -458,6 +463,8 @@ struct kvm_trace_rec {
 #define KVM_S390_INITIAL_RESET    _IO(KVMIO,  0x97)
 #define KVM_GET_MP_STATE          _IOR(KVMIO,  0x98, struct kvm_mp_state)
 #define KVM_SET_MP_STATE          _IOW(KVMIO,  0x99, struct kvm_mp_state)
+/* Available with KVM_CAP_NMI */
+#define KVM_NMI                   _IO(KVMIO,  0x9a)
 
 #define KVM_TRC_INJ_VIRQ         (KVM_TRC_HANDLER + 0x02)
 #define KVM_TRC_REDELIVER_EVT    (KVM_TRC_HANDLER + 0x03)
-- 
cgit v1.2.3


From e19e30effac03f5a005a8e42ed941a2a5dc62654 Mon Sep 17 00:00:00 2001
From: Sheng Yang <sheng@linux.intel.com>
Date: Mon, 20 Oct 2008 16:07:10 +0800
Subject: KVM: IRQ ACK notifier should be used with in-kernel irqchip

Also remove unnecessary parameter of unregister irq ack notifier.

Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 include/linux/kvm_host.h | 3 +--
 virt/kvm/irq_comm.c      | 8 ++++++--
 virt/kvm/kvm_main.c      | 2 +-
 3 files changed, 8 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index bb92be2153bc..3a0fb77d1f6a 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -316,8 +316,7 @@ void kvm_set_irq(struct kvm *kvm, int irq_source_id, int irq, int level);
 void kvm_notify_acked_irq(struct kvm *kvm, unsigned gsi);
 void kvm_register_irq_ack_notifier(struct kvm *kvm,
 				   struct kvm_irq_ack_notifier *kian);
-void kvm_unregister_irq_ack_notifier(struct kvm *kvm,
-				     struct kvm_irq_ack_notifier *kian);
+void kvm_unregister_irq_ack_notifier(struct kvm_irq_ack_notifier *kian);
 int kvm_request_irq_source_id(struct kvm *kvm);
 void kvm_free_irq_source_id(struct kvm *kvm, int irq_source_id);
 
diff --git a/virt/kvm/irq_comm.c b/virt/kvm/irq_comm.c
index 55ad76ee2d09..9fbbdea3d1d5 100644
--- a/virt/kvm/irq_comm.c
+++ b/virt/kvm/irq_comm.c
@@ -58,12 +58,16 @@ void kvm_notify_acked_irq(struct kvm *kvm, unsigned gsi)
 void kvm_register_irq_ack_notifier(struct kvm *kvm,
 				   struct kvm_irq_ack_notifier *kian)
 {
+	/* Must be called with in-kernel IRQ chip, otherwise it's nonsense */
+	ASSERT(irqchip_in_kernel(kvm));
+	ASSERT(kian);
 	hlist_add_head(&kian->link, &kvm->arch.irq_ack_notifier_list);
 }
 
-void kvm_unregister_irq_ack_notifier(struct kvm *kvm,
-				     struct kvm_irq_ack_notifier *kian)
+void kvm_unregister_irq_ack_notifier(struct kvm_irq_ack_notifier *kian)
 {
+	if (!kian)
+		return;
 	hlist_del(&kian->link);
 }
 
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index a87f45edfae8..4f43abe198e4 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -143,7 +143,7 @@ static void kvm_free_assigned_device(struct kvm *kvm,
 	if (irqchip_in_kernel(kvm) && assigned_dev->irq_requested)
 		free_irq(assigned_dev->host_irq, (void *)assigned_dev);
 
-	kvm_unregister_irq_ack_notifier(kvm, &assigned_dev->ack_notifier);
+	kvm_unregister_irq_ack_notifier(&assigned_dev->ack_notifier);
 	kvm_free_irq_source_id(kvm, assigned_dev->irq_source_id);
 
 	if (cancel_work_sync(&assigned_dev->interrupt_work))
-- 
cgit v1.2.3


From 4f906c19ae29397409bedabf7bbe5cb42ad90332 Mon Sep 17 00:00:00 2001
From: Sheng Yang <sheng@linux.intel.com>
Date: Mon, 24 Nov 2008 14:32:51 +0800
Subject: KVM: Replace irq_requested with more generic irq_requested_type

Separate guest irq type and host irq type, for we can support guest using INTx
with host using MSI (but not opposite combination).

Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 include/linux/kvm_host.h | 4 +++-
 virt/kvm/kvm_main.c      | 9 +++++----
 2 files changed, 8 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 3a0fb77d1f6a..c3d4b96a08fa 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -307,7 +307,9 @@ struct kvm_assigned_dev_kernel {
 	int host_devfn;
 	int host_irq;
 	int guest_irq;
-	int irq_requested;
+#define KVM_ASSIGNED_DEV_GUEST_INTX	(1 << 0)
+#define KVM_ASSIGNED_DEV_HOST_INTX	(1 << 8)
+	unsigned long irq_requested_type;
 	int irq_source_id;
 	struct pci_dev *dev;
 	struct kvm *kvm;
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index ef2f03cf42c0..638de47e167d 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -140,7 +140,7 @@ static void kvm_free_assigned_device(struct kvm *kvm,
 				     struct kvm_assigned_dev_kernel
 				     *assigned_dev)
 {
-	if (irqchip_in_kernel(kvm) && assigned_dev->irq_requested)
+	if (irqchip_in_kernel(kvm) && assigned_dev->irq_requested_type)
 		free_irq(assigned_dev->host_irq, (void *)assigned_dev);
 
 	kvm_unregister_irq_ack_notifier(&assigned_dev->ack_notifier);
@@ -180,7 +180,7 @@ static int assigned_device_update_intx(struct kvm *kvm,
 			struct kvm_assigned_dev_kernel *adev,
 			struct kvm_assigned_irq *airq)
 {
-	if (adev->irq_requested) {
+	if (adev->irq_requested_type & KVM_ASSIGNED_DEV_GUEST_INTX) {
 		adev->guest_irq = airq->guest_irq;
 		adev->ack_notifier.gsi = airq->guest_irq;
 		return 0;
@@ -207,7 +207,8 @@ static int assigned_device_update_intx(struct kvm *kvm,
 			return -EIO;
 	}
 
-	adev->irq_requested = true;
+	adev->irq_requested_type = KVM_ASSIGNED_DEV_GUEST_INTX |
+				   KVM_ASSIGNED_DEV_HOST_INTX;
 	return 0;
 }
 
@@ -227,7 +228,7 @@ static int kvm_vm_ioctl_assign_irq(struct kvm *kvm,
 		return -EINVAL;
 	}
 
-	if (!match->irq_requested) {
+	if (!match->irq_requested_type) {
 		INIT_WORK(&match->interrupt_work,
 				kvm_assigned_dev_interrupt_work_handler);
 		if (irqchip_in_kernel(kvm)) {
-- 
cgit v1.2.3


From 0937c48d075ddd59ae2c12a6fa8308b9c7a63753 Mon Sep 17 00:00:00 2001
From: Sheng Yang <sheng@linux.intel.com>
Date: Mon, 24 Nov 2008 14:32:53 +0800
Subject: KVM: Add fields for MSI device assignment

Prepared for kvm_arch_assigned_device_msi_dispatch().

Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 include/linux/kvm.h      | 7 +++++++
 include/linux/kvm_host.h | 4 ++++
 2 files changed, 11 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 44fd7fa0af2b..bb283c388a24 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -507,10 +507,17 @@ struct kvm_assigned_irq {
 	__u32 guest_irq;
 	__u32 flags;
 	union {
+		struct {
+			__u32 addr_lo;
+			__u32 addr_hi;
+			__u32 data;
+		} guest_msi;
 		__u32 reserved[12];
 	};
 };
 
 #define KVM_DEV_ASSIGN_ENABLE_IOMMU	(1 << 0)
 
+#define KVM_DEV_IRQ_ASSIGN_ENABLE_MSI	(1 << 0)
+
 #endif
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index c3d4b96a08fa..8091a4d90ddf 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -16,6 +16,7 @@
 #include <linux/mm.h>
 #include <linux/preempt.h>
 #include <linux/marker.h>
+#include <linux/msi.h>
 #include <asm/signal.h>
 
 #include <linux/kvm.h>
@@ -307,8 +308,11 @@ struct kvm_assigned_dev_kernel {
 	int host_devfn;
 	int host_irq;
 	int guest_irq;
+	struct msi_msg guest_msi;
 #define KVM_ASSIGNED_DEV_GUEST_INTX	(1 << 0)
+#define KVM_ASSIGNED_DEV_GUEST_MSI	(1 << 1)
 #define KVM_ASSIGNED_DEV_HOST_INTX	(1 << 8)
+#define KVM_ASSIGNED_DEV_HOST_MSI	(1 << 9)
 	unsigned long irq_requested_type;
 	int irq_source_id;
 	struct pci_dev *dev;
-- 
cgit v1.2.3


From 6b9cc7fd469869bed38831c5adac3f59dc25eaf5 Mon Sep 17 00:00:00 2001
From: Sheng Yang <sheng@linux.intel.com>
Date: Mon, 24 Nov 2008 14:32:56 +0800
Subject: KVM: Enable MSI for device assignment

We enable guest MSI and host MSI support in this patch. The userspace want to
enable MSI should set KVM_DEV_IRQ_ASSIGN_ENABLE_MSI in the assigned_irq's flag.
Function would return -ENOTTY if can't enable MSI, userspace shouldn't set MSI
Enable bit when KVM_ASSIGN_IRQ return -ENOTTY with
KVM_DEV_IRQ_ASSIGN_ENABLE_MSI.

Userspace can tell the support of MSI device from #ifdef KVM_CAP_DEVICE_MSI.

Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 include/linux/kvm.h |  3 ++
 virt/kvm/kvm_main.c | 81 +++++++++++++++++++++++++++++++++++++++++++++++++----
 2 files changed, 78 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index bb283c388a24..0997e6f5490c 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -392,6 +392,9 @@ struct kvm_trace_rec {
 #endif
 #define KVM_CAP_IOMMU 18
 #define KVM_CAP_NMI 19
+#if defined(CONFIG_X86)
+#define KVM_CAP_DEVICE_MSI 20
+#endif
 
 /*
  * ioctls for VM fds
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 228c1d18a614..bf36ae9ae7df 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -159,9 +159,15 @@ static void kvm_assigned_dev_interrupt_work_handler(struct work_struct *work)
 	 * finer-grained lock, update this
 	 */
 	mutex_lock(&assigned_dev->kvm->lock);
-	kvm_set_irq(assigned_dev->kvm,
-		    assigned_dev->irq_source_id,
-		    assigned_dev->guest_irq, 1);
+	if (assigned_dev->irq_requested_type & KVM_ASSIGNED_DEV_GUEST_INTX)
+		kvm_set_irq(assigned_dev->kvm,
+			    assigned_dev->irq_source_id,
+			    assigned_dev->guest_irq, 1);
+	else if (assigned_dev->irq_requested_type &
+				KVM_ASSIGNED_DEV_GUEST_MSI) {
+		assigned_device_msi_dispatch(assigned_dev);
+		enable_irq(assigned_dev->host_irq);
+	}
 	mutex_unlock(&assigned_dev->kvm->lock);
 	kvm_put_kvm(assigned_dev->kvm);
 }
@@ -197,6 +203,8 @@ static void kvm_free_assigned_device(struct kvm *kvm,
 {
 	if (irqchip_in_kernel(kvm) && assigned_dev->irq_requested_type)
 		free_irq(assigned_dev->host_irq, (void *)assigned_dev);
+	if (assigned_dev->irq_requested_type & KVM_ASSIGNED_DEV_HOST_MSI)
+		pci_disable_msi(assigned_dev->dev);
 
 	kvm_unregister_irq_ack_notifier(&assigned_dev->ack_notifier);
 	kvm_free_irq_source_id(kvm, assigned_dev->irq_source_id);
@@ -242,6 +250,11 @@ static int assigned_device_update_intx(struct kvm *kvm,
 		return 0;
 
 	if (irqchip_in_kernel(kvm)) {
+		if (adev->irq_requested_type & KVM_ASSIGNED_DEV_HOST_MSI) {
+			free_irq(adev->host_irq, (void *)kvm);
+			pci_disable_msi(adev->dev);
+		}
+
 		if (!capable(CAP_SYS_RAWIO))
 			return -EPERM;
 
@@ -265,6 +278,41 @@ static int assigned_device_update_intx(struct kvm *kvm,
 	return 0;
 }
 
+#ifdef CONFIG_X86
+static int assigned_device_update_msi(struct kvm *kvm,
+			struct kvm_assigned_dev_kernel *adev,
+			struct kvm_assigned_irq *airq)
+{
+	int r;
+
+	/* x86 don't care upper address of guest msi message addr */
+	adev->guest_msi.address_lo = airq->guest_msi.addr_lo;
+	adev->guest_msi.data = airq->guest_msi.data;
+	adev->ack_notifier.gsi = -1;
+
+	if (adev->irq_requested_type & KVM_ASSIGNED_DEV_HOST_MSI)
+		return 0;
+
+	if (irqchip_in_kernel(kvm)) {
+		if (adev->irq_requested_type & KVM_ASSIGNED_DEV_HOST_INTX)
+			free_irq(adev->host_irq, (void *)adev);
+
+		r = pci_enable_msi(adev->dev);
+		if (r)
+			return r;
+
+		adev->host_irq = adev->dev->irq;
+		if (request_irq(adev->host_irq, kvm_assigned_dev_intr, 0,
+				"kvm_assigned_msi_device", (void *)adev))
+			return -EIO;
+	}
+
+	adev->irq_requested_type = KVM_ASSIGNED_DEV_GUEST_MSI |
+				   KVM_ASSIGNED_DEV_HOST_MSI;
+	return 0;
+}
+#endif
+
 static int kvm_vm_ioctl_assign_irq(struct kvm *kvm,
 				   struct kvm_assigned_irq
 				   *assigned_irq)
@@ -301,9 +349,30 @@ static int kvm_vm_ioctl_assign_irq(struct kvm *kvm,
 		}
 	}
 
-	r = assigned_device_update_intx(kvm, match, assigned_irq);
-	if (r)
-		goto out_release;
+	if (assigned_irq->flags & KVM_DEV_IRQ_ASSIGN_ENABLE_MSI) {
+#ifdef CONFIG_X86
+		r = assigned_device_update_msi(kvm, match, assigned_irq);
+		if (r) {
+			printk(KERN_WARNING "kvm: failed to enable "
+					"MSI device!\n");
+			goto out_release;
+		}
+#else
+		r = -ENOTTY;
+#endif
+	} else if (assigned_irq->host_irq == 0 && match->dev->irq == 0) {
+		/* Host device IRQ 0 means don't support INTx */
+		printk(KERN_WARNING "kvm: wait device to enable MSI!\n");
+		r = 0;
+	} else {
+		/* Non-sharing INTx mode */
+		r = assigned_device_update_intx(kvm, match, assigned_irq);
+		if (r) {
+			printk(KERN_WARNING "kvm: failed to enable "
+					"INTx device!\n");
+			goto out_release;
+		}
+	}
 
 	mutex_unlock(&kvm->lock);
 	return r;
-- 
cgit v1.2.3


From 1a811b6167089bcdb84284f2dc9fd0b4d0f1899d Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@redhat.com>
Date: Mon, 8 Dec 2008 18:25:27 +0200
Subject: KVM: Advertise the bug in memory region destruction as fixed

Userspace might need to act differently.

Signed-off-by: Avi Kivity <avi@redhat.com>
---
 include/linux/kvm.h |  2 ++
 virt/kvm/kvm_main.c | 13 ++++++++++++-
 2 files changed, 14 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 0997e6f5490c..48807767e726 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -395,6 +395,8 @@ struct kvm_trace_rec {
 #if defined(CONFIG_X86)
 #define KVM_CAP_DEVICE_MSI 20
 #endif
+/* Bug in KVM_SET_USER_MEMORY_REGION fixed: */
+#define KVM_CAP_DESTROY_MEMORY_REGION_WORKS 21
 
 /*
  * ioctls for VM fds
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index e7644b90667e..e066eb125e55 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -1905,6 +1905,17 @@ static int kvm_dev_ioctl_create_vm(void)
 	return fd;
 }
 
+static long kvm_dev_ioctl_check_extension_generic(long arg)
+{
+	switch (arg) {
+	case KVM_CAP_DESTROY_MEMORY_REGION_WORKS:
+		return 1;
+	default:
+		break;
+	}
+	return kvm_dev_ioctl_check_extension(arg);
+}
+
 static long kvm_dev_ioctl(struct file *filp,
 			  unsigned int ioctl, unsigned long arg)
 {
@@ -1924,7 +1935,7 @@ static long kvm_dev_ioctl(struct file *filp,
 		r = kvm_dev_ioctl_create_vm();
 		break;
 	case KVM_CHECK_EXTENSION:
-		r = kvm_dev_ioctl_check_extension(arg);
+		r = kvm_dev_ioctl_check_extension_generic(arg);
 		break;
 	case KVM_GET_VCPU_MMAP_SIZE:
 		r = -EINVAL;
-- 
cgit v1.2.3


From defaf1587c5d7dff828f6f11c8941e5bcef00f50 Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Tue, 2 Dec 2008 12:16:33 +0000
Subject: KVM: fix handling of ACK from shared guest IRQ

If an assigned device shares a guest irq with an emulated
device then we currently interpret an ack generated by the
emulated device as originating from the assigned device
leading to e.g. "Unbalanced enable for IRQ 4347" from the
enable_irq() in kvm_assigned_dev_ack_irq().

The fix is fairly simple - don't enable the physical device
irq unless it was previously disabled.

Of course, this can still lead to a situation where a
non-assigned device ACK can cause the physical device irq to
be reenabled before the device was serviced. However, being
level sensitive, the interrupt will merely be regenerated.

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 include/linux/kvm_host.h |  1 +
 virt/kvm/kvm_main.c      | 15 ++++++++++++++-
 2 files changed, 15 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 8091a4d90ddf..eafabd5c66b2 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -307,6 +307,7 @@ struct kvm_assigned_dev_kernel {
 	int host_busnr;
 	int host_devfn;
 	int host_irq;
+	bool host_irq_disabled;
 	int guest_irq;
 	struct msi_msg guest_msi;
 #define KVM_ASSIGNED_DEV_GUEST_INTX	(1 << 0)
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index eb70ca6c7145..fc6127cbea1f 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -170,6 +170,7 @@ static void kvm_assigned_dev_interrupt_work_handler(struct work_struct *work)
 				KVM_ASSIGNED_DEV_GUEST_MSI) {
 		assigned_device_msi_dispatch(assigned_dev);
 		enable_irq(assigned_dev->host_irq);
+		assigned_dev->host_irq_disabled = false;
 	}
 	mutex_unlock(&assigned_dev->kvm->lock);
 	kvm_put_kvm(assigned_dev->kvm);
@@ -181,8 +182,12 @@ static irqreturn_t kvm_assigned_dev_intr(int irq, void *dev_id)
 		(struct kvm_assigned_dev_kernel *) dev_id;
 
 	kvm_get_kvm(assigned_dev->kvm);
+
 	schedule_work(&assigned_dev->interrupt_work);
+
 	disable_irq_nosync(irq);
+	assigned_dev->host_irq_disabled = true;
+
 	return IRQ_HANDLED;
 }
 
@@ -196,8 +201,16 @@ static void kvm_assigned_dev_ack_irq(struct kvm_irq_ack_notifier *kian)
 
 	dev = container_of(kian, struct kvm_assigned_dev_kernel,
 			   ack_notifier);
+
 	kvm_set_irq(dev->kvm, dev->irq_source_id, dev->guest_irq, 0);
-	enable_irq(dev->host_irq);
+
+	/* The guest irq may be shared so this ack may be
+	 * from another device.
+	 */
+	if (dev->host_irq_disabled) {
+		enable_irq(dev->host_irq);
+		dev->host_irq_disabled = false;
+	}
 }
 
 static void kvm_free_assigned_irq(struct kvm *kvm,
-- 
cgit v1.2.3


From 4531220b71f0399e71cda0c4cf749e7281a7416a Mon Sep 17 00:00:00 2001
From: Jan Kiszka <jan.kiszka@siemens.com>
Date: Thu, 11 Dec 2008 16:54:54 +0100
Subject: KVM: x86: Rework user space NMI injection as KVM_CAP_USER_NMI

There is no point in doing the ready_for_nmi_injection/
request_nmi_window dance with user space. First, we don't do this for
in-kernel irqchip anyway, while the code path is the same as for user
space irqchip mode. And second, there is nothing to loose if a pending
NMI is overwritten by another one (in contrast to IRQs where we have to
save the number). Actually, there is even the risk of raising spurious
NMIs this way because the reason for the held-back NMI might already be
handled while processing the first one.

Therefore this patch creates a simplified user space NMI injection
interface, exporting it under KVM_CAP_USER_NMI and dropping the old
KVM_CAP_NMI capability. And this time we also take care to provide the
interface only on archs supporting NMIs via KVM (right now only x86).

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/x86/kvm/vmx.c  | 24 ++----------------------
 arch/x86/kvm/x86.c  | 28 ++--------------------------
 include/linux/kvm.h | 11 +++++------
 3 files changed, 9 insertions(+), 54 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 487e1dcdce33..6259d7467648 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -2498,15 +2498,13 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu,
 	}
 	if (vcpu->arch.nmi_injected) {
 		vmx_inject_nmi(vcpu);
-		if (vcpu->arch.nmi_pending || kvm_run->request_nmi_window)
+		if (vcpu->arch.nmi_pending)
 			enable_nmi_window(vcpu);
 		else if (vcpu->arch.irq_summary
 			 || kvm_run->request_interrupt_window)
 			enable_irq_window(vcpu);
 		return;
 	}
-	if (!vcpu->arch.nmi_window_open || kvm_run->request_nmi_window)
-		enable_nmi_window(vcpu);
 
 	if (vcpu->arch.interrupt_window_open) {
 		if (vcpu->arch.irq_summary && !vcpu->arch.interrupt.pending)
@@ -3040,14 +3038,6 @@ static int handle_nmi_window(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control);
 	++vcpu->stat.nmi_window_exits;
 
-	/*
-	 * If the user space waits to inject a NMI, exit as soon as possible
-	 */
-	if (kvm_run->request_nmi_window && !vcpu->arch.nmi_pending) {
-		kvm_run->exit_reason = KVM_EXIT_NMI_WINDOW_OPEN;
-		return 0;
-	}
-
 	return 1;
 }
 
@@ -3162,7 +3152,7 @@ static int kvm_handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
 			vmx->soft_vnmi_blocked = 0;
 			vcpu->arch.nmi_window_open = 1;
 		} else if (vmx->vnmi_blocked_time > 1000000000LL &&
-		    (kvm_run->request_nmi_window || vcpu->arch.nmi_pending)) {
+			   vcpu->arch.nmi_pending) {
 			/*
 			 * This CPU don't support us in finding the end of an
 			 * NMI-blocked window if the guest runs with IRQs
@@ -3175,16 +3165,6 @@ static int kvm_handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu)
 			vmx->soft_vnmi_blocked = 0;
 			vmx->vcpu.arch.nmi_window_open = 1;
 		}
-
-		/*
-		 * If the user space waits to inject an NNI, exit ASAP
-		 */
-		if (vcpu->arch.nmi_window_open && kvm_run->request_nmi_window
-		    && !vcpu->arch.nmi_pending) {
-			kvm_run->exit_reason = KVM_EXIT_NMI_WINDOW_OPEN;
-			++vcpu->stat.nmi_window_exits;
-			return 0;
-		}
 	}
 
 	if (exit_reason < kvm_vmx_max_exit_handlers
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 10302d3bd415..0e6aa8141dcd 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2887,37 +2887,18 @@ static int dm_request_for_irq_injection(struct kvm_vcpu *vcpu,
 		(kvm_x86_ops->get_rflags(vcpu) & X86_EFLAGS_IF));
 }
 
-/*
- * Check if userspace requested a NMI window, and that the NMI window
- * is open.
- *
- * No need to exit to userspace if we already have a NMI queued.
- */
-static int dm_request_for_nmi_injection(struct kvm_vcpu *vcpu,
-					struct kvm_run *kvm_run)
-{
-	return (!vcpu->arch.nmi_pending &&
-		kvm_run->request_nmi_window &&
-		vcpu->arch.nmi_window_open);
-}
-
 static void post_kvm_run_save(struct kvm_vcpu *vcpu,
 			      struct kvm_run *kvm_run)
 {
 	kvm_run->if_flag = (kvm_x86_ops->get_rflags(vcpu) & X86_EFLAGS_IF) != 0;
 	kvm_run->cr8 = kvm_get_cr8(vcpu);
 	kvm_run->apic_base = kvm_get_apic_base(vcpu);
-	if (irqchip_in_kernel(vcpu->kvm)) {
+	if (irqchip_in_kernel(vcpu->kvm))
 		kvm_run->ready_for_interrupt_injection = 1;
-		kvm_run->ready_for_nmi_injection = 1;
-	} else {
+	else
 		kvm_run->ready_for_interrupt_injection =
 					(vcpu->arch.interrupt_window_open &&
 					 vcpu->arch.irq_summary == 0);
-		kvm_run->ready_for_nmi_injection =
-					(vcpu->arch.nmi_window_open &&
-					 vcpu->arch.nmi_pending == 0);
-	}
 }
 
 static void vapic_enter(struct kvm_vcpu *vcpu)
@@ -3093,11 +3074,6 @@ static int __vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		}
 
 		if (r > 0) {
-			if (dm_request_for_nmi_injection(vcpu, kvm_run)) {
-				r = -EINTR;
-				kvm_run->exit_reason = KVM_EXIT_NMI;
-				++vcpu->stat.request_nmi_exits;
-			}
 			if (dm_request_for_irq_injection(vcpu, kvm_run)) {
 				r = -EINTR;
 				kvm_run->exit_reason = KVM_EXIT_INTR;
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 48807767e726..35525ac63337 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -84,21 +84,18 @@ struct kvm_irqchip {
 #define KVM_EXIT_S390_RESET       14
 #define KVM_EXIT_DCR              15
 #define KVM_EXIT_NMI              16
-#define KVM_EXIT_NMI_WINDOW_OPEN  17
 
 /* for KVM_RUN, returned by mmap(vcpu_fd, offset=0) */
 struct kvm_run {
 	/* in */
 	__u8 request_interrupt_window;
-	__u8 request_nmi_window;
-	__u8 padding1[6];
+	__u8 padding1[7];
 
 	/* out */
 	__u32 exit_reason;
 	__u8 ready_for_interrupt_injection;
 	__u8 if_flag;
-	__u8 ready_for_nmi_injection;
-	__u8 padding2;
+	__u8 padding2[2];
 
 	/* in (pre_kvm_run), out (post_kvm_run) */
 	__u64 cr8;
@@ -391,12 +388,14 @@ struct kvm_trace_rec {
 #define KVM_CAP_DEVICE_ASSIGNMENT 17
 #endif
 #define KVM_CAP_IOMMU 18
-#define KVM_CAP_NMI 19
 #if defined(CONFIG_X86)
 #define KVM_CAP_DEVICE_MSI 20
 #endif
 /* Bug in KVM_SET_USER_MEMORY_REGION fixed: */
 #define KVM_CAP_DESTROY_MEMORY_REGION_WORKS 21
+#if defined(CONFIG_X86)
+#define KVM_CAP_USER_NMI 22
+#endif
 
 /*
  * ioctls for VM fds
-- 
cgit v1.2.3


From b30f8af3358b5c66be223e3a9f3d11b3d02b4a8f Mon Sep 17 00:00:00 2001
From: Jarkko Lavinen <jarkko.lavinen@nokia.com>
Date: Mon, 17 Nov 2008 14:35:21 +0200
Subject: mmc: Add 8-bit bus width support

Signed-off-by: Jarkko Lavinen <jarkko.lavinen@nokia.com>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/core/mmc.c   | 18 ++++++++++++++----
 include/linux/mmc/host.h |  2 ++
 2 files changed, 16 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c
index fdd7c760be8c..c232d11a7ed4 100644
--- a/drivers/mmc/core/mmc.c
+++ b/drivers/mmc/core/mmc.c
@@ -434,13 +434,24 @@ static int mmc_init_card(struct mmc_host *host, u32 ocr,
 	 * Activate wide bus (if supported).
 	 */
 	if ((card->csd.mmca_vsn >= CSD_SPEC_VER_4) &&
-		(host->caps & MMC_CAP_4_BIT_DATA)) {
+	    (host->caps & (MMC_CAP_4_BIT_DATA | MMC_CAP_8_BIT_DATA))) {
+		unsigned ext_csd_bit, bus_width;
+
+		if (host->caps & MMC_CAP_8_BIT_DATA) {
+			ext_csd_bit = EXT_CSD_BUS_WIDTH_8;
+			bus_width = MMC_BUS_WIDTH_8;
+		} else {
+			ext_csd_bit = EXT_CSD_BUS_WIDTH_4;
+			bus_width = MMC_BUS_WIDTH_4;
+		}
+
 		err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
-			EXT_CSD_BUS_WIDTH, EXT_CSD_BUS_WIDTH_4);
+				 EXT_CSD_BUS_WIDTH, ext_csd_bit);
+
 		if (err)
 			goto free_card;
 
-		mmc_set_bus_width(card->host, MMC_BUS_WIDTH_4);
+		mmc_set_bus_width(card->host, bus_width);
 	}
 
 	if (!oldcard)
@@ -624,4 +635,3 @@ err:
 
 	return err;
 }
-
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index f842f234e44f..4e457256bd33 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -41,6 +41,7 @@ struct mmc_ios {
 
 #define MMC_BUS_WIDTH_1		0
 #define MMC_BUS_WIDTH_4		2
+#define MMC_BUS_WIDTH_8		3
 
 	unsigned char	timing;			/* timing specification used */
 
@@ -116,6 +117,7 @@ struct mmc_host {
 #define MMC_CAP_SDIO_IRQ	(1 << 3)	/* Can signal pending SDIO IRQs */
 #define MMC_CAP_SPI		(1 << 4)	/* Talks only SPI protocols */
 #define MMC_CAP_NEEDS_POLL	(1 << 5)	/* Needs polling for card-detection */
+#define MMC_CAP_8_BIT_DATA	(1 << 6)	/* Can the host do 8 bit transfers */
 
 	/* host specific block data */
 	unsigned int		max_seg_size;	/* see blk_queue_max_segment_size */
-- 
cgit v1.2.3


From 86e8286a0e48663e1e86a5884b30a6d05de2993a Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <avorontsov@ru.mvista.com>
Date: Wed, 26 Nov 2008 22:54:17 +0300
Subject: mmc: Add mmc_vddrange_to_ocrmask() helper function

This function sets the OCR mask bits according to provided voltage
ranges. Will be used by the mmc_spi OpenFirmware bindings.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/core/core.c  | 75 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/mmc/core.h |  2 ++
 2 files changed, 77 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index f7284b905eb3..5f288aeeb721 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -20,6 +20,7 @@
 #include <linux/err.h>
 #include <linux/leds.h>
 #include <linux/scatterlist.h>
+#include <linux/log2.h>
 
 #include <linux/mmc/card.h>
 #include <linux/mmc/host.h>
@@ -448,6 +449,80 @@ void mmc_set_bus_width(struct mmc_host *host, unsigned int width)
 	mmc_set_ios(host);
 }
 
+/**
+ * mmc_vdd_to_ocrbitnum - Convert a voltage to the OCR bit number
+ * @vdd:	voltage (mV)
+ * @low_bits:	prefer low bits in boundary cases
+ *
+ * This function returns the OCR bit number according to the provided @vdd
+ * value. If conversion is not possible a negative errno value returned.
+ *
+ * Depending on the @low_bits flag the function prefers low or high OCR bits
+ * on boundary voltages. For example,
+ * with @low_bits = true, 3300 mV translates to ilog2(MMC_VDD_32_33);
+ * with @low_bits = false, 3300 mV translates to ilog2(MMC_VDD_33_34);
+ *
+ * Any value in the [1951:1999] range translates to the ilog2(MMC_VDD_20_21).
+ */
+static int mmc_vdd_to_ocrbitnum(int vdd, bool low_bits)
+{
+	const int max_bit = ilog2(MMC_VDD_35_36);
+	int bit;
+
+	if (vdd < 1650 || vdd > 3600)
+		return -EINVAL;
+
+	if (vdd >= 1650 && vdd <= 1950)
+		return ilog2(MMC_VDD_165_195);
+
+	if (low_bits)
+		vdd -= 1;
+
+	/* Base 2000 mV, step 100 mV, bit's base 8. */
+	bit = (vdd - 2000) / 100 + 8;
+	if (bit > max_bit)
+		return max_bit;
+	return bit;
+}
+
+/**
+ * mmc_vddrange_to_ocrmask - Convert a voltage range to the OCR mask
+ * @vdd_min:	minimum voltage value (mV)
+ * @vdd_max:	maximum voltage value (mV)
+ *
+ * This function returns the OCR mask bits according to the provided @vdd_min
+ * and @vdd_max values. If conversion is not possible the function returns 0.
+ *
+ * Notes wrt boundary cases:
+ * This function sets the OCR bits for all boundary voltages, for example
+ * [3300:3400] range is translated to MMC_VDD_32_33 | MMC_VDD_33_34 |
+ * MMC_VDD_34_35 mask.
+ */
+u32 mmc_vddrange_to_ocrmask(int vdd_min, int vdd_max)
+{
+	u32 mask = 0;
+
+	if (vdd_max < vdd_min)
+		return 0;
+
+	/* Prefer high bits for the boundary vdd_max values. */
+	vdd_max = mmc_vdd_to_ocrbitnum(vdd_max, false);
+	if (vdd_max < 0)
+		return 0;
+
+	/* Prefer low bits for the boundary vdd_min values. */
+	vdd_min = mmc_vdd_to_ocrbitnum(vdd_min, true);
+	if (vdd_min < 0)
+		return 0;
+
+	/* Fill the mask, from max bit to min bit. */
+	while (vdd_max >= vdd_min)
+		mask |= 1 << vdd_max--;
+
+	return mask;
+}
+EXPORT_SYMBOL(mmc_vddrange_to_ocrmask);
+
 /*
  * Mask off any voltages we don't support and select
  * the lowest voltage
diff --git a/include/linux/mmc/core.h b/include/linux/mmc/core.h
index 143cebf0586f..7ac8b500d55c 100644
--- a/include/linux/mmc/core.h
+++ b/include/linux/mmc/core.h
@@ -151,4 +151,6 @@ static inline void mmc_claim_host(struct mmc_host *host)
 	__mmc_claim_host(host, NULL);
 }
 
+extern u32 mmc_vddrange_to_ocrmask(int vdd_min, int vdd_max);
+
 #endif
-- 
cgit v1.2.3


From 9c43df57910bbba540a6cb5c9132302a9ea5f41a Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <avorontsov@ru.mvista.com>
Date: Tue, 30 Dec 2008 18:15:28 +0300
Subject: mmc_spi: Add support for OpenFirmware bindings

The support is implemented via platform data accessors, new module
(of_mmc_spi) will be created automatically when the driver compiles
on OpenFirmware platforms. Link-time dependency will load the module
automatically.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/host/Makefile     |   3 +
 drivers/mmc/host/mmc_spi.c    |   4 +-
 drivers/mmc/host/of_mmc_spi.c | 149 ++++++++++++++++++++++++++++++++++++++++++
 include/linux/spi/mmc_spi.h   |  15 ++++-
 4 files changed, 169 insertions(+), 2 deletions(-)
 create mode 100644 drivers/mmc/host/of_mmc_spi.c

(limited to 'include/linux')

diff --git a/drivers/mmc/host/Makefile b/drivers/mmc/host/Makefile
index c794cc5ce442..f4853288bbb1 100644
--- a/drivers/mmc/host/Makefile
+++ b/drivers/mmc/host/Makefile
@@ -19,6 +19,9 @@ obj-$(CONFIG_MMC_AT91)		+= at91_mci.o
 obj-$(CONFIG_MMC_ATMELMCI)	+= atmel-mci.o
 obj-$(CONFIG_MMC_TIFM_SD)	+= tifm_sd.o
 obj-$(CONFIG_MMC_SPI)		+= mmc_spi.o
+ifeq ($(CONFIG_OF),y)
+obj-$(CONFIG_MMC_SPI)		+= of_mmc_spi.o
+endif
 obj-$(CONFIG_MMC_S3C)   	+= s3cmci.o
 obj-$(CONFIG_MMC_SDRICOH_CS)	+= sdricoh_cs.o
 obj-$(CONFIG_MMC_TMIO)		+= tmio_mmc.o
diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
index ad00e1632317..87e211df68ac 100644
--- a/drivers/mmc/host/mmc_spi.c
+++ b/drivers/mmc/host/mmc_spi.c
@@ -1285,7 +1285,7 @@ static int mmc_spi_probe(struct spi_device *spi)
 	/* Platform data is used to hook up things like card sensing
 	 * and power switching gpios.
 	 */
-	host->pdata = spi->dev.platform_data;
+	host->pdata = mmc_spi_get_pdata(spi);
 	if (host->pdata)
 		mmc->ocr_avail = host->pdata->ocr_mask;
 	if (!mmc->ocr_avail) {
@@ -1368,6 +1368,7 @@ fail_glue_init:
 
 fail_nobuf1:
 	mmc_free_host(mmc);
+	mmc_spi_put_pdata(spi);
 	dev_set_drvdata(&spi->dev, NULL);
 
 nomem:
@@ -1402,6 +1403,7 @@ static int __devexit mmc_spi_remove(struct spi_device *spi)
 
 		spi->max_speed_hz = mmc->f_max;
 		mmc_free_host(mmc);
+		mmc_spi_put_pdata(spi);
 		dev_set_drvdata(&spi->dev, NULL);
 	}
 	return 0;
diff --git a/drivers/mmc/host/of_mmc_spi.c b/drivers/mmc/host/of_mmc_spi.c
new file mode 100644
index 000000000000..fb2921f8099d
--- /dev/null
+++ b/drivers/mmc/host/of_mmc_spi.c
@@ -0,0 +1,149 @@
+/*
+ * OpenFirmware bindings for the MMC-over-SPI driver
+ *
+ * Copyright (c) MontaVista Software, Inc. 2008.
+ *
+ * Author: Anton Vorontsov <avorontsov@ru.mvista.com>
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/gpio.h>
+#include <linux/of.h>
+#include <linux/of_gpio.h>
+#include <linux/spi/spi.h>
+#include <linux/spi/mmc_spi.h>
+#include <linux/mmc/core.h>
+#include <linux/mmc/host.h>
+
+enum {
+	CD_GPIO = 0,
+	WP_GPIO,
+	NUM_GPIOS,
+};
+
+struct of_mmc_spi {
+	int gpios[NUM_GPIOS];
+	bool alow_gpios[NUM_GPIOS];
+	struct mmc_spi_platform_data pdata;
+};
+
+static struct of_mmc_spi *to_of_mmc_spi(struct device *dev)
+{
+	return container_of(dev->platform_data, struct of_mmc_spi, pdata);
+}
+
+static int of_mmc_spi_read_gpio(struct device *dev, int gpio_num)
+{
+	struct of_mmc_spi *oms = to_of_mmc_spi(dev);
+	bool active_low = oms->alow_gpios[gpio_num];
+	bool value = gpio_get_value(oms->gpios[gpio_num]);
+
+	return active_low ^ value;
+}
+
+static int of_mmc_spi_get_cd(struct device *dev)
+{
+	return of_mmc_spi_read_gpio(dev, CD_GPIO);
+}
+
+static int of_mmc_spi_get_ro(struct device *dev)
+{
+	return of_mmc_spi_read_gpio(dev, WP_GPIO);
+}
+
+struct mmc_spi_platform_data *mmc_spi_get_pdata(struct spi_device *spi)
+{
+	struct device *dev = &spi->dev;
+	struct device_node *np = dev_archdata_get_node(&dev->archdata);
+	struct of_mmc_spi *oms;
+	const u32 *voltage_ranges;
+	int num_ranges;
+	int i;
+	int ret = -EINVAL;
+
+	if (dev->platform_data || !np)
+		return dev->platform_data;
+
+	oms = kzalloc(sizeof(*oms), GFP_KERNEL);
+	if (!oms)
+		return NULL;
+
+	voltage_ranges = of_get_property(np, "voltage-ranges", &num_ranges);
+	num_ranges = num_ranges / sizeof(*voltage_ranges) / 2;
+	if (!voltage_ranges || !num_ranges) {
+		dev_err(dev, "OF: voltage-ranges unspecified\n");
+		goto err_ocr;
+	}
+
+	for (i = 0; i < num_ranges; i++) {
+		const int j = i * 2;
+		u32 mask;
+
+		mask = mmc_vddrange_to_ocrmask(voltage_ranges[j],
+					       voltage_ranges[j + 1]);
+		if (!mask) {
+			ret = -EINVAL;
+			dev_err(dev, "OF: voltage-range #%d is invalid\n", i);
+			goto err_ocr;
+		}
+		oms->pdata.ocr_mask |= mask;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(oms->gpios); i++) {
+		enum of_gpio_flags gpio_flags;
+
+		oms->gpios[i] = of_get_gpio_flags(np, i, &gpio_flags);
+		if (!gpio_is_valid(oms->gpios[i]))
+			continue;
+
+		ret = gpio_request(oms->gpios[i], dev->bus_id);
+		if (ret < 0) {
+			oms->gpios[i] = -EINVAL;
+			continue;
+		}
+
+		if (gpio_flags & OF_GPIO_ACTIVE_LOW)
+			oms->alow_gpios[i] = true;
+	}
+
+	if (gpio_is_valid(oms->gpios[CD_GPIO]))
+		oms->pdata.get_cd = of_mmc_spi_get_cd;
+	if (gpio_is_valid(oms->gpios[WP_GPIO]))
+		oms->pdata.get_ro = of_mmc_spi_get_ro;
+
+	/* We don't support interrupts yet, let's poll. */
+	oms->pdata.caps |= MMC_CAP_NEEDS_POLL;
+
+	dev->platform_data = &oms->pdata;
+	return dev->platform_data;
+err_ocr:
+	kfree(oms);
+	return NULL;
+}
+EXPORT_SYMBOL(mmc_spi_get_pdata);
+
+void mmc_spi_put_pdata(struct spi_device *spi)
+{
+	struct device *dev = &spi->dev;
+	struct device_node *np = dev_archdata_get_node(&dev->archdata);
+	struct of_mmc_spi *oms = to_of_mmc_spi(dev);
+	int i;
+
+	if (!dev->platform_data || !np)
+		return;
+
+	for (i = 0; i < ARRAY_SIZE(oms->gpios); i++) {
+		if (gpio_is_valid(oms->gpios[i]))
+			gpio_free(oms->gpios[i]);
+	}
+	kfree(oms);
+	dev->platform_data = NULL;
+}
+EXPORT_SYMBOL(mmc_spi_put_pdata);
diff --git a/include/linux/spi/mmc_spi.h b/include/linux/spi/mmc_spi.h
index a3626aedaec9..0f4eb165f254 100644
--- a/include/linux/spi/mmc_spi.h
+++ b/include/linux/spi/mmc_spi.h
@@ -1,9 +1,10 @@
 #ifndef __LINUX_SPI_MMC_SPI_H
 #define __LINUX_SPI_MMC_SPI_H
 
+#include <linux/device.h>
+#include <linux/spi/spi.h>
 #include <linux/interrupt.h>
 
-struct device;
 struct mmc_host;
 
 /* Put this in platform_data of a device being used to manage an MMC/SD
@@ -41,4 +42,16 @@ struct mmc_spi_platform_data {
 	void (*setpower)(struct device *, unsigned int maskval);
 };
 
+#ifdef CONFIG_OF
+extern struct mmc_spi_platform_data *mmc_spi_get_pdata(struct spi_device *spi);
+extern void mmc_spi_put_pdata(struct spi_device *spi);
+#else
+static inline struct mmc_spi_platform_data *
+mmc_spi_get_pdata(struct spi_device *spi)
+{
+	return spi->dev.platform_data;
+}
+static inline void mmc_spi_put_pdata(struct spi_device *spi) {}
+#endif /* CONFIG_OF */
+
 #endif /* __LINUX_SPI_MMC_SPI_H */
-- 
cgit v1.2.3


From be6d3e56a6b9b3a4ee44a0685e39e595073c6f0d Mon Sep 17 00:00:00 2001
From: Kentaro Takeda <takedakn@nttdata.co.jp>
Date: Wed, 17 Dec 2008 13:24:15 +0900
Subject: introduce new LSM hooks where vfsmount is available.

Add new LSM hooks for path-based checks.  Call them on directory-modifying
operations at the points where we still know the vfsmount involved.

Signed-off-by: Kentaro Takeda <takedakn@nttdata.co.jp>
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Toshiharu Harada <haradats@nttdata.co.jp>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c               |  36 +++++++++++++
 fs/open.c                |   5 ++
 include/linux/security.h | 137 +++++++++++++++++++++++++++++++++++++++++++++++
 net/unix/af_unix.c       |   4 ++
 security/Kconfig         |   9 ++++
 security/capability.c    |  57 ++++++++++++++++++++
 security/security.c      |  66 +++++++++++++++++++++++
 7 files changed, 314 insertions(+)

(limited to 'include/linux')

diff --git a/fs/namei.c b/fs/namei.c
index af3783fff1de..ab441af4196b 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1556,6 +1556,9 @@ int may_open(struct nameidata *nd, int acc_mode, int flag)
 		 * Refuse to truncate files with mandatory locks held on them.
 		 */
 		error = locks_verify_locked(inode);
+		if (!error)
+			error = security_path_truncate(&nd->path, 0,
+					       ATTR_MTIME|ATTR_CTIME|ATTR_OPEN);
 		if (!error) {
 			DQUOT_INIT(inode);
 
@@ -1586,7 +1589,11 @@ static int __open_namei_create(struct nameidata *nd, struct path *path,
 
 	if (!IS_POSIXACL(dir->d_inode))
 		mode &= ~current->fs->umask;
+	error = security_path_mknod(&nd->path, path->dentry, mode, 0);
+	if (error)
+		goto out_unlock;
 	error = vfs_create(dir->d_inode, path->dentry, mode, nd);
+out_unlock:
 	mutex_unlock(&dir->d_inode->i_mutex);
 	dput(nd->path.dentry);
 	nd->path.dentry = path->dentry;
@@ -1999,6 +2006,9 @@ asmlinkage long sys_mknodat(int dfd, const char __user *filename, int mode,
 	error = mnt_want_write(nd.path.mnt);
 	if (error)
 		goto out_dput;
+	error = security_path_mknod(&nd.path, dentry, mode, dev);
+	if (error)
+		goto out_drop_write;
 	switch (mode & S_IFMT) {
 		case 0: case S_IFREG:
 			error = vfs_create(nd.path.dentry->d_inode,dentry,mode,&nd);
@@ -2011,6 +2021,7 @@ asmlinkage long sys_mknodat(int dfd, const char __user *filename, int mode,
 			error = vfs_mknod(nd.path.dentry->d_inode,dentry,mode,0);
 			break;
 	}
+out_drop_write:
 	mnt_drop_write(nd.path.mnt);
 out_dput:
 	dput(dentry);
@@ -2070,7 +2081,11 @@ asmlinkage long sys_mkdirat(int dfd, const char __user *pathname, int mode)
 	error = mnt_want_write(nd.path.mnt);
 	if (error)
 		goto out_dput;
+	error = security_path_mkdir(&nd.path, dentry, mode);
+	if (error)
+		goto out_drop_write;
 	error = vfs_mkdir(nd.path.dentry->d_inode, dentry, mode);
+out_drop_write:
 	mnt_drop_write(nd.path.mnt);
 out_dput:
 	dput(dentry);
@@ -2180,7 +2195,11 @@ static long do_rmdir(int dfd, const char __user *pathname)
 	error = mnt_want_write(nd.path.mnt);
 	if (error)
 		goto exit3;
+	error = security_path_rmdir(&nd.path, dentry);
+	if (error)
+		goto exit4;
 	error = vfs_rmdir(nd.path.dentry->d_inode, dentry);
+exit4:
 	mnt_drop_write(nd.path.mnt);
 exit3:
 	dput(dentry);
@@ -2265,7 +2284,11 @@ static long do_unlinkat(int dfd, const char __user *pathname)
 		error = mnt_want_write(nd.path.mnt);
 		if (error)
 			goto exit2;
+		error = security_path_unlink(&nd.path, dentry);
+		if (error)
+			goto exit3;
 		error = vfs_unlink(nd.path.dentry->d_inode, dentry);
+exit3:
 		mnt_drop_write(nd.path.mnt);
 	exit2:
 		dput(dentry);
@@ -2346,7 +2369,11 @@ asmlinkage long sys_symlinkat(const char __user *oldname,
 	error = mnt_want_write(nd.path.mnt);
 	if (error)
 		goto out_dput;
+	error = security_path_symlink(&nd.path, dentry, from);
+	if (error)
+		goto out_drop_write;
 	error = vfs_symlink(nd.path.dentry->d_inode, dentry, from);
+out_drop_write:
 	mnt_drop_write(nd.path.mnt);
 out_dput:
 	dput(dentry);
@@ -2443,7 +2470,11 @@ asmlinkage long sys_linkat(int olddfd, const char __user *oldname,
 	error = mnt_want_write(nd.path.mnt);
 	if (error)
 		goto out_dput;
+	error = security_path_link(old_path.dentry, &nd.path, new_dentry);
+	if (error)
+		goto out_drop_write;
 	error = vfs_link(old_path.dentry, nd.path.dentry->d_inode, new_dentry);
+out_drop_write:
 	mnt_drop_write(nd.path.mnt);
 out_dput:
 	dput(new_dentry);
@@ -2679,8 +2710,13 @@ asmlinkage long sys_renameat(int olddfd, const char __user *oldname,
 	error = mnt_want_write(oldnd.path.mnt);
 	if (error)
 		goto exit5;
+	error = security_path_rename(&oldnd.path, old_dentry,
+				     &newnd.path, new_dentry);
+	if (error)
+		goto exit6;
 	error = vfs_rename(old_dir->d_inode, old_dentry,
 				   new_dir->d_inode, new_dentry);
+exit6:
 	mnt_drop_write(oldnd.path.mnt);
 exit5:
 	dput(new_dentry);
diff --git a/fs/open.c b/fs/open.c
index c0a426d5766c..1cd7d40e9991 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -272,6 +272,8 @@ static long do_sys_truncate(const char __user *pathname, loff_t length)
 		goto put_write_and_out;
 
 	error = locks_verify_truncate(inode, NULL, length);
+	if (!error)
+		error = security_path_truncate(&path, length, 0);
 	if (!error) {
 		DQUOT_INIT(inode);
 		error = do_truncate(path.dentry, length, 0, NULL);
@@ -328,6 +330,9 @@ static long do_sys_ftruncate(unsigned int fd, loff_t length, int small)
 		goto out_putf;
 
 	error = locks_verify_truncate(inode, file, length);
+	if (!error)
+		error = security_path_truncate(&file->f_path, length,
+					       ATTR_MTIME|ATTR_CTIME);
 	if (!error)
 		error = do_truncate(dentry, length, ATTR_MTIME|ATTR_CTIME, file);
 out_putf:
diff --git a/include/linux/security.h b/include/linux/security.h
index 3416cb85e77b..b92b5e453f64 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -335,17 +335,37 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	@dir contains the inode structure of the parent directory of the new link.
  *	@new_dentry contains the dentry structure for the new link.
  *	Return 0 if permission is granted.
+ * @path_link:
+ *	Check permission before creating a new hard link to a file.
+ *	@old_dentry contains the dentry structure for an existing link
+ *	to the file.
+ *	@new_dir contains the path structure of the parent directory of
+ *	the new link.
+ *	@new_dentry contains the dentry structure for the new link.
+ *	Return 0 if permission is granted.
  * @inode_unlink:
  *	Check the permission to remove a hard link to a file.
  *	@dir contains the inode structure of parent directory of the file.
  *	@dentry contains the dentry structure for file to be unlinked.
  *	Return 0 if permission is granted.
+ * @path_unlink:
+ *	Check the permission to remove a hard link to a file.
+ *	@dir contains the path structure of parent directory of the file.
+ *	@dentry contains the dentry structure for file to be unlinked.
+ *	Return 0 if permission is granted.
  * @inode_symlink:
  *	Check the permission to create a symbolic link to a file.
  *	@dir contains the inode structure of parent directory of the symbolic link.
  *	@dentry contains the dentry structure of the symbolic link.
  *	@old_name contains the pathname of file.
  *	Return 0 if permission is granted.
+ * @path_symlink:
+ *	Check the permission to create a symbolic link to a file.
+ *	@dir contains the path structure of parent directory of
+ *	the symbolic link.
+ *	@dentry contains the dentry structure of the symbolic link.
+ *	@old_name contains the pathname of file.
+ *	Return 0 if permission is granted.
  * @inode_mkdir:
  *	Check permissions to create a new directory in the existing directory
  *	associated with inode strcture @dir.
@@ -353,11 +373,25 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	@dentry contains the dentry structure of new directory.
  *	@mode contains the mode of new directory.
  *	Return 0 if permission is granted.
+ * @path_mkdir:
+ *	Check permissions to create a new directory in the existing directory
+ *	associated with path strcture @path.
+ *	@dir containst the path structure of parent of the directory
+ *	to be created.
+ *	@dentry contains the dentry structure of new directory.
+ *	@mode contains the mode of new directory.
+ *	Return 0 if permission is granted.
  * @inode_rmdir:
  *	Check the permission to remove a directory.
  *	@dir contains the inode structure of parent of the directory to be removed.
  *	@dentry contains the dentry structure of directory to be removed.
  *	Return 0 if permission is granted.
+ * @path_rmdir:
+ *	Check the permission to remove a directory.
+ *	@dir contains the path structure of parent of the directory to be
+ *	removed.
+ *	@dentry contains the dentry structure of directory to be removed.
+ *	Return 0 if permission is granted.
  * @inode_mknod:
  *	Check permissions when creating a special file (or a socket or a fifo
  *	file created via the mknod system call).  Note that if mknod operation
@@ -368,6 +402,15 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	@mode contains the mode of the new file.
  *	@dev contains the device number.
  *	Return 0 if permission is granted.
+ * @path_mknod:
+ *	Check permissions when creating a file. Note that this hook is called
+ *	even if mknod operation is being done for a regular file.
+ *	@dir contains the path structure of parent of the new file.
+ *	@dentry contains the dentry structure of the new file.
+ *	@mode contains the mode of the new file.
+ *	@dev contains the undecoded device number. Use new_decode_dev() to get
+ *	the decoded device number.
+ *	Return 0 if permission is granted.
  * @inode_rename:
  *	Check for permission to rename a file or directory.
  *	@old_dir contains the inode structure for parent of the old link.
@@ -375,6 +418,13 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	@new_dir contains the inode structure for parent of the new link.
  *	@new_dentry contains the dentry structure of the new link.
  *	Return 0 if permission is granted.
+ * @path_rename:
+ *	Check for permission to rename a file or directory.
+ *	@old_dir contains the path structure for parent of the old link.
+ *	@old_dentry contains the dentry structure of the old link.
+ *	@new_dir contains the path structure for parent of the new link.
+ *	@new_dentry contains the dentry structure of the new link.
+ *	Return 0 if permission is granted.
  * @inode_readlink:
  *	Check the permission to read the symbolic link.
  *	@dentry contains the dentry structure for the file link.
@@ -403,6 +453,12 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts)
  *	@dentry contains the dentry structure for the file.
  *	@attr is the iattr structure containing the new file attributes.
  *	Return 0 if permission is granted.
+ * @path_truncate:
+ *	Check permission before truncating a file.
+ *	@path contains the path structure for the file.
+ *	@length is the new length of the file.
+ *	@time_attrs is the flags passed to do_truncate().
+ *	Return 0 if permission is granted.
  * @inode_getattr:
  *	Check permission before obtaining file attributes.
  *	@mnt is the vfsmount where the dentry was looked up
@@ -1331,6 +1387,22 @@ struct security_operations {
 				   struct super_block *newsb);
 	int (*sb_parse_opts_str) (char *options, struct security_mnt_opts *opts);
 
+#ifdef CONFIG_SECURITY_PATH
+	int (*path_unlink) (struct path *dir, struct dentry *dentry);
+	int (*path_mkdir) (struct path *dir, struct dentry *dentry, int mode);
+	int (*path_rmdir) (struct path *dir, struct dentry *dentry);
+	int (*path_mknod) (struct path *dir, struct dentry *dentry, int mode,
+			   unsigned int dev);
+	int (*path_truncate) (struct path *path, loff_t length,
+			      unsigned int time_attrs);
+	int (*path_symlink) (struct path *dir, struct dentry *dentry,
+			     const char *old_name);
+	int (*path_link) (struct dentry *old_dentry, struct path *new_dir,
+			  struct dentry *new_dentry);
+	int (*path_rename) (struct path *old_dir, struct dentry *old_dentry,
+			    struct path *new_dir, struct dentry *new_dentry);
+#endif
+
 	int (*inode_alloc_security) (struct inode *inode);
 	void (*inode_free_security) (struct inode *inode);
 	int (*inode_init_security) (struct inode *inode, struct inode *dir,
@@ -2705,6 +2777,71 @@ static inline void security_skb_classify_flow(struct sk_buff *skb, struct flowi
 
 #endif	/* CONFIG_SECURITY_NETWORK_XFRM */
 
+#ifdef CONFIG_SECURITY_PATH
+int security_path_unlink(struct path *dir, struct dentry *dentry);
+int security_path_mkdir(struct path *dir, struct dentry *dentry, int mode);
+int security_path_rmdir(struct path *dir, struct dentry *dentry);
+int security_path_mknod(struct path *dir, struct dentry *dentry, int mode,
+			unsigned int dev);
+int security_path_truncate(struct path *path, loff_t length,
+			   unsigned int time_attrs);
+int security_path_symlink(struct path *dir, struct dentry *dentry,
+			  const char *old_name);
+int security_path_link(struct dentry *old_dentry, struct path *new_dir,
+		       struct dentry *new_dentry);
+int security_path_rename(struct path *old_dir, struct dentry *old_dentry,
+			 struct path *new_dir, struct dentry *new_dentry);
+#else	/* CONFIG_SECURITY_PATH */
+static inline int security_path_unlink(struct path *dir, struct dentry *dentry)
+{
+	return 0;
+}
+
+static inline int security_path_mkdir(struct path *dir, struct dentry *dentry,
+				      int mode)
+{
+	return 0;
+}
+
+static inline int security_path_rmdir(struct path *dir, struct dentry *dentry)
+{
+	return 0;
+}
+
+static inline int security_path_mknod(struct path *dir, struct dentry *dentry,
+				      int mode, unsigned int dev)
+{
+	return 0;
+}
+
+static inline int security_path_truncate(struct path *path, loff_t length,
+					 unsigned int time_attrs)
+{
+	return 0;
+}
+
+static inline int security_path_symlink(struct path *dir, struct dentry *dentry,
+					const char *old_name)
+{
+	return 0;
+}
+
+static inline int security_path_link(struct dentry *old_dentry,
+				     struct path *new_dir,
+				     struct dentry *new_dentry)
+{
+	return 0;
+}
+
+static inline int security_path_rename(struct path *old_dir,
+				       struct dentry *old_dentry,
+				       struct path *new_dir,
+				       struct dentry *new_dentry)
+{
+	return 0;
+}
+#endif	/* CONFIG_SECURITY_PATH */
+
 #ifdef CONFIG_KEYS
 #ifdef CONFIG_SECURITY
 
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index c6250d0055d2..d1b89820ab4f 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -836,7 +836,11 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 		err = mnt_want_write(nd.path.mnt);
 		if (err)
 			goto out_mknod_dput;
+		err = security_path_mknod(&nd.path, dentry, mode, 0);
+		if (err)
+			goto out_mknod_drop_write;
 		err = vfs_mknod(nd.path.dentry->d_inode, dentry, mode, 0);
+out_mknod_drop_write:
 		mnt_drop_write(nd.path.mnt);
 		if (err)
 			goto out_mknod_dput;
diff --git a/security/Kconfig b/security/Kconfig
index d9f47ce7e207..9438535d7fd0 100644
--- a/security/Kconfig
+++ b/security/Kconfig
@@ -81,6 +81,15 @@ config SECURITY_NETWORK_XFRM
 	  IPSec.
 	  If you are unsure how to answer this question, answer N.
 
+config SECURITY_PATH
+	bool "Security hooks for pathname based access control"
+	depends on SECURITY
+	help
+	  This enables the security hooks for pathname based access control.
+	  If enabled, a security module can use these hooks to
+	  implement pathname based access controls.
+	  If you are unsure how to answer this question, answer N.
+
 config SECURITY_FILE_CAPABILITIES
 	bool "File POSIX Capabilities"
 	default n
diff --git a/security/capability.c b/security/capability.c
index 2dce66fcb992..c545bd1300b5 100644
--- a/security/capability.c
+++ b/security/capability.c
@@ -263,6 +263,53 @@ static void cap_inode_getsecid(const struct inode *inode, u32 *secid)
 	*secid = 0;
 }
 
+#ifdef CONFIG_SECURITY_PATH
+static int cap_path_mknod(struct path *dir, struct dentry *dentry, int mode,
+			  unsigned int dev)
+{
+	return 0;
+}
+
+static int cap_path_mkdir(struct path *dir, struct dentry *dentry, int mode)
+{
+	return 0;
+}
+
+static int cap_path_rmdir(struct path *dir, struct dentry *dentry)
+{
+	return 0;
+}
+
+static int cap_path_unlink(struct path *dir, struct dentry *dentry)
+{
+	return 0;
+}
+
+static int cap_path_symlink(struct path *dir, struct dentry *dentry,
+			    const char *old_name)
+{
+	return 0;
+}
+
+static int cap_path_link(struct dentry *old_dentry, struct path *new_dir,
+			 struct dentry *new_dentry)
+{
+	return 0;
+}
+
+static int cap_path_rename(struct path *old_path, struct dentry *old_dentry,
+			   struct path *new_path, struct dentry *new_dentry)
+{
+	return 0;
+}
+
+static int cap_path_truncate(struct path *path, loff_t length,
+			     unsigned int time_attrs)
+{
+	return 0;
+}
+#endif
+
 static int cap_file_permission(struct file *file, int mask)
 {
 	return 0;
@@ -883,6 +930,16 @@ void security_fixup_ops(struct security_operations *ops)
 	set_to_cap_if_null(ops, inode_setsecurity);
 	set_to_cap_if_null(ops, inode_listsecurity);
 	set_to_cap_if_null(ops, inode_getsecid);
+#ifdef CONFIG_SECURITY_PATH
+	set_to_cap_if_null(ops, path_mknod);
+	set_to_cap_if_null(ops, path_mkdir);
+	set_to_cap_if_null(ops, path_rmdir);
+	set_to_cap_if_null(ops, path_unlink);
+	set_to_cap_if_null(ops, path_symlink);
+	set_to_cap_if_null(ops, path_link);
+	set_to_cap_if_null(ops, path_rename);
+	set_to_cap_if_null(ops, path_truncate);
+#endif
 	set_to_cap_if_null(ops, file_permission);
 	set_to_cap_if_null(ops, file_alloc_security);
 	set_to_cap_if_null(ops, file_free_security);
diff --git a/security/security.c b/security/security.c
index d85dbb37c972..678d4d07b852 100644
--- a/security/security.c
+++ b/security/security.c
@@ -355,6 +355,72 @@ int security_inode_init_security(struct inode *inode, struct inode *dir,
 }
 EXPORT_SYMBOL(security_inode_init_security);
 
+#ifdef CONFIG_SECURITY_PATH
+int security_path_mknod(struct path *path, struct dentry *dentry, int mode,
+			unsigned int dev)
+{
+	if (unlikely(IS_PRIVATE(path->dentry->d_inode)))
+		return 0;
+	return security_ops->path_mknod(path, dentry, mode, dev);
+}
+EXPORT_SYMBOL(security_path_mknod);
+
+int security_path_mkdir(struct path *path, struct dentry *dentry, int mode)
+{
+	if (unlikely(IS_PRIVATE(path->dentry->d_inode)))
+		return 0;
+	return security_ops->path_mkdir(path, dentry, mode);
+}
+
+int security_path_rmdir(struct path *path, struct dentry *dentry)
+{
+	if (unlikely(IS_PRIVATE(path->dentry->d_inode)))
+		return 0;
+	return security_ops->path_rmdir(path, dentry);
+}
+
+int security_path_unlink(struct path *path, struct dentry *dentry)
+{
+	if (unlikely(IS_PRIVATE(path->dentry->d_inode)))
+		return 0;
+	return security_ops->path_unlink(path, dentry);
+}
+
+int security_path_symlink(struct path *path, struct dentry *dentry,
+			  const char *old_name)
+{
+	if (unlikely(IS_PRIVATE(path->dentry->d_inode)))
+		return 0;
+	return security_ops->path_symlink(path, dentry, old_name);
+}
+
+int security_path_link(struct dentry *old_dentry, struct path *new_dir,
+		       struct dentry *new_dentry)
+{
+	if (unlikely(IS_PRIVATE(old_dentry->d_inode)))
+		return 0;
+	return security_ops->path_link(old_dentry, new_dir, new_dentry);
+}
+
+int security_path_rename(struct path *old_dir, struct dentry *old_dentry,
+			 struct path *new_dir, struct dentry *new_dentry)
+{
+	if (unlikely(IS_PRIVATE(old_dentry->d_inode) ||
+		     (new_dentry->d_inode && IS_PRIVATE(new_dentry->d_inode))))
+		return 0;
+	return security_ops->path_rename(old_dir, old_dentry, new_dir,
+					 new_dentry);
+}
+
+int security_path_truncate(struct path *path, loff_t length,
+			   unsigned int time_attrs)
+{
+	if (unlikely(IS_PRIVATE(path->dentry->d_inode)))
+		return 0;
+	return security_ops->path_truncate(path, length, time_attrs);
+}
+#endif
+
 int security_inode_create(struct inode *dir, struct dentry *dentry, int mode)
 {
 	if (unlikely(IS_PRIVATE(dir)))
-- 
cgit v1.2.3


From c2452f32786159ed85f0e4b21fec09258f822fc8 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 1 Dec 2008 09:33:43 +0100
Subject: shrink struct dentry

struct dentry is one of the most critical structures in the kernel. So it's
sad to see it going neglected.

With CONFIG_PROFILING turned on (which is probably the common case at least
for distros and kernel developers), sizeof(struct dcache) == 208 here
(64-bit). This gives 19 objects per slab.

I packed d_mounted into a hole, and took another 4 bytes off the inline
name length to take the padding out from the end of the structure. This
shinks it to 200 bytes. I could have gone the other way and increased the
length to 40, but I'm aiming for a magic number, read on...

I then got rid of the d_cookie pointer. This shrinks it to 192 bytes. Rant:
why was this ever a good idea? The cookie system should increase its hash
size or use a tree or something if lookups are a problem. Also the "fast
dcookie lookups" in oprofile should be moved into the dcookie code -- how
can oprofile possibly care about the dcookie_mutex? It gets dropped after
get_dcookie() returns so it can't be providing any sort of protection.

At 192 bytes, 21 objects fit into a 4K page, saving about 3MB on my system
with ~140 000 entries allocated. 192 is also a multiple of 64, so we get
nice cacheline alignment on 64 and 32 byte line systems -- any given dentry
will now require 3 cachelines to touch all fields wheras previously it
would require 4.

I know the inline name size was chosen quite carefully, however with the
reduction in cacheline footprint, it should actually be just about as fast
to do a name lookup for a 36 character name as it was before the patch (and
faster for other sizes). The memory footprint savings for names which are
<= 32 or > 36 bytes long should more than make up for the memory cost for
33-36 byte names.

Performance is a feature...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 arch/powerpc/oprofile/cell/spu_task_sync.c |  2 +-
 drivers/oprofile/buffer_sync.c             |  2 +-
 fs/dcache.c                                |  4 ----
 fs/dcookies.c                              | 28 +++++++++++++++++++---------
 include/linux/dcache.h                     | 21 ++++++++++++++-------
 5 files changed, 35 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/arch/powerpc/oprofile/cell/spu_task_sync.c b/arch/powerpc/oprofile/cell/spu_task_sync.c
index 2949126d28d1..6b793aeda72e 100644
--- a/arch/powerpc/oprofile/cell/spu_task_sync.c
+++ b/arch/powerpc/oprofile/cell/spu_task_sync.c
@@ -297,7 +297,7 @@ static inline unsigned long fast_get_dcookie(struct path *path)
 {
 	unsigned long cookie;
 
-	if (path->dentry->d_cookie)
+	if (path->dentry->d_flags & DCACHE_COOKIE)
 		return (unsigned long)path->dentry;
 	get_dcookie(path, &cookie);
 	return cookie;
diff --git a/drivers/oprofile/buffer_sync.c b/drivers/oprofile/buffer_sync.c
index 737bd9484822..65e8294a9e29 100644
--- a/drivers/oprofile/buffer_sync.c
+++ b/drivers/oprofile/buffer_sync.c
@@ -200,7 +200,7 @@ static inline unsigned long fast_get_dcookie(struct path *path)
 {
 	unsigned long cookie;
 
-	if (path->dentry->d_cookie)
+	if (path->dentry->d_flags & DCACHE_COOKIE)
 		return (unsigned long)path->dentry;
 	get_dcookie(path, &cookie);
 	return cookie;
diff --git a/fs/dcache.c b/fs/dcache.c
index a1d86c7f3e66..fd244c7a7cc0 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -34,7 +34,6 @@
 #include <linux/bootmem.h>
 #include "internal.h"
 
-
 int sysctl_vfs_cache_pressure __read_mostly = 100;
 EXPORT_SYMBOL_GPL(sysctl_vfs_cache_pressure);
 
@@ -948,9 +947,6 @@ struct dentry *d_alloc(struct dentry * parent, const struct qstr *name)
 	dentry->d_op = NULL;
 	dentry->d_fsdata = NULL;
 	dentry->d_mounted = 0;
-#ifdef CONFIG_PROFILING
-	dentry->d_cookie = NULL;
-#endif
 	INIT_HLIST_NODE(&dentry->d_hash);
 	INIT_LIST_HEAD(&dentry->d_lru);
 	INIT_LIST_HEAD(&dentry->d_subdirs);
diff --git a/fs/dcookies.c b/fs/dcookies.c
index 855d4b1d619a..180e9fec4ad8 100644
--- a/fs/dcookies.c
+++ b/fs/dcookies.c
@@ -93,10 +93,15 @@ static struct dcookie_struct *alloc_dcookie(struct path *path)
 {
 	struct dcookie_struct *dcs = kmem_cache_alloc(dcookie_cache,
 							GFP_KERNEL);
+	struct dentry *d;
 	if (!dcs)
 		return NULL;
 
-	path->dentry->d_cookie = dcs;
+	d = path->dentry;
+	spin_lock(&d->d_lock);
+	d->d_flags |= DCACHE_COOKIE;
+	spin_unlock(&d->d_lock);
+
 	dcs->path = *path;
 	path_get(path);
 	hash_dcookie(dcs);
@@ -119,14 +124,14 @@ int get_dcookie(struct path *path, unsigned long *cookie)
 		goto out;
 	}
 
-	dcs = path->dentry->d_cookie;
-
-	if (!dcs)
+	if (path->dentry->d_flags & DCACHE_COOKIE) {
+		dcs = find_dcookie((unsigned long)path->dentry);
+	} else {
 		dcs = alloc_dcookie(path);
-
-	if (!dcs) {
-		err = -ENOMEM;
-		goto out;
+		if (!dcs) {
+			err = -ENOMEM;
+			goto out;
+		}
 	}
 
 	*cookie = dcookie_value(dcs);
@@ -251,7 +256,12 @@ out_kmem:
 
 static void free_dcookie(struct dcookie_struct * dcs)
 {
-	dcs->path.dentry->d_cookie = NULL;
+	struct dentry *d = dcs->path.dentry;
+
+	spin_lock(&d->d_lock);
+	d->d_flags &= ~DCACHE_COOKIE;
+	spin_unlock(&d->d_lock);
+
 	path_put(&dcs->path);
 	kmem_cache_free(dcookie_cache, dcs);
 }
diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index a37359d0bad1..c66d22487bf8 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -75,14 +75,22 @@ full_name_hash(const unsigned char *name, unsigned int len)
 	return end_name_hash(hash);
 }
 
-struct dcookie_struct;
-
-#define DNAME_INLINE_LEN_MIN 36
+/*
+ * Try to keep struct dentry aligned on 64 byte cachelines (this will
+ * give reasonable cacheline footprint with larger lines without the
+ * large memory footprint increase).
+ */
+#ifdef CONFIG_64BIT
+#define DNAME_INLINE_LEN_MIN 32 /* 192 bytes */
+#else
+#define DNAME_INLINE_LEN_MIN 40 /* 128 bytes */
+#endif
 
 struct dentry {
 	atomic_t d_count;
 	unsigned int d_flags;		/* protected by d_lock */
 	spinlock_t d_lock;		/* per dentry lock */
+	int d_mounted;
 	struct inode *d_inode;		/* Where the name belongs to - NULL is
 					 * negative */
 	/*
@@ -107,10 +115,7 @@ struct dentry {
 	struct dentry_operations *d_op;
 	struct super_block *d_sb;	/* The root of the dentry tree */
 	void *d_fsdata;			/* fs-specific data */
-#ifdef CONFIG_PROFILING
-	struct dcookie_struct *d_cookie; /* cookie, if any */
-#endif
-	int d_mounted;
+
 	unsigned char d_iname[DNAME_INLINE_LEN_MIN];	/* small names */
 };
 
@@ -177,6 +182,8 @@ d_iput:		no		no		no       yes
 
 #define DCACHE_INOTIFY_PARENT_WATCHED	0x0020 /* Parent inode is watched */
 
+#define DCACHE_COOKIE		0x0040	/* For use by dcookie subsystem */
+
 extern spinlock_t dcache_lock;
 extern seqlock_t rename_lock;
 
-- 
cgit v1.2.3


From dded4f4d5048e64a01cf52eed4d27c8cb2600525 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@medozas.de>
Date: Mon, 1 Dec 2008 14:34:50 -0800
Subject: include: linux/fs.h: put declarations in __KERNEL__

include/linux/fs.h contains externs for a bunch of variables.  That obviously
belongs under ifdef __KERNEL__.

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/fs.h | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/fs.h b/include/linux/fs.h
index 001ded4845b4..c5e4c5c74034 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -21,7 +21,6 @@
 
 /* Fixed constants first: */
 #undef NR_OPEN
-extern int sysctl_nr_open;
 #define INR_OPEN 1024		/* Initial setting for nfile rlimits */
 
 #define BLOCK_SIZE_BITS 10
@@ -38,21 +37,13 @@ struct files_stat_struct {
 	int nr_free_files;	/* read only */
 	int max_files;		/* tunable */
 };
-extern struct files_stat_struct files_stat;
-extern int get_max_files(void);
 
 struct inodes_stat_t {
 	int nr_inodes;
 	int nr_unused;
 	int dummy[5];		/* padding for sysctl ABI compatibility */
 };
-extern struct inodes_stat_t inodes_stat;
 
-extern int leases_enable, lease_break_time;
-
-#ifdef CONFIG_DNOTIFY
-extern int dir_notify_enable;
-#endif
 
 #define NR_FILE  8192	/* this can well be larger on a larger system */
 
@@ -330,6 +321,15 @@ extern void __init inode_init(void);
 extern void __init inode_init_early(void);
 extern void __init files_init(unsigned long);
 
+extern struct files_stat_struct files_stat;
+extern int get_max_files(void);
+extern int sysctl_nr_open;
+extern struct inodes_stat_t inodes_stat;
+extern int leases_enable, lease_break_time;
+#ifdef CONFIG_DNOTIFY
+extern int dir_notify_enable;
+#endif
+
 struct buffer_head;
 typedef int (get_block_t)(struct inode *inode, sector_t iblock,
 			struct buffer_head *bh_result, int create);
-- 
cgit v1.2.3


From 035146851cfa2fe24c1d9dc7637bb009ad06b2f7 Mon Sep 17 00:00:00 2001
From: Duane Griffin <duaneg@dghda.com>
Date: Fri, 19 Dec 2008 20:47:11 +0000
Subject: vfs: introduce helper function to safely NUL-terminate symlinks

A number of filesystems were potentially triggering kernel bugs due to
corrupted symlink names on disk. This function helps safely terminate
the names.

Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Duane Griffin <duaneg@dghda.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/namei.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/namei.h b/include/linux/namei.h
index 99eb80306dc5..fc2e03579877 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -94,4 +94,9 @@ static inline char *nd_get_link(struct nameidata *nd)
 	return nd->saved_names[nd->depth];
 }
 
+static inline void nd_terminate_link(void *name, size_t len, size_t maxlen)
+{
+	((char *) name)[min(len, maxlen)] = '\0';
+}
+
 #endif /* _LINUX_NAMEI_H */
-- 
cgit v1.2.3


From 3fb64190aa3c23c10e6e9fd0124ac030115c99bf Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Fri, 24 Oct 2008 09:58:10 +0200
Subject: pass a struct path * to may_open

No need for the nameidata in may_open - a struct path is enough.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c         | 14 +++++++-------
 fs/nfsctl.c        |  5 +++--
 include/linux/fs.h |  2 +-
 3 files changed, 11 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/fs/namei.c b/fs/namei.c
index d4d0b59ed2cc..5cc0dc95a7a5 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1487,9 +1487,9 @@ int vfs_create(struct inode *dir, struct dentry *dentry, int mode,
 	return error;
 }
 
-int may_open(struct nameidata *nd, int acc_mode, int flag)
+int may_open(struct path *path, int acc_mode, int flag)
 {
-	struct dentry *dentry = nd->path.dentry;
+	struct dentry *dentry = path->dentry;
 	struct inode *inode = dentry->d_inode;
 	int error;
 
@@ -1510,13 +1510,13 @@ int may_open(struct nameidata *nd, int acc_mode, int flag)
 	if (S_ISFIFO(inode->i_mode) || S_ISSOCK(inode->i_mode)) {
 	    	flag &= ~O_TRUNC;
 	} else if (S_ISBLK(inode->i_mode) || S_ISCHR(inode->i_mode)) {
-		if (nd->path.mnt->mnt_flags & MNT_NODEV)
+		if (path->mnt->mnt_flags & MNT_NODEV)
 			return -EACCES;
 
 		flag &= ~O_TRUNC;
 	}
 
-	error = vfs_permission(nd, acc_mode);
+	error = inode_permission(inode, acc_mode);
 	if (error)
 		return error;
 	/*
@@ -1551,7 +1551,7 @@ int may_open(struct nameidata *nd, int acc_mode, int flag)
 		 */
 		error = locks_verify_locked(inode);
 		if (!error)
-			error = security_path_truncate(&nd->path, 0,
+			error = security_path_truncate(path, 0,
 					       ATTR_MTIME|ATTR_CTIME|ATTR_OPEN);
 		if (!error) {
 			DQUOT_INIT(inode);
@@ -1594,7 +1594,7 @@ out_unlock:
 	if (error)
 		return error;
 	/* Don't check for write permission, don't truncate */
-	return may_open(nd, 0, flag & ~O_TRUNC);
+	return may_open(&nd->path, 0, flag & ~O_TRUNC);
 }
 
 /*
@@ -1780,7 +1780,7 @@ ok:
 		if (error)
 			goto exit;
 	}
-	error = may_open(&nd, acc_mode, flag);
+	error = may_open(&nd.path, acc_mode, flag);
 	if (error) {
 		if (will_write)
 			mnt_drop_write(nd.path.mnt);
diff --git a/fs/nfsctl.c b/fs/nfsctl.c
index b1acbd6ab6fb..b27451909dff 100644
--- a/fs/nfsctl.c
+++ b/fs/nfsctl.c
@@ -38,9 +38,10 @@ static struct file *do_open(char *name, int flags)
 		return ERR_PTR(error);
 
 	if (flags == O_RDWR)
-		error = may_open(&nd,MAY_READ|MAY_WRITE,FMODE_READ|FMODE_WRITE);
+		error = may_open(&nd.path, MAY_READ|MAY_WRITE,
+					   FMODE_READ|FMODE_WRITE);
 	else
-		error = may_open(&nd, MAY_WRITE, FMODE_WRITE);
+		error = may_open(&nd.path, MAY_WRITE, FMODE_WRITE);
 
 	if (!error)
 		return dentry_open(nd.path.dentry, nd.path.mnt, flags,
diff --git a/include/linux/fs.h b/include/linux/fs.h
index c5e4c5c74034..3468df5a06e0 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1869,7 +1869,7 @@ extern void free_write_pipe(struct file *);
 
 extern struct file *do_filp_open(int dfd, const char *pathname,
 		int open_flag, int mode);
-extern int may_open(struct nameidata *, int, int);
+extern int may_open(struct path *, int, int);
 
 extern int kernel_read(struct file *, unsigned long, char *, unsigned long);
 extern struct file * open_exec(const char *);
-- 
cgit v1.2.3


From cb23beb55100171646e69e248fb45f10db6e99a4 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Fri, 24 Oct 2008 09:59:29 +0200
Subject: kill vfs_permission

With all the nameidata removal there's no point anymore for this helper.
Of the three callers left two will go away with the next lookup series
anyway.

Also add proper kerneldoc to inode_permission as this is the main
permission check routine now.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/exec.c          |  5 +++--
 fs/namei.c         | 31 +++++++++++++------------------
 include/linux/fs.h |  1 -
 3 files changed, 16 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/fs/exec.c b/fs/exec.c
index 02d2e120542d..dfbf7009fbe7 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -127,7 +127,8 @@ asmlinkage long sys_uselib(const char __user * library)
 	if (nd.path.mnt->mnt_flags & MNT_NOEXEC)
 		goto exit;
 
-	error = vfs_permission(&nd, MAY_READ | MAY_EXEC | MAY_OPEN);
+	error = inode_permission(nd.path.dentry->d_inode,
+				 MAY_READ | MAY_EXEC | MAY_OPEN);
 	if (error)
 		goto exit;
 
@@ -680,7 +681,7 @@ struct file *open_exec(const char *name)
 	if (nd.path.mnt->mnt_flags & MNT_NOEXEC)
 		goto out_path_put;
 
-	err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN);
+	err = inode_permission(nd.path.dentry->d_inode, MAY_EXEC | MAY_OPEN);
 	if (err)
 		goto out_path_put;
 
diff --git a/fs/namei.c b/fs/namei.c
index 5cc0dc95a7a5..3f88e043d459 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -226,6 +226,16 @@ int generic_permission(struct inode *inode, int mask,
 	return -EACCES;
 }
 
+/**
+ * inode_permission  -  check for access rights to a given inode
+ * @inode:	inode to check permission on
+ * @mask:	right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC)
+ *
+ * Used to check for read/write/execute permissions on an inode.
+ * We use "fsuid" for this, letting us set arbitrary permissions
+ * for filesystem access without changing the "normal" uids which
+ * are used for other things.
+ */
 int inode_permission(struct inode *inode, int mask)
 {
 	int retval;
@@ -263,21 +273,6 @@ int inode_permission(struct inode *inode, int mask)
 			mask & (MAY_READ|MAY_WRITE|MAY_EXEC|MAY_APPEND));
 }
 
-/**
- * vfs_permission  -  check for access rights to a given path
- * @nd:		lookup result that describes the path
- * @mask:	right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC)
- *
- * Used to check for read/write/execute permissions on a path.
- * We use "fsuid" for this, letting us set arbitrary permissions
- * for filesystem access without changing the "normal" uids which
- * are used for other things.
- */
-int vfs_permission(struct nameidata *nd, int mask)
-{
-	return inode_permission(nd->path.dentry->d_inode, mask);
-}
-
 /**
  * file_permission  -  check for additional access rights to a given file
  * @file:	file to check access rights for
@@ -288,7 +283,7 @@ int vfs_permission(struct nameidata *nd, int mask)
  *
  * Note:
  *	Do not use this function in new code.  All access checks should
- *	be done using vfs_permission().
+ *	be done using inode_permission().
  */
 int file_permission(struct file *file, int mask)
 {
@@ -853,7 +848,8 @@ static int __link_path_walk(const char *name, struct nameidata *nd)
 		nd->flags |= LOOKUP_CONTINUE;
 		err = exec_permission_lite(inode);
 		if (err == -EAGAIN)
-			err = vfs_permission(nd, MAY_EXEC);
+			err = inode_permission(nd->path.dentry->d_inode,
+					       MAY_EXEC);
  		if (err)
 			break;
 
@@ -2882,7 +2878,6 @@ EXPORT_SYMBOL(path_lookup);
 EXPORT_SYMBOL(kern_path);
 EXPORT_SYMBOL(vfs_path_lookup);
 EXPORT_SYMBOL(inode_permission);
-EXPORT_SYMBOL(vfs_permission);
 EXPORT_SYMBOL(file_permission);
 EXPORT_SYMBOL(unlock_rename);
 EXPORT_SYMBOL(vfs_create);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 3468df5a06e0..fd615986a41c 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1212,7 +1212,6 @@ extern void unlock_super(struct super_block *);
 /*
  * VFS helper functions..
  */
-extern int vfs_permission(struct nameidata *, int);
 extern int vfs_create(struct inode *, struct dentry *, int, struct nameidata *);
 extern int vfs_mkdir(struct inode *, struct dentry *, int);
 extern int vfs_mknod(struct inode *, struct dentry *, int, dev_t);
-- 
cgit v1.2.3


From 18d8fda7c3c9439be04d7ea2e82da2513b121acb Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Fri, 26 Dec 2008 00:35:37 -0500
Subject: take init_fs to saner place

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 arch/alpha/kernel/init_task.c     | 1 -
 arch/arm/kernel/init_task.c       | 1 -
 arch/avr32/kernel/init_task.c     | 1 -
 arch/blackfin/kernel/init_task.c  | 1 -
 arch/cris/kernel/process.c        | 1 -
 arch/frv/kernel/init_task.c       | 1 -
 arch/h8300/kernel/init_task.c     | 1 -
 arch/ia64/kernel/init_task.c      | 1 -
 arch/m32r/kernel/init_task.c      | 1 -
 arch/m68k/kernel/process.c        | 1 -
 arch/m68knommu/kernel/init_task.c | 1 -
 arch/mips/kernel/init_task.c      | 1 -
 arch/mn10300/kernel/init_task.c   | 1 -
 arch/parisc/kernel/init_task.c    | 1 -
 arch/powerpc/kernel/init_task.c   | 1 -
 arch/s390/kernel/init_task.c      | 1 -
 arch/sh/kernel/init_task.c        | 1 -
 arch/sparc/kernel/init_task.c     | 1 -
 arch/um/kernel/init_task.c        | 1 -
 arch/x86/kernel/init_task.c       | 1 -
 arch/xtensa/kernel/init_task.c    | 1 -
 fs/namei.c                        | 7 +++++++
 include/linux/fs_struct.h         | 6 ------
 include/linux/init_task.h         | 1 +
 24 files changed, 8 insertions(+), 27 deletions(-)

(limited to 'include/linux')

diff --git a/arch/alpha/kernel/init_task.c b/arch/alpha/kernel/init_task.c
index 1f762189fa64..c2938e574a40 100644
--- a/arch/alpha/kernel/init_task.c
+++ b/arch/alpha/kernel/init_task.c
@@ -8,7 +8,6 @@
 #include <asm/uaccess.h>
 
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/arm/kernel/init_task.c b/arch/arm/kernel/init_task.c
index 0bbf80625395..e859af349467 100644
--- a/arch/arm/kernel/init_task.c
+++ b/arch/arm/kernel/init_task.c
@@ -12,7 +12,6 @@
 
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/avr32/kernel/init_task.c b/arch/avr32/kernel/init_task.c
index 44058469c6ec..993d56ee3cf3 100644
--- a/arch/avr32/kernel/init_task.c
+++ b/arch/avr32/kernel/init_task.c
@@ -13,7 +13,6 @@
 
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/blackfin/kernel/init_task.c b/arch/blackfin/kernel/init_task.c
index 6bdba7b21109..2c228c020978 100644
--- a/arch/blackfin/kernel/init_task.c
+++ b/arch/blackfin/kernel/init_task.c
@@ -33,7 +33,6 @@
 #include <linux/mqueue.h>
 #include <linux/fs.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 
diff --git a/arch/cris/kernel/process.c b/arch/cris/kernel/process.c
index 5933656db5a2..60816e876455 100644
--- a/arch/cris/kernel/process.c
+++ b/arch/cris/kernel/process.c
@@ -37,7 +37,6 @@
  * setup.
  */
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/frv/kernel/init_task.c b/arch/frv/kernel/init_task.c
index e2198815b630..29429a8b7f6a 100644
--- a/arch/frv/kernel/init_task.c
+++ b/arch/frv/kernel/init_task.c
@@ -10,7 +10,6 @@
 #include <asm/pgtable.h>
 
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/h8300/kernel/init_task.c b/arch/h8300/kernel/init_task.c
index 93a4899e46c2..cb5dc552da97 100644
--- a/arch/h8300/kernel/init_task.c
+++ b/arch/h8300/kernel/init_task.c
@@ -12,7 +12,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/ia64/kernel/init_task.c b/arch/ia64/kernel/init_task.c
index 9d7e1c66faf4..5b0e830c6f33 100644
--- a/arch/ia64/kernel/init_task.c
+++ b/arch/ia64/kernel/init_task.c
@@ -17,7 +17,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/m32r/kernel/init_task.c b/arch/m32r/kernel/init_task.c
index 0d658dbb6766..016885c6f260 100644
--- a/arch/m32r/kernel/init_task.c
+++ b/arch/m32r/kernel/init_task.c
@@ -11,7 +11,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/m68k/kernel/process.c b/arch/m68k/kernel/process.c
index 3042c2bc8c58..632ce016014d 100644
--- a/arch/m68k/kernel/process.c
+++ b/arch/m68k/kernel/process.c
@@ -40,7 +40,6 @@
  * alignment requirements and potentially different initial
  * setup.
  */
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/m68knommu/kernel/init_task.c b/arch/m68knommu/kernel/init_task.c
index 344c01aede08..fe282de1d596 100644
--- a/arch/m68knommu/kernel/init_task.c
+++ b/arch/m68knommu/kernel/init_task.c
@@ -12,7 +12,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/mips/kernel/init_task.c b/arch/mips/kernel/init_task.c
index d72487ad7c15..149cd914526e 100644
--- a/arch/mips/kernel/init_task.c
+++ b/arch/mips/kernel/init_task.c
@@ -9,7 +9,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/mn10300/kernel/init_task.c b/arch/mn10300/kernel/init_task.c
index af16f6e5c918..5ac3566f8c98 100644
--- a/arch/mn10300/kernel/init_task.c
+++ b/arch/mn10300/kernel/init_task.c
@@ -18,7 +18,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/parisc/kernel/init_task.c b/arch/parisc/kernel/init_task.c
index f5941c086551..1e25a45d64c1 100644
--- a/arch/parisc/kernel/init_task.c
+++ b/arch/parisc/kernel/init_task.c
@@ -34,7 +34,6 @@
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/powerpc/kernel/init_task.c b/arch/powerpc/kernel/init_task.c
index 4c85b8d56478..688b329800bd 100644
--- a/arch/powerpc/kernel/init_task.c
+++ b/arch/powerpc/kernel/init_task.c
@@ -7,7 +7,6 @@
 #include <linux/mqueue.h>
 #include <asm/uaccess.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/s390/kernel/init_task.c b/arch/s390/kernel/init_task.c
index e80716843619..7db95c0b8693 100644
--- a/arch/s390/kernel/init_task.c
+++ b/arch/s390/kernel/init_task.c
@@ -16,7 +16,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/sh/kernel/init_task.c b/arch/sh/kernel/init_task.c
index b151a25cb14d..80c35ff71d56 100644
--- a/arch/sh/kernel/init_task.c
+++ b/arch/sh/kernel/init_task.c
@@ -7,7 +7,6 @@
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct pt_regs fake_swapper_regs;
diff --git a/arch/sparc/kernel/init_task.c b/arch/sparc/kernel/init_task.c
index 62126e4cec54..f28cb8278e98 100644
--- a/arch/sparc/kernel/init_task.c
+++ b/arch/sparc/kernel/init_task.c
@@ -8,7 +8,6 @@
 #include <asm/pgtable.h>
 #include <asm/uaccess.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/um/kernel/init_task.c b/arch/um/kernel/init_task.c
index 910eda8fca18..806d381947bf 100644
--- a/arch/um/kernel/init_task.c
+++ b/arch/um/kernel/init_task.c
@@ -10,7 +10,6 @@
 #include "linux/mqueue.h"
 #include "asm/uaccess.h"
 
-static struct fs_struct init_fs = INIT_FS;
 struct mm_struct init_mm = INIT_MM(init_mm);
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
diff --git a/arch/x86/kernel/init_task.c b/arch/x86/kernel/init_task.c
index d39918076bb4..df3bf269beab 100644
--- a/arch/x86/kernel/init_task.c
+++ b/arch/x86/kernel/init_task.c
@@ -10,7 +10,6 @@
 #include <asm/pgtable.h>
 #include <asm/desc.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/arch/xtensa/kernel/init_task.c b/arch/xtensa/kernel/init_task.c
index 3df469dbe814..e07f5c9fcd35 100644
--- a/arch/xtensa/kernel/init_task.c
+++ b/arch/xtensa/kernel/init_task.c
@@ -21,7 +21,6 @@
 
 #include <asm/uaccess.h>
 
-static struct fs_struct init_fs = INIT_FS;
 static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
 static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
 struct mm_struct init_mm = INIT_MM(init_mm);
diff --git a/fs/namei.c b/fs/namei.c
index 3f88e043d459..e203691b9d12 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2893,3 +2893,10 @@ EXPORT_SYMBOL(vfs_symlink);
 EXPORT_SYMBOL(vfs_unlink);
 EXPORT_SYMBOL(dentry_unhash);
 EXPORT_SYMBOL(generic_readlink);
+
+/* to be mentioned only in INIT_TASK */
+struct fs_struct init_fs = {
+	.count		= ATOMIC_INIT(1),
+	.lock		= RW_LOCK_UNLOCKED,
+	.umask		= 0022,
+};
diff --git a/include/linux/fs_struct.h b/include/linux/fs_struct.h
index 9e5a06e78d02..a97c053d3a9a 100644
--- a/include/linux/fs_struct.h
+++ b/include/linux/fs_struct.h
@@ -10,12 +10,6 @@ struct fs_struct {
 	struct path root, pwd;
 };
 
-#define INIT_FS {				\
-	.count		= ATOMIC_INIT(1),	\
-	.lock		= RW_LOCK_UNLOCKED,	\
-	.umask		= 0022, \
-}
-
 extern struct kmem_cache *fs_cachep;
 
 extern void exit_fs(struct task_struct *);
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 959f5522d10a..2f3c2d4ef73b 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -12,6 +12,7 @@
 #include <net/net_namespace.h>
 
 extern struct files_struct init_files;
+extern struct fs_struct init_fs;
 
 #define INIT_KIOCTX(name, which_mm) \
 {							\
-- 
cgit v1.2.3


From b6b3fdead251d432f32f2cfce2a893ab8a658110 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Wed, 10 Dec 2008 09:35:45 -0800
Subject: filp_cachep can be static in fs/file_table.c

Instead of creating the "filp" kmem_cache in vfs_caches_init(),
we can do it a litle be later in files_init(), so that filp_cachep
is static to fs/file_table.c

Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/dcache.c             |  6 ------
 fs/file_table.c         | 10 +++++++++-
 include/linux/fdtable.h |  2 --
 3 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/fs/dcache.c b/fs/dcache.c
index bdb3f50248a7..e88c23b85a32 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -2314,9 +2314,6 @@ static void __init dcache_init(void)
 /* SLAB cache for __getname() consumers */
 struct kmem_cache *names_cachep __read_mostly;
 
-/* SLAB cache for file structures */
-struct kmem_cache *filp_cachep __read_mostly;
-
 EXPORT_SYMBOL(d_genocide);
 
 void __init vfs_caches_init_early(void)
@@ -2338,9 +2335,6 @@ void __init vfs_caches_init(unsigned long mempages)
 	names_cachep = kmem_cache_create("names_cache", PATH_MAX, 0,
 			SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
 
-	filp_cachep = kmem_cache_create("filp", sizeof(struct file), 0,
-			SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
-
 	dcache_init();
 	inode_init();
 	files_init(mempages);
diff --git a/fs/file_table.c b/fs/file_table.c
index 0fbcacc3ea75..bbeeac6efa1a 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -32,6 +32,9 @@ struct files_stat_struct files_stat = {
 /* public. Not pretty! */
 __cacheline_aligned_in_smp DEFINE_SPINLOCK(files_lock);
 
+/* SLAB cache for file structures */
+static struct kmem_cache *filp_cachep __read_mostly;
+
 static struct percpu_counter nr_files __cacheline_aligned_in_smp;
 
 static inline void file_free_rcu(struct rcu_head *head)
@@ -397,7 +400,12 @@ too_bad:
 void __init files_init(unsigned long mempages)
 { 
 	int n; 
-	/* One file with associated inode and dcache is very roughly 1K. 
+
+	filp_cachep = kmem_cache_create("filp", sizeof(struct file), 0,
+			SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL);
+
+	/*
+	 * One file with associated inode and dcache is very roughly 1K.
 	 * Per default don't use more than 10% of our memory for files. 
 	 */ 
 
diff --git a/include/linux/fdtable.h b/include/linux/fdtable.h
index 4aab6f12cfab..09d6c5bbdddd 100644
--- a/include/linux/fdtable.h
+++ b/include/linux/fdtable.h
@@ -57,8 +57,6 @@ struct files_struct {
 
 #define files_fdtable(files) (rcu_dereference((files)->fdt))
 
-extern struct kmem_cache *filp_cachep;
-
 struct file_operations;
 struct vfsmount;
 struct dentry;
-- 
cgit v1.2.3


From 6badd79bd002788aaec27b50a74ab69ef65ab8ee Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Fri, 26 Dec 2008 00:57:40 -0500
Subject: kill ->dir_notify()

Remove the hopelessly misguided ->dir_notify().  The only instance (cifs)
has been broken by design from the very beginning; the objects it creates
are never destroyed, keep references to struct file they can outlive, nothing
that could possibly evict them exists on close(2) path *and* no locking
whatsoever is done to prevent races with close(), should the previous, er,
deficiencies someday be dealt with.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 Documentation/filesystems/Locking |   2 -
 Documentation/filesystems/vfs.txt |   3 -
 fs/bad_inode.c                    |   6 --
 fs/cifs/Makefile                  |   2 +-
 fs/cifs/cifsfs.c                  |   7 ---
 fs/cifs/cifsfs.h                  |   1 -
 fs/cifs/fcntl.c                   | 118 --------------------------------------
 fs/dnotify.c                      |   3 -
 include/linux/fs.h                |   1 -
 9 files changed, 1 insertion(+), 142 deletions(-)
 delete mode 100644 fs/cifs/fcntl.c

(limited to 'include/linux')

diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking
index 23d2f4460deb..ccec55394380 100644
--- a/Documentation/filesystems/Locking
+++ b/Documentation/filesystems/Locking
@@ -394,7 +394,6 @@ prototypes:
 	unsigned long (*get_unmapped_area)(struct file *, unsigned long,
 			unsigned long, unsigned long, unsigned long);
 	int (*check_flags)(int);
-	int (*dir_notify)(struct file *, unsigned long);
 };
 
 locking rules:
@@ -424,7 +423,6 @@ sendfile:		no
 sendpage:		no
 get_unmapped_area:	no
 check_flags:		no
-dir_notify:		no
 
 ->llseek() locking has moved from llseek to the individual llseek
 implementations.  If your fs is not using generic_file_llseek, you
diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt
index 041cb771d505..ef19afa186a9 100644
--- a/Documentation/filesystems/vfs.txt
+++ b/Documentation/filesystems/vfs.txt
@@ -733,7 +733,6 @@ struct file_operations {
 	ssize_t (*sendpage) (struct file *, struct page *, int, size_t, loff_t *, int);
 	unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
 	int (*check_flags)(int);
-	int (*dir_notify)(struct file *filp, unsigned long arg);
 	int (*flock) (struct file *, int, struct file_lock *);
 	ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, size_t, unsigned int);
 	ssize_t (*splice_read)(struct file *, struct pipe_inode_info *, size_t, unsigned int);
@@ -800,8 +799,6 @@ otherwise noted.
 
   check_flags: called by the fcntl(2) system call for F_SETFL command
 
-  dir_notify: called by the fcntl(2) system call for F_NOTIFY command
-
   flock: called by the flock(2) system call
 
   splice_write: called by the VFS to splice data from a pipe to a file. This
diff --git a/fs/bad_inode.c b/fs/bad_inode.c
index 5f1538c03b1b..a05287a23f62 100644
--- a/fs/bad_inode.c
+++ b/fs/bad_inode.c
@@ -132,11 +132,6 @@ static int bad_file_check_flags(int flags)
 	return -EIO;
 }
 
-static int bad_file_dir_notify(struct file *file, unsigned long arg)
-{
-	return -EIO;
-}
-
 static int bad_file_flock(struct file *filp, int cmd, struct file_lock *fl)
 {
 	return -EIO;
@@ -179,7 +174,6 @@ static const struct file_operations bad_file_ops =
 	.sendpage	= bad_file_sendpage,
 	.get_unmapped_area = bad_file_get_unmapped_area,
 	.check_flags	= bad_file_check_flags,
-	.dir_notify	= bad_file_dir_notify,
 	.flock		= bad_file_flock,
 	.splice_write	= bad_file_splice_write,
 	.splice_read	= bad_file_splice_read,
diff --git a/fs/cifs/Makefile b/fs/cifs/Makefile
index 6ba43fb346fb..9948c0030e86 100644
--- a/fs/cifs/Makefile
+++ b/fs/cifs/Makefile
@@ -5,7 +5,7 @@ obj-$(CONFIG_CIFS) += cifs.o
 
 cifs-y := cifsfs.o cifssmb.o cifs_debug.o connect.o dir.o file.o inode.o \
 	  link.o misc.o netmisc.o smbdes.o smbencrypt.o transport.o asn1.o \
-	  md4.o md5.o cifs_unicode.o nterr.o xattr.o cifsencrypt.o fcntl.o \
+	  md4.o md5.o cifs_unicode.o nterr.o xattr.o cifsencrypt.o \
 	  readdir.o ioctl.o sess.o export.o cifsacl.o
 
 cifs-$(CONFIG_CIFS_UPCALL) += cifs_spnego.o
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 0005a194a75c..13ea53251dcf 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -747,7 +747,6 @@ const struct file_operations cifs_file_ops = {
 #endif /* CONFIG_CIFS_POSIX */
 
 #ifdef CONFIG_CIFS_EXPERIMENTAL
-	.dir_notify = cifs_dir_notify,
 	.setlease = cifs_setlease,
 #endif /* CONFIG_CIFS_EXPERIMENTAL */
 };
@@ -768,7 +767,6 @@ const struct file_operations cifs_file_direct_ops = {
 #endif /* CONFIG_CIFS_POSIX */
 	.llseek = cifs_llseek,
 #ifdef CONFIG_CIFS_EXPERIMENTAL
-	.dir_notify = cifs_dir_notify,
 	.setlease = cifs_setlease,
 #endif /* CONFIG_CIFS_EXPERIMENTAL */
 };
@@ -789,7 +787,6 @@ const struct file_operations cifs_file_nobrl_ops = {
 #endif /* CONFIG_CIFS_POSIX */
 
 #ifdef CONFIG_CIFS_EXPERIMENTAL
-	.dir_notify = cifs_dir_notify,
 	.setlease = cifs_setlease,
 #endif /* CONFIG_CIFS_EXPERIMENTAL */
 };
@@ -809,7 +806,6 @@ const struct file_operations cifs_file_direct_nobrl_ops = {
 #endif /* CONFIG_CIFS_POSIX */
 	.llseek = cifs_llseek,
 #ifdef CONFIG_CIFS_EXPERIMENTAL
-	.dir_notify = cifs_dir_notify,
 	.setlease = cifs_setlease,
 #endif /* CONFIG_CIFS_EXPERIMENTAL */
 };
@@ -818,9 +814,6 @@ const struct file_operations cifs_dir_ops = {
 	.readdir = cifs_readdir,
 	.release = cifs_closedir,
 	.read    = generic_read_dir,
-#ifdef CONFIG_CIFS_EXPERIMENTAL
-	.dir_notify = cifs_dir_notify,
-#endif /* CONFIG_CIFS_EXPERIMENTAL */
 	.unlocked_ioctl  = cifs_ioctl,
 	.llseek = generic_file_llseek,
 };
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index 2ce04c73d74e..7ac481841f87 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -76,7 +76,6 @@ extern int cifs_file_mmap(struct file * , struct vm_area_struct *);
 extern const struct file_operations cifs_dir_ops;
 extern int cifs_dir_open(struct inode *inode, struct file *file);
 extern int cifs_readdir(struct file *file, void *direntry, filldir_t filldir);
-extern int cifs_dir_notify(struct file *, unsigned long arg);
 
 /* Functions related to dir entries */
 extern struct dentry_operations cifs_dentry_ops;
diff --git a/fs/cifs/fcntl.c b/fs/cifs/fcntl.c
deleted file mode 100644
index 5a57581eb4b2..000000000000
--- a/fs/cifs/fcntl.c
+++ /dev/null
@@ -1,118 +0,0 @@
-/*
- *   fs/cifs/fcntl.c
- *
- *   vfs operations that deal with the file control API
- *
- *   Copyright (C) International Business Machines  Corp., 2003,2004
- *   Author(s): Steve French (sfrench@us.ibm.com)
- *
- *   This library is free software; you can redistribute it and/or modify
- *   it under the terms of the GNU Lesser General Public License as published
- *   by the Free Software Foundation; either version 2.1 of the License, or
- *   (at your option) any later version.
- *
- *   This library is distributed in the hope that it will be useful,
- *   but WITHOUT ANY WARRANTY; without even the implied warranty of
- *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
- *   the GNU Lesser General Public License for more details.
- *
- *   You should have received a copy of the GNU Lesser General Public License
- *   along with this library; if not, write to the Free Software
- *   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- */
-#include <linux/fs.h>
-#include <linux/stat.h>
-#include <linux/fcntl.h>
-#include "cifsglob.h"
-#include "cifsproto.h"
-#include "cifs_unicode.h"
-#include "cifs_debug.h"
-#include "cifsfs.h"
-
-static __u32 convert_to_cifs_notify_flags(unsigned long fcntl_notify_flags)
-{
-	__u32 cifs_ntfy_flags = 0;
-
-	/* No way on Linux VFS to ask to monitor xattr
-	changes (and no stream support either */
-	if (fcntl_notify_flags & DN_ACCESS)
-		cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_LAST_ACCESS;
-	if (fcntl_notify_flags & DN_MODIFY) {
-		/* What does this mean on directories? */
-		cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_LAST_WRITE |
-			FILE_NOTIFY_CHANGE_SIZE;
-	}
-	if (fcntl_notify_flags & DN_CREATE) {
-		cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_CREATION |
-			FILE_NOTIFY_CHANGE_LAST_WRITE;
-	}
-	if (fcntl_notify_flags & DN_DELETE)
-		cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_LAST_WRITE;
-	if (fcntl_notify_flags & DN_RENAME) {
-		/* BB review this - checking various server behaviors */
-		cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_DIR_NAME |
-			FILE_NOTIFY_CHANGE_FILE_NAME;
-	}
-	if (fcntl_notify_flags & DN_ATTRIB) {
-		cifs_ntfy_flags |= FILE_NOTIFY_CHANGE_SECURITY |
-			FILE_NOTIFY_CHANGE_ATTRIBUTES;
-	}
-/*	if (fcntl_notify_flags & DN_MULTISHOT) {
-		cifs_ntfy_flags |= ;
-	} */ /* BB fixme - not sure how to handle this with CIFS yet */
-
-	return cifs_ntfy_flags;
-}
-
-int cifs_dir_notify(struct file *file, unsigned long arg)
-{
-	int xid;
-	int rc = -EINVAL;
-	int oplock = 0;
-	struct cifs_sb_info *cifs_sb;
-	struct cifsTconInfo *pTcon;
-	char *full_path = NULL;
-	__u32 filter = FILE_NOTIFY_CHANGE_NAME | FILE_NOTIFY_CHANGE_ATTRIBUTES;
-	__u16 netfid;
-
-	if (experimEnabled == 0)
-		return 0;
-
-	xid = GetXid();
-	cifs_sb = CIFS_SB(file->f_path.dentry->d_sb);
-	pTcon = cifs_sb->tcon;
-
-	full_path = build_path_from_dentry(file->f_path.dentry);
-
-	if (full_path == NULL) {
-		rc = -ENOMEM;
-	} else {
-		cFYI(1, ("dir notify on file %s Arg 0x%lx", full_path, arg));
-		rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN,
-			GENERIC_READ | SYNCHRONIZE, 0 /* create options */,
-			&netfid, &oplock, NULL, cifs_sb->local_nls,
-			cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR);
-		/* BB fixme - add this handle to a notify handle list */
-		if (rc) {
-			cFYI(1, ("Could not open directory for notify"));
-		} else {
-			filter = convert_to_cifs_notify_flags(arg);
-			if (filter != 0) {
-				rc = CIFSSMBNotify(xid, pTcon,
-					0 /* no subdirs */, netfid,
-					filter, file, arg & DN_MULTISHOT,
-					cifs_sb->local_nls);
-			} else {
-				rc = -EINVAL;
-			}
-			/* BB add code to close file eventually (at unmount
-			it would close automatically but may be a way
-			to do it easily when inode freed or when
-			notify info is cleared/changed */
-			cFYI(1, ("notify rc %d", rc));
-		}
-	}
-
-	FreeXid(xid);
-	return rc;
-}
diff --git a/fs/dnotify.c b/fs/dnotify.c
index 676073b8dda5..b0aa2cde80bd 100644
--- a/fs/dnotify.c
+++ b/fs/dnotify.c
@@ -115,9 +115,6 @@ int fcntl_dirnotify(int fd, struct file *filp, unsigned long arg)
 	dn->dn_next = inode->i_dnotify;
 	inode->i_dnotify = dn;
 	spin_unlock(&inode->i_lock);
-
-	if (filp->f_op && filp->f_op->dir_notify)
-		return filp->f_op->dir_notify(filp, arg);
 	return 0;
 
 out_free:
diff --git a/include/linux/fs.h b/include/linux/fs.h
index fd615986a41c..be16ce01fb1b 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1309,7 +1309,6 @@ struct file_operations {
 	ssize_t (*sendpage) (struct file *, struct page *, int, size_t, loff_t *, int);
 	unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
 	int (*check_flags)(int);
-	int (*dir_notify)(struct file *filp, unsigned long arg);
 	int (*flock) (struct file *, int, struct file_lock *);
 	ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int);
 	ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int);
-- 
cgit v1.2.3


From 261bca86ed4f7f391d1938167624e78da61dcc6b Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 30 Dec 2008 01:48:21 -0500
Subject: nfsd/create race fixes, infrastructure

new helpers - insert_inode_locked() and insert_inode_locked4().
Hash new inode, making sure that there's no such inode in icache
already.  If there is and it does not end up unhashed (as would
happen if we have nfsd trying to resolve a bogus fhandle), fail.
Otherwise insert our inode into hash and succeed.

In either case have i_state set to new+locked; cleanup ends up
being simpler with such calling conventions.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/inode.c         | 59 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/fs.h |  2 ++
 2 files changed, 61 insertions(+)

(limited to 'include/linux')

diff --git a/fs/inode.c b/fs/inode.c
index 098a2443196f..7de1cda92489 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -1032,6 +1032,65 @@ struct inode *iget_locked(struct super_block *sb, unsigned long ino)
 
 EXPORT_SYMBOL(iget_locked);
 
+int insert_inode_locked(struct inode *inode)
+{
+	struct super_block *sb = inode->i_sb;
+	ino_t ino = inode->i_ino;
+	struct hlist_head *head = inode_hashtable + hash(sb, ino);
+	struct inode *old;
+
+	inode->i_state |= I_LOCK|I_NEW;
+	while (1) {
+		spin_lock(&inode_lock);
+		old = find_inode_fast(sb, head, ino);
+		if (likely(!old)) {
+			hlist_add_head(&inode->i_hash, head);
+			spin_unlock(&inode_lock);
+			return 0;
+		}
+		__iget(old);
+		spin_unlock(&inode_lock);
+		wait_on_inode(old);
+		if (unlikely(!hlist_unhashed(&old->i_hash))) {
+			iput(old);
+			return -EBUSY;
+		}
+		iput(old);
+	}
+}
+
+EXPORT_SYMBOL(insert_inode_locked);
+
+int insert_inode_locked4(struct inode *inode, unsigned long hashval,
+		int (*test)(struct inode *, void *), void *data)
+{
+	struct super_block *sb = inode->i_sb;
+	struct hlist_head *head = inode_hashtable + hash(sb, hashval);
+	struct inode *old;
+
+	inode->i_state |= I_LOCK|I_NEW;
+
+	while (1) {
+		spin_lock(&inode_lock);
+		old = find_inode(sb, head, test, data);
+		if (likely(!old)) {
+			hlist_add_head(&inode->i_hash, head);
+			spin_unlock(&inode_lock);
+			return 0;
+		}
+		__iget(old);
+		spin_unlock(&inode_lock);
+		wait_on_inode(old);
+		if (unlikely(!hlist_unhashed(&old->i_hash))) {
+			iput(old);
+			return -EBUSY;
+		}
+		iput(old);
+	}
+}
+
+EXPORT_SYMBOL(insert_inode_locked4);
+
 /**
  *	__insert_inode_hash - hash an inode
  *	@inode: unhashed inode
diff --git a/include/linux/fs.h b/include/linux/fs.h
index be16ce01fb1b..e2170ee21e18 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1902,6 +1902,8 @@ extern struct inode *ilookup(struct super_block *sb, unsigned long ino);
 
 extern struct inode * iget5_locked(struct super_block *, unsigned long, int (*test)(struct inode *, void *), int (*set)(struct inode *, void *), void *);
 extern struct inode * iget_locked(struct super_block *, unsigned long);
+extern int insert_inode_locked4(struct inode *, unsigned long, int (*test)(struct inode *, void *), void *);
+extern int insert_inode_locked(struct inode *);
 extern void unlock_new_inode(struct inode *);
 
 extern void __iget(struct inode * inode);
-- 
cgit v1.2.3


From ab53d472e785e51fdfc08fc1d66252c1153e6c0f Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 1 Jan 2009 10:12:19 +1030
Subject: bitmap: find_last_bit()

Impact: New API

As the name suggests.  For the moment everyone uses the generic one.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 include/linux/bitops.h | 13 ++++++++++++-
 lib/Kconfig            |  4 ++++
 lib/Makefile           |  1 +
 lib/find_last_bit.c    | 45 +++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 62 insertions(+), 1 deletion(-)
 create mode 100644 lib/find_last_bit.c

(limited to 'include/linux')

diff --git a/include/linux/bitops.h b/include/linux/bitops.h
index 024f2b027244..61829139795a 100644
--- a/include/linux/bitops.h
+++ b/include/linux/bitops.h
@@ -134,9 +134,20 @@ extern unsigned long find_first_bit(const unsigned long *addr,
  */
 extern unsigned long find_first_zero_bit(const unsigned long *addr,
 					 unsigned long size);
-
 #endif /* CONFIG_GENERIC_FIND_FIRST_BIT */
 
+#ifdef CONFIG_GENERIC_FIND_LAST_BIT
+/**
+ * find_last_bit - find the last set bit in a memory region
+ * @addr: The address to start the search at
+ * @size: The maximum size to search
+ *
+ * Returns the bit number of the first set bit, or size.
+ */
+extern unsigned long find_last_bit(const unsigned long *addr,
+				   unsigned long size);
+#endif /* CONFIG_GENERIC_FIND_LAST_BIT */
+
 #ifdef CONFIG_GENERIC_FIND_NEXT_BIT
 
 /**
diff --git a/lib/Kconfig b/lib/Kconfig
index 2ba43c4a5b07..fc5f5ee50bc2 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -13,6 +13,10 @@ config GENERIC_FIND_FIRST_BIT
 config GENERIC_FIND_NEXT_BIT
 	bool
 
+config GENERIC_FIND_LAST_BIT
+	bool
+	default y
+
 config CRC_CCITT
 	tristate "CRC-CCITT functions"
 	help
diff --git a/lib/Makefile b/lib/Makefile
index 80fe8a3ec12a..32b0e64ded27 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -37,6 +37,7 @@ lib-$(CONFIG_RWSEM_GENERIC_SPINLOCK) += rwsem-spinlock.o
 lib-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem.o
 lib-$(CONFIG_GENERIC_FIND_FIRST_BIT) += find_next_bit.o
 lib-$(CONFIG_GENERIC_FIND_NEXT_BIT) += find_next_bit.o
+lib-$(CONFIG_GENERIC_FIND_LAST_BIT) += find_last_bit.o
 obj-$(CONFIG_GENERIC_HWEIGHT) += hweight.o
 obj-$(CONFIG_LOCK_KERNEL) += kernel_lock.o
 obj-$(CONFIG_PLIST) += plist.o
diff --git a/lib/find_last_bit.c b/lib/find_last_bit.c
new file mode 100644
index 000000000000..5d202e36bdd8
--- /dev/null
+++ b/lib/find_last_bit.c
@@ -0,0 +1,45 @@
+/* find_last_bit.c: fallback find next bit implementation
+ *
+ * Copyright (C) 2008 IBM Corporation
+ * Written by Rusty Russell <rusty@rustcorp.com.au>
+ * (Inspired by David Howell's find_next_bit implementation)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <asm/types.h>
+#include <asm/byteorder.h>
+
+unsigned long find_last_bit(const unsigned long *addr, unsigned long size)
+{
+	unsigned long words;
+	unsigned long tmp;
+
+	/* Start at final word. */
+	words = size / BITS_PER_LONG;
+
+	/* Partial final word? */
+	if (size & (BITS_PER_LONG-1)) {
+		tmp = (addr[words] & (~0UL >> (BITS_PER_LONG
+					 - (size & (BITS_PER_LONG-1)))));
+		if (tmp)
+			goto found;
+	}
+
+	while (words) {
+		tmp = addr[--words];
+		if (tmp) {
+found:
+			return words * BITS_PER_LONG + __fls(tmp);
+		}
+	}
+
+	/* Not found */
+	return size;
+}
+EXPORT_SYMBOL(find_last_bit);
-- 
cgit v1.2.3


From 6b954823c24f04ed026a8517f6bab5abda279db8 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 1 Jan 2009 10:12:25 +1030
Subject: cpumask: convert kernel time functions

Impact: Use new APIs

Convert kernel/time functions to use struct cpumask *.

Note the ugly bitmap declarations in tick-broadcast.c.  These should
be cpumask_var_t, but there was no obvious initialization function to
put the alloc_cpumask_var() calls in.  This was safe.

(Eventually 'struct cpumask' will be undefined for CONFIG_CPUMASK_OFFSTACK,
so we use a bitmap here to show we really mean it).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
---
 include/linux/tick.h         |   4 +-
 kernel/time/clocksource.c    |   6 +--
 kernel/time/tick-broadcast.c | 113 ++++++++++++++++++++++---------------------
 kernel/time/tick-common.c    |   6 +--
 4 files changed, 66 insertions(+), 63 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/tick.h b/include/linux/tick.h
index b6ec8189ac0c..469b82d88b3b 100644
--- a/include/linux/tick.h
+++ b/include/linux/tick.h
@@ -84,10 +84,10 @@ static inline void tick_cancel_sched_timer(int cpu) { }
 
 # ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST
 extern struct tick_device *tick_get_broadcast_device(void);
-extern cpumask_t *tick_get_broadcast_mask(void);
+extern struct cpumask *tick_get_broadcast_mask(void);
 
 #  ifdef CONFIG_TICK_ONESHOT
-extern cpumask_t *tick_get_broadcast_oneshot_mask(void);
+extern struct cpumask *tick_get_broadcast_oneshot_mask(void);
 #  endif
 
 # endif /* BROADCAST */
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index 9ed2eec97526..32141b15d63e 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -148,7 +148,7 @@ static void clocksource_watchdog(unsigned long data)
 		int next_cpu = next_cpu_nr(raw_smp_processor_id(), cpu_online_map);
 
 		if (next_cpu >= nr_cpu_ids)
-			next_cpu = first_cpu(cpu_online_map);
+			next_cpu = cpumask_first(cpu_online_mask);
 		watchdog_timer.expires += WATCHDOG_INTERVAL;
 		add_timer_on(&watchdog_timer, next_cpu);
 	}
@@ -173,7 +173,7 @@ static void clocksource_check_watchdog(struct clocksource *cs)
 			watchdog_last = watchdog->read();
 			watchdog_timer.expires = jiffies + WATCHDOG_INTERVAL;
 			add_timer_on(&watchdog_timer,
-				     first_cpu(cpu_online_map));
+				     cpumask_first(cpu_online_mask));
 		}
 	} else {
 		if (cs->flags & CLOCK_SOURCE_IS_CONTINUOUS)
@@ -195,7 +195,7 @@ static void clocksource_check_watchdog(struct clocksource *cs)
 				watchdog_timer.expires =
 					jiffies + WATCHDOG_INTERVAL;
 				add_timer_on(&watchdog_timer,
-					     first_cpu(cpu_online_map));
+					     cpumask_first(cpu_online_mask));
 			}
 		}
 	}
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index 9590af2327be..356fac57a182 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -28,7 +28,9 @@
  */
 
 struct tick_device tick_broadcast_device;
-static cpumask_t tick_broadcast_mask;
+/* FIXME: Use cpumask_var_t. */
+static DECLARE_BITMAP(tick_broadcast_mask, NR_CPUS);
+static DECLARE_BITMAP(tmpmask, NR_CPUS);
 static DEFINE_SPINLOCK(tick_broadcast_lock);
 static int tick_broadcast_force;
 
@@ -46,9 +48,9 @@ struct tick_device *tick_get_broadcast_device(void)
 	return &tick_broadcast_device;
 }
 
-cpumask_t *tick_get_broadcast_mask(void)
+struct cpumask *tick_get_broadcast_mask(void)
 {
-	return &tick_broadcast_mask;
+	return to_cpumask(tick_broadcast_mask);
 }
 
 /*
@@ -72,7 +74,7 @@ int tick_check_broadcast_device(struct clock_event_device *dev)
 
 	clockevents_exchange_device(NULL, dev);
 	tick_broadcast_device.evtdev = dev;
-	if (!cpus_empty(tick_broadcast_mask))
+	if (!cpumask_empty(tick_get_broadcast_mask()))
 		tick_broadcast_start_periodic(dev);
 	return 1;
 }
@@ -104,7 +106,7 @@ int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu)
 	 */
 	if (!tick_device_is_functional(dev)) {
 		dev->event_handler = tick_handle_periodic;
-		cpu_set(cpu, tick_broadcast_mask);
+		cpumask_set_cpu(cpu, tick_get_broadcast_mask());
 		tick_broadcast_start_periodic(tick_broadcast_device.evtdev);
 		ret = 1;
 	} else {
@@ -116,7 +118,7 @@ int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu)
 		if (!(dev->features & CLOCK_EVT_FEAT_C3STOP)) {
 			int cpu = smp_processor_id();
 
-			cpu_clear(cpu, tick_broadcast_mask);
+			cpumask_clear_cpu(cpu, tick_get_broadcast_mask());
 			tick_broadcast_clear_oneshot(cpu);
 		}
 	}
@@ -125,9 +127,9 @@ int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu)
 }
 
 /*
- * Broadcast the event to the cpus, which are set in the mask
+ * Broadcast the event to the cpus, which are set in the mask (mangled).
  */
-static void tick_do_broadcast(cpumask_t mask)
+static void tick_do_broadcast(struct cpumask *mask)
 {
 	int cpu = smp_processor_id();
 	struct tick_device *td;
@@ -135,22 +137,21 @@ static void tick_do_broadcast(cpumask_t mask)
 	/*
 	 * Check, if the current cpu is in the mask
 	 */
-	if (cpu_isset(cpu, mask)) {
-		cpu_clear(cpu, mask);
+	if (cpumask_test_cpu(cpu, mask)) {
+		cpumask_clear_cpu(cpu, mask);
 		td = &per_cpu(tick_cpu_device, cpu);
 		td->evtdev->event_handler(td->evtdev);
 	}
 
-	if (!cpus_empty(mask)) {
+	if (!cpumask_empty(mask)) {
 		/*
 		 * It might be necessary to actually check whether the devices
 		 * have different broadcast functions. For now, just use the
 		 * one of the first device. This works as long as we have this
 		 * misfeature only on x86 (lapic)
 		 */
-		cpu = first_cpu(mask);
-		td = &per_cpu(tick_cpu_device, cpu);
-		td->evtdev->broadcast(&mask);
+		td = &per_cpu(tick_cpu_device, cpumask_first(mask));
+		td->evtdev->broadcast(mask);
 	}
 }
 
@@ -160,12 +161,11 @@ static void tick_do_broadcast(cpumask_t mask)
  */
 static void tick_do_periodic_broadcast(void)
 {
-	cpumask_t mask;
-
 	spin_lock(&tick_broadcast_lock);
 
-	cpus_and(mask, cpu_online_map, tick_broadcast_mask);
-	tick_do_broadcast(mask);
+	cpumask_and(to_cpumask(tmpmask),
+		    cpu_online_mask, tick_get_broadcast_mask());
+	tick_do_broadcast(to_cpumask(tmpmask));
 
 	spin_unlock(&tick_broadcast_lock);
 }
@@ -228,13 +228,13 @@ static void tick_do_broadcast_on_off(void *why)
 	if (!tick_device_is_functional(dev))
 		goto out;
 
-	bc_stopped = cpus_empty(tick_broadcast_mask);
+	bc_stopped = cpumask_empty(tick_get_broadcast_mask());
 
 	switch (*reason) {
 	case CLOCK_EVT_NOTIFY_BROADCAST_ON:
 	case CLOCK_EVT_NOTIFY_BROADCAST_FORCE:
-		if (!cpu_isset(cpu, tick_broadcast_mask)) {
-			cpu_set(cpu, tick_broadcast_mask);
+		if (!cpumask_test_cpu(cpu, tick_get_broadcast_mask())) {
+			cpumask_set_cpu(cpu, tick_get_broadcast_mask());
 			if (tick_broadcast_device.mode ==
 			    TICKDEV_MODE_PERIODIC)
 				clockevents_shutdown(dev);
@@ -244,8 +244,8 @@ static void tick_do_broadcast_on_off(void *why)
 		break;
 	case CLOCK_EVT_NOTIFY_BROADCAST_OFF:
 		if (!tick_broadcast_force &&
-		    cpu_isset(cpu, tick_broadcast_mask)) {
-			cpu_clear(cpu, tick_broadcast_mask);
+		    cpumask_test_cpu(cpu, tick_get_broadcast_mask())) {
+			cpumask_clear_cpu(cpu, tick_get_broadcast_mask());
 			if (tick_broadcast_device.mode ==
 			    TICKDEV_MODE_PERIODIC)
 				tick_setup_periodic(dev, 0);
@@ -253,7 +253,7 @@ static void tick_do_broadcast_on_off(void *why)
 		break;
 	}
 
-	if (cpus_empty(tick_broadcast_mask)) {
+	if (cpumask_empty(tick_get_broadcast_mask())) {
 		if (!bc_stopped)
 			clockevents_shutdown(bc);
 	} else if (bc_stopped) {
@@ -272,7 +272,7 @@ out:
  */
 void tick_broadcast_on_off(unsigned long reason, int *oncpu)
 {
-	if (!cpu_isset(*oncpu, cpu_online_map))
+	if (!cpumask_test_cpu(*oncpu, cpu_online_mask))
 		printk(KERN_ERR "tick-broadcast: ignoring broadcast for "
 		       "offline CPU #%d\n", *oncpu);
 	else
@@ -303,10 +303,10 @@ void tick_shutdown_broadcast(unsigned int *cpup)
 	spin_lock_irqsave(&tick_broadcast_lock, flags);
 
 	bc = tick_broadcast_device.evtdev;
-	cpu_clear(cpu, tick_broadcast_mask);
+	cpumask_clear_cpu(cpu, tick_get_broadcast_mask());
 
 	if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC) {
-		if (bc && cpus_empty(tick_broadcast_mask))
+		if (bc && cpumask_empty(tick_get_broadcast_mask()))
 			clockevents_shutdown(bc);
 	}
 
@@ -342,10 +342,10 @@ int tick_resume_broadcast(void)
 
 		switch (tick_broadcast_device.mode) {
 		case TICKDEV_MODE_PERIODIC:
-			if(!cpus_empty(tick_broadcast_mask))
+			if (!cpumask_empty(tick_get_broadcast_mask()))
 				tick_broadcast_start_periodic(bc);
-			broadcast = cpu_isset(smp_processor_id(),
-					      tick_broadcast_mask);
+			broadcast = cpumask_test_cpu(smp_processor_id(),
+						     tick_get_broadcast_mask());
 			break;
 		case TICKDEV_MODE_ONESHOT:
 			broadcast = tick_resume_broadcast_oneshot(bc);
@@ -360,14 +360,15 @@ int tick_resume_broadcast(void)
 
 #ifdef CONFIG_TICK_ONESHOT
 
-static cpumask_t tick_broadcast_oneshot_mask;
+/* FIXME: use cpumask_var_t. */
+static DECLARE_BITMAP(tick_broadcast_oneshot_mask, NR_CPUS);
 
 /*
- * Debugging: see timer_list.c
+ * Exposed for debugging: see timer_list.c
  */
-cpumask_t *tick_get_broadcast_oneshot_mask(void)
+struct cpumask *tick_get_broadcast_oneshot_mask(void)
 {
-	return &tick_broadcast_oneshot_mask;
+	return to_cpumask(tick_broadcast_oneshot_mask);
 }
 
 static int tick_broadcast_set_event(ktime_t expires, int force)
@@ -389,7 +390,7 @@ int tick_resume_broadcast_oneshot(struct clock_event_device *bc)
  */
 void tick_check_oneshot_broadcast(int cpu)
 {
-	if (cpu_isset(cpu, tick_broadcast_oneshot_mask)) {
+	if (cpumask_test_cpu(cpu, to_cpumask(tick_broadcast_oneshot_mask))) {
 		struct tick_device *td = &per_cpu(tick_cpu_device, cpu);
 
 		clockevents_set_mode(td->evtdev, CLOCK_EVT_MODE_ONESHOT);
@@ -402,7 +403,6 @@ void tick_check_oneshot_broadcast(int cpu)
 static void tick_handle_oneshot_broadcast(struct clock_event_device *dev)
 {
 	struct tick_device *td;
-	cpumask_t mask;
 	ktime_t now, next_event;
 	int cpu;
 
@@ -410,13 +410,13 @@ static void tick_handle_oneshot_broadcast(struct clock_event_device *dev)
 again:
 	dev->next_event.tv64 = KTIME_MAX;
 	next_event.tv64 = KTIME_MAX;
-	mask = CPU_MASK_NONE;
+	cpumask_clear(to_cpumask(tmpmask));
 	now = ktime_get();
 	/* Find all expired events */
-	for_each_cpu_mask_nr(cpu, tick_broadcast_oneshot_mask) {
+	for_each_cpu(cpu, tick_get_broadcast_oneshot_mask()) {
 		td = &per_cpu(tick_cpu_device, cpu);
 		if (td->evtdev->next_event.tv64 <= now.tv64)
-			cpu_set(cpu, mask);
+			cpumask_set_cpu(cpu, to_cpumask(tmpmask));
 		else if (td->evtdev->next_event.tv64 < next_event.tv64)
 			next_event.tv64 = td->evtdev->next_event.tv64;
 	}
@@ -424,7 +424,7 @@ again:
 	/*
 	 * Wakeup the cpus which have an expired event.
 	 */
-	tick_do_broadcast(mask);
+	tick_do_broadcast(to_cpumask(tmpmask));
 
 	/*
 	 * Two reasons for reprogram:
@@ -476,15 +476,16 @@ void tick_broadcast_oneshot_control(unsigned long reason)
 		goto out;
 
 	if (reason == CLOCK_EVT_NOTIFY_BROADCAST_ENTER) {
-		if (!cpu_isset(cpu, tick_broadcast_oneshot_mask)) {
-			cpu_set(cpu, tick_broadcast_oneshot_mask);
+		if (!cpumask_test_cpu(cpu, tick_get_broadcast_oneshot_mask())) {
+			cpumask_set_cpu(cpu, tick_get_broadcast_oneshot_mask());
 			clockevents_set_mode(dev, CLOCK_EVT_MODE_SHUTDOWN);
 			if (dev->next_event.tv64 < bc->next_event.tv64)
 				tick_broadcast_set_event(dev->next_event, 1);
 		}
 	} else {
-		if (cpu_isset(cpu, tick_broadcast_oneshot_mask)) {
-			cpu_clear(cpu, tick_broadcast_oneshot_mask);
+		if (cpumask_test_cpu(cpu, tick_get_broadcast_oneshot_mask())) {
+			cpumask_clear_cpu(cpu,
+					  tick_get_broadcast_oneshot_mask());
 			clockevents_set_mode(dev, CLOCK_EVT_MODE_ONESHOT);
 			if (dev->next_event.tv64 != KTIME_MAX)
 				tick_program_event(dev->next_event, 1);
@@ -502,10 +503,11 @@ out:
  */
 static void tick_broadcast_clear_oneshot(int cpu)
 {
-	cpu_clear(cpu, tick_broadcast_oneshot_mask);
+	cpumask_clear_cpu(cpu, tick_get_broadcast_oneshot_mask());
 }
 
-static void tick_broadcast_init_next_event(cpumask_t *mask, ktime_t expires)
+static void tick_broadcast_init_next_event(struct cpumask *mask,
+					   ktime_t expires)
 {
 	struct tick_device *td;
 	int cpu;
@@ -526,7 +528,6 @@ void tick_broadcast_setup_oneshot(struct clock_event_device *bc)
 	if (bc->event_handler != tick_handle_oneshot_broadcast) {
 		int was_periodic = bc->mode == CLOCK_EVT_MODE_PERIODIC;
 		int cpu = smp_processor_id();
-		cpumask_t mask;
 
 		bc->event_handler = tick_handle_oneshot_broadcast;
 		clockevents_set_mode(bc, CLOCK_EVT_MODE_ONESHOT);
@@ -540,13 +541,15 @@ void tick_broadcast_setup_oneshot(struct clock_event_device *bc)
 		 * oneshot_mask bits for those and program the
 		 * broadcast device to fire.
 		 */
-		mask = tick_broadcast_mask;
-		cpu_clear(cpu, mask);
-		cpus_or(tick_broadcast_oneshot_mask,
-			tick_broadcast_oneshot_mask, mask);
-
-		if (was_periodic && !cpus_empty(mask)) {
-			tick_broadcast_init_next_event(&mask, tick_next_period);
+		cpumask_copy(to_cpumask(tmpmask), tick_get_broadcast_mask());
+		cpumask_clear_cpu(cpu, to_cpumask(tmpmask));
+		cpumask_or(tick_get_broadcast_oneshot_mask(),
+			   tick_get_broadcast_oneshot_mask(),
+			   to_cpumask(tmpmask));
+
+		if (was_periodic && !cpumask_empty(to_cpumask(tmpmask))) {
+			tick_broadcast_init_next_event(to_cpumask(tmpmask),
+						       tick_next_period);
 			tick_broadcast_set_event(tick_next_period, 1);
 		} else
 			bc->next_event.tv64 = KTIME_MAX;
@@ -585,7 +588,7 @@ void tick_shutdown_broadcast_oneshot(unsigned int *cpup)
 	 * Clear the broadcast mask flag for the dead cpu, but do not
 	 * stop the broadcast device!
 	 */
-	cpu_clear(cpu, tick_broadcast_oneshot_mask);
+	cpumask_clear_cpu(cpu, tick_get_broadcast_oneshot_mask());
 
 	spin_unlock_irqrestore(&tick_broadcast_lock, flags);
 }
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index f8372be74122..63e05d423a09 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -254,7 +254,7 @@ static int tick_check_new_device(struct clock_event_device *newdev)
 		curdev = NULL;
 	}
 	clockevents_exchange_device(curdev, newdev);
-	tick_setup_device(td, newdev, cpu, &cpumask_of_cpu(cpu));
+	tick_setup_device(td, newdev, cpu, cpumask_of(cpu));
 	if (newdev->features & CLOCK_EVT_FEAT_ONESHOT)
 		tick_oneshot_notify();
 
@@ -299,9 +299,9 @@ static void tick_shutdown(unsigned int *cpup)
 	}
 	/* Transfer the do_timer job away from this cpu */
 	if (*cpup == tick_do_timer_cpu) {
-		int cpu = first_cpu(cpu_online_map);
+		int cpu = cpumask_first(cpu_online_mask);
 
-		tick_do_timer_cpu = (cpu != NR_CPUS) ? cpu :
+		tick_do_timer_cpu = (cpu < nr_cpu_ids) ? cpu :
 			TICK_DO_TIMER_NONE;
 	}
 	spin_unlock_irqrestore(&tick_device_lock, flags);
-- 
cgit v1.2.3


From d036e67b40f52bdd95392390108defbac7e53837 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 1 Jan 2009 10:12:26 +1030
Subject: cpumask: convert kernel/irq

Impact: Reduce stack usage, use new cpumask API.  ALPHA mod!

Main change is that irq_default_affinity becomes a cpumask_var_t, so
treat it as a pointer (this effects alpha).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 arch/alpha/kernel/irq.c   |  3 ++-
 include/linux/interrupt.h |  2 +-
 kernel/irq/manage.c       | 11 +++++++++--
 kernel/irq/proc.c         | 32 +++++++++++++++++++++-----------
 4 files changed, 33 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/arch/alpha/kernel/irq.c b/arch/alpha/kernel/irq.c
index d0f1620007f7..703731accda6 100644
--- a/arch/alpha/kernel/irq.c
+++ b/arch/alpha/kernel/irq.c
@@ -50,7 +50,8 @@ int irq_select_affinity(unsigned int irq)
 	if (!irq_desc[irq].chip->set_affinity || irq_user_affinity[irq])
 		return 1;
 
-	while (!cpu_possible(cpu) || !cpu_isset(cpu, irq_default_affinity))
+	while (!cpu_possible(cpu) ||
+	       !cpumask_test_cpu(cpu, irq_default_affinity))
 		cpu = (cpu < (NR_CPUS-1) ? cpu + 1 : 0);
 	last_cpu = cpu;
 
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index dfaee6bd265b..91f1ef8e5810 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -109,7 +109,7 @@ extern void enable_irq(unsigned int irq);
 
 #if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS)
 
-extern cpumask_t irq_default_affinity;
+extern cpumask_var_t irq_default_affinity;
 
 extern int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask);
 extern int irq_can_set_affinity(unsigned int irq);
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 61c4a9b62165..cd0cd8dcb345 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -16,8 +16,15 @@
 #include "internals.h"
 
 #ifdef CONFIG_SMP
+cpumask_var_t irq_default_affinity;
 
-cpumask_t irq_default_affinity = CPU_MASK_ALL;
+static int init_irq_default_affinity(void)
+{
+	alloc_cpumask_var(&irq_default_affinity, GFP_KERNEL);
+	cpumask_setall(irq_default_affinity);
+	return 0;
+}
+core_initcall(init_irq_default_affinity);
 
 /**
  *	synchronize_irq - wait for pending IRQ handlers (on other CPUs)
@@ -127,7 +134,7 @@ int do_irq_select_affinity(unsigned int irq, struct irq_desc *desc)
 			desc->status &= ~IRQ_AFFINITY_SET;
 	}
 
-	cpumask_and(&desc->affinity, cpu_online_mask, &irq_default_affinity);
+	cpumask_and(&desc->affinity, cpu_online_mask, irq_default_affinity);
 set_affinity:
 	desc->chip->set_affinity(irq, &desc->affinity);
 
diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c
index d2c0e5ee53c5..2abd3a7716ed 100644
--- a/kernel/irq/proc.c
+++ b/kernel/irq/proc.c
@@ -20,7 +20,7 @@ static struct proc_dir_entry *root_irq_dir;
 static int irq_affinity_proc_show(struct seq_file *m, void *v)
 {
 	struct irq_desc *desc = irq_to_desc((long)m->private);
-	cpumask_t *mask = &desc->affinity;
+	const struct cpumask *mask = &desc->affinity;
 
 #ifdef CONFIG_GENERIC_PENDING_IRQ
 	if (desc->status & IRQ_MOVE_PENDING)
@@ -93,7 +93,7 @@ static const struct file_operations irq_affinity_proc_fops = {
 
 static int default_affinity_show(struct seq_file *m, void *v)
 {
-	seq_cpumask(m, &irq_default_affinity);
+	seq_cpumask(m, irq_default_affinity);
 	seq_putc(m, '\n');
 	return 0;
 }
@@ -101,27 +101,37 @@ static int default_affinity_show(struct seq_file *m, void *v)
 static ssize_t default_affinity_write(struct file *file,
 		const char __user *buffer, size_t count, loff_t *ppos)
 {
-	cpumask_t new_value;
+	cpumask_var_t new_value;
 	int err;
 
-	err = cpumask_parse_user(buffer, count, &new_value);
+	if (!alloc_cpumask_var(&new_value, GFP_KERNEL))
+		return -ENOMEM;
+
+	err = cpumask_parse_user(buffer, count, new_value);
 	if (err)
-		return err;
+		goto out;
 
-	if (!is_affinity_mask_valid(new_value))
-		return -EINVAL;
+	if (!is_affinity_mask_valid(new_value)) {
+		err = -EINVAL;
+		goto out;
+	}
 
 	/*
 	 * Do not allow disabling IRQs completely - it's a too easy
 	 * way to make the system unusable accidentally :-) At least
 	 * one online CPU still has to be targeted.
 	 */
-	if (!cpus_intersects(new_value, cpu_online_map))
-		return -EINVAL;
+	if (!cpumask_intersects(new_value, cpu_online_mask)) {
+		err = -EINVAL;
+		goto out;
+	}
 
-	irq_default_affinity = new_value;
+	cpumask_copy(irq_default_affinity, new_value);
+	err = count;
 
-	return count;
+out:
+	free_cpumask_var(new_value);
+	return err;
 }
 
 static int default_affinity_open(struct inode *inode, struct file *file)
-- 
cgit v1.2.3


From bd232f97b30f6bb630efa136a777647545db3039 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 1 Jan 2009 10:12:26 +1030
Subject: cpumask: convert RCU implementations

Impact: use new cpumask API.

rcu_ctrlblk contains a cpumask, and it's highly optimized so I don't want
a cpumask_var_t (ie. a pointer) for the CONFIG_CPUMASK_OFFSTACK case.  It
could use a dangling bitmap, and be allocated in __rcu_init to save memory,
but for the moment we use a bitmap.

(Eventually 'struct cpumask' will be undefined for CONFIG_CPUMASK_OFFSTACK,
so we use a bitmap here to show we really mean it).

We remove on-stack cpumasks, using cpumask_var_t for
rcu_torture_shuffle_tasks() and for_each_cpu_and in force_quiescent_state().

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 include/linux/rcuclassic.h |  4 ++--
 kernel/rcuclassic.c        | 32 +++++++++++++++++---------------
 kernel/rcupreempt.c        | 19 ++++++++++---------
 kernel/rcutorture.c        | 27 +++++++++++++++------------
 4 files changed, 44 insertions(+), 38 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rcuclassic.h b/include/linux/rcuclassic.h
index 301dda829e37..f3f697df1d71 100644
--- a/include/linux/rcuclassic.h
+++ b/include/linux/rcuclassic.h
@@ -59,8 +59,8 @@ struct rcu_ctrlblk {
 	int	signaled;
 
 	spinlock_t	lock	____cacheline_internodealigned_in_smp;
-	cpumask_t	cpumask; /* CPUs that need to switch in order    */
-				 /* for current batch to proceed.        */
+	DECLARE_BITMAP(cpumask, NR_CPUS); /* CPUs that need to switch for */
+					  /* current batch to proceed.     */
 } ____cacheline_internodealigned_in_smp;
 
 /* Is batch a before batch b ? */
diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c
index e503a002f330..0ff9b05706a6 100644
--- a/kernel/rcuclassic.c
+++ b/kernel/rcuclassic.c
@@ -63,14 +63,14 @@ static struct rcu_ctrlblk rcu_ctrlblk = {
 	.completed = -300,
 	.pending = -300,
 	.lock = __SPIN_LOCK_UNLOCKED(&rcu_ctrlblk.lock),
-	.cpumask = CPU_MASK_NONE,
+	.cpumask = CPU_BITS_NONE,
 };
 static struct rcu_ctrlblk rcu_bh_ctrlblk = {
 	.cur = -300,
 	.completed = -300,
 	.pending = -300,
 	.lock = __SPIN_LOCK_UNLOCKED(&rcu_bh_ctrlblk.lock),
-	.cpumask = CPU_MASK_NONE,
+	.cpumask = CPU_BITS_NONE,
 };
 
 DEFINE_PER_CPU(struct rcu_data, rcu_data) = { 0L };
@@ -85,7 +85,6 @@ static void force_quiescent_state(struct rcu_data *rdp,
 			struct rcu_ctrlblk *rcp)
 {
 	int cpu;
-	cpumask_t cpumask;
 	unsigned long flags;
 
 	set_need_resched();
@@ -96,10 +95,10 @@ static void force_quiescent_state(struct rcu_data *rdp,
 		 * Don't send IPI to itself. With irqs disabled,
 		 * rdp->cpu is the current cpu.
 		 *
-		 * cpu_online_map is updated by the _cpu_down()
+		 * cpu_online_mask is updated by the _cpu_down()
 		 * using __stop_machine(). Since we're in irqs disabled
 		 * section, __stop_machine() is not exectuting, hence
-		 * the cpu_online_map is stable.
+		 * the cpu_online_mask is stable.
 		 *
 		 * However,  a cpu might have been offlined _just_ before
 		 * we disabled irqs while entering here.
@@ -107,13 +106,14 @@ static void force_quiescent_state(struct rcu_data *rdp,
 		 * notification, leading to the offlined cpu's bit
 		 * being set in the rcp->cpumask.
 		 *
-		 * Hence cpumask = (rcp->cpumask & cpu_online_map) to prevent
+		 * Hence cpumask = (rcp->cpumask & cpu_online_mask) to prevent
 		 * sending smp_reschedule() to an offlined CPU.
 		 */
-		cpus_and(cpumask, rcp->cpumask, cpu_online_map);
-		cpu_clear(rdp->cpu, cpumask);
-		for_each_cpu_mask_nr(cpu, cpumask)
-			smp_send_reschedule(cpu);
+		for_each_cpu_and(cpu,
+				  to_cpumask(rcp->cpumask), cpu_online_mask) {
+			if (cpu != rdp->cpu)
+				smp_send_reschedule(cpu);
+		}
 	}
 	spin_unlock_irqrestore(&rcp->lock, flags);
 }
@@ -193,7 +193,7 @@ static void print_other_cpu_stall(struct rcu_ctrlblk *rcp)
 
 	printk(KERN_ERR "INFO: RCU detected CPU stalls:");
 	for_each_possible_cpu(cpu) {
-		if (cpu_isset(cpu, rcp->cpumask))
+		if (cpumask_test_cpu(cpu, to_cpumask(rcp->cpumask)))
 			printk(" %d", cpu);
 	}
 	printk(" (detected by %d, t=%ld jiffies)\n",
@@ -221,7 +221,8 @@ static void check_cpu_stall(struct rcu_ctrlblk *rcp)
 	long delta;
 
 	delta = jiffies - rcp->jiffies_stall;
-	if (cpu_isset(smp_processor_id(), rcp->cpumask) && delta >= 0) {
+	if (cpumask_test_cpu(smp_processor_id(), to_cpumask(rcp->cpumask)) &&
+		delta >= 0) {
 
 		/* We haven't checked in, so go dump stack. */
 		print_cpu_stall(rcp);
@@ -393,7 +394,8 @@ static void rcu_start_batch(struct rcu_ctrlblk *rcp)
 		 * unnecessarily.
 		 */
 		smp_mb();
-		cpus_andnot(rcp->cpumask, cpu_online_map, nohz_cpu_mask);
+		cpumask_andnot(to_cpumask(rcp->cpumask),
+			       cpu_online_mask, &nohz_cpu_mask);
 
 		rcp->signaled = 0;
 	}
@@ -406,8 +408,8 @@ static void rcu_start_batch(struct rcu_ctrlblk *rcp)
  */
 static void cpu_quiet(int cpu, struct rcu_ctrlblk *rcp)
 {
-	cpu_clear(cpu, rcp->cpumask);
-	if (cpus_empty(rcp->cpumask)) {
+	cpumask_clear_cpu(cpu, to_cpumask(rcp->cpumask));
+	if (cpumask_empty(to_cpumask(rcp->cpumask))) {
 		/* batch completed ! */
 		rcp->completed = rcp->cur;
 		rcu_start_batch(rcp);
diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c
index 04982659875a..f9dc8f3720f6 100644
--- a/kernel/rcupreempt.c
+++ b/kernel/rcupreempt.c
@@ -164,7 +164,8 @@ static char *rcu_try_flip_state_names[] =
 	{ "idle", "waitack", "waitzero", "waitmb" };
 #endif /* #ifdef CONFIG_RCU_TRACE */
 
-static cpumask_t rcu_cpu_online_map __read_mostly = CPU_MASK_NONE;
+static DECLARE_BITMAP(rcu_cpu_online_map, NR_CPUS) __read_mostly
+	= CPU_BITS_NONE;
 
 /*
  * Enum and per-CPU flag to determine when each CPU has seen
@@ -758,7 +759,7 @@ rcu_try_flip_idle(void)
 
 	/* Now ask each CPU for acknowledgement of the flip. */
 
-	for_each_cpu_mask_nr(cpu, rcu_cpu_online_map) {
+	for_each_cpu(cpu, to_cpumask(rcu_cpu_online_map)) {
 		per_cpu(rcu_flip_flag, cpu) = rcu_flipped;
 		dyntick_save_progress_counter(cpu);
 	}
@@ -776,7 +777,7 @@ rcu_try_flip_waitack(void)
 	int cpu;
 
 	RCU_TRACE_ME(rcupreempt_trace_try_flip_a1);
-	for_each_cpu_mask_nr(cpu, rcu_cpu_online_map)
+	for_each_cpu(cpu, to_cpumask(rcu_cpu_online_map))
 		if (rcu_try_flip_waitack_needed(cpu) &&
 		    per_cpu(rcu_flip_flag, cpu) != rcu_flip_seen) {
 			RCU_TRACE_ME(rcupreempt_trace_try_flip_ae1);
@@ -808,7 +809,7 @@ rcu_try_flip_waitzero(void)
 	/* Check to see if the sum of the "last" counters is zero. */
 
 	RCU_TRACE_ME(rcupreempt_trace_try_flip_z1);
-	for_each_cpu_mask_nr(cpu, rcu_cpu_online_map)
+	for_each_cpu(cpu, to_cpumask(rcu_cpu_online_map))
 		sum += RCU_DATA_CPU(cpu)->rcu_flipctr[lastidx];
 	if (sum != 0) {
 		RCU_TRACE_ME(rcupreempt_trace_try_flip_ze1);
@@ -823,7 +824,7 @@ rcu_try_flip_waitzero(void)
 	smp_mb();  /*  ^^^^^^^^^^^^ */
 
 	/* Call for a memory barrier from each CPU. */
-	for_each_cpu_mask_nr(cpu, rcu_cpu_online_map) {
+	for_each_cpu(cpu, to_cpumask(rcu_cpu_online_map)) {
 		per_cpu(rcu_mb_flag, cpu) = rcu_mb_needed;
 		dyntick_save_progress_counter(cpu);
 	}
@@ -843,7 +844,7 @@ rcu_try_flip_waitmb(void)
 	int cpu;
 
 	RCU_TRACE_ME(rcupreempt_trace_try_flip_m1);
-	for_each_cpu_mask_nr(cpu, rcu_cpu_online_map)
+	for_each_cpu(cpu, to_cpumask(rcu_cpu_online_map))
 		if (rcu_try_flip_waitmb_needed(cpu) &&
 		    per_cpu(rcu_mb_flag, cpu) != rcu_mb_done) {
 			RCU_TRACE_ME(rcupreempt_trace_try_flip_me1);
@@ -1032,7 +1033,7 @@ void rcu_offline_cpu(int cpu)
 	RCU_DATA_CPU(cpu)->rcu_flipctr[0] = 0;
 	RCU_DATA_CPU(cpu)->rcu_flipctr[1] = 0;
 
-	cpu_clear(cpu, rcu_cpu_online_map);
+	cpumask_clear_cpu(cpu, to_cpumask(rcu_cpu_online_map));
 
 	spin_unlock_irqrestore(&rcu_ctrlblk.fliplock, flags);
 
@@ -1072,7 +1073,7 @@ void __cpuinit rcu_online_cpu(int cpu)
 	struct rcu_data *rdp;
 
 	spin_lock_irqsave(&rcu_ctrlblk.fliplock, flags);
-	cpu_set(cpu, rcu_cpu_online_map);
+	cpumask_set_cpu(cpu, to_cpumask(rcu_cpu_online_map));
 	spin_unlock_irqrestore(&rcu_ctrlblk.fliplock, flags);
 
 	/*
@@ -1430,7 +1431,7 @@ void __init __rcu_init(void)
 	 * We don't need protection against CPU-Hotplug here
 	 * since
 	 * a) If a CPU comes online while we are iterating over the
-	 *    cpu_online_map below, we would only end up making a
+	 *    cpu_online_mask below, we would only end up making a
 	 *    duplicate call to rcu_online_cpu() which sets the corresponding
 	 *    CPU's mask in the rcu_cpu_online_map.
 	 *
diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c
index b31065522104..3245b40952c6 100644
--- a/kernel/rcutorture.c
+++ b/kernel/rcutorture.c
@@ -868,49 +868,52 @@ static int rcu_idle_cpu;	/* Force all torture tasks off this CPU */
  */
 static void rcu_torture_shuffle_tasks(void)
 {
-	cpumask_t tmp_mask;
+	cpumask_var_t tmp_mask;
 	int i;
 
-	cpus_setall(tmp_mask);
+	if (!alloc_cpumask_var(&tmp_mask, GFP_KERNEL))
+		BUG();
+
+	cpumask_setall(tmp_mask);
 	get_online_cpus();
 
 	/* No point in shuffling if there is only one online CPU (ex: UP) */
-	if (num_online_cpus() == 1) {
-		put_online_cpus();
-		return;
-	}
+	if (num_online_cpus() == 1)
+		goto out;
 
 	if (rcu_idle_cpu != -1)
-		cpu_clear(rcu_idle_cpu, tmp_mask);
+		cpumask_clear_cpu(rcu_idle_cpu, tmp_mask);
 
-	set_cpus_allowed_ptr(current, &tmp_mask);
+	set_cpus_allowed_ptr(current, tmp_mask);
 
 	if (reader_tasks) {
 		for (i = 0; i < nrealreaders; i++)
 			if (reader_tasks[i])
 				set_cpus_allowed_ptr(reader_tasks[i],
-						     &tmp_mask);
+						     tmp_mask);
 	}
 
 	if (fakewriter_tasks) {
 		for (i = 0; i < nfakewriters; i++)
 			if (fakewriter_tasks[i])
 				set_cpus_allowed_ptr(fakewriter_tasks[i],
-						     &tmp_mask);
+						     tmp_mask);
 	}
 
 	if (writer_task)
-		set_cpus_allowed_ptr(writer_task, &tmp_mask);
+		set_cpus_allowed_ptr(writer_task, tmp_mask);
 
 	if (stats_task)
-		set_cpus_allowed_ptr(stats_task, &tmp_mask);
+		set_cpus_allowed_ptr(stats_task, tmp_mask);
 
 	if (rcu_idle_cpu == -1)
 		rcu_idle_cpu = num_online_cpus() - 1;
 	else
 		rcu_idle_cpu--;
 
+out:
 	put_online_cpus();
+	free_cpumask_var(tmp_mask);
 }
 
 /* Shuffle tasks across CPUs, with the intent of allowing each CPU in the
-- 
cgit v1.2.3


From 41c7bb9588904eb060a95bcad47bd3804a1ece25 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 1 Jan 2009 10:12:28 +1030
Subject: cpumask: convert rest of files in kernel/

Impact: Reduce stack usage, use new cpumask API.

Mainly changing cpumask_t to 'struct cpumask' and similar simple API
conversion.  Two conversions worth mentioning:

1) we use cpumask_any_but to avoid a temporary in kernel/softlockup.c,
2) Use cpumask_var_t in taskstats_user_cmd().

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
---
 include/linux/stop_machine.h |  6 +++---
 kernel/power/poweroff.c      |  2 +-
 kernel/softlockup.c          |  6 ++----
 kernel/stop_machine.c        |  8 ++++----
 kernel/taskstats.c           | 39 ++++++++++++++++++++++++---------------
 5 files changed, 34 insertions(+), 27 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/stop_machine.h b/include/linux/stop_machine.h
index faf1519b5adc..74d59a641362 100644
--- a/include/linux/stop_machine.h
+++ b/include/linux/stop_machine.h
@@ -23,7 +23,7 @@
  *
  * This can be thought of as a very heavy write lock, equivalent to
  * grabbing every spinlock in the kernel. */
-int stop_machine(int (*fn)(void *), void *data, const cpumask_t *cpus);
+int stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus);
 
 /**
  * __stop_machine: freeze the machine on all CPUs and run this function
@@ -34,11 +34,11 @@ int stop_machine(int (*fn)(void *), void *data, const cpumask_t *cpus);
  * Description: This is a special version of the above, which assumes cpus
  * won't come or go while it's being called.  Used by hotplug cpu.
  */
-int __stop_machine(int (*fn)(void *), void *data, const cpumask_t *cpus);
+int __stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus);
 #else
 
 static inline int stop_machine(int (*fn)(void *), void *data,
-			       const cpumask_t *cpus)
+			       const struct cpumask *cpus)
 {
 	int ret;
 	local_irq_disable();
diff --git a/kernel/power/poweroff.c b/kernel/power/poweroff.c
index 72016f051477..97890831e1b5 100644
--- a/kernel/power/poweroff.c
+++ b/kernel/power/poweroff.c
@@ -27,7 +27,7 @@ static DECLARE_WORK(poweroff_work, do_poweroff);
 static void handle_poweroff(int key, struct tty_struct *tty)
 {
 	/* run sysrq poweroff on boot cpu */
-	schedule_work_on(first_cpu(cpu_online_map), &poweroff_work);
+	schedule_work_on(cpumask_first(cpu_online_mask), &poweroff_work);
 }
 
 static struct sysrq_key_op	sysrq_poweroff_op = {
diff --git a/kernel/softlockup.c b/kernel/softlockup.c
index 492f0c72fec5..d9188c66278a 100644
--- a/kernel/softlockup.c
+++ b/kernel/softlockup.c
@@ -310,10 +310,8 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
 	case CPU_DOWN_PREPARE:
 	case CPU_DOWN_PREPARE_FROZEN:
 		if (hotcpu == check_cpu) {
-			cpumask_t temp_cpu_online_map = cpu_online_map;
-
-			cpu_clear(hotcpu, temp_cpu_online_map);
-			check_cpu = cpumask_any(&temp_cpu_online_map);
+			/* Pick any other online cpu. */
+			check_cpu = cpumask_any_but(cpu_online_mask, hotcpu);
 		}
 		break;
 
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index 24e8ceacc388..286c41722e8c 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -69,10 +69,10 @@ static void stop_cpu(struct work_struct *unused)
 	int err;
 
 	if (!active_cpus) {
-		if (cpu == first_cpu(cpu_online_map))
+		if (cpu == cpumask_first(cpu_online_mask))
 			smdata = &active;
 	} else {
-		if (cpu_isset(cpu, *active_cpus))
+		if (cpumask_test_cpu(cpu, active_cpus))
 			smdata = &active;
 	}
 	/* Simple state machine */
@@ -109,7 +109,7 @@ static int chill(void *unused)
 	return 0;
 }
 
-int __stop_machine(int (*fn)(void *), void *data, const cpumask_t *cpus)
+int __stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus)
 {
 	struct work_struct *sm_work;
 	int i, ret;
@@ -142,7 +142,7 @@ int __stop_machine(int (*fn)(void *), void *data, const cpumask_t *cpus)
 	return ret;
 }
 
-int stop_machine(int (*fn)(void *), void *data, const cpumask_t *cpus)
+int stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus)
 {
 	int ret;
 
diff --git a/kernel/taskstats.c b/kernel/taskstats.c
index 6d7dc4ec4aa5..888adbcca30c 100644
--- a/kernel/taskstats.c
+++ b/kernel/taskstats.c
@@ -290,18 +290,17 @@ ret:
 	return;
 }
 
-static int add_del_listener(pid_t pid, cpumask_t *maskp, int isadd)
+static int add_del_listener(pid_t pid, const struct cpumask *mask, int isadd)
 {
 	struct listener_list *listeners;
 	struct listener *s, *tmp;
 	unsigned int cpu;
-	cpumask_t mask = *maskp;
 
-	if (!cpus_subset(mask, cpu_possible_map))
+	if (!cpumask_subset(mask, cpu_possible_mask))
 		return -EINVAL;
 
 	if (isadd == REGISTER) {
-		for_each_cpu_mask_nr(cpu, mask) {
+		for_each_cpu(cpu, mask) {
 			s = kmalloc_node(sizeof(struct listener), GFP_KERNEL,
 					 cpu_to_node(cpu));
 			if (!s)
@@ -320,7 +319,7 @@ static int add_del_listener(pid_t pid, cpumask_t *maskp, int isadd)
 
 	/* Deregister or cleanup */
 cleanup:
-	for_each_cpu_mask_nr(cpu, mask) {
+	for_each_cpu(cpu, mask) {
 		listeners = &per_cpu(listener_array, cpu);
 		down_write(&listeners->sem);
 		list_for_each_entry_safe(s, tmp, &listeners->list, list) {
@@ -335,7 +334,7 @@ cleanup:
 	return 0;
 }
 
-static int parse(struct nlattr *na, cpumask_t *mask)
+static int parse(struct nlattr *na, struct cpumask *mask)
 {
 	char *data;
 	int len;
@@ -428,23 +427,33 @@ err:
 
 static int taskstats_user_cmd(struct sk_buff *skb, struct genl_info *info)
 {
-	int rc = 0;
+	int rc;
 	struct sk_buff *rep_skb;
 	struct taskstats *stats;
 	size_t size;
-	cpumask_t mask;
+	cpumask_var_t mask;
+
+	if (!alloc_cpumask_var(&mask, GFP_KERNEL))
+		return -ENOMEM;
 
-	rc = parse(info->attrs[TASKSTATS_CMD_ATTR_REGISTER_CPUMASK], &mask);
+	rc = parse(info->attrs[TASKSTATS_CMD_ATTR_REGISTER_CPUMASK], mask);
 	if (rc < 0)
-		return rc;
-	if (rc == 0)
-		return add_del_listener(info->snd_pid, &mask, REGISTER);
+		goto free_return_rc;
+	if (rc == 0) {
+		rc = add_del_listener(info->snd_pid, mask, REGISTER);
+		goto free_return_rc;
+	}
 
-	rc = parse(info->attrs[TASKSTATS_CMD_ATTR_DEREGISTER_CPUMASK], &mask);
+	rc = parse(info->attrs[TASKSTATS_CMD_ATTR_DEREGISTER_CPUMASK], mask);
 	if (rc < 0)
+		goto free_return_rc;
+	if (rc == 0) {
+		rc = add_del_listener(info->snd_pid, mask, DEREGISTER);
+free_return_rc:
+		free_cpumask_var(mask);
 		return rc;
-	if (rc == 0)
-		return add_del_listener(info->snd_pid, &mask, DEREGISTER);
+	}
+	free_cpumask_var(mask);
 
 	/*
 	 * Size includes space for nested attributes
-- 
cgit v1.2.3


From 8c384cdee3e04d6194a2c2b192b624754f990835 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 1 Jan 2009 10:12:30 +1030
Subject: cpumask: CONFIG_DISABLE_OBSOLETE_CPUMASK_FUNCTIONS

Impact: new debug CONFIG options

This helps find unconverted code.  It currently breaks compile horribly,
but we never wanted a flag day so that's expected.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 include/linux/cpumask.h | 11 ++++++++++-
 lib/Kconfig             |  4 ++++
 2 files changed, 14 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 7c178a6baae3..9f315382610b 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -144,6 +144,7 @@
 typedef struct cpumask { DECLARE_BITMAP(bits, NR_CPUS); } cpumask_t;
 extern cpumask_t _unused_cpumask_arg_;
 
+#ifndef CONFIG_DISABLE_OBSOLETE_CPUMASK_FUNCTIONS
 #define cpu_set(cpu, dst) __cpu_set((cpu), &(dst))
 static inline void __cpu_set(int cpu, volatile cpumask_t *dstp)
 {
@@ -267,6 +268,7 @@ static inline void __cpus_shift_left(cpumask_t *dstp,
 {
 	bitmap_shift_left(dstp->bits, srcp->bits, n, nbits);
 }
+#endif /* !CONFIG_DISABLE_OBSOLETE_CPUMASK_FUNCTIONS */
 
 /**
  * to_cpumask - convert an NR_CPUS bitmap to a struct cpumask *
@@ -304,6 +306,7 @@ static inline const struct cpumask *get_cpu_mask(unsigned int cpu)
 	return to_cpumask(p);
 }
 
+#ifndef CONFIG_DISABLE_OBSOLETE_CPUMASK_FUNCTIONS
 /*
  * In cases where we take the address of the cpumask immediately,
  * gcc optimizes it out (it's a constant) and there's no huge stack
@@ -389,19 +392,22 @@ static inline void __cpus_fold(cpumask_t *dstp, const cpumask_t *origp,
 {
 	bitmap_fold(dstp->bits, origp->bits, sz, nbits);
 }
+#endif /* !CONFIG_DISABLE_OBSOLETE_CPUMASK_FUNCTIONS */
 
 #if NR_CPUS == 1
 
 #define nr_cpu_ids		1
+#ifndef CONFIG_DISABLE_OBSOLETE_CPUMASK_FUNCTIONS
 #define first_cpu(src)		({ (void)(src); 0; })
 #define next_cpu(n, src)	({ (void)(src); 1; })
 #define any_online_cpu(mask)	0
 #define for_each_cpu_mask(cpu, mask)	\
 	for ((cpu) = 0; (cpu) < 1; (cpu)++, (void)mask)
-
+#endif /* !CONFIG_DISABLE_OBSOLETE_CPUMASK_FUNCTIONS */
 #else /* NR_CPUS > 1 */
 
 extern int nr_cpu_ids;
+#ifndef CONFIG_DISABLE_OBSOLETE_CPUMASK_FUNCTIONS
 int __first_cpu(const cpumask_t *srcp);
 int __next_cpu(int n, const cpumask_t *srcp);
 int __any_online_cpu(const cpumask_t *mask);
@@ -413,8 +419,10 @@ int __any_online_cpu(const cpumask_t *mask);
 	for ((cpu) = -1;				\
 		(cpu) = next_cpu((cpu), (mask)),	\
 		(cpu) < NR_CPUS; )
+#endif /* !CONFIG_DISABLE_OBSOLETE_CPUMASK_FUNCTIONS */
 #endif
 
+#ifndef CONFIG_DISABLE_OBSOLETE_CPUMASK_FUNCTIONS
 #if NR_CPUS <= 64
 
 #define next_cpu_nr(n, src)		next_cpu(n, src)
@@ -432,6 +440,7 @@ int __next_cpu_nr(int n, const cpumask_t *srcp);
 		(cpu) < nr_cpu_ids; )
 
 #endif /* NR_CPUS > 64 */
+#endif /* !CONFIG_DISABLE_OBSOLETE_CPUMASK_FUNCTIONS */
 
 /*
  * The following particular system cpumasks and operations manage
diff --git a/lib/Kconfig b/lib/Kconfig
index fc5f5ee50bc2..03c2c24b9083 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -170,4 +170,8 @@ config CPUMASK_OFFSTACK
 	  them on the stack.  This is a bit more expensive, but avoids
 	  stack overflow.
 
+config DISABLE_OBSOLETE_CPUMASK_FUNCTIONS
+       bool "Disable obsolete cpumask functions" if DEBUG_PER_CPU_MAPS
+       depends on EXPERIMENTAL && BROKEN
+
 endmenu
-- 
cgit v1.2.3


From ebdab07dad3d3a008e519b0a028e1e1ad5ecaef0 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 2 Jan 2009 16:12:48 +0100
Subject: ide: move sysfs support to ide-sysfs.c

While at it:
- media_string() -> ide_media_string()

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/Makefile    |   2 +-
 drivers/ide/ide-probe.c |  52 --------------------
 drivers/ide/ide-sysfs.c | 125 ++++++++++++++++++++++++++++++++++++++++++++++++
 drivers/ide/ide.c       |  72 +---------------------------
 include/linux/ide.h     |   4 ++
 5 files changed, 132 insertions(+), 123 deletions(-)
 create mode 100644 drivers/ide/ide-sysfs.c

(limited to 'include/linux')

diff --git a/drivers/ide/Makefile b/drivers/ide/Makefile
index 177e3f8523ed..410728992e6a 100644
--- a/drivers/ide/Makefile
+++ b/drivers/ide/Makefile
@@ -5,7 +5,7 @@
 EXTRA_CFLAGS				+= -Idrivers/ide
 
 ide-core-y += ide.o ide-ioctls.o ide-io.o ide-iops.o ide-lib.o ide-probe.o \
-	      ide-taskfile.o ide-pm.o ide-park.o ide-pio-blacklist.o
+	      ide-taskfile.o ide-pm.o ide-park.o ide-pio-blacklist.o ide-sysfs.o
 
 # core IDE code
 ide-core-$(CONFIG_IDE_TIMINGS)		+= ide-timings.o
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 91f5faee7404..f9efd069edc2 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1420,58 +1420,6 @@ static void ide_port_cable_detect(ide_hwif_t *hwif)
 	}
 }
 
-static ssize_t store_delete_devices(struct device *portdev,
-				    struct device_attribute *attr,
-				    const char *buf, size_t n)
-{
-	ide_hwif_t *hwif = dev_get_drvdata(portdev);
-
-	if (strncmp(buf, "1", n))
-		return -EINVAL;
-
-	ide_port_unregister_devices(hwif);
-
-	return n;
-};
-
-static DEVICE_ATTR(delete_devices, S_IWUSR, NULL, store_delete_devices);
-
-static ssize_t store_scan(struct device *portdev,
-			  struct device_attribute *attr,
-			  const char *buf, size_t n)
-{
-	ide_hwif_t *hwif = dev_get_drvdata(portdev);
-
-	if (strncmp(buf, "1", n))
-		return -EINVAL;
-
-	ide_port_unregister_devices(hwif);
-	ide_port_scan(hwif);
-
-	return n;
-};
-
-static DEVICE_ATTR(scan, S_IWUSR, NULL, store_scan);
-
-static struct device_attribute *ide_port_attrs[] = {
-	&dev_attr_delete_devices,
-	&dev_attr_scan,
-	NULL
-};
-
-static int ide_sysfs_register_port(ide_hwif_t *hwif)
-{
-	int i, uninitialized_var(rc);
-
-	for (i = 0; ide_port_attrs[i]; i++) {
-		rc = device_create_file(hwif->portdev, ide_port_attrs[i]);
-		if (rc)
-			break;
-	}
-
-	return rc;
-}
-
 static unsigned int ide_indexes;
 
 /**
diff --git a/drivers/ide/ide-sysfs.c b/drivers/ide/ide-sysfs.c
new file mode 100644
index 000000000000..883ffacaf45a
--- /dev/null
+++ b/drivers/ide/ide-sysfs.c
@@ -0,0 +1,125 @@
+#include <linux/kernel.h>
+#include <linux/ide.h>
+
+char *ide_media_string(ide_drive_t *drive)
+{
+	switch (drive->media) {
+	case ide_disk:
+		return "disk";
+	case ide_cdrom:
+		return "cdrom";
+	case ide_tape:
+		return "tape";
+	case ide_floppy:
+		return "floppy";
+	case ide_optical:
+		return "optical";
+	default:
+		return "UNKNOWN";
+	}
+}
+
+static ssize_t media_show(struct device *dev, struct device_attribute *attr,
+			  char *buf)
+{
+	ide_drive_t *drive = to_ide_device(dev);
+	return sprintf(buf, "%s\n", ide_media_string(drive));
+}
+
+static ssize_t drivename_show(struct device *dev, struct device_attribute *attr,
+			      char *buf)
+{
+	ide_drive_t *drive = to_ide_device(dev);
+	return sprintf(buf, "%s\n", drive->name);
+}
+
+static ssize_t modalias_show(struct device *dev, struct device_attribute *attr,
+			     char *buf)
+{
+	ide_drive_t *drive = to_ide_device(dev);
+	return sprintf(buf, "ide:m-%s\n", ide_media_string(drive));
+}
+
+static ssize_t model_show(struct device *dev, struct device_attribute *attr,
+			  char *buf)
+{
+	ide_drive_t *drive = to_ide_device(dev);
+	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_PROD]);
+}
+
+static ssize_t firmware_show(struct device *dev, struct device_attribute *attr,
+			     char *buf)
+{
+	ide_drive_t *drive = to_ide_device(dev);
+	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_FW_REV]);
+}
+
+static ssize_t serial_show(struct device *dev, struct device_attribute *attr,
+			   char *buf)
+{
+	ide_drive_t *drive = to_ide_device(dev);
+	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_SERNO]);
+}
+
+struct device_attribute ide_dev_attrs[] = {
+	__ATTR_RO(media),
+	__ATTR_RO(drivename),
+	__ATTR_RO(modalias),
+	__ATTR_RO(model),
+	__ATTR_RO(firmware),
+	__ATTR(serial, 0400, serial_show, NULL),
+	__ATTR(unload_heads, 0644, ide_park_show, ide_park_store),
+	__ATTR_NULL
+};
+
+static ssize_t store_delete_devices(struct device *portdev,
+				    struct device_attribute *attr,
+				    const char *buf, size_t n)
+{
+	ide_hwif_t *hwif = dev_get_drvdata(portdev);
+
+	if (strncmp(buf, "1", n))
+		return -EINVAL;
+
+	ide_port_unregister_devices(hwif);
+
+	return n;
+};
+
+static DEVICE_ATTR(delete_devices, S_IWUSR, NULL, store_delete_devices);
+
+static ssize_t store_scan(struct device *portdev,
+			  struct device_attribute *attr,
+			  const char *buf, size_t n)
+{
+	ide_hwif_t *hwif = dev_get_drvdata(portdev);
+
+	if (strncmp(buf, "1", n))
+		return -EINVAL;
+
+	ide_port_unregister_devices(hwif);
+	ide_port_scan(hwif);
+
+	return n;
+};
+
+static DEVICE_ATTR(scan, S_IWUSR, NULL, store_scan);
+
+static struct device_attribute *ide_port_attrs[] = {
+	&dev_attr_delete_devices,
+	&dev_attr_scan,
+	NULL
+};
+
+int ide_sysfs_register_port(ide_hwif_t *hwif)
+{
+	int i, uninitialized_var(rc);
+
+	for (i = 0; ide_port_attrs[i]; i++) {
+		rc = device_create_file(hwif->portdev, ide_port_attrs[i]);
+		if (rc)
+			break;
+	}
+
+	return rc;
+}
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index f0f09f702e9c..46a2d4ca812b 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -440,81 +440,13 @@ static int ide_bus_match(struct device *dev, struct device_driver *drv)
 	return 1;
 }
 
-static char *media_string(ide_drive_t *drive)
-{
-	switch (drive->media) {
-	case ide_disk:
-		return "disk";
-	case ide_cdrom:
-		return "cdrom";
-	case ide_tape:
-		return "tape";
-	case ide_floppy:
-		return "floppy";
-	case ide_optical:
-		return "optical";
-	default:
-		return "UNKNOWN";
-	}
-}
-
-static ssize_t media_show(struct device *dev, struct device_attribute *attr, char *buf)
-{
-	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "%s\n", media_string(drive));
-}
-
-static ssize_t drivename_show(struct device *dev, struct device_attribute *attr, char *buf)
-{
-	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "%s\n", drive->name);
-}
-
-static ssize_t modalias_show(struct device *dev, struct device_attribute *attr, char *buf)
-{
-	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "ide:m-%s\n", media_string(drive));
-}
-
-static ssize_t model_show(struct device *dev, struct device_attribute *attr,
-			  char *buf)
-{
-	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_PROD]);
-}
-
-static ssize_t firmware_show(struct device *dev, struct device_attribute *attr,
-			     char *buf)
-{
-	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_FW_REV]);
-}
-
-static ssize_t serial_show(struct device *dev, struct device_attribute *attr,
-			   char *buf)
-{
-	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_SERNO]);
-}
-
-static struct device_attribute ide_dev_attrs[] = {
-	__ATTR_RO(media),
-	__ATTR_RO(drivename),
-	__ATTR_RO(modalias),
-	__ATTR_RO(model),
-	__ATTR_RO(firmware),
-	__ATTR(serial, 0400, serial_show, NULL),
-	__ATTR(unload_heads, 0644, ide_park_show, ide_park_store),
-	__ATTR_NULL
-};
-
 static int ide_uevent(struct device *dev, struct kobj_uevent_env *env)
 {
 	ide_drive_t *drive = to_ide_device(dev);
 
-	add_uevent_var(env, "MEDIA=%s", media_string(drive));
+	add_uevent_var(env, "MEDIA=%s", ide_media_string(drive));
 	add_uevent_var(env, "DRIVENAME=%s", drive->name);
-	add_uevent_var(env, "MODALIAS=ide:m-%s", media_string(drive));
+	add_uevent_var(env, "MODALIAS=ide:m-%s", ide_media_string(drive));
 	return 0;
 }
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index e99c56de7f56..62fccaea3110 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1533,6 +1533,7 @@ void ide_unregister_region(struct gendisk *);
 void ide_undecoded_slave(ide_drive_t *);
 
 void ide_port_apply_params(ide_hwif_t *);
+int ide_sysfs_register_port(ide_hwif_t *);
 
 struct ide_host *ide_host_alloc(const struct ide_port_info *, hw_regs_t **);
 void ide_host_free(struct ide_host *);
@@ -1627,6 +1628,9 @@ extern struct mutex ide_cfg_mtx;
 
 #define local_irq_set(flags)	do { local_save_flags((flags)); local_irq_enable_in_hardirq(); } while (0)
 
+char *ide_media_string(ide_drive_t *);
+
+extern struct device_attribute ide_dev_attrs[];
 extern struct bus_type ide_bus_type;
 extern struct class *ide_port_class;
 
-- 
cgit v1.2.3


From 295f00042aaf6b553b5f37348f89bab463d4a469 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 2 Jan 2009 16:12:48 +0100
Subject: ide: don't execute the next queued command from the hard-IRQ context
 (v2)

* Tell the block layer that we are not done handling requests by using
  blk_plug_device() in ide_do_request() (request handling function)
  and ide_timer_expiry() (timeout handler) if the queue is not empty.

* Remove optimization which directly calls ide_do_request() for the next
  queued command from the ide_intr() (IRQ handler) and ide_timer_expiry().

* Remove no longer needed IRQ masking from ide_do_request() - in case of
  IDE ports needing serialization disable_irq_nosync()/enable_irq() was
  used for the (possibly shared) IRQ of the other IDE port.

* Put the misplaced comment in the right place in ide_do_request().

* Drop no longer needed 'int masked_irq' argument from ide_do_request().

* Merge ide_do_request() into do_ide_request().

* Remove no longer needed IDE_NO_IRQ define.

While at it:

* Don't use HWGROUP() macro in do_ide_request().

* Use __func__ in ide_intr().

This patch reduces IRQ hadling latency for IDE and improves the system-wide
handling of shared IRQs (which should result in more timeout resistant and
stable IDE systems).  It also makes it possible to do some further changes
later (i.e. replace some busy-waiting delays with sleeping equivalents).

v2:
Changes per review from Elias Oltmanns:
- fix wrong goto statement in 'if (startstop == ide_stopped)' block
- use spin_unlock_irq()
- don't use obsolete HWIF() macro

Cc: Elias Oltmanns <eo@nebensachen.de>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c | 67 +++++++++++++++++++++++-----------------------------
 include/linux/ide.h  |  7 ------
 2 files changed, 30 insertions(+), 44 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index ecacc008fdaf..23754bc5e595 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -778,8 +778,10 @@ repeat:
  * the driver.  This makes the driver much more friendlier to shared IRQs
  * than previous designs, while remaining 100% (?) SMP safe and capable.
  */
-static void ide_do_request (ide_hwgroup_t *hwgroup, int masked_irq)
+void do_ide_request(struct request_queue *q)
 {
+	ide_drive_t	*orig_drive = q->queuedata;
+	ide_hwgroup_t	*hwgroup = orig_drive->hwif->hwgroup;
 	ide_drive_t	*drive;
 	ide_hwif_t	*hwif;
 	struct request	*rq;
@@ -837,10 +839,14 @@ static void ide_do_request (ide_hwgroup_t *hwgroup, int masked_irq)
 			}
 
 			/* no more work for this hwgroup (for now) */
-			return;
+			goto plug_device;
 		}
-	again:
-		hwif = HWIF(drive);
+
+		if (drive != orig_drive)
+			goto plug_device;
+again:
+		hwif = drive->hwif;
+
 		if (hwif != hwgroup->hwif) {
 			/*
 			 * set nIEN for previous hwif, drives in the
@@ -888,41 +894,26 @@ static void ide_do_request (ide_hwgroup_t *hwgroup, int masked_irq)
 				goto again;
 			/* We clear busy, there should be no pending ATA command at this point. */
 			hwgroup->busy = 0;
-			break;
+			goto plug_device;
 		}
 
 		hwgroup->rq = rq;
 
-		/*
-		 * Some systems have trouble with IDE IRQs arriving while
-		 * the driver is still setting things up.  So, here we disable
-		 * the IRQ used by this interface while the request is being started.
-		 * This may look bad at first, but pretty much the same thing
-		 * happens anyway when any interrupt comes in, IDE or otherwise
-		 *  -- the kernel masks the IRQ while it is being handled.
-		 */
-		if (masked_irq != IDE_NO_IRQ && hwif->irq != masked_irq)
-			disable_irq_nosync(hwif->irq);
-		spin_unlock(&hwgroup->lock);
-		local_irq_enable_in_hardirq();
-			/* allow other IRQs while we start this request */
+		spin_unlock_irq(&hwgroup->lock);
 		startstop = start_request(drive, rq);
 		spin_lock_irq(&hwgroup->lock);
-		if (masked_irq != IDE_NO_IRQ && hwif->irq != masked_irq)
-			enable_irq(hwif->irq);
-		if (startstop == ide_stopped)
+
+		if (startstop == ide_stopped) {
 			hwgroup->busy = 0;
+			if (!elv_queue_empty(orig_drive->queue))
+				blk_plug_device(orig_drive->queue);
+		}
 	}
-}
+	return;
 
-/*
- * Passes the stuff to ide_do_request
- */
-void do_ide_request(struct request_queue *q)
-{
-	ide_drive_t *drive = q->queuedata;
-
-	ide_do_request(HWGROUP(drive), IDE_NO_IRQ);
+plug_device:
+	if (!elv_queue_empty(orig_drive->queue))
+		blk_plug_device(orig_drive->queue);
 }
 
 /*
@@ -1074,11 +1065,13 @@ void ide_timer_expiry (unsigned long data)
 			drive->service_time = jiffies - drive->service_start;
 			spin_lock_irq(&hwgroup->lock);
 			enable_irq(hwif->irq);
-			if (startstop == ide_stopped)
+			if (startstop == ide_stopped) {
 				hwgroup->busy = 0;
+				if (!elv_queue_empty(drive->queue))
+					blk_plug_device(drive->queue);
+			}
 		}
 	}
-	ide_do_request(hwgroup, IDE_NO_IRQ);
 	spin_unlock_irqrestore(&hwgroup->lock, flags);
 }
 
@@ -1271,11 +1264,11 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 	if (startstop == ide_stopped) {
 		if (hwgroup->handler == NULL) {	/* paranoia */
 			hwgroup->busy = 0;
-			ide_do_request(hwgroup, hwif->irq);
-		} else {
-			printk(KERN_ERR "%s: ide_intr: huh? expected NULL handler "
-				"on exit\n", drive->name);
-		}
+			if (!elv_queue_empty(drive->queue))
+				blk_plug_device(drive->queue);
+		} else
+			printk(KERN_ERR "%s: %s: huh? expected NULL handler "
+					"on exit\n", __func__, drive->name);
 	}
 out_handled:
 	irq_ret = IRQ_HANDLED;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 62fccaea3110..968ca8f60531 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -32,13 +32,6 @@
 # define SUPPORT_VLB_SYNC 1
 #endif
 
-/*
- * Used to indicate "no IRQ", should be a value that cannot be an IRQ
- * number.
- */
- 
-#define IDE_NO_IRQ		(-1)
-
 typedef unsigned char	byte;	/* used everywhere */
 
 /*
-- 
cgit v1.2.3


From 631de3708d595d153e8a510a3608689290f4c0ed Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 2 Jan 2009 16:12:50 +0100
Subject: ide: add ide_[un]lock_hwgroup() helpers

Add ide_[un]lock_hwgroup() inline helpers for obtaining exclusive
access to the given hwgroup and update the core code accordingly.

[ This change besides making code saner results in more efficient
  use of ide_{get,release}_lock(). ]

Cc: Michael Schmitz <schmitz@biophys.uni-duesseldorf.de>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Elias Oltmanns <eo@nebensachen.de>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c   | 32 +++++++++++---------------------
 drivers/ide/ide-park.c |  2 +-
 include/linux/ide.h    | 20 ++++++++++++++++++++
 3 files changed, 32 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index c60512196f61..ab480042757a 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -790,10 +790,7 @@ void do_ide_request(struct request_queue *q)
 	/* caller must own hwgroup->lock */
 	BUG_ON(!irqs_disabled());
 
-	while (!hwgroup->busy) {
-		hwgroup->busy = 1;
-		/* for atari only */
-		ide_get_lock(ide_intr, hwgroup);
+	while (!ide_lock_hwgroup(hwgroup)) {
 		drive = choose_drive(hwgroup);
 		if (drive == NULL) {
 			int sleeping = 0;
@@ -825,17 +822,10 @@ void do_ide_request(struct request_queue *q)
 				hwgroup->sleeping = 1;
 				hwgroup->req_gen_timer = hwgroup->req_gen;
 				mod_timer(&hwgroup->timer, sleep);
-				/* we purposely leave hwgroup->busy==1
+				/* we purposely leave hwgroup locked
 				 * while sleeping */
-			} else {
-				/* Ugly, but how can we sleep for the lock
-				 * otherwise? perhaps from tq_disk?
-				 */
-
-				/* for atari only */
-				ide_release_lock();
-				hwgroup->busy = 0;
-			}
+			} else
+				ide_unlock_hwgroup(hwgroup);
 
 			/* no more work for this hwgroup (for now) */
 			goto plug_device;
@@ -865,7 +855,7 @@ void do_ide_request(struct request_queue *q)
 		 */
 		rq = elv_next_request(drive->queue);
 		if (!rq) {
-			hwgroup->busy = 0;
+			ide_unlock_hwgroup(hwgroup);
 			break;
 		}
 
@@ -885,8 +875,8 @@ void do_ide_request(struct request_queue *q)
 		if ((drive->dev_flags & IDE_DFLAG_BLOCKED) &&
 		    blk_pm_request(rq) == 0 &&
 		    (rq->cmd_flags & REQ_PREEMPT) == 0) {
-			/* We clear busy, there should be no pending ATA command at this point. */
-			hwgroup->busy = 0;
+			/* there should be no pending command at this point */
+			ide_unlock_hwgroup(hwgroup);
 			goto plug_device;
 		}
 
@@ -897,7 +887,7 @@ void do_ide_request(struct request_queue *q)
 		spin_lock_irq(&hwgroup->lock);
 
 		if (startstop == ide_stopped) {
-			hwgroup->busy = 0;
+			ide_unlock_hwgroup(hwgroup);
 			if (!elv_queue_empty(orig_drive->queue))
 				blk_plug_device(orig_drive->queue);
 		}
@@ -1001,7 +991,7 @@ void ide_timer_expiry (unsigned long data)
 		 */
 		if (hwgroup->sleeping) {
 			hwgroup->sleeping = 0;
-			hwgroup->busy = 0;
+			ide_unlock_hwgroup(hwgroup);
 		}
 	} else {
 		ide_drive_t *drive = hwgroup->drive;
@@ -1056,7 +1046,7 @@ void ide_timer_expiry (unsigned long data)
 			spin_lock_irq(&hwgroup->lock);
 			enable_irq(hwif->irq);
 			if (startstop == ide_stopped) {
-				hwgroup->busy = 0;
+				ide_unlock_hwgroup(hwgroup);
 				if (!elv_queue_empty(drive->queue))
 					blk_plug_device(drive->queue);
 			}
@@ -1249,7 +1239,7 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 	drive->service_time = jiffies - drive->service_start;
 	if (startstop == ide_stopped) {
 		if (hwgroup->handler == NULL) {	/* paranoia */
-			hwgroup->busy = 0;
+			ide_unlock_hwgroup(hwgroup);
 			if (!elv_queue_empty(drive->queue))
 				blk_plug_device(drive->queue);
 		} else
diff --git a/drivers/ide/ide-park.c b/drivers/ide/ide-park.c
index 63d01c55f865..44c6787f8aeb 100644
--- a/drivers/ide/ide-park.c
+++ b/drivers/ide/ide-park.c
@@ -22,7 +22,7 @@ static void issue_park_cmd(ide_drive_t *drive, unsigned long timeout)
 		if (reset_timer && hwgroup->sleeping &&
 		    del_timer(&hwgroup->timer)) {
 			hwgroup->sleeping = 0;
-			hwgroup->busy = 0;
+			ide_unlock_hwgroup(hwgroup);
 			blk_start_queueing(q);
 		}
 		spin_unlock_irq(&hwgroup->lock);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 968ca8f60531..f408d6123f14 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1280,6 +1280,26 @@ extern void ide_stall_queue(ide_drive_t *drive, unsigned long timeout);
 
 extern void ide_timer_expiry(unsigned long);
 extern irqreturn_t ide_intr(int irq, void *dev_id);
+
+static inline int ide_lock_hwgroup(ide_hwgroup_t *hwgroup)
+{
+	if (hwgroup->busy)
+		return 1;
+
+	hwgroup->busy = 1;
+	/* for atari only */
+	ide_get_lock(ide_intr, hwgroup);
+
+	return 0;
+}
+
+static inline void ide_unlock_hwgroup(ide_hwgroup_t *hwgroup)
+{
+	/* for atari only */
+	ide_release_lock();
+	hwgroup->busy = 0;
+}
+
 extern void do_ide_request(struct request_queue *);
 
 void ide_init_disk(struct gendisk *, ide_drive_t *);
-- 
cgit v1.2.3


From 201bffa46466b4afdf7d29db8eca3fa5decb39c8 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Fri, 2 Jan 2009 16:12:50 +0100
Subject: ide: use per-device request queue locks (v2)

* Move hack for flush requests from choose_drive() to do_ide_request().

* Add ide_plug_device() helper and convert core IDE code from using
  per-hwgroup lock as a request lock to use the ->queue_lock instead.

* Remove no longer needed:
  - choose_drive() function
  - WAKEUP() macro
  - 'sleeping' flag from ide_hwif_t
  - 'service_{start,time}' fields from ide_drive_t

This patch results in much simpler and more maintainable code
(besides being a scalability improvement).

v2:
* Fixes/improvements based on review from Elias:
  - take as many requests off the queue as possible
  - remove now redundant BUG_ON()

Cc: Elias Oltmanns <eo@nebensachen.de>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c    | 214 +++++++++++++++---------------------------------
 drivers/ide/ide-park.c  |  13 +--
 drivers/ide/ide-probe.c |   3 +-
 include/linux/ide.h     |   4 -
 4 files changed, 77 insertions(+), 157 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index ab480042757a..bb3248abf47d 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -667,85 +667,10 @@ void ide_stall_queue (ide_drive_t *drive, unsigned long timeout)
 	drive->sleep = timeout + jiffies;
 	drive->dev_flags |= IDE_DFLAG_SLEEPING;
 }
-
 EXPORT_SYMBOL(ide_stall_queue);
 
-#define WAKEUP(drive)	((drive)->service_start + 2 * (drive)->service_time)
-
-/**
- *	choose_drive		-	select a drive to service
- *	@hwgroup: hardware group to select on
- *
- *	choose_drive() selects the next drive which will be serviced.
- *	This is necessary because the IDE layer can't issue commands
- *	to both drives on the same cable, unlike SCSI.
- */
- 
-static inline ide_drive_t *choose_drive (ide_hwgroup_t *hwgroup)
-{
-	ide_drive_t *drive, *best;
-
-repeat:	
-	best = NULL;
-	drive = hwgroup->drive;
-
-	/*
-	 * drive is doing pre-flush, ordered write, post-flush sequence. even
-	 * though that is 3 requests, it must be seen as a single transaction.
-	 * we must not preempt this drive until that is complete
-	 */
-	if (blk_queue_flushing(drive->queue)) {
-		/*
-		 * small race where queue could get replugged during
-		 * the 3-request flush cycle, just yank the plug since
-		 * we want it to finish asap
-		 */
-		blk_remove_plug(drive->queue);
-		return drive;
-	}
-
-	do {
-		u8 dev_s = !!(drive->dev_flags & IDE_DFLAG_SLEEPING);
-		u8 best_s = (best && !!(best->dev_flags & IDE_DFLAG_SLEEPING));
-
-		if ((dev_s == 0 || time_after_eq(jiffies, drive->sleep)) &&
-		    !elv_queue_empty(drive->queue)) {
-			if (best == NULL ||
-			    (dev_s && (best_s == 0 || time_before(drive->sleep, best->sleep))) ||
-			    (best_s == 0 && time_before(WAKEUP(drive), WAKEUP(best)))) {
-				if (!blk_queue_plugged(drive->queue))
-					best = drive;
-			}
-		}
-	} while ((drive = drive->next) != hwgroup->drive);
-
-	if (best && (best->dev_flags & IDE_DFLAG_NICE1) &&
-	    (best->dev_flags & IDE_DFLAG_SLEEPING) == 0 &&
-	    best != hwgroup->drive && best->service_time > WAIT_MIN_SLEEP) {
-		long t = (signed long)(WAKEUP(best) - jiffies);
-		if (t >= WAIT_MIN_SLEEP) {
-		/*
-		 * We *may* have some time to spare, but first let's see if
-		 * someone can potentially benefit from our nice mood today..
-		 */
-			drive = best->next;
-			do {
-				if ((drive->dev_flags & IDE_DFLAG_SLEEPING) == 0
-				 && time_before(jiffies - best->service_time, WAKEUP(drive))
-				 && time_before(WAKEUP(drive), jiffies + t))
-				{
-					ide_stall_queue(best, min_t(long, t, 10 * WAIT_MIN_SLEEP));
-					goto repeat;
-				}
-			} while ((drive = drive->next) != best);
-		}
-	}
-	return best;
-}
-
 /*
  * Issue a new request to a drive from hwgroup
- * Caller must have already done spin_lock_irqsave(&hwgroup->lock, ..);
  *
  * A hwgroup is a serialized group of IDE interfaces.  Usually there is
  * exactly one hwif (interface) per hwgroup, but buggy controllers (eg. CMD640)
@@ -757,8 +682,7 @@ repeat:
  * possibly along with many other devices.  This is especially common in
  * PCI-based systems with off-board IDE controller cards.
  *
- * The IDE driver uses a per-hwgroup spinlock to protect
- * access to the request queues, and to protect the hwgroup->busy flag.
+ * The IDE driver uses a per-hwgroup lock to protect the hwgroup->busy flag.
  *
  * The first thread into the driver for a particular hwgroup sets the
  * hwgroup->busy flag to indicate that this hwgroup is now active,
@@ -780,61 +704,38 @@ repeat:
  */
 void do_ide_request(struct request_queue *q)
 {
-	ide_drive_t	*orig_drive = q->queuedata;
-	ide_hwgroup_t	*hwgroup = orig_drive->hwif->hwgroup;
-	ide_drive_t	*drive;
-	ide_hwif_t	*hwif;
+	ide_drive_t	*drive = q->queuedata;
+	ide_hwif_t	*hwif = drive->hwif;
+	ide_hwgroup_t	*hwgroup = hwif->hwgroup;
 	struct request	*rq;
 	ide_startstop_t	startstop;
 
-	/* caller must own hwgroup->lock */
-	BUG_ON(!irqs_disabled());
-
-	while (!ide_lock_hwgroup(hwgroup)) {
-		drive = choose_drive(hwgroup);
-		if (drive == NULL) {
-			int sleeping = 0;
-			unsigned long sleep = 0; /* shut up, gcc */
-			hwgroup->rq = NULL;
-			drive = hwgroup->drive;
-			do {
-				if ((drive->dev_flags & IDE_DFLAG_SLEEPING) &&
-				    (sleeping == 0 ||
-				     time_before(drive->sleep, sleep))) {
-					sleeping = 1;
-					sleep = drive->sleep;
-				}
-			} while ((drive = drive->next) != hwgroup->drive);
-			if (sleeping) {
+	/*
+	 * drive is doing pre-flush, ordered write, post-flush sequence. even
+	 * though that is 3 requests, it must be seen as a single transaction.
+	 * we must not preempt this drive until that is complete
+	 */
+	if (blk_queue_flushing(q))
 		/*
-		 * Take a short snooze, and then wake up this hwgroup again.
-		 * This gives other hwgroups on the same a chance to
-		 * play fairly with us, just in case there are big differences
-		 * in relative throughputs.. don't want to hog the cpu too much.
+		 * small race where queue could get replugged during
+		 * the 3-request flush cycle, just yank the plug since
+		 * we want it to finish asap
 		 */
-				if (time_before(sleep, jiffies + WAIT_MIN_SLEEP))
-					sleep = jiffies + WAIT_MIN_SLEEP;
-#if 1
-				if (timer_pending(&hwgroup->timer))
-					printk(KERN_CRIT "ide_set_handler: timer already active\n");
-#endif
-				/* so that ide_timer_expiry knows what to do */
-				hwgroup->sleeping = 1;
-				hwgroup->req_gen_timer = hwgroup->req_gen;
-				mod_timer(&hwgroup->timer, sleep);
-				/* we purposely leave hwgroup locked
-				 * while sleeping */
-			} else
-				ide_unlock_hwgroup(hwgroup);
+		blk_remove_plug(q);
 
-			/* no more work for this hwgroup (for now) */
-			goto plug_device;
-		}
+	spin_unlock_irq(q->queue_lock);
+	spin_lock_irq(&hwgroup->lock);
 
-		if (drive != orig_drive)
-			goto plug_device;
+	if (!ide_lock_hwgroup(hwgroup)) {
+repeat:
+		hwgroup->rq = NULL;
 
-		hwif = drive->hwif;
+		if (drive->dev_flags & IDE_DFLAG_SLEEPING) {
+			if (time_before(drive->sleep, jiffies)) {
+				ide_unlock_hwgroup(hwgroup);
+				goto plug_device;
+			}
+		}
 
 		if (hwif != hwgroup->hwif) {
 			/*
@@ -847,16 +748,20 @@ void do_ide_request(struct request_queue *q)
 		hwgroup->hwif = hwif;
 		hwgroup->drive = drive;
 		drive->dev_flags &= ~(IDE_DFLAG_SLEEPING | IDE_DFLAG_PARKED);
-		drive->service_start = jiffies;
 
+		spin_unlock_irq(&hwgroup->lock);
+		spin_lock_irq(q->queue_lock);
 		/*
 		 * we know that the queue isn't empty, but this can happen
 		 * if the q->prep_rq_fn() decides to kill a request
 		 */
 		rq = elv_next_request(drive->queue);
+		spin_unlock_irq(q->queue_lock);
+		spin_lock_irq(&hwgroup->lock);
+
 		if (!rq) {
 			ide_unlock_hwgroup(hwgroup);
-			break;
+			goto out;
 		}
 
 		/*
@@ -886,17 +791,21 @@ void do_ide_request(struct request_queue *q)
 		startstop = start_request(drive, rq);
 		spin_lock_irq(&hwgroup->lock);
 
-		if (startstop == ide_stopped) {
-			ide_unlock_hwgroup(hwgroup);
-			if (!elv_queue_empty(orig_drive->queue))
-				blk_plug_device(orig_drive->queue);
-		}
-	}
+		if (startstop == ide_stopped)
+			goto repeat;
+	} else
+		goto plug_device;
+out:
+	spin_unlock_irq(&hwgroup->lock);
+	spin_lock_irq(q->queue_lock);
 	return;
 
 plug_device:
-	if (!elv_queue_empty(orig_drive->queue))
-		blk_plug_device(orig_drive->queue);
+	spin_unlock_irq(&hwgroup->lock);
+	spin_lock_irq(q->queue_lock);
+
+	if (!elv_queue_empty(q))
+		blk_plug_device(q);
 }
 
 /*
@@ -957,6 +866,17 @@ out:
 	return ret;
 }
 
+static void ide_plug_device(ide_drive_t *drive)
+{
+	struct request_queue *q = drive->queue;
+	unsigned long flags;
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	if (!elv_queue_empty(q))
+		blk_plug_device(q);
+	spin_unlock_irqrestore(q->queue_lock, flags);
+}
+
 /**
  *	ide_timer_expiry	-	handle lack of an IDE interrupt
  *	@data: timer callback magic (hwgroup)
@@ -974,10 +894,12 @@ out:
 void ide_timer_expiry (unsigned long data)
 {
 	ide_hwgroup_t	*hwgroup = (ide_hwgroup_t *) data;
+	ide_drive_t	*uninitialized_var(drive);
 	ide_handler_t	*handler;
 	ide_expiry_t	*expiry;
 	unsigned long	flags;
 	unsigned long	wait = -1;
+	int		plug_device = 0;
 
 	spin_lock_irqsave(&hwgroup->lock, flags);
 
@@ -989,12 +911,8 @@ void ide_timer_expiry (unsigned long data)
 		 * or we were "sleeping" to give other devices a chance.
 		 * Either way, we don't really want to complain about anything.
 		 */
-		if (hwgroup->sleeping) {
-			hwgroup->sleeping = 0;
-			ide_unlock_hwgroup(hwgroup);
-		}
 	} else {
-		ide_drive_t *drive = hwgroup->drive;
+		drive = hwgroup->drive;
 		if (!drive) {
 			printk(KERN_ERR "ide_timer_expiry: hwgroup->drive was NULL\n");
 			hwgroup->handler = NULL;
@@ -1042,17 +960,18 @@ void ide_timer_expiry (unsigned long data)
 					ide_error(drive, "irq timeout",
 						  hwif->tp_ops->read_status(hwif));
 			}
-			drive->service_time = jiffies - drive->service_start;
 			spin_lock_irq(&hwgroup->lock);
 			enable_irq(hwif->irq);
 			if (startstop == ide_stopped) {
 				ide_unlock_hwgroup(hwgroup);
-				if (!elv_queue_empty(drive->queue))
-					blk_plug_device(drive->queue);
+				plug_device = 1;
 			}
 		}
 	}
 	spin_unlock_irqrestore(&hwgroup->lock, flags);
+
+	if (plug_device)
+		ide_plug_device(drive);
 }
 
 /**
@@ -1146,10 +1065,11 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 	unsigned long flags;
 	ide_hwgroup_t *hwgroup = (ide_hwgroup_t *)dev_id;
 	ide_hwif_t *hwif = hwgroup->hwif;
-	ide_drive_t *drive;
+	ide_drive_t *uninitialized_var(drive);
 	ide_handler_t *handler;
 	ide_startstop_t startstop;
 	irqreturn_t irq_ret = IRQ_NONE;
+	int plug_device = 0;
 
 	spin_lock_irqsave(&hwgroup->lock, flags);
 
@@ -1236,12 +1156,10 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 	 * same irq as is currently being serviced here, and Linux
 	 * won't allow another of the same (on any CPU) until we return.
 	 */
-	drive->service_time = jiffies - drive->service_start;
 	if (startstop == ide_stopped) {
 		if (hwgroup->handler == NULL) {	/* paranoia */
 			ide_unlock_hwgroup(hwgroup);
-			if (!elv_queue_empty(drive->queue))
-				blk_plug_device(drive->queue);
+			plug_device = 1;
 		} else
 			printk(KERN_ERR "%s: %s: huh? expected NULL handler "
 					"on exit\n", __func__, drive->name);
@@ -1250,6 +1168,10 @@ out_handled:
 	irq_ret = IRQ_HANDLED;
 out:
 	spin_unlock_irqrestore(&hwgroup->lock, flags);
+
+	if (plug_device)
+		ide_plug_device(drive);
+
 	return irq_ret;
 }
 
diff --git a/drivers/ide/ide-park.c b/drivers/ide/ide-park.c
index 44c6787f8aeb..678454ac2483 100644
--- a/drivers/ide/ide-park.c
+++ b/drivers/ide/ide-park.c
@@ -16,16 +16,19 @@ static void issue_park_cmd(ide_drive_t *drive, unsigned long timeout)
 	spin_lock_irq(&hwgroup->lock);
 	if (drive->dev_flags & IDE_DFLAG_PARKED) {
 		int reset_timer = time_before(timeout, drive->sleep);
+		int start_queue = 0;
 
 		drive->sleep = timeout;
 		wake_up_all(&ide_park_wq);
-		if (reset_timer && hwgroup->sleeping &&
-		    del_timer(&hwgroup->timer)) {
-			hwgroup->sleeping = 0;
-			ide_unlock_hwgroup(hwgroup);
+		if (reset_timer && del_timer(&hwgroup->timer))
+			start_queue = 1;
+		spin_unlock_irq(&hwgroup->lock);
+
+		if (start_queue) {
+			spin_lock_irq(q->queue_lock);
 			blk_start_queueing(q);
+			spin_unlock_irq(q->queue_lock);
 		}
-		spin_unlock_irq(&hwgroup->lock);
 		return;
 	}
 	spin_unlock_irq(&hwgroup->lock);
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index f9efd069edc2..966b74c15773 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -881,8 +881,7 @@ static int ide_init_queue(ide_drive_t *drive)
 	 *	do not.
 	 */
 
-	q = blk_init_queue_node(do_ide_request, &hwif->hwgroup->lock,
-				hwif_to_node(hwif));
+	q = blk_init_queue_node(do_ide_request, NULL, hwif_to_node(hwif));
 	if (!q)
 		return 1;
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index f408d6123f14..5f86ad40ee7e 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -603,8 +603,6 @@ struct ide_drive_s {
 	unsigned long dev_flags;
 
 	unsigned long sleep;		/* sleep until this time */
-	unsigned long service_start;	/* time we started last request */
-	unsigned long service_time;	/* service time of last request */
 	unsigned long timeout;		/* max time to wait for irq */
 
 	special_t	special;	/* special action flags */
@@ -872,8 +870,6 @@ typedef struct hwgroup_s {
 
 		/* BOOL: protects all fields below */
 	volatile int busy;
-		/* BOOL: wake us up on timer expiry */
-	unsigned int sleeping	: 1;
 		/* BOOL: polling active & poll_timeout field valid */
 	unsigned int polling	: 1;
 
-- 
cgit v1.2.3


From bf64741fe89280bd81a9e3a1beadec1570861848 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Fri, 2 Jan 2009 16:12:50 +0100
Subject: ide: make IDE_AFLAG_.. numbering continuous again

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/linux/ide.h | 48 ++++++++++++++++++++++++------------------------
 1 file changed, 24 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ide.h b/include/linux/ide.h
index 5f86ad40ee7e..eb4c01f7f253 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -473,53 +473,53 @@ enum {
 
 	/* ide-cd */
 	/* Drive cannot eject the disc. */
-	IDE_AFLAG_NO_EJECT		= (1 << 3),
+	IDE_AFLAG_NO_EJECT		= (1 << 1),
 	/* Drive is a pre ATAPI 1.2 drive. */
-	IDE_AFLAG_PRE_ATAPI12		= (1 << 4),
+	IDE_AFLAG_PRE_ATAPI12		= (1 << 2),
 	/* TOC addresses are in BCD. */
-	IDE_AFLAG_TOCADDR_AS_BCD	= (1 << 5),
+	IDE_AFLAG_TOCADDR_AS_BCD	= (1 << 3),
 	/* TOC track numbers are in BCD. */
-	IDE_AFLAG_TOCTRACKS_AS_BCD	= (1 << 6),
+	IDE_AFLAG_TOCTRACKS_AS_BCD	= (1 << 4),
 	/*
 	 * Drive does not provide data in multiples of SECTOR_SIZE
 	 * when more than one interrupt is needed.
 	 */
-	IDE_AFLAG_LIMIT_NFRAMES		= (1 << 7),
+	IDE_AFLAG_LIMIT_NFRAMES		= (1 << 5),
 	/* Saved TOC information is current. */
-	IDE_AFLAG_TOC_VALID		= (1 << 9),
+	IDE_AFLAG_TOC_VALID		= (1 << 6),
 	/* We think that the drive door is locked. */
-	IDE_AFLAG_DOOR_LOCKED		= (1 << 10),
+	IDE_AFLAG_DOOR_LOCKED		= (1 << 7),
 	/* SET_CD_SPEED command is unsupported. */
-	IDE_AFLAG_NO_SPEED_SELECT	= (1 << 11),
-	IDE_AFLAG_VERTOS_300_SSD	= (1 << 12),
-	IDE_AFLAG_VERTOS_600_ESD	= (1 << 13),
-	IDE_AFLAG_SANYO_3CD		= (1 << 14),
-	IDE_AFLAG_FULL_CAPS_PAGE	= (1 << 15),
-	IDE_AFLAG_PLAY_AUDIO_OK		= (1 << 16),
-	IDE_AFLAG_LE_SPEED_FIELDS	= (1 << 17),
+	IDE_AFLAG_NO_SPEED_SELECT	= (1 << 8),
+	IDE_AFLAG_VERTOS_300_SSD	= (1 << 9),
+	IDE_AFLAG_VERTOS_600_ESD	= (1 << 10),
+	IDE_AFLAG_SANYO_3CD		= (1 << 11),
+	IDE_AFLAG_FULL_CAPS_PAGE	= (1 << 12),
+	IDE_AFLAG_PLAY_AUDIO_OK		= (1 << 13),
+	IDE_AFLAG_LE_SPEED_FIELDS	= (1 << 14),
 
 	/* ide-floppy */
 	/* Avoid commands not supported in Clik drive */
-	IDE_AFLAG_CLIK_DRIVE		= (1 << 19),
+	IDE_AFLAG_CLIK_DRIVE		= (1 << 15),
 	/* Requires BH algorithm for packets */
-	IDE_AFLAG_ZIP_DRIVE		= (1 << 20),
+	IDE_AFLAG_ZIP_DRIVE		= (1 << 16),
 	/* Supports format progress report */
-	IDE_AFLAG_SRFP			= (1 << 22),
+	IDE_AFLAG_SRFP			= (1 << 17),
 
 	/* ide-tape */
-	IDE_AFLAG_IGNORE_DSC		= (1 << 23),
+	IDE_AFLAG_IGNORE_DSC		= (1 << 18),
 	/* 0 When the tape position is unknown */
-	IDE_AFLAG_ADDRESS_VALID		= (1 <<	24),
+	IDE_AFLAG_ADDRESS_VALID		= (1 <<	19),
 	/* Device already opened */
-	IDE_AFLAG_BUSY			= (1 << 25),
+	IDE_AFLAG_BUSY			= (1 << 20),
 	/* Attempt to auto-detect the current user block size */
-	IDE_AFLAG_DETECT_BS		= (1 << 26),
+	IDE_AFLAG_DETECT_BS		= (1 << 21),
 	/* Currently on a filemark */
-	IDE_AFLAG_FILEMARK		= (1 << 27),
+	IDE_AFLAG_FILEMARK		= (1 << 22),
 	/* 0 = no tape is loaded, so we don't rewind after ejecting */
-	IDE_AFLAG_MEDIUM_PRESENT	= (1 << 28),
+	IDE_AFLAG_MEDIUM_PRESENT	= (1 << 23),
 
-	IDE_AFLAG_NO_AUTOCLOSE		= (1 << 29),
+	IDE_AFLAG_NO_AUTOCLOSE		= (1 << 24),
 };
 
 /* device flags */
-- 
cgit v1.2.3


From 392de1d53dd40e2eebee3a0a26aa647a3865ca78 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Fri, 2 Jan 2009 16:12:52 +0100
Subject: ide-atapi: accomodate transfer length calculation for ide-cd

... by factoring it out of ide_cd_do_request() into a helper, as suggested by
Bart.

There should be no functionality change resulting from this patch.

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
[bart: BLK_DEV_IDECD needs to select IDE_ATAPI now]
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/Kconfig     |  1 +
 drivers/ide/ide-atapi.c | 15 ++++++++++++++-
 drivers/ide/ide-cd.c    |  4 ++--
 include/linux/ide.h     |  2 ++
 4 files changed, 19 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index 937945e471df..4ee85fcf9aaf 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -137,6 +137,7 @@ config BLK_DEV_DELKIN
 
 config BLK_DEV_IDECD
 	tristate "Include IDE/ATAPI CDROM support"
+	select IDE_ATAPI
 	---help---
 	  If you have a CD-ROM drive using the ATAPI protocol, say Y. ATAPI is
 	  a newer protocol used by IDE CD-ROM and TAPE drives, similar to the
diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index 74273fdc8827..8884877bd2b5 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -252,6 +252,18 @@ int ide_scsi_expiry(ide_drive_t *drive)
 }
 EXPORT_SYMBOL_GPL(ide_scsi_expiry);
 
+int ide_cd_get_xferlen(struct request *rq)
+{
+	if (blk_fs_request(rq))
+		return 32768;
+	else if (blk_sense_request(rq) || blk_pc_request(rq) ||
+			 rq->cmd_type == REQ_TYPE_ATA_PC)
+		return rq->data_len;
+	else
+		return 0;
+}
+EXPORT_SYMBOL_GPL(ide_cd_get_xferlen);
+
 /*
  * This is the usual interrupt handler which will be called during a packet
  * command.  We will transfer some of the data (as requested by the drive)
@@ -551,7 +563,7 @@ ide_startstop_t ide_issue_pc(ide_drive_t *drive, unsigned int timeout,
 	struct ide_atapi_pc *pc = drive->pc;
 	ide_hwif_t *hwif = drive->hwif;
 	u32 tf_flags;
-	u16 bcount = 0;
+	u16 bcount;
 	u8 scsi = !!(drive->dev_flags & IDE_DFLAG_SCSI);
 
 	/* We haven't transferred any data yet */
@@ -560,6 +572,7 @@ ide_startstop_t ide_issue_pc(ide_drive_t *drive, unsigned int timeout,
 
 	if (dev_is_idecd(drive)) {
 		tf_flags = IDE_TFLAG_OUT_NSECT | IDE_TFLAG_OUT_LBAL;
+		bcount = ide_cd_get_xferlen(hwif->hwgroup->rq);
 	} else if (scsi) {
 		tf_flags = 0;
 		bcount = min(pc->req_xfer, 63 * 1024);
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 65e5513758b0..8d3c7714682e 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -1214,8 +1214,9 @@ static ide_startstop_t ide_cd_do_request(ide_drive_t *drive, struct request *rq,
 		      __func__, rq->cmd[0], rq->cmd_type,
 		      (unsigned long long)block);
 
+	xferlen = ide_cd_get_xferlen(rq);
+
 	if (blk_fs_request(rq)) {
-		xferlen = 32768;
 		fn = cdrom_start_rw_cont;
 
 		if (cdrom_start_rw(drive, rq) == ide_stopped)
@@ -1225,7 +1226,6 @@ static ide_startstop_t ide_cd_do_request(ide_drive_t *drive, struct request *rq,
 			return ide_stopped;
 	} else if (blk_sense_request(rq) || blk_pc_request(rq) ||
 		   rq->cmd_type == REQ_TYPE_ATA_PC) {
-		xferlen = rq->data_len;
 		fn = cdrom_do_newpc_cont;
 
 		if (!rq->timeout)
diff --git a/include/linux/ide.h b/include/linux/ide.h
index eb4c01f7f253..e35ff6827897 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1254,6 +1254,8 @@ static inline unsigned long ide_scsi_get_timeout(struct ide_atapi_pc *pc)
 
 int ide_scsi_expiry(ide_drive_t *);
 
+int ide_cd_get_xferlen(struct request *);
+
 ide_startstop_t ide_issue_pc(ide_drive_t *, unsigned int, ide_expiry_t *);
 
 ide_startstop_t do_rw_taskfile(ide_drive_t *, ide_task_t *);
-- 
cgit v1.2.3


From 4cad085efbce8dcc5006b0d1034089758b4fc7ba Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@gmail.com>
Date: Fri, 2 Jan 2009 16:12:53 +0100
Subject: ide-cd: move cdrom_timer_expiry to ide-atapi.c

- cdrom_timer_expiry -> ide_cd_expiry
- remove expiry-arg to ide_issue_pc as it is redundant now
- ide_debug_log -> debug_log

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-atapi.c  | 41 ++++++++++++++++++++++++++++++++++++++---
 drivers/ide/ide-cd.c     | 38 +++-----------------------------------
 drivers/ide/ide-cd.h     |  4 ----
 drivers/ide/ide-floppy.c |  2 +-
 drivers/ide/ide-tape.c   |  2 +-
 include/linux/ide.h      |  4 +++-
 6 files changed, 46 insertions(+), 45 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index 8c5cf68fbd79..c110329ccb13 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -3,6 +3,7 @@
  */
 
 #include <linux/kernel.h>
+#include <linux/cdrom.h>
 #include <linux/delay.h>
 #include <linux/ide.h>
 #include <scsi/scsi.h>
@@ -252,6 +253,38 @@ int ide_scsi_expiry(ide_drive_t *drive)
 }
 EXPORT_SYMBOL_GPL(ide_scsi_expiry);
 
+int ide_cd_expiry(ide_drive_t *drive)
+{
+	struct request *rq = HWGROUP(drive)->rq;
+	unsigned long wait = 0;
+
+	debug_log("%s: rq->cmd[0]: 0x%x\n", __func__, rq->cmd[0]);
+
+	/*
+	 * Some commands are *slow* and normally take a long time to complete.
+	 * Usually we can use the ATAPI "disconnect" to bypass this, but not all
+	 * commands/drives support that. Let ide_timer_expiry keep polling us
+	 * for these.
+	 */
+	switch (rq->cmd[0]) {
+	case GPCMD_BLANK:
+	case GPCMD_FORMAT_UNIT:
+	case GPCMD_RESERVE_RZONE_TRACK:
+	case GPCMD_CLOSE_TRACK:
+	case GPCMD_FLUSH_CACHE:
+		wait = ATAPI_WAIT_PC;
+		break;
+	default:
+		if (!(rq->cmd_flags & REQ_QUIET))
+			printk(KERN_INFO "cmd 0x%x timed out\n",
+					 rq->cmd[0]);
+		wait = 0;
+		break;
+	}
+	return wait;
+}
+EXPORT_SYMBOL_GPL(ide_cd_expiry);
+
 int ide_cd_get_xferlen(struct request *rq)
 {
 	if (blk_fs_request(rq))
@@ -562,11 +595,11 @@ static ide_startstop_t ide_transfer_pc(ide_drive_t *drive)
 	return ide_started;
 }
 
-ide_startstop_t ide_issue_pc(ide_drive_t *drive, unsigned int timeout,
-			     ide_expiry_t *expiry)
+ide_startstop_t ide_issue_pc(ide_drive_t *drive, unsigned int timeout)
 {
 	struct ide_atapi_pc *pc = drive->pc;
 	ide_hwif_t *hwif = drive->hwif;
+	ide_expiry_t *expiry = NULL;
 	u32 tf_flags;
 	u16 bcount;
 	u8 scsi = !!(drive->dev_flags & IDE_DFLAG_SCSI);
@@ -578,9 +611,11 @@ ide_startstop_t ide_issue_pc(ide_drive_t *drive, unsigned int timeout,
 	if (dev_is_idecd(drive)) {
 		tf_flags = IDE_TFLAG_OUT_NSECT | IDE_TFLAG_OUT_LBAL;
 		bcount = ide_cd_get_xferlen(hwif->hwgroup->rq);
+		expiry = ide_cd_expiry;
 	} else if (scsi) {
 		tf_flags = 0;
 		bcount = min(pc->req_xfer, 63 * 1024);
+		expiry = ide_scsi_expiry;
 	} else {
 		tf_flags = IDE_TFLAG_OUT_DEVICE;
 		bcount = ((drive->media == ide_tape) ?
@@ -613,7 +648,7 @@ ide_startstop_t ide_issue_pc(ide_drive_t *drive, unsigned int timeout,
 		if (drive->dma)
 			drive->waiting_for_dma = 0;
 		ide_execute_command(drive, ATA_CMD_PACKET, ide_transfer_pc,
-				    timeout, NULL);
+				    timeout, expiry);
 		return ide_started;
 	} else {
 		ide_execute_pkt_cmd(drive);
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 8d3c7714682e..105e4d855e6e 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -511,38 +511,6 @@ end_request:
 	return 1;
 }
 
-static int cdrom_timer_expiry(ide_drive_t *drive)
-{
-	struct request *rq = HWGROUP(drive)->rq;
-	unsigned long wait = 0;
-
-	ide_debug_log(IDE_DBG_RQ, "Call %s: rq->cmd[0]: 0x%x\n", __func__,
-		      rq->cmd[0]);
-
-	/*
-	 * Some commands are *slow* and normally take a long time to complete.
-	 * Usually we can use the ATAPI "disconnect" to bypass this, but not all
-	 * commands/drives support that. Let ide_timer_expiry keep polling us
-	 * for these.
-	 */
-	switch (rq->cmd[0]) {
-	case GPCMD_BLANK:
-	case GPCMD_FORMAT_UNIT:
-	case GPCMD_RESERVE_RZONE_TRACK:
-	case GPCMD_CLOSE_TRACK:
-	case GPCMD_FLUSH_CACHE:
-		wait = ATAPI_WAIT_PC;
-		break;
-	default:
-		if (!(rq->cmd_flags & REQ_QUIET))
-			printk(KERN_INFO PFX "cmd 0x%x timed out\n",
-					 rq->cmd[0]);
-		wait = 0;
-		break;
-	}
-	return wait;
-}
-
 /*
  * Set up the device registers for transferring a packet command on DEV,
  * expecting to later transfer XFERLEN bytes.  HANDLER is the routine
@@ -574,7 +542,7 @@ static ide_startstop_t cdrom_start_packet_command(ide_drive_t *drive,
 
 		/* packet command */
 		ide_execute_command(drive, ATA_CMD_PACKET, handler,
-				    ATAPI_WAIT_PC, cdrom_timer_expiry);
+				    ATAPI_WAIT_PC, ide_cd_expiry);
 		return ide_started;
 	} else {
 		ide_execute_pkt_cmd(drive);
@@ -621,7 +589,7 @@ static ide_startstop_t cdrom_transfer_packet_command(ide_drive_t *drive,
 	}
 
 	/* arm the interrupt handler */
-	ide_set_handler(drive, handler, rq->timeout, cdrom_timer_expiry);
+	ide_set_handler(drive, handler, rq->timeout, ide_cd_expiry);
 
 	/* ATAPI commands get padded out to 12 bytes minimum */
 	cmd_len = COMMAND_SIZE(rq->cmd[0]);
@@ -1088,7 +1056,7 @@ static ide_startstop_t cdrom_newpc_intr(ide_drive_t *drive)
 	} else {
 		timeout = ATAPI_WAIT_PC;
 		if (!blk_fs_request(rq))
-			expiry = cdrom_timer_expiry;
+			expiry = ide_cd_expiry;
 	}
 
 	ide_set_handler(drive, cdrom_newpc_intr, timeout, expiry);
diff --git a/drivers/ide/ide-cd.h b/drivers/ide/ide-cd.h
index 389faa42eaa1..bf676b262181 100644
--- a/drivers/ide/ide-cd.h
+++ b/drivers/ide/ide-cd.h
@@ -16,10 +16,6 @@
 #define ide_debug_log(lvl, fmt, args...) do {} while (0)
 #endif
 
-/*
- * typical timeout for packet command
- */
-#define ATAPI_WAIT_PC		(60 * HZ)
 #define ATAPI_WAIT_WRITE_BUSY	(10 * HZ)
 
 /************************************************************************/
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 1f07f3818938..fdec729d0e49 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -197,7 +197,7 @@ static ide_startstop_t idefloppy_issue_pc(ide_drive_t *drive,
 
 	pc->retries++;
 
-	return ide_issue_pc(drive, WAIT_FLOPPY_CMD, NULL);
+	return ide_issue_pc(drive, WAIT_FLOPPY_CMD);
 }
 
 void ide_floppy_create_read_capacity_cmd(struct ide_atapi_pc *pc)
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index a2d470eb2b55..ac9e29a4991f 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -694,7 +694,7 @@ static ide_startstop_t idetape_issue_pc(ide_drive_t *drive,
 
 	pc->retries++;
 
-	return ide_issue_pc(drive, WAIT_TAPE_CMD, NULL);
+	return ide_issue_pc(drive, WAIT_TAPE_CMD);
 }
 
 /* A mode sense command is used to "sense" tape parameters. */
diff --git a/include/linux/ide.h b/include/linux/ide.h
index e35ff6827897..e20e0b5c1739 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -396,6 +396,7 @@ enum {
  * This is used for several packet commands (not for READ/WRITE commands).
  */
 #define IDE_PC_BUFFER_SIZE	256
+#define ATAPI_WAIT_PC		(60 * HZ)
 
 struct ide_atapi_pc {
 	/* actual packet bytes */
@@ -1253,10 +1254,11 @@ static inline unsigned long ide_scsi_get_timeout(struct ide_atapi_pc *pc)
 }
 
 int ide_scsi_expiry(ide_drive_t *);
+int ide_cd_expiry(ide_drive_t *);
 
 int ide_cd_get_xferlen(struct request *);
 
-ide_startstop_t ide_issue_pc(ide_drive_t *, unsigned int, ide_expiry_t *);
+ide_startstop_t ide_issue_pc(ide_drive_t *, unsigned int);
 
 ide_startstop_t do_rw_taskfile(ide_drive_t *, ide_task_t *);
 
-- 
cgit v1.2.3


From 5d655a03b847fbe5353a8a74bbeb75e18708dca3 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Fri, 2 Jan 2009 16:12:54 +0100
Subject: ide-atapi: remove ide-scsi remnants from ide_pc_intr()

As a result, remove now unused ide_scsi_get_timeout and ide_scsi_expiry.

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-atapi.c | 68 ++++++++++---------------------------------------
 include/linux/ide.h     |  6 -----
 2 files changed, 13 insertions(+), 61 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index f5bf405c36aa..7a04509bf962 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -240,19 +240,6 @@ void ide_retry_pc(ide_drive_t *drive, struct gendisk *disk)
 }
 EXPORT_SYMBOL_GPL(ide_retry_pc);
 
-int ide_scsi_expiry(ide_drive_t *drive)
-{
-	struct ide_atapi_pc *pc = drive->pc;
-
-	debug_log("%s called for %lu at %lu\n", __func__,
-		  pc->scsi_cmd->serial_number, jiffies);
-
-	pc->flags |= PC_FLAG_TIMEDOUT;
-
-	return 0; /* we do not want the IDE subsystem to retry */
-}
-EXPORT_SYMBOL_GPL(ide_scsi_expiry);
-
 int ide_cd_expiry(ide_drive_t *drive)
 {
 	struct request *rq = HWGROUP(drive)->rq;
@@ -309,21 +296,14 @@ static ide_startstop_t ide_pc_intr(ide_drive_t *drive)
 	struct request *rq = hwif->hwgroup->rq;
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	xfer_func_t *xferfunc;
-	ide_expiry_t *expiry;
 	unsigned int timeout, temp;
 	u16 bcount;
-	u8 stat, ireason, scsi = !!(drive->dev_flags & IDE_DFLAG_SCSI), dsc = 0;
+	u8 stat, ireason, dsc = 0;
 
 	debug_log("Enter %s - interrupt handler\n", __func__);
 
-	if (scsi) {
-		timeout = ide_scsi_get_timeout(pc);
-		expiry = ide_scsi_expiry;
-	} else {
-		timeout = (drive->media == ide_floppy) ? WAIT_FLOPPY_CMD
-						       : WAIT_TAPE_CMD;
-		expiry = NULL;
-	}
+	timeout = (drive->media == ide_floppy) ? WAIT_FLOPPY_CMD
+					       : WAIT_TAPE_CMD;
 
 	if (pc->flags & PC_FLAG_TIMEDOUT) {
 		drive->pc_callback(drive, 0);
@@ -335,8 +315,8 @@ static ide_startstop_t ide_pc_intr(ide_drive_t *drive)
 
 	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
 		if (hwif->dma_ops->dma_end(drive) ||
-		    (drive->media == ide_tape && !scsi && (stat & ATA_ERR))) {
-			if (drive->media == ide_floppy && !scsi)
+		    (drive->media == ide_tape && (stat & ATA_ERR))) {
+			if (drive->media == ide_floppy)
 				printk(KERN_ERR "%s: DMA %s error\n",
 					drive->name, rq_data_dir(pc->rq)
 						     ? "write" : "read");
@@ -358,7 +338,7 @@ static ide_startstop_t ide_pc_intr(ide_drive_t *drive)
 
 		local_irq_enable_in_hardirq();
 
-		if (drive->media == ide_tape && !scsi &&
+		if (drive->media == ide_tape &&
 		    (stat & ATA_ERR) && rq->cmd[0] == REQUEST_SENSE)
 			stat &= ~ATA_ERR;
 
@@ -366,11 +346,8 @@ static ide_startstop_t ide_pc_intr(ide_drive_t *drive)
 			/* Error detected */
 			debug_log("%s: I/O error\n", drive->name);
 
-			if (drive->media != ide_tape || scsi) {
+			if (drive->media != ide_tape)
 				pc->rq->errors++;
-				if (scsi)
-					goto cmd_finished;
-			}
 
 			if (rq->cmd[0] == REQUEST_SENSE) {
 				printk(KERN_ERR "%s: I/O error in request sense"
@@ -386,7 +363,6 @@ static ide_startstop_t ide_pc_intr(ide_drive_t *drive)
 			/* queued, but not started */
 			return ide_stopped;
 		}
-cmd_finished:
 		pc->error = 0;
 
 		if ((pc->flags & PC_FLAG_WAIT_FOR_DSC) && (stat & ATA_DSC) == 0)
@@ -433,25 +409,8 @@ cmd_finished:
 						"us more data than expected - "
 						"discarding data\n",
 						drive->name);
-				if (scsi)
-					temp = pc->buf_size - pc->xferred;
-				else
-					temp = 0;
-				if (temp) {
-					if (pc->sg)
-						drive->pc_io_buffers(drive, pc,
-								     temp, 0);
-					else
-						tp_ops->input_data(drive, NULL,
-							pc->cur_pos, temp);
-					printk(KERN_ERR "%s: transferred %d of "
-							"%d bytes\n",
-							drive->name,
-							temp, bcount);
-				}
-				pc->xferred += temp;
-				pc->cur_pos += temp;
-				ide_pad_transfer(drive, 0, bcount - temp);
+
+				ide_pad_transfer(drive, 0, bcount);
 				goto next_irq;
 			}
 			debug_log("The device wants to send us more data than "
@@ -461,14 +420,13 @@ cmd_finished:
 	} else
 		xferfunc = tp_ops->output_data;
 
-	if ((drive->media == ide_floppy && !scsi && !pc->buf) ||
-	    (drive->media == ide_tape && !scsi && pc->bh) ||
-	    (scsi && pc->sg)) {
+	if ((drive->media == ide_floppy && !pc->buf) ||
+	    (drive->media == ide_tape && pc->bh)) {
 		int done = drive->pc_io_buffers(drive, pc, bcount,
 				  !!(pc->flags & PC_FLAG_WRITING));
 
 		/* FIXME: don't do partial completions */
-		if (drive->media == ide_floppy && !scsi)
+		if (drive->media == ide_floppy)
 			ide_end_request(drive, 1, done >> 9);
 	} else
 		xferfunc(drive, NULL, pc->cur_pos, bcount);
@@ -481,7 +439,7 @@ cmd_finished:
 		  rq->cmd[0], bcount);
 next_irq:
 	/* And set the interrupt handler again */
-	ide_set_handler(drive, ide_pc_intr, timeout, expiry);
+	ide_set_handler(drive, ide_pc_intr, timeout, NULL);
 	return ide_started;
 }
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index e20e0b5c1739..257524ee1af2 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1248,12 +1248,6 @@ int ide_set_media_lock(ide_drive_t *, struct gendisk *, int);
 void ide_create_request_sense_cmd(ide_drive_t *, struct ide_atapi_pc *);
 void ide_retry_pc(ide_drive_t *, struct gendisk *);
 
-static inline unsigned long ide_scsi_get_timeout(struct ide_atapi_pc *pc)
-{
-	return max_t(unsigned long, WAIT_CMD, pc->timeout - jiffies);
-}
-
-int ide_scsi_expiry(ide_drive_t *);
 int ide_cd_expiry(ide_drive_t *);
 
 int ide_cd_get_xferlen(struct request *);
-- 
cgit v1.2.3


From 5317464dccd0c03026d60f1e9968de4f9cd23f69 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Fri, 2 Jan 2009 16:12:54 +0100
Subject: ide: remove the last ide-scsi remnants

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-atapi.c  |  3 +--
 drivers/ide/ide-io.c     |  3 ---
 drivers/ide/ide-ioctls.c |  3 +--
 drivers/ide/ide-probe.c  |  2 --
 include/linux/ide.h      | 28 +++++++++++++---------------
 5 files changed, 15 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index 7a04509bf962..d412bd2bd7fd 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -17,8 +17,7 @@
 
 static inline int dev_is_idecd(ide_drive_t *drive)
 {
-	return (drive->media == ide_cdrom || drive->media == ide_optical) &&
-		!(drive->dev_flags & IDE_DFLAG_SCSI);
+	return drive->media == ide_cdrom || drive->media == ide_optical;
 }
 
 /*
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index bb3248abf47d..1c36a8e83d36 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -426,9 +426,6 @@ void ide_map_sg(ide_drive_t *drive, struct request *rq)
 	ide_hwif_t *hwif = drive->hwif;
 	struct scatterlist *sg = hwif->sg_table;
 
-	if (hwif->sg_mapped)	/* needed by ide-scsi */
-		return;
-
 	if (rq->cmd_type != REQ_TYPE_ATA_TASKFILE) {
 		hwif->sg_nents = blk_rq_map_sg(drive->queue, rq, sg);
 	} else {
diff --git a/drivers/ide/ide-ioctls.c b/drivers/ide/ide-ioctls.c
index 28232c64c346..1be263eb9c07 100644
--- a/drivers/ide/ide-ioctls.c
+++ b/drivers/ide/ide-ioctls.c
@@ -95,8 +95,7 @@ static int ide_set_nice_ioctl(ide_drive_t *drive, unsigned long arg)
 		return -EPERM;
 
 	if (((arg >> IDE_NICE_DSC_OVERLAP) & 1) &&
-	    (drive->media != ide_tape ||
-	     (drive->dev_flags & IDE_DFLAG_SCSI)))
+	    (drive->media != ide_tape))
 		return -EPERM;
 
 	if ((arg >> IDE_NICE_DSC_OVERLAP) & 1)
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 966b74c15773..c5adb7b9c5b5 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1141,8 +1141,6 @@ static struct kobject *ata_probe(dev_t dev, int *part, void *data)
 
 	if (drive->media == ide_disk)
 		request_module("ide-disk");
-	if (drive->dev_flags & IDE_DFLAG_SCSI)
-		request_module("ide-scsi");
 	if (drive->media == ide_cdrom || drive->media == ide_optical)
 		request_module("ide-cd");
 	if (drive->media == ide_tape)
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 257524ee1af2..ad57a4492941 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -559,28 +559,26 @@ enum {
 	IDE_DFLAG_NODMA			= (1 << 16),
 	/* powermanagment told us not to do anything, so sleep nicely */
 	IDE_DFLAG_BLOCKED		= (1 << 17),
-	/* ide-scsi emulation */
-	IDE_DFLAG_SCSI			= (1 << 18),
 	/* sleeping & sleep field valid */
-	IDE_DFLAG_SLEEPING		= (1 << 19),
-	IDE_DFLAG_POST_RESET		= (1 << 20),
-	IDE_DFLAG_UDMA33_WARNED		= (1 << 21),
-	IDE_DFLAG_LBA48			= (1 << 22),
+	IDE_DFLAG_SLEEPING		= (1 << 18),
+	IDE_DFLAG_POST_RESET		= (1 << 19),
+	IDE_DFLAG_UDMA33_WARNED		= (1 << 20),
+	IDE_DFLAG_LBA48			= (1 << 21),
 	/* status of write cache */
-	IDE_DFLAG_WCACHE		= (1 << 23),
+	IDE_DFLAG_WCACHE		= (1 << 22),
 	/* used for ignoring ATA_DF */
-	IDE_DFLAG_NOWERR		= (1 << 24),
+	IDE_DFLAG_NOWERR		= (1 << 23),
 	/* retrying in PIO */
-	IDE_DFLAG_DMA_PIO_RETRY		= (1 << 25),
-	IDE_DFLAG_LBA			= (1 << 26),
+	IDE_DFLAG_DMA_PIO_RETRY		= (1 << 24),
+	IDE_DFLAG_LBA			= (1 << 25),
 	/* don't unload heads */
-	IDE_DFLAG_NO_UNLOAD		= (1 << 27),
+	IDE_DFLAG_NO_UNLOAD		= (1 << 26),
 	/* heads unloaded, please don't reset port */
-	IDE_DFLAG_PARKED		= (1 << 28),
-	IDE_DFLAG_MEDIA_CHANGED		= (1 << 29),
+	IDE_DFLAG_PARKED		= (1 << 27),
+	IDE_DFLAG_MEDIA_CHANGED		= (1 << 28),
 	/* write protect */
-	IDE_DFLAG_WP			= (1 << 30),
-	IDE_DFLAG_FORMAT_IN_PROGRESS	= (1 << 31),
+	IDE_DFLAG_WP			= (1 << 29),
+	IDE_DFLAG_FORMAT_IN_PROGRESS	= (1 << 30),
 };
 
 struct ide_drive_s {
-- 
cgit v1.2.3


From 28ad91db77755f1c49d79652de11b28ee2cfbf03 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@googlemail.com>
Date: Fri, 2 Jan 2009 16:12:56 +0100
Subject: ide-atapi: remove timeout arg to ide_issue_pc

There should be no functionality change resulting from this patch.

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-atapi.c  | 7 ++++++-
 drivers/ide/ide-floppy.c | 2 +-
 drivers/ide/ide-tape.c   | 2 +-
 include/linux/ide.h      | 2 +-
 4 files changed, 9 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index fa7a70a3f24c..c470dbb155ca 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -562,11 +562,12 @@ static ide_startstop_t ide_transfer_pc(ide_drive_t *drive)
 	return ide_started;
 }
 
-ide_startstop_t ide_issue_pc(ide_drive_t *drive, unsigned int timeout)
+ide_startstop_t ide_issue_pc(ide_drive_t *drive)
 {
 	struct ide_atapi_pc *pc;
 	ide_hwif_t *hwif = drive->hwif;
 	ide_expiry_t *expiry = NULL;
+	unsigned int timeout;
 	u32 tf_flags;
 	u16 bcount;
 
@@ -574,6 +575,7 @@ ide_startstop_t ide_issue_pc(ide_drive_t *drive, unsigned int timeout)
 		tf_flags = IDE_TFLAG_OUT_NSECT | IDE_TFLAG_OUT_LBAL;
 		bcount = ide_cd_get_xferlen(hwif->hwgroup->rq);
 		expiry = ide_cd_expiry;
+		timeout = ATAPI_WAIT_PC;
 
 		if (drive->dma)
 			drive->dma = !hwif->dma_ops->dma_setup(drive);
@@ -600,6 +602,9 @@ ide_startstop_t ide_issue_pc(ide_drive_t *drive, unsigned int timeout)
 
 		if (!drive->dma)
 			pc->flags &= ~PC_FLAG_DMA_OK;
+
+		timeout = (drive->media == ide_floppy) ? WAIT_FLOPPY_CMD
+						       : WAIT_TAPE_CMD;
 	}
 
 	ide_pktcmd_tf_load(drive, tf_flags, bcount, drive->dma);
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index fdec729d0e49..0a48e2dc53a2 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -197,7 +197,7 @@ static ide_startstop_t idefloppy_issue_pc(ide_drive_t *drive,
 
 	pc->retries++;
 
-	return ide_issue_pc(drive, WAIT_FLOPPY_CMD);
+	return ide_issue_pc(drive);
 }
 
 void ide_floppy_create_read_capacity_cmd(struct ide_atapi_pc *pc)
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index ac9e29a4991f..5d2aa22cd6e4 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -694,7 +694,7 @@ static ide_startstop_t idetape_issue_pc(ide_drive_t *drive,
 
 	pc->retries++;
 
-	return ide_issue_pc(drive, WAIT_TAPE_CMD);
+	return ide_issue_pc(drive);
 }
 
 /* A mode sense command is used to "sense" tape parameters. */
diff --git a/include/linux/ide.h b/include/linux/ide.h
index ad57a4492941..db5ef8ae1ab9 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1250,7 +1250,7 @@ int ide_cd_expiry(ide_drive_t *);
 
 int ide_cd_get_xferlen(struct request *);
 
-ide_startstop_t ide_issue_pc(ide_drive_t *, unsigned int);
+ide_startstop_t ide_issue_pc(ide_drive_t *);
 
 ide_startstop_t do_rw_taskfile(ide_drive_t *, ide_task_t *);
 
-- 
cgit v1.2.3


From 56c451f4b583ccdf80c9e676179c9cb49de86745 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Thu, 18 Dec 2008 14:49:37 +0900
Subject: [SCSI] block: fix the partial mappings with struct rq_map_data

This fixes bio_copy_user_iov to properly handle the partial mappings
with struct rq_map_data (which only sg uses for now but st and osst
will shortly). It adds the offset member to struct rq_map_data and
changes blk_rq_map_user to update it so that bio_copy_user_iov can add
an appropriate page frame via bio_add_pc_page().

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 block/blk-map.c        |  3 +++
 drivers/scsi/sg.c      |  1 +
 fs/bio.c               | 12 +++++++++---
 include/linux/blkdev.h |  1 +
 4 files changed, 14 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-map.c b/block/blk-map.c
index 2990447f45e9..c7e55b23a2bc 100644
--- a/block/blk-map.c
+++ b/block/blk-map.c
@@ -150,6 +150,9 @@ int blk_rq_map_user(struct request_queue *q, struct request *rq,
 			bio = rq->bio;
 		bytes_read += ret;
 		ubuf += ret;
+
+		if (map_data)
+			map_data->offset += ret;
 	}
 
 	if (!bio_flagged(bio, BIO_USER_MAPPED))
diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c
index 5103855242ae..7d0b3d9ee43b 100644
--- a/drivers/scsi/sg.c
+++ b/drivers/scsi/sg.c
@@ -1669,6 +1669,7 @@ static int sg_start_req(Sg_request *srp, unsigned char *cmd)
 		md->pages = req_schp->pages;
 		md->page_order = req_schp->page_order;
 		md->nr_entries = req_schp->k_use_sg;
+		md->offset = 0;
 	}
 
 	if (iov_count)
diff --git a/fs/bio.c b/fs/bio.c
index 356e7423b923..13be075806b6 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -788,6 +788,7 @@ struct bio *bio_copy_user_iov(struct request_queue *q,
 	int i, ret;
 	int nr_pages = 0;
 	unsigned int len = 0;
+	unsigned int offset = map_data ? map_data->offset & ~PAGE_MASK : 0;
 
 	for (i = 0; i < iov_count; i++) {
 		unsigned long uaddr;
@@ -814,12 +815,16 @@ struct bio *bio_copy_user_iov(struct request_queue *q,
 	bio->bi_rw |= (!write_to_vm << BIO_RW);
 
 	ret = 0;
-	i = 0;
-	if (map_data)
+
+	if (map_data) {
 		nr_pages = 1 << map_data->page_order;
+		i = map_data->offset / PAGE_SIZE;
+	}
 	while (len) {
 		unsigned int bytes = PAGE_SIZE;
 
+		bytes -= offset;
+
 		if (bytes > len)
 			bytes = len;
 
@@ -841,10 +846,11 @@ struct bio *bio_copy_user_iov(struct request_queue *q,
 			}
 		}
 
-		if (bio_add_pc_page(q, bio, page, bytes, 0) < bytes)
+		if (bio_add_pc_page(q, bio, page, bytes, offset) < bytes)
 			break;
 
 		len -= bytes;
+		offset = 0;
 	}
 
 	if (ret)
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 7035cec583b6..811e5342c452 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -690,6 +690,7 @@ struct rq_map_data {
 	struct page **pages;
 	int page_order;
 	int nr_entries;
+	unsigned long offset;
 };
 
 struct req_iterator {
-- 
cgit v1.2.3


From 97ae77a1cd332c7b011d71315c8faabce6840c72 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Thu, 18 Dec 2008 14:49:38 +0900
Subject: [SCSI] block: make blk_rq_map_user take a NULL user-space buffer for
 WRITE

The commit 818827669d85b84241696ffef2de485db46b0b5e (block: make
blk_rq_map_user take a NULL user-space buffer) extended
blk_rq_map_user to accept a NULL user-space buffer with a READ
command. It was necessary to convert sg to use the block layer mapping
API.

This patch extends blk_rq_map_user again for a WRITE command. It is
necessary to convert st and osst drivers to use the block layer
apping API.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 block/blk-map.c        | 16 +++++++---------
 drivers/scsi/sg.c      |  1 +
 fs/bio.c               |  2 +-
 include/linux/blkdev.h |  1 +
 4 files changed, 10 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-map.c b/block/blk-map.c
index c7e55b23a2bc..f103729b462f 100644
--- a/block/blk-map.c
+++ b/block/blk-map.c
@@ -42,7 +42,7 @@ static int __blk_rq_unmap_user(struct bio *bio)
 
 static int __blk_rq_map_user(struct request_queue *q, struct request *rq,
 			     struct rq_map_data *map_data, void __user *ubuf,
-			     unsigned int len, int null_mapped, gfp_t gfp_mask)
+			     unsigned int len, gfp_t gfp_mask)
 {
 	unsigned long uaddr;
 	struct bio *bio, *orig_bio;
@@ -63,7 +63,7 @@ static int __blk_rq_map_user(struct request_queue *q, struct request *rq,
 	if (IS_ERR(bio))
 		return PTR_ERR(bio);
 
-	if (null_mapped)
+	if (map_data && map_data->null_mapped)
 		bio->bi_flags |= (1 << BIO_NULL_MAPPED);
 
 	orig_bio = bio;
@@ -114,17 +114,15 @@ int blk_rq_map_user(struct request_queue *q, struct request *rq,
 {
 	unsigned long bytes_read = 0;
 	struct bio *bio = NULL;
-	int ret, null_mapped = 0;
+	int ret;
 
 	if (len > (q->max_hw_sectors << 9))
 		return -EINVAL;
 	if (!len)
 		return -EINVAL;
-	if (!ubuf) {
-		if (!map_data || rq_data_dir(rq) != READ)
-			return -EINVAL;
-		null_mapped = 1;
-	}
+
+	if (!ubuf && (!map_data || !map_data->null_mapped))
+		return -EINVAL;
 
 	while (bytes_read != len) {
 		unsigned long map_len, end, start;
@@ -143,7 +141,7 @@ int blk_rq_map_user(struct request_queue *q, struct request *rq,
 			map_len -= PAGE_SIZE;
 
 		ret = __blk_rq_map_user(q, rq, map_data, ubuf, map_len,
-					null_mapped, gfp_mask);
+					gfp_mask);
 		if (ret < 0)
 			goto unmap_rq;
 		if (!bio)
diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c
index 7d0b3d9ee43b..8f0bd3f7a59f 100644
--- a/drivers/scsi/sg.c
+++ b/drivers/scsi/sg.c
@@ -1670,6 +1670,7 @@ static int sg_start_req(Sg_request *srp, unsigned char *cmd)
 		md->page_order = req_schp->page_order;
 		md->nr_entries = req_schp->k_use_sg;
 		md->offset = 0;
+		md->null_mapped = hp->dxferp ? 0 : 1;
 	}
 
 	if (iov_count)
diff --git a/fs/bio.c b/fs/bio.c
index 13be075806b6..062299acbccd 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -859,7 +859,7 @@ struct bio *bio_copy_user_iov(struct request_queue *q,
 	/*
 	 * success
 	 */
-	if (!write_to_vm) {
+	if (!write_to_vm && (!map_data || !map_data->null_mapped)) {
 		ret = __bio_copy_iov(bio, bio->bi_io_vec, iov, iov_count, 0, 0);
 		if (ret)
 			goto cleanup;
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 811e5342c452..044467ef7b11 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -691,6 +691,7 @@ struct rq_map_data {
 	int page_order;
 	int nr_entries;
 	unsigned long offset;
+	int null_mapped;
 };
 
 struct req_iterator {
-- 
cgit v1.2.3


From f153b82121b0366fe0e5f9553545cce237335175 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Fri, 2 Jan 2009 09:23:03 -0800
Subject: Sanitize gcc version header includes

 - include the gcc version-dependent header files from the generic gcc
   header file, rather than the other way around (iow: don't make the
   non-gcc header file have to know about gcc versions)

 - don't include compiler-gcc4.h for gcc 5 (for whenever it gets
   released).  That's just confusing and made us do odd things in the
   gcc4 header file (testing that we really had version 4!)

 - generate the name from the __GNUC__ version directly, rather than
   having a mess of #if conditionals.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/compiler-gcc.h  | 5 +++++
 include/linux/compiler-gcc3.h | 3 ---
 include/linux/compiler-gcc4.h | 5 +----
 include/linux/compiler.h      | 8 ++------
 4 files changed, 8 insertions(+), 13 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h
index 5c8351b859f0..af40f8eb86f0 100644
--- a/include/linux/compiler-gcc.h
+++ b/include/linux/compiler-gcc.h
@@ -61,3 +61,8 @@
 #define  noinline			__attribute__((noinline))
 #define __attribute_const__		__attribute__((__const__))
 #define __maybe_unused			__attribute__((unused))
+
+#define __gcc_header(x) #x
+#define _gcc_header(x) __gcc_header(linux/compiler-gcc##x.h)
+#define gcc_header(x) _gcc_header(x)
+#include gcc_header(__GNUC__)
diff --git a/include/linux/compiler-gcc3.h b/include/linux/compiler-gcc3.h
index e5eb795f78a1..2befe6513ce4 100644
--- a/include/linux/compiler-gcc3.h
+++ b/include/linux/compiler-gcc3.h
@@ -2,9 +2,6 @@
 #error "Please don't include <linux/compiler-gcc3.h> directly, include <linux/compiler.h> instead."
 #endif
 
-/* These definitions are for GCC v3.x.  */
-#include <linux/compiler-gcc.h>
-
 #if __GNUC_MINOR__ >= 3
 # define __used			__attribute__((__used__))
 #else
diff --git a/include/linux/compiler-gcc4.h b/include/linux/compiler-gcc4.h
index 974f5b7bb205..aa426214331b 100644
--- a/include/linux/compiler-gcc4.h
+++ b/include/linux/compiler-gcc4.h
@@ -2,9 +2,6 @@
 #error "Please don't include <linux/compiler-gcc4.h> directly, include <linux/compiler.h> instead."
 #endif
 
-/* These definitions are for GCC v4.x.  */
-#include <linux/compiler-gcc.h>
-
 #define __used			__attribute__((__used__))
 #define __must_check 		__attribute__((warn_unused_result))
 #define __compiler_offsetof(a,b) __builtin_offsetof(a,b)
@@ -16,7 +13,7 @@
  */
 #define uninitialized_var(x) x = x
 
-#if !(__GNUC__ == 4 && __GNUC_MINOR__ < 3)
+#if __GNUC_MINOR__ >= 3
 /* Mark functions as cold. gcc will assume any path leading to a call
    to them will be unlikely.  This means a lot of manual unlikely()s
    are unnecessary now for any paths leading to the usual suspects
diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index ea7c6be354b7..d95da1020f1c 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -36,12 +36,8 @@ extern void __chk_io_ptr(const volatile void __iomem *);
 
 #ifdef __KERNEL__
 
-#if __GNUC__ >= 4
-# include <linux/compiler-gcc4.h>
-#elif __GNUC__ == 3 && __GNUC_MINOR__ >= 2
-# include <linux/compiler-gcc3.h>
-#else
-# error Sorry, your compiler is too old/not recognized.
+#ifdef __GNUC__
+#include <linux/compiler-gcc.h>
 #endif
 
 #define notrace __attribute__((no_instrument_function))
-- 
cgit v1.2.3


From f9d14250071eda9972e4c9cea745a11185952114 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Fri, 2 Jan 2009 09:29:43 -0800
Subject: Disallow gcc versions 4.1.{0,1}

These compiler versions are known to miscompile __weak functions and
thus generate kernels that don't necessarily work correctly.  If a weak
function is int he same compilation unit as a caller, gcc may end up
inlining it, and thus binding the weak function too early.

See

    http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27781

for details.

Cc: Adrian Bunk <bunk@kernel.org>
Cc: Helge Deller <deller@gmx.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/compiler-gcc4.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/compiler-gcc4.h b/include/linux/compiler-gcc4.h
index aa426214331b..09992718f9e8 100644
--- a/include/linux/compiler-gcc4.h
+++ b/include/linux/compiler-gcc4.h
@@ -2,6 +2,11 @@
 #error "Please don't include <linux/compiler-gcc4.h> directly, include <linux/compiler.h> instead."
 #endif
 
+/* GCC 4.1.[01] miscompiles __weak */
+#if __GNUC_MINOR__ == 1 && __GNUC_PATCHLEVEL__ <= 1
+# error Your version of gcc miscompiles the __weak directive
+#endif
+
 #define __used			__attribute__((__used__))
 #define __must_check 		__attribute__((warn_unused_result))
 #define __compiler_offsetof(a,b) __builtin_offsetof(a,b)
-- 
cgit v1.2.3


From a88a69c91256418c5907c2f1f8a0ec0a36f9e6cc Mon Sep 17 00:00:00 2001
From: Joe Peterson <joe@skyrush.com>
Date: Fri, 2 Jan 2009 13:40:53 +0000
Subject: n_tty: Fix loss of echoed characters and remove bkl from n_tty

Fixes the loss of echoed (and other ldisc-generated characters) when
the tty is stopped or when the driver output buffer is full (happens
frequently for input during continuous program output, such as ^C)
and removes the Big Kernel Lock from the N_TTY line discipline.

Adds an "echo buffer" to the N_TTY line discipline that handles all
ldisc-generated output (including echoed characters).  Along with the
loss of characters, this also fixes the associated loss of sync between
tty output and the ldisc state when characters cannot be immediately
written to the tty driver.

The echo buffer stores (in addition to characters) state operations that need
to be done at the time of character output (like management of the column
position).  This allows echo to cooperate correctly with program output,
since the ldisc state remains consistent with actual characters written.

Since the echo buffer code now isolates the tty column state code
to the process_out* and process_echoes functions, we can remove the
Big Kernel Lock (BKL) and replace it with mutex locks.

Highlights are:

* Handles echo (and other ldisc output) when tty driver buffer is full
  - continuous program output can block echo
* Saves echo when tty is in stopped state (e.g. ^S)
  - (e.g.: ^Q will correctly cause held characters to be released for output)
* Control character pairs (e.g. "^C") are treated atomically and not
  split up by interleaved program output
* Line discipline state is kept consistent with characters sent to
  the tty driver
* Remove the big kernel lock (BKL) from N_TTY line discipline

Signed-off-by: Joe Peterson <joe@skyrush.com>
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/n_tty.c  | 736 +++++++++++++++++++++++++++++++++++++++-----------
 drivers/char/tty_io.c |   6 +-
 drivers/char/vt.c     |   2 +-
 include/linux/tty.h   |   6 +
 4 files changed, 594 insertions(+), 156 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/n_tty.c b/drivers/char/n_tty.c
index efbfe9612658..a9bc5764fe75 100644
--- a/drivers/char/n_tty.c
+++ b/drivers/char/n_tty.c
@@ -62,6 +62,17 @@
 #define TTY_THRESHOLD_THROTTLE		128 /* now based on remaining room */
 #define TTY_THRESHOLD_UNTHROTTLE 	128
 
+/*
+ * Special byte codes used in the echo buffer to represent operations
+ * or special handling of characters.  Bytes in the echo buffer that
+ * are not part of such special blocks are treated as normal character
+ * codes.
+ */
+#define ECHO_OP_START 0xff
+#define ECHO_OP_MOVE_BACK_COL 0x80
+#define ECHO_OP_SET_CANON_COL 0x81
+#define ECHO_OP_ERASE_TAB 0x82
+
 static inline unsigned char *alloc_buf(void)
 {
 	gfp_t prio = in_interrupt() ? GFP_ATOMIC : GFP_KERNEL;
@@ -169,6 +180,7 @@ static void check_unthrottle(struct tty_struct *tty)
  *
  *	Locking: tty_read_lock for read fields.
  */
+
 static void reset_buffer_flags(struct tty_struct *tty)
 {
 	unsigned long flags;
@@ -176,6 +188,11 @@ static void reset_buffer_flags(struct tty_struct *tty)
 	spin_lock_irqsave(&tty->read_lock, flags);
 	tty->read_head = tty->read_tail = tty->read_cnt = 0;
 	spin_unlock_irqrestore(&tty->read_lock, flags);
+
+	mutex_lock(&tty->echo_lock);
+	tty->echo_pos = tty->echo_cnt = tty->echo_overrun = 0;
+	mutex_unlock(&tty->echo_lock);
+
 	tty->canon_head = tty->canon_data = tty->erasing = 0;
 	memset(&tty->read_flags, 0, sizeof tty->read_flags);
 	n_tty_set_room(tty);
@@ -266,89 +283,116 @@ static inline int is_continuation(unsigned char c, struct tty_struct *tty)
 }
 
 /**
- *	opost			-	output post processor
+ *	do_output_char			-	output one character
  *	@c: character (or partial unicode symbol)
  *	@tty: terminal device
+ *	@space: space available in tty driver write buffer
  *
- *	Perform OPOST processing.  Returns -1 when the output device is
- *	full and the character must be retried. Note that Linux currently
- *	ignores TABDLY, CRDLY, VTDLY, FFDLY and NLDLY. They simply aren't
- *	relevant in the world today. If you ever need them, add them here.
+ *	This is a helper function that handles one output character
+ *	(including special characters like TAB, CR, LF, etc.),
+ *	putting the results in the tty driver's write buffer.
+ *
+ *	Note that Linux currently ignores TABDLY, CRDLY, VTDLY, FFDLY
+ *	and NLDLY.  They simply aren't relevant in the world today.
+ *	If you ever need them, add them here.
  *
- *	Called from both the receive and transmit sides and can be called
- *	re-entrantly. Relies on lock_kernel() for tty->column state.
+ *	Returns the number of bytes of buffer space used or -1 if
+ *	no space left.
+ *
+ *	Locking: should be called under the output_lock to protect
+ *		 the column state and space left in the buffer
  */
 
-static int opost(unsigned char c, struct tty_struct *tty)
+static int do_output_char(unsigned char c, struct tty_struct *tty, int space)
 {
-	int	space, spaces;
+	int	spaces;
 
-	space = tty_write_room(tty);
 	if (!space)
 		return -1;
-
-	lock_kernel();
-	if (O_OPOST(tty)) {
-		switch (c) {
-		case '\n':
-			if (O_ONLRET(tty))
-				tty->column = 0;
-			if (O_ONLCR(tty)) {
-				if (space < 2) {
-					unlock_kernel();
-					return -1;
-				}
-				tty_put_char(tty, '\r');
-				tty->column = 0;
-			}
-			tty->canon_column = tty->column;
-			break;
-		case '\r':
-			if (O_ONOCR(tty) && tty->column == 0) {
-				unlock_kernel();
-				return 0;
-			}
-			if (O_OCRNL(tty)) {
-				c = '\n';
-				if (O_ONLRET(tty))
-					tty->canon_column = tty->column = 0;
-				break;
-			}
+	
+	switch (c) {
+	case '\n':
+		if (O_ONLRET(tty))
+			tty->column = 0;
+		if (O_ONLCR(tty)) {
+			if (space < 2)
+				return -1;
 			tty->canon_column = tty->column = 0;
+			tty_put_char(tty, '\r');
+			tty_put_char(tty, c);
+			return 2;
+		}
+		tty->canon_column = tty->column;
+		break;
+	case '\r':
+		if (O_ONOCR(tty) && tty->column == 0)
+			return 0;
+		if (O_OCRNL(tty)) {
+			c = '\n';
+			if (O_ONLRET(tty))
+				tty->canon_column = tty->column = 0;
 			break;
-		case '\t':
-			spaces = 8 - (tty->column & 7);
-			if (O_TABDLY(tty) == XTABS) {
-				if (space < spaces) {
-					unlock_kernel();
-					return -1;
-				}
-				tty->column += spaces;
-				tty->ops->write(tty, "        ", spaces);
-				unlock_kernel();
-				return 0;
-			}
+		}
+		tty->canon_column = tty->column = 0;
+		break;
+	case '\t':
+		spaces = 8 - (tty->column & 7);
+		if (O_TABDLY(tty) == XTABS) {
+			if (space < spaces)
+				return -1;
 			tty->column += spaces;
-			break;
-		case '\b':
-			if (tty->column > 0)
-				tty->column--;
-			break;
-		default:
-			if (O_OLCUC(tty))
-				c = toupper(c);
-			if (!iscntrl(c) && !is_continuation(c, tty))
-				tty->column++;
-			break;
+			tty->ops->write(tty, "        ", spaces);
+			return spaces;
 		}
+		tty->column += spaces;
+		break;
+	case '\b':
+		if (tty->column > 0)
+			tty->column--;
+		break;
+	default:
+		if (O_OLCUC(tty))
+			c = toupper(c);
+		if (!iscntrl(c) && !is_continuation(c, tty))
+			tty->column++;
+		break;
 	}
+
 	tty_put_char(tty, c);
-	unlock_kernel();
-	return 0;
+	return 1;
+}
+
+/**
+ *	process_output			-	output post processor
+ *	@c: character (or partial unicode symbol)
+ *	@tty: terminal device
+ *
+ *	Perform OPOST processing.  Returns -1 when the output device is
+ *	full and the character must be retried.
+ *
+ *	Locking: output_lock to protect column state and space left
+ *		 (also, this is called from n_tty_write under the
+ *		  tty layer write lock)
+ */
+
+static int process_output(unsigned char c, struct tty_struct *tty)
+{
+	int	space, retval;
+
+	mutex_lock(&tty->output_lock);
+
+	space = tty_write_room(tty);
+	retval = do_output_char(c, tty, space);
+
+	mutex_unlock(&tty->output_lock);
+	if (retval < 0)
+		return -1;
+	else
+		return 0;
 }
 
 /**
- *	opost_block		-	block postprocess
+ *	process_output_block		-	block post processor
  *	@tty: terminal device
  *	@inbuf: user buffer
  *	@nr: number of bytes
@@ -358,24 +402,29 @@ static int opost(unsigned char c, struct tty_struct *tty)
  *	the simple cases normally found and helps to generate blocks of
  *	symbols for the console driver and thus improve performance.
  *
- *	Called from n_tty_write under the tty layer write lock. Relies
- *	on lock_kernel for the tty->column state.
+ *	Locking: output_lock to protect column state and space left
+ *		 (also, this is called from n_tty_write under the
+ *		  tty layer write lock)
  */
 
-static ssize_t opost_block(struct tty_struct *tty,
-		       const unsigned char *buf, unsigned int nr)
+static ssize_t process_output_block(struct tty_struct *tty,
+				    const unsigned char *buf, unsigned int nr)
 {
 	int	space;
 	int 	i;
 	const unsigned char *cp;
 
+	mutex_lock(&tty->output_lock);
+
 	space = tty_write_room(tty);
 	if (!space)
+	{
+		mutex_unlock(&tty->output_lock);
 		return 0;
+	}
 	if (nr > space)
 		nr = space;
 
-	lock_kernel();
 	for (i = 0, cp = buf; i < nr; i++, cp++) {
 		switch (*cp) {
 		case '\n':
@@ -407,46 +456,393 @@ static ssize_t opost_block(struct tty_struct *tty,
 		}
 	}
 break_out:
-	if (tty->ops->flush_chars)
-		tty->ops->flush_chars(tty);
 	i = tty->ops->write(tty, buf, i);
-	unlock_kernel();
+
+	mutex_unlock(&tty->output_lock);
 	return i;
 }
 
+/**
+ *	process_echoes	-	write pending echo characters
+ *	@tty: terminal device
+ *
+ *	Write previously buffered echo (and other ldisc-generated)
+ *	characters to the tty.
+ *
+ *	Characters generated by the ldisc (including echoes) need to
+ *	be buffered because the driver's write buffer can fill during
+ *	heavy program output.  Echoing straight to the driver will
+ *	often fail under these conditions, causing lost characters and
+ *	resulting mismatches of ldisc state information.
+ *
+ *	Since the ldisc state must represent the characters actually sent
+ *	to the driver at the time of the write, operations like certain
+ *	changes in column state are also saved in the buffer and executed
+ *	here.
+ *
+ *	A circular fifo buffer is used so that the most recent characters
+ *	are prioritized.  Also, when control characters are echoed with a
+ *	prefixed "^", the pair is treated atomically and thus not separated.
+ *
+ *	Locking: output_lock to protect column state and space left,
+ *		 echo_lock to protect the echo buffer
+ */
+
+static void process_echoes(struct tty_struct *tty)
+{
+	int	space, nr;
+	unsigned char c;
+	unsigned char *cp, *buf_end;
+
+	if (!tty->echo_cnt)
+		return;
+
+	mutex_lock(&tty->output_lock);
+	mutex_lock(&tty->echo_lock);
+
+	space = tty_write_room(tty);
+
+	buf_end = tty->echo_buf + N_TTY_BUF_SIZE;
+	cp = tty->echo_buf + tty->echo_pos;
+	nr = tty->echo_cnt;
+	while (nr > 0) {
+		c = *cp;
+		if (c == ECHO_OP_START) {
+			unsigned char op;
+			unsigned char *opp;
+			int no_space_left = 0;
+
+			/*
+			 * If the buffer byte is the start of a multi-byte
+			 * operation, get the next byte, which is either the
+			 * op code or a control character value.
+			 */
+			opp = cp + 1;
+			if (opp == buf_end)
+				opp -= N_TTY_BUF_SIZE;
+			op = *opp;
+			
+			switch (op) {
+				unsigned int num_chars, num_bs;
+
+			case ECHO_OP_ERASE_TAB:
+				if (++opp == buf_end)
+					opp -= N_TTY_BUF_SIZE;
+				num_chars = *opp;
+
+				/*
+				 * Determine how many columns to go back
+				 * in order to erase the tab.
+				 * This depends on the number of columns
+				 * used by other characters within the tab
+				 * area.  If this (modulo 8) count is from
+				 * the start of input rather than from a
+				 * previous tab, we offset by canon column.
+				 * Otherwise, tab spacing is normal.
+				 */
+				if (!(num_chars & 0x80))
+					num_chars += tty->canon_column;
+				num_bs = 8 - (num_chars & 7);
+
+				if (num_bs > space) {
+					no_space_left = 1;
+					break;
+				}
+				space -= num_bs;
+				while (num_bs--) {
+					tty_put_char(tty, '\b');
+					if (tty->column > 0)
+						tty->column--;
+				}
+				cp += 3;
+				nr -= 3;
+				break;
+
+			case ECHO_OP_SET_CANON_COL:
+				tty->canon_column = tty->column;
+				cp += 2;
+				nr -= 2;
+				break;
+
+			case ECHO_OP_MOVE_BACK_COL:
+				if (tty->column > 0)
+					tty->column--;
+				cp += 2;
+				nr -= 2;
+				break;
+
+			case ECHO_OP_START:
+				/* This is an escaped echo op start code */
+				if (!space) {
+					no_space_left = 1;
+					break;
+				}
+				tty_put_char(tty, ECHO_OP_START);
+				tty->column++;
+				space--;
+				cp += 2;
+				nr -= 2;
+				break;
+
+			default:
+				if (iscntrl(op)) {
+					if (L_ECHOCTL(tty)) {
+						/*
+						 * Ensure there is enough space
+						 * for the whole ctrl pair.
+						 */
+						if (space < 2) {
+							no_space_left = 1;
+							break;
+						}
+						tty_put_char(tty, '^');
+						tty_put_char(tty, op ^ 0100);
+						tty->column += 2;
+						space -= 2;
+					} else {
+						if (!space) {
+							no_space_left = 1;
+							break;
+						}
+						tty_put_char(tty, op);
+						space--;
+					}
+				}
+				/*
+				 * If above falls through, this was an
+				 * undefined op.
+				 */
+				cp += 2;
+				nr -= 2;
+			}
+
+			if (no_space_left)
+				break;
+		} else {
+			int retval;
+
+			if ((retval = do_output_char(c, tty, space)) < 0)
+				break;
+			space -= retval;
+			cp += 1;
+			nr -= 1;
+		}
+
+		/* When end of circular buffer reached, wrap around */
+		if (cp >= buf_end)
+			cp -= N_TTY_BUF_SIZE;
+	}
+
+	if (nr == 0) {
+		tty->echo_pos = 0;
+		tty->echo_cnt = 0;
+		tty->echo_overrun = 0;
+	} else {
+		int num_processed = tty->echo_cnt - nr;
+		tty->echo_pos += num_processed;
+		tty->echo_pos &= N_TTY_BUF_SIZE - 1;
+		tty->echo_cnt = nr;
+		if (num_processed > 0)
+			tty->echo_overrun = 0;
+	}
+
+	mutex_unlock(&tty->echo_lock);
+	mutex_unlock(&tty->output_lock);
+
+	if (tty->ops->flush_chars)
+		tty->ops->flush_chars(tty);
+}
+
+/**
+ *	add_echo_byte	-	add a byte to the echo buffer
+ *	@c: unicode byte to echo
+ *	@tty: terminal device
+ *
+ *	Add a character or operation byte to the echo buffer.
+ *
+ *	Should be called under the echo lock to protect the echo buffer.
+ */
+
+static void add_echo_byte(unsigned char c, struct tty_struct *tty)
+{
+	int	new_byte_pos;
+
+	if (tty->echo_cnt == N_TTY_BUF_SIZE) {
+		/* Circular buffer is already at capacity */
+		new_byte_pos = tty->echo_pos;
+
+		/*
+		 * Since the buffer start position needs to be advanced,
+		 * be sure to step by a whole operation byte group.
+		 */
+		if (tty->echo_buf[tty->echo_pos] == ECHO_OP_START)
+		{
+			if (tty->echo_buf[(tty->echo_pos + 1) &
+					  (N_TTY_BUF_SIZE - 1)] ==
+						ECHO_OP_ERASE_TAB) {
+				tty->echo_pos += 3;
+				tty->echo_cnt -= 2;
+			} else {
+				tty->echo_pos += 2;
+				tty->echo_cnt -= 1;
+			}
+		} else {
+			tty->echo_pos++;
+		}
+		tty->echo_pos &= N_TTY_BUF_SIZE - 1;
+
+		tty->echo_overrun = 1;
+	} else {
+		new_byte_pos = tty->echo_pos + tty->echo_cnt;
+		new_byte_pos &= N_TTY_BUF_SIZE - 1;
+		tty->echo_cnt++;
+	}
+
+	tty->echo_buf[new_byte_pos] = c;
+}
+
+/**
+ *	echo_move_back_col	-	add operation to move back a column
+ *	@tty: terminal device
+ *
+ *	Add an operation to the echo buffer to move back one column.
+ *
+ *	Locking: echo_lock to protect the echo buffer
+ */
+
+static void echo_move_back_col(struct tty_struct *tty)
+{
+	mutex_lock(&tty->echo_lock);
+
+	add_echo_byte(ECHO_OP_START, tty);
+	add_echo_byte(ECHO_OP_MOVE_BACK_COL, tty);
+
+	mutex_unlock(&tty->echo_lock);
+}
+
+/**
+ *	echo_set_canon_col	-	add operation to set the canon column
+ *	@tty: terminal device
+ *
+ *	Add an operation to the echo buffer to set the canon column
+ *	to the current column.
+ *
+ *	Locking: echo_lock to protect the echo buffer
+ */
+
+static void echo_set_canon_col(struct tty_struct *tty)
+{
+	mutex_lock(&tty->echo_lock);
+
+	add_echo_byte(ECHO_OP_START, tty);
+	add_echo_byte(ECHO_OP_SET_CANON_COL, tty);
+
+	mutex_unlock(&tty->echo_lock);
+}
+
+/**
+ *	echo_erase_tab	-	add operation to erase a tab
+ *	@num_chars: number of character columns already used
+ *	@after_tab: true if num_chars starts after a previous tab
+ *	@tty: terminal device
+ *
+ *	Add an operation to the echo buffer to erase a tab.
+ *
+ *	Called by the eraser function, which knows how many character
+ *	columns have been used since either a previous tab or the start
+ *	of input.  This information will be used later, along with
+ *	canon column (if applicable), to go back the correct number
+ *	of columns.
+ *
+ *	Locking: echo_lock to protect the echo buffer
+ */
+
+static void echo_erase_tab(unsigned int num_chars, int after_tab,
+			   struct tty_struct *tty)
+{
+	mutex_lock(&tty->echo_lock);
+
+	add_echo_byte(ECHO_OP_START, tty);
+	add_echo_byte(ECHO_OP_ERASE_TAB, tty);
+
+	/* We only need to know this modulo 8 (tab spacing) */
+	num_chars &= 7;
+
+	/* Set the high bit as a flag if num_chars is after a previous tab */
+	if (after_tab)
+		num_chars |= 0x80;
+	
+	add_echo_byte(num_chars, tty);
+
+	mutex_unlock(&tty->echo_lock);
+}
+
+/**
+ *	echo_char_raw	-	echo a character raw
+ *	@c: unicode byte to echo
+ *	@tty: terminal device
+ *
+ *	Echo user input back onto the screen. This must be called only when
+ *	L_ECHO(tty) is true. Called from the driver receive_buf path.
+ *
+ *	This variant does not treat control characters specially.
+ *
+ *	Locking: echo_lock to protect the echo buffer
+ */
+
+static void echo_char_raw(unsigned char c, struct tty_struct *tty)
+{
+	mutex_lock(&tty->echo_lock);
+
+	if (c == ECHO_OP_START) {
+		add_echo_byte(ECHO_OP_START, tty);
+		add_echo_byte(ECHO_OP_START, tty);
+	} else {
+		add_echo_byte(c, tty);
+	}
+
+	mutex_unlock(&tty->echo_lock);
+}
 
 /**
- *	echo_char	-	echo characters
+ *	echo_char	-	echo a character
  *	@c: unicode byte to echo
  *	@tty: terminal device
  *
  *	Echo user input back onto the screen. This must be called only when
  *	L_ECHO(tty) is true. Called from the driver receive_buf path.
  *
- *	Relies on BKL for tty column locking
+ *	This variant tags control characters to be possibly echoed as
+ *	as "^X" (where X is the letter representing the control char).
+ *
+ *	Locking: echo_lock to protect the echo buffer
  */
 
 static void echo_char(unsigned char c, struct tty_struct *tty)
 {
-	if (L_ECHOCTL(tty) && iscntrl(c) && c != '\t') {
-		tty_put_char(tty, '^');
-		tty_put_char(tty, c ^ 0100);
-		tty->column += 2;
-	} else
-		opost(c, tty);
+	mutex_lock(&tty->echo_lock);
+
+	if (c == ECHO_OP_START) {
+		add_echo_byte(ECHO_OP_START, tty);
+		add_echo_byte(ECHO_OP_START, tty);
+	} else {
+		if (iscntrl(c) && c != '\t')
+			add_echo_byte(ECHO_OP_START, tty);
+		add_echo_byte(c, tty);
+	}
+
+	mutex_unlock(&tty->echo_lock);
 }
 
 /**
- *	finsh_erasing		-	complete erase
+ *	finish_erasing		-	complete erase
  *	@tty: tty doing the erase
- *
- *	Relies on BKL for tty column locking
  */
+
 static inline void finish_erasing(struct tty_struct *tty)
 {
 	if (tty->erasing) {
-		tty_put_char(tty, '/');
-		tty->column++;
+		echo_char_raw('/', tty);
 		tty->erasing = 0;
 	}
 }
@@ -460,7 +856,7 @@ static inline void finish_erasing(struct tty_struct *tty)
  *	present in the stream from the driver layer. Handles the complexities
  *	of UTF-8 multibyte symbols.
  *
- *	Locking: read_lock for tty buffers, BKL for column/erasing state
+ *	Locking: read_lock for tty buffers
  */
 
 static void eraser(unsigned char c, struct tty_struct *tty)
@@ -471,7 +867,7 @@ static void eraser(unsigned char c, struct tty_struct *tty)
 
 	/* FIXME: locking needed ? */
 	if (tty->read_head == tty->canon_head) {
-		/* opost('\a', tty); */		/* what do you think? */
+		/* echo_char_raw('\a', tty); */ /* what do you think? */
 		return;
 	}
 	if (c == ERASE_CHAR(tty))
@@ -497,7 +893,7 @@ static void eraser(unsigned char c, struct tty_struct *tty)
 			echo_char(KILL_CHAR(tty), tty);
 			/* Add a newline if ECHOK is on and ECHOKE is off. */
 			if (L_ECHOK(tty))
-				opost('\n', tty);
+				echo_char_raw('\n', tty);
 			return;
 		}
 		kill_type = KILL;
@@ -533,67 +929,62 @@ static void eraser(unsigned char c, struct tty_struct *tty)
 		if (L_ECHO(tty)) {
 			if (L_ECHOPRT(tty)) {
 				if (!tty->erasing) {
-					tty_put_char(tty, '\\');
-					tty->column++;
+					echo_char_raw('\\', tty);
 					tty->erasing = 1;
 				}
 				/* if cnt > 1, output a multi-byte character */
 				echo_char(c, tty);
 				while (--cnt > 0) {
 					head = (head+1) & (N_TTY_BUF_SIZE-1);
-					tty_put_char(tty, tty->read_buf[head]);
+					echo_char_raw(tty->read_buf[head], tty);
+					echo_move_back_col(tty);
 				}
 			} else if (kill_type == ERASE && !L_ECHOE(tty)) {
 				echo_char(ERASE_CHAR(tty), tty);
 			} else if (c == '\t') {
-				unsigned int col = tty->canon_column;
-				unsigned long tail = tty->canon_head;
-
-				/* Find the column of the last char. */
-				while (tail != tty->read_head) {
+				unsigned int num_chars = 0;
+				int after_tab = 0;
+				unsigned long tail = tty->read_head;
+
+				/*
+				 * Count the columns used for characters
+				 * since the start of input or after a
+				 * previous tab.
+				 * This info is used to go back the correct
+				 * number of columns.
+				 */
+				while (tail != tty->canon_head) {
+					tail = (tail-1) & (N_TTY_BUF_SIZE-1);
 					c = tty->read_buf[tail];
-					if (c == '\t')
-						col = (col | 7) + 1;
+					if (c == '\t') {
+						after_tab = 1;
+						break;
+					}
 					else if (iscntrl(c)) {
 						if (L_ECHOCTL(tty))
-							col += 2;
-					} else if (!is_continuation(c, tty))
-						col++;
-					tail = (tail+1) & (N_TTY_BUF_SIZE-1);
-				}
-
-				/* should never happen */
-				if (tty->column > 0x80000000)
-					tty->column = 0;
-
-				/* Now backup to that column. */
-				while (tty->column > col) {
-					/* Can't use opost here. */
-					tty_put_char(tty, '\b');
-					if (tty->column > 0)
-						tty->column--;
+							num_chars += 2;
+					} else if (!is_continuation(c, tty)) {
+						num_chars++;
+					}
 				}
+				echo_erase_tab(num_chars, after_tab, tty);
 			} else {
 				if (iscntrl(c) && L_ECHOCTL(tty)) {
-					tty_put_char(tty, '\b');
-					tty_put_char(tty, ' ');
-					tty_put_char(tty, '\b');
-					if (tty->column > 0)
-						tty->column--;
+					echo_char_raw('\b', tty);
+					echo_char_raw(' ', tty);
+					echo_char_raw('\b', tty);
 				}
 				if (!iscntrl(c) || L_ECHOCTL(tty)) {
-					tty_put_char(tty, '\b');
-					tty_put_char(tty, ' ');
-					tty_put_char(tty, '\b');
-					if (tty->column > 0)
-						tty->column--;
+					echo_char_raw('\b', tty);
+					echo_char_raw(' ', tty);
+					echo_char_raw('\b', tty);
 				}
 			}
 		}
 		if (kill_type == ERASE)
 			break;
 	}
-	if (tty->read_head == tty->canon_head)
+	if (tty->read_head == tty->canon_head && L_ECHO(tty))
 		finish_erasing(tty);
 }
 
@@ -724,14 +1115,18 @@ static inline void n_tty_receive_char(struct tty_struct *tty, unsigned char c)
 		c=tolower(c);
 
 	if (tty->stopped && !tty->flow_stopped && I_IXON(tty) &&
-	    ((I_IXANY(tty) && c != START_CHAR(tty) && c != STOP_CHAR(tty)) ||
-	     c == INTR_CHAR(tty) || c == QUIT_CHAR(tty) || c == SUSP_CHAR(tty)))
+	    I_IXANY(tty) && c != START_CHAR(tty) && c != STOP_CHAR(tty) &&
+	    c != INTR_CHAR(tty) && c != QUIT_CHAR(tty) && c != SUSP_CHAR(tty)) {
 		start_tty(tty);
+		process_echoes(tty);
+	}
 
 	if (tty->closing) {
 		if (I_IXON(tty)) {
-			if (c == START_CHAR(tty))
+			if (c == START_CHAR(tty)) {
 				start_tty(tty);
+				process_echoes(tty);
+			}
 			else if (c == STOP_CHAR(tty))
 				stop_tty(tty);
 		}
@@ -745,17 +1140,20 @@ static inline void n_tty_receive_char(struct tty_struct *tty, unsigned char c)
 	 * up.
 	 */
 	if (!test_bit(c, tty->process_char_map) || tty->lnext) {
-		finish_erasing(tty);
 		tty->lnext = 0;
 		if (L_ECHO(tty)) {
+			finish_erasing(tty);
 			if (tty->read_cnt >= N_TTY_BUF_SIZE-1) {
-				tty_put_char(tty, '\a'); /* beep if no space */
+				/* beep if no space */
+				echo_char_raw('\a', tty);
+				process_echoes(tty);
 				return;
 			}
 			/* Record the column of first canon char. */
 			if (tty->canon_head == tty->read_head)
-				tty->canon_column = tty->column;
+				echo_set_canon_col(tty);
 			echo_char(c, tty);
+			process_echoes(tty);
 		}
 		if (I_PARMRK(tty) && c == (unsigned char) '\377')
 			put_tty_queue(c, tty);
@@ -766,6 +1164,7 @@ static inline void n_tty_receive_char(struct tty_struct *tty, unsigned char c)
 	if (I_IXON(tty)) {
 		if (c == START_CHAR(tty)) {
 			start_tty(tty);
+			process_echoes(tty);
 			return;
 		}
 		if (c == STOP_CHAR(tty)) {
@@ -786,7 +1185,6 @@ static inline void n_tty_receive_char(struct tty_struct *tty, unsigned char c)
 		if (c == SUSP_CHAR(tty)) {
 send_signal:
 			/*
-			 * Echo character, and then send the signal.
 			 * Note that we do not use isig() here because we want
 			 * the order to be:
 			 * 1) flush, 2) echo, 3) signal
@@ -795,8 +1193,12 @@ send_signal:
 				n_tty_flush_buffer(tty);
 				tty_driver_flush_buffer(tty);
 			}
-			if (L_ECHO(tty))
+			if (I_IXON(tty))
+				start_tty(tty);
+			if (L_ECHO(tty)) {
 				echo_char(c, tty);
+				process_echoes(tty);
+			}
 			if (tty->pgrp)
 				kill_pgrp(tty->pgrp, signal, 1);
 			return;
@@ -815,6 +1217,7 @@ send_signal:
 		if (c == ERASE_CHAR(tty) || c == KILL_CHAR(tty) ||
 		    (c == WERASE_CHAR(tty) && L_IEXTEN(tty))) {
 			eraser(c, tty);
+			process_echoes(tty);
 			return;
 		}
 		if (c == LNEXT_CHAR(tty) && L_IEXTEN(tty)) {
@@ -822,8 +1225,9 @@ send_signal:
 			if (L_ECHO(tty)) {
 				finish_erasing(tty);
 				if (L_ECHOCTL(tty)) {
-					tty_put_char(tty, '^');
-					tty_put_char(tty, '\b');
+					echo_char_raw('^', tty);
+					echo_char_raw('\b', tty);
+					process_echoes(tty);
 				}
 			}
 			return;
@@ -834,18 +1238,20 @@ send_signal:
 
 			finish_erasing(tty);
 			echo_char(c, tty);
-			opost('\n', tty);
+			echo_char_raw('\n', tty);
 			while (tail != tty->read_head) {
 				echo_char(tty->read_buf[tail], tty);
 				tail = (tail+1) & (N_TTY_BUF_SIZE-1);
 			}
+			process_echoes(tty);
 			return;
 		}
 		if (c == '\n') {
 			if (L_ECHO(tty) || L_ECHONL(tty)) {
 				if (tty->read_cnt >= N_TTY_BUF_SIZE-1)
-					tty_put_char(tty, '\a');
-				opost('\n', tty);
+					echo_char_raw('\a', tty);
+				echo_char_raw('\n', tty);
+				process_echoes(tty);
 			}
 			goto handle_newline;
 		}
@@ -862,11 +1268,12 @@ send_signal:
 			 */
 			if (L_ECHO(tty)) {
 				if (tty->read_cnt >= N_TTY_BUF_SIZE-1)
-					tty_put_char(tty, '\a');
+					echo_char_raw('\a', tty);
 				/* Record the column of first canon char. */
 				if (tty->canon_head == tty->read_head)
-					tty->canon_column = tty->column;
+					echo_set_canon_col(tty);
 				echo_char(c, tty);
+				process_echoes(tty);
 			}
 			/*
 			 * XXX does PARMRK doubling happen for
@@ -889,20 +1296,23 @@ handle_newline:
 		}
 	}
 
-	finish_erasing(tty);
 	if (L_ECHO(tty)) {
+		finish_erasing(tty);
 		if (tty->read_cnt >= N_TTY_BUF_SIZE-1) {
-			tty_put_char(tty, '\a'); /* beep if no space */
+			/* beep if no space */
+			echo_char_raw('\a', tty);
+			process_echoes(tty);
 			return;
 		}
 		if (c == '\n')
-			opost('\n', tty);
+			echo_char_raw('\n', tty);
 		else {
 			/* Record the column of first canon char. */
 			if (tty->canon_head == tty->read_head)
-				tty->canon_column = tty->column;
+				echo_set_canon_col(tty);
 			echo_char(c, tty);
 		}
+		process_echoes(tty);
 	}
 
 	if (I_PARMRK(tty) && c == (unsigned char) '\377')
@@ -923,6 +1333,9 @@ handle_newline:
 
 static void n_tty_write_wakeup(struct tty_struct *tty)
 {
+	/* Write out any echoed characters that are still pending */
+	process_echoes(tty);
+	
 	if (tty->fasync) {
 		set_bit(TTY_DO_WRITE_WAKEUP, &tty->flags);
 		kill_fasync(&tty->fasync, SIGIO, POLL_OUT);
@@ -1134,6 +1547,10 @@ static void n_tty_close(struct tty_struct *tty)
 		free_buf(tty->read_buf);
 		tty->read_buf = NULL;
 	}
+	if (tty->echo_buf) {
+		free_buf(tty->echo_buf);
+		tty->echo_buf = NULL;
+	}
 }
 
 /**
@@ -1151,13 +1568,19 @@ static int n_tty_open(struct tty_struct *tty)
 	if (!tty)
 		return -EINVAL;
 
-	/* This one is ugly. Currently a malloc failure here can panic */
+	/* These are ugly. Currently a malloc failure here can panic */
 	if (!tty->read_buf) {
 		tty->read_buf = alloc_buf();
 		if (!tty->read_buf)
 			return -ENOMEM;
 	}
+	if (!tty->echo_buf) {
+		tty->echo_buf = alloc_buf();
+		if (!tty->echo_buf)
+			return -ENOMEM;
+	}
 	memset(tty->read_buf, 0, N_TTY_BUF_SIZE);
+	memset(tty->echo_buf, 0, N_TTY_BUF_SIZE);
 	reset_buffer_flags(tty);
 	tty->column = 0;
 	n_tty_set_termios(tty, NULL);
@@ -1487,16 +1910,23 @@ do_it_again:
  *	@buf: userspace buffer pointer
  *	@nr: size of I/O
  *
- *	Write function of the terminal device. This is serialized with
+ *	Write function of the terminal device.  This is serialized with
  *	respect to other write callers but not to termios changes, reads
- *	and other such events. We must be careful with N_TTY as the receive
- *	code will echo characters, thus calling driver write methods.
+ *	and other such events.  Since the receive code will echo characters,
+ *	thus calling driver write methods, the output_lock is used in
+ *	the output processing functions called here as well as in the
+ *	echo processing function to protect the column state and space
+ *	left in the buffer.
  *
  *	This code must be sure never to sleep through a hangup.
+ *
+ *	Locking: output_lock to protect column state and space left
+ *		 (note that the process_output*() functions take this
+ *		  lock themselves)
  */
 
 static ssize_t n_tty_write(struct tty_struct *tty, struct file *file,
-			  const unsigned char *buf, size_t nr)
+			   const unsigned char *buf, size_t nr)
 {
 	const unsigned char *b = buf;
 	DECLARE_WAITQUEUE(wait, current);
@@ -1510,6 +1940,9 @@ static ssize_t n_tty_write(struct tty_struct *tty, struct file *file,
 			return retval;
 	}
 
+	/* Write out any echoed characters that are still pending */
+	process_echoes(tty);
+	
 	add_wait_queue(&tty->write_wait, &wait);
 	while (1) {
 		set_current_state(TASK_INTERRUPTIBLE);
@@ -1523,7 +1956,7 @@ static ssize_t n_tty_write(struct tty_struct *tty, struct file *file,
 		}
 		if (O_OPOST(tty) && !(test_bit(TTY_HW_COOK_OUT, &tty->flags))) {
 			while (nr > 0) {
-				ssize_t num = opost_block(tty, b, nr);
+				ssize_t num = process_output_block(tty, b, nr);
 				if (num < 0) {
 					if (num == -EAGAIN)
 						break;
@@ -1535,7 +1968,7 @@ static ssize_t n_tty_write(struct tty_struct *tty, struct file *file,
 				if (nr == 0)
 					break;
 				c = *b;
-				if (opost(c, tty) < 0)
+				if (process_output(c, tty) < 0)
 					break;
 				b++; nr--;
 			}
@@ -1663,4 +2096,3 @@ struct tty_ldisc_ops tty_ldisc_N_TTY = {
 	.receive_buf     = n_tty_receive_buf,
 	.write_wakeup    = n_tty_write_wakeup
 };
-
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index db15f9ba7c0b..d8d240c8a25a 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -1111,9 +1111,7 @@ void tty_write_message(struct tty_struct *tty, char *msg)
  *		Locks the line discipline as required
  *		Writes to the tty driver are serialized by the atomic_write_lock
  *	and are then processed in chunks to the device. The line discipline
- *	write method will not be involked in parallel for each device
- *		The line discipline write method is called under the big
- *	kernel lock for historical reasons. New code should not rely on this.
+ *	write method will not be invoked in parallel for each device.
  */
 
 static ssize_t tty_write(struct file *file, const char __user *buf,
@@ -2785,6 +2783,8 @@ void initialize_tty_struct(struct tty_struct *tty,
 	INIT_WORK(&tty->hangup_work, do_tty_hangup);
 	mutex_init(&tty->atomic_read_lock);
 	mutex_init(&tty->atomic_write_lock);
+	mutex_init(&tty->output_lock);
+	mutex_init(&tty->echo_lock);
 	spin_lock_init(&tty->read_lock);
 	spin_lock_init(&tty->ctrl_lock);
 	INIT_LIST_HEAD(&tty->tty_files);
diff --git a/drivers/char/vt.c b/drivers/char/vt.c
index 008176edbd64..639e126b2bff 100644
--- a/drivers/char/vt.c
+++ b/drivers/char/vt.c
@@ -2679,7 +2679,7 @@ static int con_write_room(struct tty_struct *tty)
 {
 	if (tty->stopped)
 		return 0;
-	return 4096;		/* No limit, really; we're not buffering */
+	return 32768;		/* No limit, really; we're not buffering */
 }
 
 static int con_chars_in_buffer(struct tty_struct *tty)
diff --git a/include/linux/tty.h b/include/linux/tty.h
index 3f4954c55e53..dfc77ded198a 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -253,6 +253,7 @@ struct tty_struct {
 	unsigned int column;
 	unsigned char lnext:1, erasing:1, raw:1, real_raw:1, icanon:1;
 	unsigned char closing:1;
+	unsigned char echo_overrun:1;
 	unsigned short minimum_to_wake;
 	unsigned long overrun_time;
 	int num_overrun;
@@ -262,11 +263,16 @@ struct tty_struct {
 	int read_tail;
 	int read_cnt;
 	unsigned long read_flags[N_TTY_BUF_SIZE/(8*sizeof(unsigned long))];
+	unsigned char *echo_buf;
+	unsigned int echo_pos;
+	unsigned int echo_cnt;
 	int canon_data;
 	unsigned long canon_head;
 	unsigned int canon_column;
 	struct mutex atomic_read_lock;
 	struct mutex atomic_write_lock;
+	struct mutex output_lock;
+	struct mutex echo_lock;
 	unsigned char *write_buf;
 	int write_cnt;
 	spinlock_t read_lock;
-- 
cgit v1.2.3


From fc6f6238226e6d1248e1967eae2bf556eaf3ac17 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Fri, 2 Jan 2009 13:43:17 +0000
Subject: pty: simplify resize

We have special case logic for resizing pty/tty pairs. We also have a per
driver resize method so for the pty case we should use it.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/hvc_console.c |  2 +-
 drivers/char/pty.c         | 54 +++++++++++++++++++++++++++++++++++++++++++++-
 drivers/char/tty_io.c      | 31 ++++++++++----------------
 drivers/char/vt.c          | 14 ++++++------
 include/linux/tty.h        |  3 +--
 include/linux/tty_driver.h |  6 ++----
 6 files changed, 74 insertions(+), 36 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/hvc_console.c b/drivers/char/hvc_console.c
index 0587b66d6fc7..5a8a4c28c867 100644
--- a/drivers/char/hvc_console.c
+++ b/drivers/char/hvc_console.c
@@ -529,7 +529,7 @@ static void hvc_set_winsz(struct work_struct *work)
 	tty = tty_kref_get(hp->tty);
 	spin_unlock_irqrestore(&hp->lock, hvc_flags);
 
-	tty_do_resize(tty, tty, &ws);
+	tty_do_resize(tty, &ws);
 	tty_kref_put(tty);
 }
 
diff --git a/drivers/char/pty.c b/drivers/char/pty.c
index 6d4582712b1f..b5daaaa9007e 100644
--- a/drivers/char/pty.c
+++ b/drivers/char/pty.c
@@ -230,6 +230,55 @@ static void pty_set_termios(struct tty_struct *tty,
 	tty->termios->c_cflag |= (CS8 | CREAD);
 }
 
+/**
+ *	pty_do_resize		-	resize event
+ *	@tty: tty being resized
+ *	@real_tty: real tty (not the same as tty if using a pty/tty pair)
+ *	@rows: rows (character)
+ *	@cols: cols (character)
+ *
+ *	Update the termios variables and send the neccessary signals to
+ *	peform a terminal resize correctly
+ */
+
+int pty_resize(struct tty_struct *tty,  struct winsize *ws)
+{
+	struct pid *pgrp, *rpgrp;
+	unsigned long flags;
+	struct tty_struct *pty = tty->link;
+
+	/* For a PTY we need to lock the tty side */
+	mutex_lock(&tty->termios_mutex);
+	if (!memcmp(ws, &tty->winsize, sizeof(*ws)))
+		goto done;
+
+	/* Get the PID values and reference them so we can
+	   avoid holding the tty ctrl lock while sending signals.
+	   We need to lock these individually however. */
+
+	spin_lock_irqsave(&tty->ctrl_lock, flags);
+	pgrp = get_pid(tty->pgrp);
+	spin_unlock_irqrestore(&tty->ctrl_lock, flags);
+
+	spin_lock_irqsave(&pty->ctrl_lock, flags);
+	rpgrp = get_pid(pty->pgrp);
+	spin_unlock_irqrestore(&pty->ctrl_lock, flags);
+
+	if (pgrp)
+		kill_pgrp(pgrp, SIGWINCH, 1);
+	if (rpgrp != pgrp && rpgrp)
+		kill_pgrp(rpgrp, SIGWINCH, 1);
+
+	put_pid(pgrp);
+	put_pid(rpgrp);
+
+	tty->winsize = *ws;
+	pty->winsize = *ws;	/* Never used so will go away soon */
+done:
+	mutex_unlock(&tty->termios_mutex);
+	return 0;
+}
+
 static int pty_install(struct tty_driver *driver, struct tty_struct *tty)
 {
 	struct tty_struct *o_tty;
@@ -290,6 +339,7 @@ static const struct tty_operations pty_ops = {
 	.chars_in_buffer = pty_chars_in_buffer,
 	.unthrottle = pty_unthrottle,
 	.set_termios = pty_set_termios,
+	.resize = pty_resize
 };
 
 /* Traditional BSD devices */
@@ -319,6 +369,7 @@ static const struct tty_operations pty_ops_bsd = {
 	.unthrottle = pty_unthrottle,
 	.set_termios = pty_set_termios,
 	.ioctl = pty_bsd_ioctl,
+	.resize = pty_resize
 };
 
 static void __init legacy_pty_init(void)
@@ -561,7 +612,8 @@ static const struct tty_operations ptm_unix98_ops = {
 	.unthrottle = pty_unthrottle,
 	.set_termios = pty_set_termios,
 	.ioctl = pty_unix98_ioctl,
-	.shutdown = pty_unix98_shutdown
+	.shutdown = pty_unix98_shutdown,
+	.resize = pty_resize
 };
 
 static const struct tty_operations pty_unix98_ops = {
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index 2a15af65dd11..d33e5ab06177 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -2048,7 +2048,6 @@ static int tiocgwinsz(struct tty_struct *tty, struct winsize __user *arg)
 /**
  *	tty_do_resize		-	resize event
  *	@tty: tty being resized
- *	@real_tty: real tty (not the same as tty if using a pty/tty pair)
  *	@rows: rows (character)
  *	@cols: cols (character)
  *
@@ -2056,41 +2055,34 @@ static int tiocgwinsz(struct tty_struct *tty, struct winsize __user *arg)
  *	peform a terminal resize correctly
  */
 
-int tty_do_resize(struct tty_struct *tty, struct tty_struct *real_tty,
-					struct winsize *ws)
+int tty_do_resize(struct tty_struct *tty, struct winsize *ws)
 {
-	struct pid *pgrp, *rpgrp;
+	struct pid *pgrp;
 	unsigned long flags;
 
-	/* For a PTY we need to lock the tty side */
-	mutex_lock(&real_tty->termios_mutex);
-	if (!memcmp(ws, &real_tty->winsize, sizeof(*ws)))
+	/* Lock the tty */
+	mutex_lock(&tty->termios_mutex);
+	if (!memcmp(ws, &tty->winsize, sizeof(*ws)))
 		goto done;
 	/* Get the PID values and reference them so we can
 	   avoid holding the tty ctrl lock while sending signals */
 	spin_lock_irqsave(&tty->ctrl_lock, flags);
 	pgrp = get_pid(tty->pgrp);
-	rpgrp = get_pid(real_tty->pgrp);
 	spin_unlock_irqrestore(&tty->ctrl_lock, flags);
 
 	if (pgrp)
 		kill_pgrp(pgrp, SIGWINCH, 1);
-	if (rpgrp != pgrp && rpgrp)
-		kill_pgrp(rpgrp, SIGWINCH, 1);
-
 	put_pid(pgrp);
-	put_pid(rpgrp);
 
 	tty->winsize = *ws;
-	real_tty->winsize = *ws;
 done:
-	mutex_unlock(&real_tty->termios_mutex);
+	mutex_unlock(&tty->termios_mutex);
 	return 0;
 }
 
 /**
  *	tiocswinsz		-	implement window size set ioctl
- *	@tty; tty
+ *	@tty; tty side of tty
  *	@arg: user buffer for result
  *
  *	Copies the user idea of the window size to the kernel. Traditionally
@@ -2103,17 +2095,16 @@ done:
  *	then calls into the default method.
  */
 
-static int tiocswinsz(struct tty_struct *tty, struct tty_struct *real_tty,
-	struct winsize __user *arg)
+static int tiocswinsz(struct tty_struct *tty, struct winsize __user *arg)
 {
 	struct winsize tmp_ws;
 	if (copy_from_user(&tmp_ws, arg, sizeof(*arg)))
 		return -EFAULT;
 
 	if (tty->ops->resize)
-		return tty->ops->resize(tty, real_tty, &tmp_ws);
+		return tty->ops->resize(tty, &tmp_ws);
 	else
-		return tty_do_resize(tty, real_tty, &tmp_ws);
+		return tty_do_resize(tty, &tmp_ws);
 }
 
 /**
@@ -2538,7 +2529,7 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case TIOCGWINSZ:
 		return tiocgwinsz(real_tty, p);
 	case TIOCSWINSZ:
-		return tiocswinsz(tty, real_tty, p);
+		return tiocswinsz(real_tty, p);
 	case TIOCCONS:
 		return real_tty != tty ? -EINVAL : tioccons(file);
 	case FIONBIO:
diff --git a/drivers/char/vt.c b/drivers/char/vt.c
index 639e126b2bff..80014213fb53 100644
--- a/drivers/char/vt.c
+++ b/drivers/char/vt.c
@@ -819,8 +819,8 @@ static inline int resize_screen(struct vc_data *vc, int width, int height,
  *	ctrl_lock of the tty IFF a tty is passed.
  */
 
-static int vc_do_resize(struct tty_struct *tty, struct tty_struct *real_tty,
-		struct vc_data *vc, unsigned int cols, unsigned int lines)
+static int vc_do_resize(struct tty_struct *tty, struct vc_data *vc,
+				unsigned int cols, unsigned int lines)
 {
 	unsigned long old_origin, new_origin, new_scr_end, rlth, rrem, err = 0;
 	unsigned int old_cols, old_rows, old_row_size, old_screen_size;
@@ -932,7 +932,7 @@ static int vc_do_resize(struct tty_struct *tty, struct tty_struct *real_tty,
 		ws.ws_row = vc->vc_rows;
 		ws.ws_col = vc->vc_cols;
 		ws.ws_ypixel = vc->vc_scan_lines;
-		tty_do_resize(tty, real_tty, &ws);
+		tty_do_resize(tty, &ws);
 	}
 
 	if (CON_IS_VISIBLE(vc))
@@ -954,13 +954,12 @@ static int vc_do_resize(struct tty_struct *tty, struct tty_struct *real_tty,
 
 int vc_resize(struct vc_data *vc, unsigned int cols, unsigned int rows)
 {
-	return vc_do_resize(vc->vc_tty, vc->vc_tty, vc, cols, rows);
+	return vc_do_resize(vc->vc_tty, vc, cols, rows);
 }
 
 /**
  *	vt_resize		-	resize a VT
  *	@tty: tty to resize
- *	@real_tty: tty if a pty/tty pair
  *	@ws: winsize attributes
  *
  *	Resize a virtual terminal. This is called by the tty layer as we
@@ -971,14 +970,13 @@ int vc_resize(struct vc_data *vc, unsigned int cols, unsigned int rows)
  *	termios_mutex and the tty ctrl_lock in that order.
  */
 
-int vt_resize(struct tty_struct *tty, struct tty_struct *real_tty,
-	struct winsize *ws)
+int vt_resize(struct tty_struct *tty, struct winsize *ws)
 {
 	struct vc_data *vc = tty->driver_data;
 	int ret;
 
 	acquire_console_sem();
-	ret = vc_do_resize(tty, real_tty, vc, ws->ws_col, ws->ws_row);
+	ret = vc_do_resize(tty, vc, ws->ws_col, ws->ws_row);
 	release_console_sem();
 	return ret;
 }
diff --git a/include/linux/tty.h b/include/linux/tty.h
index dfc77ded198a..f88169787a5f 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -360,8 +360,7 @@ extern int tty_write_room(struct tty_struct *tty);
 extern void tty_driver_flush_buffer(struct tty_struct *tty);
 extern void tty_throttle(struct tty_struct *tty);
 extern void tty_unthrottle(struct tty_struct *tty);
-extern int tty_do_resize(struct tty_struct *tty, struct tty_struct *real_tty,
-						struct winsize *ws);
+extern int tty_do_resize(struct tty_struct *tty, struct winsize *ws);
 extern void tty_shutdown(struct tty_struct *tty);
 extern void tty_free_termios(struct tty_struct *tty);
 extern int is_current_pgrp_orphaned(void);
diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h
index 78416b901589..08e088334dba 100644
--- a/include/linux/tty_driver.h
+++ b/include/linux/tty_driver.h
@@ -196,8 +196,7 @@
  *	Optional: If not provided then the write method is called under
  *	the atomic write lock to keep it serialized with the ldisc.
  *
- * int (*resize)(struct tty_struct *tty, struct tty_struct *real_tty,
- *				unsigned int rows, unsigned int cols);
+ * int (*resize)(struct tty_struct *tty, struct winsize *ws)
  *
  *	Called when a termios request is issued which changes the
  *	requested terminal geometry.
@@ -258,8 +257,7 @@ struct tty_operations {
 	int (*tiocmget)(struct tty_struct *tty, struct file *file);
 	int (*tiocmset)(struct tty_struct *tty, struct file *file,
 			unsigned int set, unsigned int clear);
-	int (*resize)(struct tty_struct *tty, struct tty_struct *real_tty,
-				struct winsize *ws);
+	int (*resize)(struct tty_struct *tty, struct winsize *ws);
 	int (*set_termiox)(struct tty_struct *tty, struct termiox *tnew);
 #ifdef CONFIG_CONSOLE_POLL
 	int (*poll_init)(struct tty_driver *driver, int line, char *options);
-- 
cgit v1.2.3


From 975a1a7d887048d4afc9201383e11b7af991866b Mon Sep 17 00:00:00 2001
From: Russell King <rmk+lkml@arm.linux.org.uk>
Date: Fri, 2 Jan 2009 13:44:27 +0000
Subject: And here's a patch (to be applied on top of the last) which prevents

this happening again by making use of 'const'.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/8250_pci.c | 37 +++++++++++++++++++++----------------
 include/linux/8250_pci.h  |  2 +-
 2 files changed, 22 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/serial/8250_pci.c b/drivers/serial/8250_pci.c
index 057b532ccaad..0b794138f686 100644
--- a/drivers/serial/8250_pci.c
+++ b/drivers/serial/8250_pci.c
@@ -42,7 +42,8 @@ struct pci_serial_quirk {
 	u32	subvendor;
 	u32	subdevice;
 	int	(*init)(struct pci_dev *dev);
-	int	(*setup)(struct serial_private *, struct pciserial_board *,
+	int	(*setup)(struct serial_private *,
+			 const struct pciserial_board *,
 			 struct uart_port *, int);
 	void	(*exit)(struct pci_dev *dev);
 };
@@ -107,7 +108,7 @@ setup_port(struct serial_private *priv, struct uart_port *port,
  * ADDI-DATA GmbH communication cards <info@addi-data.com>
  */
 static int addidata_apci7800_setup(struct serial_private *priv,
-				struct pciserial_board *board,
+				const struct pciserial_board *board,
 				struct uart_port *port, int idx)
 {
 	unsigned int bar = 0, offset = board->first_offset;
@@ -134,7 +135,7 @@ static int addidata_apci7800_setup(struct serial_private *priv,
  * Not that ugly ;) -- HW
  */
 static int
-afavlab_setup(struct serial_private *priv, struct pciserial_board *board,
+afavlab_setup(struct serial_private *priv, const struct pciserial_board *board,
 	      struct uart_port *port, int idx)
 {
 	unsigned int bar, offset = board->first_offset;
@@ -188,8 +189,9 @@ static int pci_hp_diva_init(struct pci_dev *dev)
  * some serial ports are supposed to be hidden on certain models.
  */
 static int
-pci_hp_diva_setup(struct serial_private *priv, struct pciserial_board *board,
-	      struct uart_port *port, int idx)
+pci_hp_diva_setup(struct serial_private *priv,
+		const struct pciserial_board *board,
+		struct uart_port *port, int idx)
 {
 	unsigned int offset = board->first_offset;
 	unsigned int bar = FL_GET_BASE(board->flags);
@@ -306,7 +308,7 @@ static void __devexit pci_plx9050_exit(struct pci_dev *dev)
 
 /* SBS Technologies Inc. PMC-OCTPRO and P-OCTAL cards */
 static int
-sbs_setup(struct serial_private *priv, struct pciserial_board *board,
+sbs_setup(struct serial_private *priv, const struct pciserial_board *board,
 		struct uart_port *port, int idx)
 {
 	unsigned int bar, offset = board->first_offset;
@@ -463,7 +465,7 @@ static int pci_siig_init(struct pci_dev *dev)
 }
 
 static int pci_siig_setup(struct serial_private *priv,
-			  struct pciserial_board *board,
+			  const struct pciserial_board *board,
 			  struct uart_port *port, int idx)
 {
 	unsigned int bar = FL_GET_BASE(board->flags) + idx, offset = 0;
@@ -534,7 +536,8 @@ static int pci_timedia_init(struct pci_dev *dev)
  * Ugh, this is ugly as all hell --- TYT
  */
 static int
-pci_timedia_setup(struct serial_private *priv, struct pciserial_board *board,
+pci_timedia_setup(struct serial_private *priv,
+		  const struct pciserial_board *board,
 		  struct uart_port *port, int idx)
 {
 	unsigned int bar = 0, offset = board->first_offset;
@@ -568,7 +571,7 @@ pci_timedia_setup(struct serial_private *priv, struct pciserial_board *board,
  */
 static int
 titan_400l_800l_setup(struct serial_private *priv,
-		      struct pciserial_board *board,
+		      const struct pciserial_board *board,
 		      struct uart_port *port, int idx)
 {
 	unsigned int bar, offset = board->first_offset;
@@ -770,7 +773,8 @@ static int pci_oxsemi_tornado_init(struct pci_dev *dev)
 }
 
 static int
-pci_default_setup(struct serial_private *priv, struct pciserial_board *board,
+pci_default_setup(struct serial_private *priv,
+		  const struct pciserial_board *board,
 		  struct uart_port *port, int idx)
 {
 	unsigned int bar, offset = board->first_offset, maxnr;
@@ -1099,7 +1103,7 @@ static struct pci_serial_quirk *find_quirk(struct pci_dev *dev)
 }
 
 static inline int get_pci_irq(struct pci_dev *dev,
-				struct pciserial_board *board)
+				const struct pciserial_board *board)
 {
 	if (board->flags & FL_NOIRQ)
 		return 0;
@@ -1894,8 +1898,8 @@ serial_pci_guess_board(struct pci_dev *dev, struct pciserial_board *board)
 }
 
 static inline int
-serial_pci_matches(struct pciserial_board *board,
-		   struct pciserial_board *guessed)
+serial_pci_matches(const struct pciserial_board *board,
+		   const struct pciserial_board *guessed)
 {
 	return
 	    board->num_ports == guessed->num_ports &&
@@ -1906,7 +1910,7 @@ serial_pci_matches(struct pciserial_board *board,
 }
 
 struct serial_private *
-pciserial_init_ports(struct pci_dev *dev, struct pciserial_board *board)
+pciserial_init_ports(struct pci_dev *dev, const struct pciserial_board *board)
 {
 	struct uart_port serial_port;
 	struct serial_private *priv;
@@ -2039,7 +2043,8 @@ static int __devinit
 pciserial_init_one(struct pci_dev *dev, const struct pci_device_id *ent)
 {
 	struct serial_private *priv;
-	struct pciserial_board *board, tmp;
+	const struct pciserial_board *board;
+	struct pciserial_board tmp;
 	int rc;
 
 	if (ent->driver_data >= ARRAY_SIZE(pci_boards)) {
@@ -2066,7 +2071,7 @@ pciserial_init_one(struct pci_dev *dev, const struct pci_device_id *ent)
 		 * We matched one of our class entries.  Try to
 		 * determine the parameters of this board.
 		 */
-		rc = serial_pci_guess_board(dev, board);
+		rc = serial_pci_guess_board(dev, &tmp);
 		if (rc)
 			goto disable;
 	} else {
diff --git a/include/linux/8250_pci.h b/include/linux/8250_pci.h
index 3209dd46ea7d..b24ff086a662 100644
--- a/include/linux/8250_pci.h
+++ b/include/linux/8250_pci.h
@@ -31,7 +31,7 @@ struct pciserial_board {
 struct serial_private;
 
 struct serial_private *
-pciserial_init_ports(struct pci_dev *dev, struct pciserial_board *board);
+pciserial_init_ports(struct pci_dev *dev, const struct pciserial_board *board);
 void pciserial_remove_ports(struct serial_private *priv);
 void pciserial_suspend_ports(struct serial_private *priv);
 void pciserial_resume_ports(struct serial_private *priv);
-- 
cgit v1.2.3


From c9b3976e3fec266be25c5001a70aa0a890b6c476 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Fri, 2 Jan 2009 13:44:56 +0000
Subject: tty: Fix PPP hang under load

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/tty_ldisc.c | 30 +++++++++++++++++++++---------
 include/linux/tty.h      |  1 +
 2 files changed, 22 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/tty_ldisc.c b/drivers/char/tty_ldisc.c
index f307f135cbfb..7a84b406a952 100644
--- a/drivers/char/tty_ldisc.c
+++ b/drivers/char/tty_ldisc.c
@@ -316,8 +316,7 @@ struct tty_ldisc *tty_ldisc_ref_wait(struct tty_struct *tty)
 {
 	/* wait_event is a macro */
 	wait_event(tty_ldisc_wait, tty_ldisc_try(tty));
-	if (tty->ldisc.refcount == 0)
-		printk(KERN_ERR "tty_ldisc_ref_wait\n");
+	WARN_ON(tty->ldisc.refcount == 0);
 	return &tty->ldisc;
 }
 
@@ -376,15 +375,17 @@ EXPORT_SYMBOL_GPL(tty_ldisc_deref);
  *	@tty: terminal to activate ldisc on
  *
  *	Set the TTY_LDISC flag when the line discipline can be called
- *	again. Do necessary wakeups for existing sleepers.
+ *	again. Do necessary wakeups for existing sleepers. Clear the LDISC
+ *	changing flag to indicate any ldisc change is now over.
  *
- *	Note: nobody should set this bit except via this function. Clearing
- *	directly is allowed.
+ *	Note: nobody should set the TTY_LDISC bit except via this function.
+ *	Clearing directly is allowed.
  */
 
 void tty_ldisc_enable(struct tty_struct *tty)
 {
 	set_bit(TTY_LDISC, &tty->flags);
+	clear_bit(TTY_LDISC_CHANGING, &tty->flags);
 	wake_up(&tty_ldisc_wait);
 }
 
@@ -496,7 +497,14 @@ restart:
 	 *	reference to the line discipline. The TTY_LDISC bit
 	 *	prevents anyone taking a reference once it is clear.
 	 *	We need the lock to avoid racing reference takers.
+	 *
+	 *	We must clear the TTY_LDISC bit here to avoid a livelock
+	 *	with a userspace app continually trying to use the tty in
+	 *	parallel to the change and re-referencing the tty.
 	 */
+	clear_bit(TTY_LDISC, &tty->flags);
+	if (o_tty)
+		clear_bit(TTY_LDISC, &o_tty->flags);
 
 	spin_lock_irqsave(&tty_ldisc_lock, flags);
 	if (tty->ldisc.refcount || (o_tty && o_tty->ldisc.refcount)) {
@@ -528,7 +536,7 @@ restart:
 	 *	If the TTY_LDISC bit is set, then we are racing against
 	 *	another ldisc change
 	 */
-	if (!test_bit(TTY_LDISC, &tty->flags)) {
+	if (test_bit(TTY_LDISC_CHANGING, &tty->flags)) {
 		struct tty_ldisc *ld;
 		spin_unlock_irqrestore(&tty_ldisc_lock, flags);
 		tty_ldisc_put(new_ldisc.ops);
@@ -536,10 +544,14 @@ restart:
 		tty_ldisc_deref(ld);
 		goto restart;
 	}
-
-	clear_bit(TTY_LDISC, &tty->flags);
+	/*
+	 *	This flag is used to avoid two parallel ldisc changes. Once
+	 *	open and close are fine grained locked this may work better
+	 *	as a mutex shared with the open/close/hup paths
+	 */
+	set_bit(TTY_LDISC_CHANGING, &tty->flags);
 	if (o_tty)
-		clear_bit(TTY_LDISC, &o_tty->flags);
+		set_bit(TTY_LDISC_CHANGING, &o_tty->flags);
 	spin_unlock_irqrestore(&tty_ldisc_lock, flags);
 	
 	/*
diff --git a/include/linux/tty.h b/include/linux/tty.h
index f88169787a5f..bbbeaef99626 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -301,6 +301,7 @@ struct tty_struct {
 #define TTY_PUSH 		6	/* n_tty private */
 #define TTY_CLOSING 		7	/* ->close() in progress */
 #define TTY_LDISC 		9	/* Line discipline attached */
+#define TTY_LDISC_CHANGING 	10	/* Line discipline changing */
 #define TTY_HW_COOK_OUT 	14	/* Hardware can do output cooking */
 #define TTY_HW_COOK_IN 		15	/* Hardware can do input cooking */
 #define TTY_PTY_LOCK 		16	/* pty private */
-- 
cgit v1.2.3


From 31f35939d1d9bcfb3099b32c67b896d2792603f9 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Fri, 2 Jan 2009 13:45:05 +0000
Subject: tty_port: Add a port level carrier detect operation

This is the first step to generalising the various pieces of waiting logic
duplicated in all sorts of serial drivers.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/esp.c             | 61 +++++++++++++++++++++++----------------
 drivers/char/generic_serial.c  | 43 ++++++++++++++--------------
 drivers/char/isicom.c          | 51 +++++++++++++++++++++------------
 drivers/char/istallion.c       | 28 ++++++++++++------
 drivers/char/moxa.c            | 26 +++++++++++++----
 drivers/char/mxser.c           | 54 +++++++++++++++++++++--------------
 drivers/char/rio/rio_linux.c   | 18 ++++++------
 drivers/char/riscom8.c         | 65 ++++++++++++++++++++++++++----------------
 drivers/char/rocket.c          | 40 +++++++++++++++++---------
 drivers/char/ser_a2232.c       | 19 ++++++------
 drivers/char/stallion.c        | 28 +++++++++++++-----
 drivers/char/sx.c              | 27 +++++++++---------
 drivers/char/synclink.c        | 61 ++++++++++++++++++++++++++-------------
 drivers/char/synclink_gt.c     | 48 ++++++++++++++++++++-----------
 drivers/char/synclinkmp.c      | 55 ++++++++++++++++++++++-------------
 drivers/char/tty_port.c        | 17 +++++++++++
 drivers/char/vme_scc.c         | 15 ++++++----
 include/linux/generic_serial.h |  1 -
 include/linux/tty.h            |  9 ++++++
 19 files changed, 427 insertions(+), 239 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/esp.c b/drivers/char/esp.c
index 7f077c0097f6..45ec263ec012 100644
--- a/drivers/char/esp.c
+++ b/drivers/char/esp.c
@@ -2054,6 +2054,15 @@ static void esp_hangup(struct tty_struct *tty)
 	wake_up_interruptible(&info->port.open_wait);
 }
 
+static int esp_carrier_raised(struct tty_port *port)
+{
+	struct esp_struct *info = container_of(port, struct esp_struct, port);
+	serial_out(info, UART_ESI_CMD1, ESI_GET_UART_STAT);
+	if (serial_in(info, UART_ESI_STAT2) & UART_MSR_DCD)
+		return 1;
+	return 0;
+}
+
 /*
  * ------------------------------------------------------------
  * esp_open() and friends
@@ -2066,17 +2075,19 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	int		retval;
 	int		do_clocal = 0;
 	unsigned long	flags;
+	int		cd;
+	struct tty_port *port = &info->port;
 
 	/*
 	 * If the device is in the middle of being closed, then block
 	 * until it's done, and then try again.
 	 */
 	if (tty_hung_up_p(filp) ||
-	    (info->port.flags & ASYNC_CLOSING)) {
-		if (info->port.flags & ASYNC_CLOSING)
-			interruptible_sleep_on(&info->port.close_wait);
+	    (port->flags & ASYNC_CLOSING)) {
+		if (port->flags & ASYNC_CLOSING)
+			interruptible_sleep_on(&port->close_wait);
 #ifdef SERIAL_DO_RESTART
-		if (info->port.flags & ASYNC_HUP_NOTIFY)
+		if (port->flags & ASYNC_HUP_NOTIFY)
 			return -EAGAIN;
 		else
 			return -ERESTARTSYS;
@@ -2091,7 +2102,7 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	 */
 	if ((filp->f_flags & O_NONBLOCK) ||
 	    (tty->flags & (1 << TTY_IO_ERROR))) {
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -2101,20 +2112,20 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	/*
 	 * Block waiting for the carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->port.count is dropped by one, so that
+	 * this loop, port->count is dropped by one, so that
 	 * rs_close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 	retval = 0;
-	add_wait_queue(&info->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 #ifdef SERIAL_DEBUG_OPEN
 	printk(KERN_DEBUG "block_til_ready before block: ttys%d, count = %d\n",
-	       info->line, info->port.count);
+	       info->line, port->count);
 #endif
 	spin_lock_irqsave(&info->lock, flags);
 	if (!tty_hung_up_p(filp))
-		info->port.count--;
-	info->port.blocked_open++;
+		port->count--;
+	port->blocked_open++;
 	while (1) {
 		if ((tty->termios->c_cflag & CBAUD)) {
 			unsigned int scratch;
@@ -2129,9 +2140,9 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 		}
 		set_current_state(TASK_INTERRUPTIBLE);
 		if (tty_hung_up_p(filp) ||
-		    !(info->port.flags & ASYNC_INITIALIZED)) {
+		    !(port->flags & ASYNC_INITIALIZED)) {
 #ifdef SERIAL_DO_RESTART
-			if (info->port.flags & ASYNC_HUP_NOTIFY)
+			if (port->flags & ASYNC_HUP_NOTIFY)
 				retval = -EAGAIN;
 			else
 				retval = -ERESTARTSYS;
@@ -2141,11 +2152,9 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 			break;
 		}
 
-		serial_out(info, UART_ESI_CMD1, ESI_GET_UART_STAT);
-		if (serial_in(info, UART_ESI_STAT2) & UART_MSR_DCD)
-			do_clocal = 1;
+		cd = tty_port_carrier_raised(port);
 
-		if (!(info->port.flags & ASYNC_CLOSING) &&
+		if (!(port->flags & ASYNC_CLOSING) &&
 		    (do_clocal))
 			break;
 		if (signal_pending(current)) {
@@ -2154,25 +2163,25 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 		}
 #ifdef SERIAL_DEBUG_OPEN
 		printk(KERN_DEBUG "block_til_ready blocking: ttys%d, count = %d\n",
-		       info->line, info->port.count);
+		       info->line, port->count);
 #endif
 		spin_unlock_irqrestore(&info->lock, flags);
 		schedule();
 		spin_lock_irqsave(&info->lock, flags);
 	}
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
 	if (!tty_hung_up_p(filp))
-		info->port.count++;
-	info->port.blocked_open--;
+		port->count++;
+	port->blocked_open--;
 	spin_unlock_irqrestore(&info->lock, flags);
 #ifdef SERIAL_DEBUG_OPEN
 	printk(KERN_DEBUG "block_til_ready after blocking: ttys%d, count = %d\n",
-	       info->line, info->port.count);
+	       info->line, port->count);
 #endif
 	if (retval)
 		return retval;
-	info->port.flags |= ASYNC_NORMAL_ACTIVE;
+	port->flags |= ASYNC_NORMAL_ACTIVE;
 	return 0;
 }
 
@@ -2329,6 +2338,10 @@ static const struct tty_operations esp_ops = {
 	.tiocmset = esp_tiocmset,
 };
 
+static const struct tty_port_operations esp_port_ops = {
+	.esp_carrier_raised,
+};
+
 /*
  * The serial driver boot-time initialization code!
  */
@@ -2415,6 +2428,8 @@ static int __init espserial_init(void)
 	offset = 0;
 
 	do {
+		tty_port_init(&info->port);
+		info->port.ops = &esp_port_ops;
 		info->io_port = esp[i] + offset;
 		info->irq = irq[i];
 		info->line = (i * 8) + (offset / 8);
@@ -2437,8 +2452,6 @@ static int __init espserial_init(void)
 		info->config.flow_off = flow_off;
 		info->config.pio_threshold = pio_threshold;
 		info->next_port = ports;
-		init_waitqueue_head(&info->port.open_wait);
-		init_waitqueue_head(&info->port.close_wait);
 		init_waitqueue_head(&info->delta_msr_wait);
 		init_waitqueue_head(&info->break_wait);
 		ports = info;
diff --git a/drivers/char/generic_serial.c b/drivers/char/generic_serial.c
index c6090f84a2e4..2356994ee010 100644
--- a/drivers/char/generic_serial.c
+++ b/drivers/char/generic_serial.c
@@ -397,7 +397,8 @@ void gs_hangup(struct tty_struct *tty)
 
 int gs_block_til_ready(void *port_, struct file * filp)
 {
-	struct gs_port *port = port_;
+	struct gs_port *gp = port_;
+	struct tty_port *port = &gp->port;
 	DECLARE_WAITQUEUE(wait, current);
 	int    retval;
 	int    do_clocal = 0;
@@ -409,16 +410,16 @@ int gs_block_til_ready(void *port_, struct file * filp)
 
 	if (!port) return 0;
 
-	tty = port->port.tty;
+	tty = port->tty;
 
 	gs_dprintk (GS_DEBUG_BTR, "Entering gs_block_till_ready.\n"); 
 	/*
 	 * If the device is in the middle of being closed, then block
 	 * until it's done, and then try again.
 	 */
-	if (tty_hung_up_p(filp) || port->port.flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&port->port.close_wait);
-		if (port->port.flags & ASYNC_HUP_NOTIFY)
+	if (tty_hung_up_p(filp) || port->flags & ASYNC_CLOSING) {
+		interruptible_sleep_on(&port->close_wait);
+		if (port->flags & ASYNC_HUP_NOTIFY)
 			return -EAGAIN;
 		else
 			return -ERESTARTSYS;
@@ -432,7 +433,7 @@ int gs_block_til_ready(void *port_, struct file * filp)
 	 */
 	if ((filp->f_flags & O_NONBLOCK) ||
 	    (tty->flags & (1 << TTY_IO_ERROR))) {
-		port->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -444,34 +445,34 @@ int gs_block_til_ready(void *port_, struct file * filp)
 	/*
 	 * Block waiting for the carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, port->port.count is dropped by one, so that
+	 * this loop, port->count is dropped by one, so that
 	 * rs_close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 	retval = 0;
 
-	add_wait_queue(&port->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 
 	gs_dprintk (GS_DEBUG_BTR, "after add waitq.\n"); 
-	spin_lock_irqsave(&port->driver_lock, flags);
+	spin_lock_irqsave(&gp->driver_lock, flags);
 	if (!tty_hung_up_p(filp)) {
-		port->port.count--;
+		port->count--;
 	}
-	spin_unlock_irqrestore(&port->driver_lock, flags);
-	port->port.blocked_open++;
+	spin_unlock_irqrestore(&gp->driver_lock, flags);
+	port->blocked_open++;
 	while (1) {
-		CD = port->rd->get_CD (port);
+		CD = tty_port_carrier_raised(port);
 		gs_dprintk (GS_DEBUG_BTR, "CD is now %d.\n", CD);
 		set_current_state (TASK_INTERRUPTIBLE);
 		if (tty_hung_up_p(filp) ||
-		    !(port->port.flags & ASYNC_INITIALIZED)) {
-			if (port->port.flags & ASYNC_HUP_NOTIFY)
+		    !(port->flags & ASYNC_INITIALIZED)) {
+			if (port->flags & ASYNC_HUP_NOTIFY)
 				retval = -EAGAIN;
 			else
 				retval = -ERESTARTSYS;
 			break;
 		}
-		if (!(port->port.flags & ASYNC_CLOSING) &&
+		if (!(port->flags & ASYNC_CLOSING) &&
 		    (do_clocal || CD))
 			break;
 		gs_dprintk (GS_DEBUG_BTR, "signal_pending is now: %d (%lx)\n", 
@@ -483,17 +484,17 @@ int gs_block_til_ready(void *port_, struct file * filp)
 		schedule();
 	}
 	gs_dprintk (GS_DEBUG_BTR, "Got out of the loop. (%d)\n",
-		    port->port.blocked_open);
+		    port->blocked_open);
 	set_current_state (TASK_RUNNING);
-	remove_wait_queue(&port->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
 	if (!tty_hung_up_p(filp)) {
-		port->port.count++;
+		port->count++;
 	}
-	port->port.blocked_open--;
+	port->blocked_open--;
 	if (retval)
 		return retval;
 
-	port->port.flags |= ASYNC_NORMAL_ACTIVE;
+	port->flags |= ASYNC_NORMAL_ACTIVE;
 	func_exit ();
 	return 0;
 }			 
diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c
index 04e4549299ba..b3da4858fd4a 100644
--- a/drivers/char/isicom.c
+++ b/drivers/char/isicom.c
@@ -830,20 +830,28 @@ static int isicom_setup_port(struct tty_struct *tty)
 	return 0;
 }
 
+static int isicom_carrier_raised(struct tty_port *port)
+{
+	struct isi_port *ip = container_of(port, struct isi_port, port);
+	return (ip->status & ISI_DCD)?1 : 0;
+}
+
 static int block_til_ready(struct tty_struct *tty, struct file *filp,
-	struct isi_port *port)
+	struct isi_port *ip)
 {
-	struct isi_board *card = port->card;
+	struct isi_board *card = ip->card;
+	struct tty_port *port = &ip->port;
 	int do_clocal = 0, retval;
 	unsigned long flags;
 	DECLARE_WAITQUEUE(wait, current);
+	int cd;
 
 	/* block if port is in the process of being closed */
 
-	if (tty_hung_up_p(filp) || port->port.flags & ASYNC_CLOSING) {
+	if (tty_hung_up_p(filp) || port->flags & ASYNC_CLOSING) {
 		pr_dbg("block_til_ready: close in progress.\n");
-		interruptible_sleep_on(&port->port.close_wait);
-		if (port->port.flags & ASYNC_HUP_NOTIFY)
+		interruptible_sleep_on(&port->close_wait);
+		if (port->flags & ASYNC_HUP_NOTIFY)
 			return -EAGAIN;
 		else
 			return -ERESTARTSYS;
@@ -854,7 +862,7 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	if ((filp->f_flags & O_NONBLOCK) ||
 			(tty->flags & (1 << TTY_IO_ERROR))) {
 		pr_dbg("block_til_ready: non-block mode.\n");
-		port->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -864,29 +872,29 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	/* block waiting for DCD to be asserted, and while
 						callout dev is busy */
 	retval = 0;
-	add_wait_queue(&port->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 
 	spin_lock_irqsave(&card->card_lock, flags);
 	if (!tty_hung_up_p(filp))
-		port->port.count--;
-	port->port.blocked_open++;
+		port->count--;
+	port->blocked_open++;
 	spin_unlock_irqrestore(&card->card_lock, flags);
 
 	while (1) {
-		raise_dtr_rts(port);
+		raise_dtr_rts(ip);
 
 		set_current_state(TASK_INTERRUPTIBLE);
-		if (tty_hung_up_p(filp) || !(port->port.flags & ASYNC_INITIALIZED)) {
-			if (port->port.flags & ASYNC_HUP_NOTIFY)
+		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)) {
+			if (port->flags & ASYNC_HUP_NOTIFY)
 				retval = -EAGAIN;
 			else
 				retval = -ERESTARTSYS;
 			break;
 		}
-		if (!(port->port.flags & ASYNC_CLOSING) &&
-				(do_clocal || (port->status & ISI_DCD))) {
+		cd = tty_port_carrier_raised(port);
+		if (!(port->flags & ASYNC_CLOSING) &&
+				(do_clocal || cd))
 			break;
-		}
 		if (signal_pending(current)) {
 			retval = -ERESTARTSYS;
 			break;
@@ -894,15 +902,15 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 		schedule();
 	}
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
 	spin_lock_irqsave(&card->card_lock, flags);
 	if (!tty_hung_up_p(filp))
-		port->port.count++;
-	port->port.blocked_open--;
+		port->count++;
+	port->blocked_open--;
 	spin_unlock_irqrestore(&card->card_lock, flags);
 	if (retval)
 		return retval;
-	port->port.flags |= ASYNC_NORMAL_ACTIVE;
+	port->flags |= ASYNC_NORMAL_ACTIVE;
 	return 0;
 }
 
@@ -1452,6 +1460,10 @@ static const struct tty_operations isicom_ops = {
 	.break_ctl		= isicom_send_break,
 };
 
+static const struct tty_port_operations isicom_port_ops = {
+	.carrier_raised		= isicom_carrier_raised,
+};
+
 static int __devinit reset_card(struct pci_dev *pdev,
 	const unsigned int card, unsigned int *signature)
 {
@@ -1794,6 +1806,7 @@ static int __init isicom_init(void)
 		spin_lock_init(&isi_card[idx].card_lock);
 		for (channel = 0; channel < 16; channel++, port++) {
 			tty_port_init(&port->port);
+			port->port.ops = &isicom_port_ops;
 			port->magic = ISICOM_MAGIC;
 			port->card = &isi_card[idx];
 			port->channel = channel;
diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c
index 4b10770fa937..c4682f9e34bb 100644
--- a/drivers/char/istallion.c
+++ b/drivers/char/istallion.c
@@ -151,7 +151,7 @@ static char	*stli_drvversion = "5.6.0";
 static char	*stli_serialname = "ttyE";
 
 static struct tty_driver	*stli_serial;
-
+static const struct tty_port_operations stli_port_ops;
 
 #define	STLI_TXBUFSIZE		4096
 
@@ -1183,6 +1183,12 @@ static int stli_setport(struct tty_struct *tty)
 
 /*****************************************************************************/
 
+static int stli_carrier_raised(struct tty_port *port)
+{
+	struct stliport *portp = container_of(port, struct stliport, port);
+	return (portp->sigs & TIOCM_CD) ? 1 : 0;
+}
+
 /*
  *	Possibly need to wait for carrier (DCD signal) to come high. Say
  *	maybe because if we are clocal then we don't need to wait...
@@ -1193,6 +1199,7 @@ static int stli_waitcarrier(struct tty_struct *tty, struct stlibrd *brdp,
 {
 	unsigned long flags;
 	int rc, doclocal;
+	struct tty_port *port = &portp->port;
 
 	rc = 0;
 	doclocal = 0;
@@ -1203,7 +1210,7 @@ static int stli_waitcarrier(struct tty_struct *tty, struct stlibrd *brdp,
 	spin_lock_irqsave(&stli_lock, flags);
 	portp->openwaitcnt++;
 	if (! tty_hung_up_p(filp))
-		portp->port.count--;
+		port->count--;
 	spin_unlock_irqrestore(&stli_lock, flags);
 
 	for (;;) {
@@ -1212,27 +1219,27 @@ static int stli_waitcarrier(struct tty_struct *tty, struct stlibrd *brdp,
 		    &portp->asig, sizeof(asysigs_t), 0)) < 0)
 			break;
 		if (tty_hung_up_p(filp) ||
-		    ((portp->port.flags & ASYNC_INITIALIZED) == 0)) {
-			if (portp->port.flags & ASYNC_HUP_NOTIFY)
+		    ((port->flags & ASYNC_INITIALIZED) == 0)) {
+			if (port->flags & ASYNC_HUP_NOTIFY)
 				rc = -EBUSY;
 			else
 				rc = -ERESTARTSYS;
 			break;
 		}
-		if (((portp->port.flags & ASYNC_CLOSING) == 0) &&
-		    (doclocal || (portp->sigs & TIOCM_CD))) {
+		if (((port->flags & ASYNC_CLOSING) == 0) &&
+		    (doclocal || tty_port_carrier_raised(port))) {
 			break;
 		}
 		if (signal_pending(current)) {
 			rc = -ERESTARTSYS;
 			break;
 		}
-		interruptible_sleep_on(&portp->port.open_wait);
+		interruptible_sleep_on(&port->open_wait);
 	}
 
 	spin_lock_irqsave(&stli_lock, flags);
 	if (! tty_hung_up_p(filp))
-		portp->port.count++;
+		port->count++;
 	portp->openwaitcnt--;
 	spin_unlock_irqrestore(&stli_lock, flags);
 
@@ -2696,6 +2703,7 @@ static int stli_initports(struct stlibrd *brdp)
 			continue;
 		}
 		tty_port_init(&portp->port);
+		portp->port.ops = &stli_port_ops;
 		portp->magic = STLI_PORTMAGIC;
 		portp->portnr = i;
 		portp->brdnr = brdp->brdnr;
@@ -4518,6 +4526,10 @@ static const struct tty_operations stli_ops = {
 	.tiocmset = stli_tiocmset,
 };
 
+static const struct tty_port_operations stli_port_ops = {
+	.carrier_raised = stli_carrier_raised,
+};
+
 /*****************************************************************************/
 /*
  *	Loadable module initialization stuff.
diff --git a/drivers/char/moxa.c b/drivers/char/moxa.c
index 12d327a2c9ba..8b0da97d5293 100644
--- a/drivers/char/moxa.c
+++ b/drivers/char/moxa.c
@@ -206,6 +206,7 @@ static void moxa_poll(unsigned long);
 static void moxa_set_tty_param(struct tty_struct *, struct ktermios *);
 static void moxa_setup_empty_event(struct tty_struct *);
 static void moxa_shut_down(struct tty_struct *);
+static int moxa_carrier_raised(struct tty_port *);
 /*
  * moxa board interface functions:
  */
@@ -405,6 +406,10 @@ static const struct tty_operations moxa_ops = {
 	.tiocmset = moxa_tiocmset,
 };
 
+static const struct tty_port_operations moxa_port_ops = {
+	.carrier_raised = moxa_carrier_raised,
+};
+
 static struct tty_driver *moxaDriver;
 static DEFINE_TIMER(moxaTimer, moxa_poll, 0, 0);
 static DEFINE_SPINLOCK(moxa_lock);
@@ -826,6 +831,7 @@ static int moxa_init_board(struct moxa_board_conf *brd, struct device *dev)
 
 	for (i = 0, p = brd->ports; i < MAX_PORTS_PER_BOARD; i++, p++) {
 		tty_port_init(&p->port);
+		p->port.ops = &moxa_port_ops;
 		p->type = PORT_16550A;
 		p->cflag = B9600 | CS8 | CREAD | CLOCAL | HUPCL;
 	}
@@ -1115,15 +1121,27 @@ static void moxa_close_port(struct tty_struct *tty)
 	tty_port_tty_set(&ch->port, NULL);
 }
 
+static int moxa_carrier_raised(struct tty_port *port)
+{
+	struct moxa_port *ch = container_of(port, struct moxa_port, port);
+	int dcd;
+
+	spin_lock_bh(&moxa_lock);
+	dcd = ch->DCDState;
+	spin_unlock_bh(&moxa_lock);
+	return dcd;
+}
+
 static int moxa_block_till_ready(struct tty_struct *tty, struct file *filp,
 			    struct moxa_port *ch)
 {
+	struct tty_port *port = &ch->port;
 	DEFINE_WAIT(wait);
 	int retval = 0;
 	u8 dcd;
 
 	while (1) {
-		prepare_to_wait(&ch->port.open_wait, &wait, TASK_INTERRUPTIBLE);
+		prepare_to_wait(&port->open_wait, &wait, TASK_INTERRUPTIBLE);
 		if (tty_hung_up_p(filp)) {
 #ifdef SERIAL_DO_RESTART
 			retval = -ERESTARTSYS;
@@ -1132,9 +1150,7 @@ static int moxa_block_till_ready(struct tty_struct *tty, struct file *filp,
 #endif
 			break;
 		}
-		spin_lock_bh(&moxa_lock);
-		dcd = ch->DCDState;
-		spin_unlock_bh(&moxa_lock);
+		dcd = tty_port_carrier_raised(port);
 		if (dcd)
 			break;
 
@@ -1144,7 +1160,7 @@ static int moxa_block_till_ready(struct tty_struct *tty, struct file *filp,
 		}
 		schedule();
 	}
-	finish_wait(&ch->port.open_wait, &wait);
+	finish_wait(&port->open_wait, &wait);
 
 	return retval;
 }
diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index 047766915411..eafbbcf355e7 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -541,13 +541,21 @@ static unsigned char mxser_get_msr(int baseaddr, int mode, int port)
 	return status;
 }
 
+static int mxser_carrier_raised(struct tty_port *port)
+{
+	struct mxser_port *mp = container_of(port, struct mxser_port, port);
+	return (inb(mp->ioaddr + UART_MSR) & UART_MSR_DCD)?1:0;
+}
+
 static int mxser_block_til_ready(struct tty_struct *tty, struct file *filp,
-		struct mxser_port *port)
+		struct mxser_port *mp)
 {
 	DECLARE_WAITQUEUE(wait, current);
 	int retval;
 	int do_clocal = 0;
 	unsigned long flags;
+	int cd;
+	struct tty_port *port = &mp->port;
 
 	/*
 	 * If non-blocking mode is set, or the port is not enabled,
@@ -555,7 +563,7 @@ static int mxser_block_til_ready(struct tty_struct *tty, struct file *filp,
 	 */
 	if ((filp->f_flags & O_NONBLOCK) ||
 			test_bit(TTY_IO_ERROR, &tty->flags)) {
-		port->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -565,34 +573,33 @@ static int mxser_block_til_ready(struct tty_struct *tty, struct file *filp,
 	/*
 	 * Block waiting for the carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, port->port.count is dropped by one, so that
+	 * this loop, port->count is dropped by one, so that
 	 * mxser_close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 	retval = 0;
-	add_wait_queue(&port->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 
-	spin_lock_irqsave(&port->slock, flags);
+	spin_lock_irqsave(&mp->slock, flags);
 	if (!tty_hung_up_p(filp))
-		port->port.count--;
-	spin_unlock_irqrestore(&port->slock, flags);
-	port->port.blocked_open++;
+		port->count--;
+	spin_unlock_irqrestore(&mp->slock, flags);
+	port->blocked_open++;
 	while (1) {
-		spin_lock_irqsave(&port->slock, flags);
-		outb(inb(port->ioaddr + UART_MCR) |
-			UART_MCR_DTR | UART_MCR_RTS, port->ioaddr + UART_MCR);
-		spin_unlock_irqrestore(&port->slock, flags);
+		spin_lock_irqsave(&mp->slock, flags);
+		outb(inb(mp->ioaddr + UART_MCR) |
+			UART_MCR_DTR | UART_MCR_RTS, mp->ioaddr + UART_MCR);
+		spin_unlock_irqrestore(&mp->slock, flags);
 		set_current_state(TASK_INTERRUPTIBLE);
-		if (tty_hung_up_p(filp) || !(port->port.flags & ASYNC_INITIALIZED)) {
-			if (port->port.flags & ASYNC_HUP_NOTIFY)
+		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)) {
+			if (port->flags & ASYNC_HUP_NOTIFY)
 				retval = -EAGAIN;
 			else
 				retval = -ERESTARTSYS;
 			break;
 		}
-		if (!(port->port.flags & ASYNC_CLOSING) &&
-				(do_clocal ||
-				(inb(port->ioaddr + UART_MSR) & UART_MSR_DCD)))
+		cd = tty_port_carrier_raised(port);
+		if (!(port->flags & ASYNC_CLOSING) && (do_clocal || cd))
 			break;
 		if (signal_pending(current)) {
 			retval = -ERESTARTSYS;
@@ -601,13 +608,13 @@ static int mxser_block_til_ready(struct tty_struct *tty, struct file *filp,
 		schedule();
 	}
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
 	if (!tty_hung_up_p(filp))
-		port->port.count++;
-	port->port.blocked_open--;
+		port->count++;
+	port->blocked_open--;
 	if (retval)
 		return retval;
-	port->port.flags |= ASYNC_NORMAL_ACTIVE;
+	port->flags |= ASYNC_NORMAL_ACTIVE;
 	return 0;
 }
 
@@ -2449,6 +2456,10 @@ static const struct tty_operations mxser_ops = {
 	.tiocmset = mxser_tiocmset,
 };
 
+struct tty_port_operations mxser_port_ops = {
+	.carrier_raised = mxser_carrier_raised,
+};
+
 /*
  * The MOXA Smartio/Industio serial driver boot-time initialization code!
  */
@@ -2482,6 +2493,7 @@ static int __devinit mxser_initbrd(struct mxser_board *brd,
 	for (i = 0; i < brd->info->nports; i++) {
 		info = &brd->ports[i];
 		tty_port_init(&info->port);
+		info->port.ops = &mxser_port_ops;
 		info->board = brd;
 		info->stop_rx = 0;
 		info->ldisc_stop_rx = 0;
diff --git a/drivers/char/rio/rio_linux.c b/drivers/char/rio/rio_linux.c
index a8f68a3f14dd..ec2afd139472 100644
--- a/drivers/char/rio/rio_linux.c
+++ b/drivers/char/rio/rio_linux.c
@@ -173,7 +173,7 @@ static void rio_disable_tx_interrupts(void *ptr);
 static void rio_enable_tx_interrupts(void *ptr);
 static void rio_disable_rx_interrupts(void *ptr);
 static void rio_enable_rx_interrupts(void *ptr);
-static int rio_get_CD(void *ptr);
+static int rio_carrier_raised(struct tty_port *port);
 static void rio_shutdown_port(void *ptr);
 static int rio_set_real_termios(void *ptr);
 static void rio_hungup(void *ptr);
@@ -224,7 +224,6 @@ static struct real_driver rio_real_driver = {
 	rio_enable_tx_interrupts,
 	rio_disable_rx_interrupts,
 	rio_enable_rx_interrupts,
-	rio_get_CD,
 	rio_shutdown_port,
 	rio_set_real_termios,
 	rio_chars_in_buffer,
@@ -476,9 +475,9 @@ static void rio_enable_rx_interrupts(void *ptr)
 
 
 /* Jeez. Isn't this simple?  */
-static int rio_get_CD(void *ptr)
+static int rio_carrier_raised(struct tty_port *port)
 {
-	struct Port *PortP = ptr;
+	struct Port *PortP = container_of(port, struct Port, gs.port);
 	int rv;
 
 	func_enter();
@@ -806,7 +805,9 @@ static void *ckmalloc(int size)
 	return p;
 }
 
-
+static const struct tty_port_operations rio_port_ops = {
+	.carrier_raised = rio_carrier_raised,
+};
 
 static int rio_init_datastructures(void)
 {
@@ -842,17 +843,14 @@ static int rio_init_datastructures(void)
 			goto free6;
 		}
 		rio_dprintk(RIO_DEBUG_INIT, "initing port %d (%d)\n", i, port->Mapped);
+		tty_port_init(&port->gs.port);
+		port->gs.port.ops = &rio_port_ops;
 		port->PortNum = i;
 		port->gs.magic = RIO_MAGIC;
 		port->gs.close_delay = HZ / 2;
 		port->gs.closing_wait = 30 * HZ;
 		port->gs.rd = &rio_real_driver;
 		spin_lock_init(&port->portSem);
-		/*
-		 * Initializing wait queue
-		 */
-		init_waitqueue_head(&port->gs.port.open_wait);
-		init_waitqueue_head(&port->gs.port.close_wait);
 	}
 #else
 	/* We could postpone initializing them to when they are configured. */
diff --git a/drivers/char/riscom8.c b/drivers/char/riscom8.c
index 2c6c8f33d6b4..6ad1c2aa2a98 100644
--- a/drivers/char/riscom8.c
+++ b/drivers/char/riscom8.c
@@ -857,23 +857,40 @@ static void rc_shutdown_port(struct tty_struct *tty,
 		rc_shutdown_board(bp);
 }
 
+static int carrier_raised(struct tty_port *port)
+{
+	struct riscom_port *p = container_of(port, struct riscom_port, port);
+	struct riscom_board *bp = port_Board(p);
+	unsigned long flags;
+	int CD;
+	
+	spin_lock_irqsave(&riscom_lock, flags);
+	rc_out(bp, CD180_CAR, port_No(p));
+	CD = rc_in(bp, CD180_MSVR) & MSVR_CD;
+	rc_out(bp, CD180_MSVR, MSVR_RTS);
+	bp->DTR &= ~(1u << port_No(p));
+	rc_out(bp, RC_DTR, bp->DTR);
+	spin_unlock_irqrestore(&riscom_lock, flags);
+	return CD;
+}
+
 static int block_til_ready(struct tty_struct *tty, struct file *filp,
-			   struct riscom_port *port)
+			   struct riscom_port *rp)
 {
 	DECLARE_WAITQUEUE(wait, current);
-	struct riscom_board *bp = port_Board(port);
 	int    retval;
 	int    do_clocal = 0;
 	int    CD;
 	unsigned long flags;
+	struct tty_port *port = &rp->port;
 
 	/*
 	 * If the device is in the middle of being closed, then block
 	 * until it's done, and then try again.
 	 */
-	if (tty_hung_up_p(filp) || port->port.flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&port->port.close_wait);
-		if (port->port.flags & ASYNC_HUP_NOTIFY)
+	if (tty_hung_up_p(filp) || port->flags & ASYNC_CLOSING) {
+		interruptible_sleep_on(&port->close_wait);
+		if (port->flags & ASYNC_HUP_NOTIFY)
 			return -EAGAIN;
 		else
 			return -ERESTARTSYS;
@@ -885,7 +902,7 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	 */
 	if ((filp->f_flags & O_NONBLOCK) ||
 	    (tty->flags & (1 << TTY_IO_ERROR))) {
-		port->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -900,37 +917,29 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	 * exit, either normal or abnormal.
 	 */
 	retval = 0;
-	add_wait_queue(&port->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 
 	spin_lock_irqsave(&riscom_lock, flags);
 
 	if (!tty_hung_up_p(filp))
-		port->port.count--;
+		port->count--;
 
 	spin_unlock_irqrestore(&riscom_lock, flags);
 
-	port->port.blocked_open++;
+	port->blocked_open++;
 	while (1) {
-		spin_lock_irqsave(&riscom_lock, flags);
-
-		rc_out(bp, CD180_CAR, port_No(port));
-		CD = rc_in(bp, CD180_MSVR) & MSVR_CD;
-		rc_out(bp, CD180_MSVR, MSVR_RTS);
-		bp->DTR &= ~(1u << port_No(port));
-		rc_out(bp, RC_DTR, bp->DTR);
-
-		spin_unlock_irqrestore(&riscom_lock, flags);
 
+		CD = tty_port_carrier_raised(port);
 		set_current_state(TASK_INTERRUPTIBLE);
 		if (tty_hung_up_p(filp) ||
-		    !(port->port.flags & ASYNC_INITIALIZED)) {
-			if (port->port.flags & ASYNC_HUP_NOTIFY)
+		    !(port->flags & ASYNC_INITIALIZED)) {
+			if (port->flags & ASYNC_HUP_NOTIFY)
 				retval = -EAGAIN;
 			else
 				retval = -ERESTARTSYS;
 			break;
 		}
-		if (!(port->port.flags & ASYNC_CLOSING) &&
+		if (!(port->flags & ASYNC_CLOSING) &&
 		    (do_clocal || CD))
 			break;
 		if (signal_pending(current)) {
@@ -940,14 +949,14 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 		schedule();
 	}
 	__set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
 	if (!tty_hung_up_p(filp))
-		port->port.count++;
-	port->port.blocked_open--;
+		port->count++;
+	port->blocked_open--;
 	if (retval)
 		return retval;
 
-	port->port.flags |= ASYNC_NORMAL_ACTIVE;
+	port->flags |= ASYNC_NORMAL_ACTIVE;
 	return 0;
 }
 
@@ -1510,6 +1519,11 @@ static const struct tty_operations riscom_ops = {
 	.break_ctl = rc_send_break,
 };
 
+static const struct tty_port_operations riscom_port_ops = {
+	.carrier_raised = carrier_raised,
+};
+
+
 static int __init rc_init_drivers(void)
 {
 	int error;
@@ -1541,6 +1555,7 @@ static int __init rc_init_drivers(void)
 	memset(rc_port, 0, sizeof(rc_port));
 	for (i = 0; i < RC_NPORT * RC_NBOARD; i++)  {
 		tty_port_init(&rc_port[i].port);
+		rc_port[i].port.ops = &riscom_port_ops;
 		rc_port[i].magic = RISCOM8_MAGIC;
 	}
 	return 0;
diff --git a/drivers/char/rocket.c b/drivers/char/rocket.c
index 584d791e84a6..4a4110e703a5 100644
--- a/drivers/char/rocket.c
+++ b/drivers/char/rocket.c
@@ -135,6 +135,7 @@ static int rcktpt_type[NUM_BOARDS];
 static int is_PCI[NUM_BOARDS];
 static rocketModel_t rocketModel[NUM_BOARDS];
 static int max_board;
+static const struct tty_port_operations rocket_port_ops;
 
 /*
  * The following arrays define the interrupt bits corresponding to each AIOP.
@@ -649,9 +650,8 @@ static void init_r_port(int board, int aiop, int chan, struct pci_dev *pci_dev)
 	info->board = board;
 	info->aiop = aiop;
 	info->chan = chan;
-	info->port.closing_wait = 3000;
-	info->port.close_delay = 50;
-	init_waitqueue_head(&info->port.open_wait);
+	tty_port_init(&info->port);
+	info->port.ops = &rocket_port_ops;
 	init_completion(&info->close_wait);
 	info->flags &= ~ROCKET_MODE_MASK;
 	switch (pc104[board][line]) {
@@ -864,11 +864,18 @@ static void configure_r_port(struct r_port *info,
 	}
 }
 
+static int carrier_raised(struct tty_port *port)
+{
+	struct r_port *info = container_of(port, struct r_port, port);
+	return (sGetChanStatusLo(&info->channel) & CD_ACT) ? 1 : 0;
+}
+
 /*  info->port.count is considered critical, protected by spinlocks.  */
 static int block_til_ready(struct tty_struct *tty, struct file *filp,
 			   struct r_port *info)
 {
 	DECLARE_WAITQUEUE(wait, current);
+	struct tty_port *port = &info->port;
 	int retval;
 	int do_clocal = 0, extra_count = 0;
 	unsigned long flags;
@@ -898,13 +905,13 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 
 	/*
 	 * Block waiting for the carrier detect and the line to become free.  While we are in
-	 * this loop, info->port.count is dropped by one, so that rp_close() knows when to free things.
+	 * this loop, port->count is dropped by one, so that rp_close() knows when to free things.
          * We restore it upon exit, either normal or abnormal.
 	 */
 	retval = 0;
-	add_wait_queue(&info->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 #ifdef ROCKET_DEBUG_OPEN
-	printk(KERN_INFO "block_til_ready before block: ttyR%d, count = %d\n", info->line, info->port.count);
+	printk(KERN_INFO "block_til_ready before block: ttyR%d, count = %d\n", info->line, port->count);
 #endif
 	spin_lock_irqsave(&info->slock, flags);
 
@@ -913,10 +920,10 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 #else
 	if (!tty_hung_up_p(filp)) {
 		extra_count = 1;
-		info->port.count--;
+		port->count--;
 	}
 #endif
-	info->port.blocked_open++;
+	port->blocked_open++;
 
 	spin_unlock_irqrestore(&info->slock, flags);
 
@@ -933,7 +940,8 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 				retval = -ERESTARTSYS;
 			break;
 		}
-		if (!(info->flags & ROCKET_CLOSING) && (do_clocal || (sGetChanStatusLo(&info->channel) & CD_ACT)))
+		if (!(info->flags & ROCKET_CLOSING) &&
+			(do_clocal || tty_port_carrier_raised(port)))
 			break;
 		if (signal_pending(current)) {
 			retval = -ERESTARTSYS;
@@ -941,24 +949,24 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 		}
 #ifdef ROCKET_DEBUG_OPEN
 		printk(KERN_INFO "block_til_ready blocking: ttyR%d, count = %d, flags=0x%0x\n",
-		     info->line, info->port.count, info->flags);
+		     info->line, port->count, info->flags);
 #endif
 		schedule();	/*  Don't hold spinlock here, will hang PC */
 	}
 	__set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
 
 	spin_lock_irqsave(&info->slock, flags);
 
 	if (extra_count)
-		info->port.count++;
-	info->port.blocked_open--;
+		port->count++;
+	port->blocked_open--;
 
 	spin_unlock_irqrestore(&info->slock, flags);
 
 #ifdef ROCKET_DEBUG_OPEN
 	printk(KERN_INFO "block_til_ready after blocking: ttyR%d, count = %d\n",
-	       info->line, info->port.count);
+	       info->line, port->count);
 #endif
 	if (retval)
 		return retval;
@@ -2371,6 +2379,10 @@ static const struct tty_operations rocket_ops = {
 	.tiocmset = rp_tiocmset,
 };
 
+static const struct tty_port_operations rocket_port_ops = {
+	.carrier_raised = carrier_raised,
+};
+
 /*
  * The module "startup" routine; it's run when the module is loaded.
  */
diff --git a/drivers/char/ser_a2232.c b/drivers/char/ser_a2232.c
index 7b0c35207d9b..0c97f34df63a 100644
--- a/drivers/char/ser_a2232.c
+++ b/drivers/char/ser_a2232.c
@@ -122,7 +122,7 @@ static void a2232_disable_tx_interrupts(void *ptr);
 static void a2232_enable_tx_interrupts(void *ptr);
 static void a2232_disable_rx_interrupts(void *ptr);
 static void a2232_enable_rx_interrupts(void *ptr);
-static int  a2232_get_CD(void *ptr);
+static int  a2232_carrier_raised(struct tty_port *port);
 static void a2232_shutdown_port(void *ptr);
 static int  a2232_set_real_termios(void *ptr);
 static int  a2232_chars_in_buffer(void *ptr);
@@ -148,7 +148,6 @@ static struct real_driver a2232_real_driver = {
         a2232_enable_tx_interrupts,
         a2232_disable_rx_interrupts,
         a2232_enable_rx_interrupts,
-        a2232_get_CD,
         a2232_shutdown_port,
         a2232_set_real_termios,
         a2232_chars_in_buffer,
@@ -260,9 +259,10 @@ static void a2232_enable_rx_interrupts(void *ptr)
 	port->disable_rx = 0;
 }
 
-static int  a2232_get_CD(void *ptr)
+static int  a2232_carrier_raised(struct tty_port *port)
 {
-	return ((struct a2232_port *) ptr)->cd_status;
+	struct a2232_port *ap = container_of(port, struct a2232_port, gs.port);
+	return ap->cd_status;
 }
 
 static void a2232_shutdown_port(void *ptr)
@@ -638,6 +638,10 @@ int ch, err, n, p;
 	return IRQ_HANDLED;
 }
 
+static const struct tty_port_operations a2232_port_ops = {
+	.carrier_raised = a2232_carrier_raised,
+};
+
 static void a2232_init_portstructs(void)
 {
 	struct a2232_port *port;
@@ -645,6 +649,8 @@ static void a2232_init_portstructs(void)
 
 	for (i = 0; i < MAX_A2232_BOARDS*NUMLINES; i++) {
 		port = a2232_ports + i;
+		tty_port_init(&port->gs.port);
+		port->gs.port.ops = &a2232_port_ops;
 		port->which_a2232 = i/NUMLINES;
 		port->which_port_on_a2232 = i%NUMLINES;
 		port->disable_rx = port->throttle_input = port->cd_status = 0;
@@ -652,11 +658,6 @@ static void a2232_init_portstructs(void)
 		port->gs.close_delay = HZ/2;
 		port->gs.closing_wait = 30 * HZ;
 		port->gs.rd = &a2232_real_driver;
-#ifdef NEW_WRITE_LOCKING
-		mutex_init(&(port->gs.port_write_mutex));
-#endif
-		init_waitqueue_head(&port->gs.port.open_wait);
-		init_waitqueue_head(&port->gs.port.close_wait);
 	}
 }
 
diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c
index 963b03fb29e5..12aecdaf61ec 100644
--- a/drivers/char/stallion.c
+++ b/drivers/char/stallion.c
@@ -130,6 +130,8 @@ static char		stl_unwanted[SC26198_RXFIFOSIZE];
 static DEFINE_MUTEX(stl_brdslock);
 static struct stlbrd		*stl_brds[STL_MAXBRDS];
 
+static const struct tty_port_operations stl_port_ops;
+
 /*
  *	Per board state flags. Used with the state field of the board struct.
  *	Not really much here!
@@ -786,6 +788,12 @@ static int stl_open(struct tty_struct *tty, struct file *filp)
 
 /*****************************************************************************/
 
+static int stl_carrier_raised(struct tty_port *port)
+{
+	struct stlport *portp = container_of(port, struct stlport, port);
+	return (portp->sigs & TIOCM_CD) ? 1 : 0;
+}
+
 /*
  *	Possibly need to wait for carrier (DCD signal) to come high. Say
  *	maybe because if we are clocal then we don't need to wait...
@@ -796,6 +804,7 @@ static int stl_waitcarrier(struct tty_struct *tty, struct stlport *portp,
 {
 	unsigned long	flags;
 	int		rc, doclocal;
+	struct tty_port *port = &portp->port;
 
 	pr_debug("stl_waitcarrier(portp=%p,filp=%p)\n", portp, filp);
 
@@ -809,32 +818,32 @@ static int stl_waitcarrier(struct tty_struct *tty, struct stlport *portp,
 
 	portp->openwaitcnt++;
 	if (! tty_hung_up_p(filp))
-		portp->port.count--;
+		port->count--;
 
 	for (;;) {
 		/* Takes brd_lock internally */
 		stl_setsignals(portp, 1, 1);
 		if (tty_hung_up_p(filp) ||
-		    ((portp->port.flags & ASYNC_INITIALIZED) == 0)) {
-			if (portp->port.flags & ASYNC_HUP_NOTIFY)
+		    ((port->flags & ASYNC_INITIALIZED) == 0)) {
+			if (port->flags & ASYNC_HUP_NOTIFY)
 				rc = -EBUSY;
 			else
 				rc = -ERESTARTSYS;
 			break;
 		}
-		if (((portp->port.flags & ASYNC_CLOSING) == 0) &&
-		    (doclocal || (portp->sigs & TIOCM_CD)))
+		if (((port->flags & ASYNC_CLOSING) == 0) &&
+		    (doclocal || tty_port_carrier_raised(port)))
 			break;
 		if (signal_pending(current)) {
 			rc = -ERESTARTSYS;
 			break;
 		}
 		/* FIXME */
-		interruptible_sleep_on(&portp->port.open_wait);
+		interruptible_sleep_on(&port->open_wait);
 	}
 
 	if (! tty_hung_up_p(filp))
-		portp->port.count++;
+		port->count++;
 	portp->openwaitcnt--;
 	spin_unlock_irqrestore(&stallion_lock, flags);
 
@@ -1776,6 +1785,7 @@ static int __devinit stl_initports(struct stlbrd *brdp, struct stlpanel *panelp)
 			break;
 		}
 		tty_port_init(&portp->port);
+		portp->port.ops = &stl_port_ops;
 		portp->magic = STL_PORTMAGIC;
 		portp->portnr = i;
 		portp->brdnr = panelp->brdnr;
@@ -2659,6 +2669,10 @@ static const struct tty_operations stl_ops = {
 	.tiocmset = stl_tiocmset,
 };
 
+static const struct tty_port_operations stl_port_ops = {
+	.carrier_raised = stl_carrier_raised,
+};
+
 /*****************************************************************************/
 /*                       CD1400 HARDWARE FUNCTIONS                           */
 /*****************************************************************************/
diff --git a/drivers/char/sx.c b/drivers/char/sx.c
index ba4e86281fbf..a71bc58abe7f 100644
--- a/drivers/char/sx.c
+++ b/drivers/char/sx.c
@@ -279,7 +279,7 @@ static void sx_disable_tx_interrupts(void *ptr);
 static void sx_enable_tx_interrupts(void *ptr);
 static void sx_disable_rx_interrupts(void *ptr);
 static void sx_enable_rx_interrupts(void *ptr);
-static int sx_get_CD(void *ptr);
+static int sx_carrier_raised(struct tty_port *port);
 static void sx_shutdown_port(void *ptr);
 static int sx_set_real_termios(void *ptr);
 static void sx_close(void *ptr);
@@ -360,7 +360,6 @@ static struct real_driver sx_real_driver = {
 	sx_enable_tx_interrupts,
 	sx_disable_rx_interrupts,
 	sx_enable_rx_interrupts,
-	sx_get_CD,
 	sx_shutdown_port,
 	sx_set_real_termios,
 	sx_chars_in_buffer,
@@ -791,7 +790,7 @@ static int sx_getsignals(struct sx_port *port)
 	sx_dprintk(SX_DEBUG_MODEMSIGNALS, "getsignals: %d/%d  (%d/%d) "
 			"%02x/%02x\n",
 			(o_stat & OP_DTR) != 0, (o_stat & OP_RTS) != 0,
-			port->c_dcd, sx_get_CD(port),
+			port->c_dcd, tty_port_carrier_raised(&port->gs.port),
 			sx_read_channel_byte(port, hi_ip),
 			sx_read_channel_byte(port, hi_state));
 
@@ -1190,7 +1189,7 @@ static inline void sx_check_modem_signals(struct sx_port *port)
 
 	hi_state = sx_read_channel_byte(port, hi_state);
 	sx_dprintk(SX_DEBUG_MODEMSIGNALS, "Checking modem signals (%d/%d)\n",
-			port->c_dcd, sx_get_CD(port));
+			port->c_dcd, tty_port_carrier_raised(&port->gs.port));
 
 	if (hi_state & ST_BREAK) {
 		hi_state &= ~ST_BREAK;
@@ -1202,11 +1201,11 @@ static inline void sx_check_modem_signals(struct sx_port *port)
 		hi_state &= ~ST_DCD;
 		sx_dprintk(SX_DEBUG_MODEMSIGNALS, "got a DCD change.\n");
 		sx_write_channel_byte(port, hi_state, hi_state);
-		c_dcd = sx_get_CD(port);
+		c_dcd = tty_port_carrier_raised(&port->gs.port);
 		sx_dprintk(SX_DEBUG_MODEMSIGNALS, "DCD is now %d\n", c_dcd);
 		if (c_dcd != port->c_dcd) {
 			port->c_dcd = c_dcd;
-			if (sx_get_CD(port)) {
+			if (tty_port_carrier_raised(&port->gs.port)) {
 				/* DCD went UP */
 				if ((sx_read_channel_byte(port, hi_hstat) !=
 						HS_IDLE_CLOSED) &&
@@ -1415,13 +1414,10 @@ static void sx_enable_rx_interrupts(void *ptr)
 }
 
 /* Jeez. Isn't this simple? */
-static int sx_get_CD(void *ptr)
+static int sx_carrier_raised(struct tty_port *port)
 {
-	struct sx_port *port = ptr;
-	func_enter2();
-
-	func_exit();
-	return ((sx_read_channel_byte(port, hi_ip) & IP_DCD) != 0);
+	struct sx_port *sp = container_of(port, struct sx_port, gs.port);
+	return ((sx_read_channel_byte(sp, hi_ip) & IP_DCD) != 0);
 }
 
 /* Jeez. Isn't this simple? */
@@ -1536,7 +1532,7 @@ static int sx_open(struct tty_struct *tty, struct file *filp)
 	}
 	/* tty->low_latency = 1; */
 
-	port->c_dcd = sx_get_CD(port);
+	port->c_dcd = sx_carrier_raised(&port->gs.port);
 	sx_dprintk(SX_DEBUG_OPEN, "at open: cd=%d\n", port->c_dcd);
 
 	func_exit();
@@ -2354,6 +2350,10 @@ static const struct tty_operations sx_ops = {
 	.tiocmset = sx_tiocmset,
 };
 
+static const struct tty_port_operations sx_port_ops = {
+	.carrier_raised = sx_carrier_raised,
+};
+
 static int sx_init_drivers(void)
 {
 	int error;
@@ -2410,6 +2410,7 @@ static int sx_init_portstructs(int nboards, int nports)
 		for (j = 0; j < boards[i].nports; j++) {
 			sx_dprintk(SX_DEBUG_INIT, "initing port %d\n", j);
 			tty_port_init(&port->gs.port);
+			port->gs.port.ops = &sx_port_ops;
 			port->gs.magic = SX_MAGIC;
 			port->gs.close_delay = HZ / 2;
 			port->gs.closing_wait = 30 * HZ;
diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c
index 500f5176b6ba..fb2e6b5e0ef1 100644
--- a/drivers/char/synclink.c
+++ b/drivers/char/synclink.c
@@ -3281,6 +3281,23 @@ static void mgsl_hangup(struct tty_struct *tty)
 	
 }	/* end of mgsl_hangup() */
 
+/*
+ * carrier_raised()
+ *
+ *	Return true if carrier is raised
+ */
+
+static int carrier_raised(struct tty_port *port)
+{
+	unsigned long flags;
+	struct mgsl_struct *info = container_of(port, struct mgsl_struct, port);
+	
+	spin_lock_irqsave(&info->irq_spinlock, flags);
+ 	usc_get_serial_signals(info);
+	spin_unlock_irqrestore(&info->irq_spinlock, flags);
+	return (info->serial_signals & SerialSignal_DCD) ? 1 : 0;
+}
+
 /* block_til_ready()
  * 
  * 	Block the current process until the specified port
@@ -3302,6 +3319,8 @@ static int block_til_ready(struct tty_struct *tty, struct file * filp,
 	bool		do_clocal = false;
 	bool		extra_count = false;
 	unsigned long	flags;
+	int		dcd;
+	struct tty_port *port = &info->port;
 	
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):block_til_ready on %s\n",
@@ -3309,7 +3328,7 @@ static int block_til_ready(struct tty_struct *tty, struct file * filp,
 
 	if (filp->f_flags & O_NONBLOCK || tty->flags & (1 << TTY_IO_ERROR)){
 		/* nonblock mode is set or port is not enabled */
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -3318,25 +3337,25 @@ static int block_til_ready(struct tty_struct *tty, struct file * filp,
 
 	/* Wait for carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->port.count is dropped by one, so that
+	 * this loop, port->count is dropped by one, so that
 	 * mgsl_close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 	 
 	retval = 0;
-	add_wait_queue(&info->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 	
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):block_til_ready before block on %s count=%d\n",
-			 __FILE__,__LINE__, tty->driver->name, info->port.count );
+			 __FILE__,__LINE__, tty->driver->name, port->count );
 
 	spin_lock_irqsave(&info->irq_spinlock, flags);
 	if (!tty_hung_up_p(filp)) {
 		extra_count = true;
-		info->port.count--;
+		port->count--;
 	}
 	spin_unlock_irqrestore(&info->irq_spinlock, flags);
-	info->port.blocked_open++;
+	port->blocked_open++;
 	
 	while (1) {
 		if (tty->termios->c_cflag & CBAUD) {
@@ -3348,20 +3367,16 @@ static int block_til_ready(struct tty_struct *tty, struct file * filp,
 		
 		set_current_state(TASK_INTERRUPTIBLE);
 		
-		if (tty_hung_up_p(filp) || !(info->port.flags & ASYNC_INITIALIZED)){
-			retval = (info->port.flags & ASYNC_HUP_NOTIFY) ?
+		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)){
+			retval = (port->flags & ASYNC_HUP_NOTIFY) ?
 					-EAGAIN : -ERESTARTSYS;
 			break;
 		}
 		
-		spin_lock_irqsave(&info->irq_spinlock,flags);
-	 	usc_get_serial_signals(info);
-		spin_unlock_irqrestore(&info->irq_spinlock,flags);
+		dcd = tty_port_carrier_raised(&info->port);
 		
- 		if (!(info->port.flags & ASYNC_CLOSING) &&
- 		    (do_clocal || (info->serial_signals & SerialSignal_DCD)) ) {
+ 		if (!(port->flags & ASYNC_CLOSING) && (do_clocal || dcd))
  			break;
-		}
 			
 		if (signal_pending(current)) {
 			retval = -ERESTARTSYS;
@@ -3370,24 +3385,24 @@ static int block_til_ready(struct tty_struct *tty, struct file * filp,
 		
 		if (debug_level >= DEBUG_LEVEL_INFO)
 			printk("%s(%d):block_til_ready blocking on %s count=%d\n",
-				 __FILE__,__LINE__, tty->driver->name, info->port.count );
+				 __FILE__,__LINE__, tty->driver->name, port->count );
 				 
 		schedule();
 	}
 	
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
 	
 	if (extra_count)
-		info->port.count++;
-	info->port.blocked_open--;
+		port->count++;
+	port->blocked_open--;
 	
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):block_til_ready after blocking on %s count=%d\n",
-			 __FILE__,__LINE__, tty->driver->name, info->port.count );
+			 __FILE__,__LINE__, tty->driver->name, port->count );
 			 
 	if (!retval)
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		
 	return retval;
 	
@@ -4304,6 +4319,11 @@ static void mgsl_add_device( struct mgsl_struct *info )
 
 }	/* end of mgsl_add_device() */
 
+static const struct tty_port_operations mgsl_port_ops = {
+	.carrier_raised = carrier_raised,
+};
+
+
 /* mgsl_allocate_device()
  * 
  * 	Allocate and initialize a device instance structure
@@ -4322,6 +4342,7 @@ static struct mgsl_struct* mgsl_allocate_device(void)
 		printk("Error can't allocate device instance data\n");
 	} else {
 		tty_port_init(&info->port);
+		info->port.ops = &mgsl_port_ops;
 		info->magic = MGSL_MAGIC;
 		INIT_WORK(&info->task, mgsl_bh_handler);
 		info->max_frame_size = 4096;
diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c
index 08911ed66494..39ccaba8ca37 100644
--- a/drivers/char/synclink_gt.c
+++ b/drivers/char/synclink_gt.c
@@ -3132,6 +3132,17 @@ static int tiocmset(struct tty_struct *tty, struct file *file,
 	return 0;
 }
 
+static int carrier_raised(struct tty_port *port)
+{
+	unsigned long flags;
+	struct slgt_info *info = container_of(port, struct slgt_info, port);
+
+	spin_lock_irqsave(&info->lock,flags);
+ 	get_signals(info);
+	spin_unlock_irqrestore(&info->lock,flags);
+	return (info->signals & SerialSignal_DCD) ? 1 : 0;
+}
+
 /*
  *  block current process until the device is ready to open
  */
@@ -3143,12 +3154,14 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	bool		do_clocal = false;
 	bool		extra_count = false;
 	unsigned long	flags;
+	int		cd;
+	struct tty_port *port = &info->port;
 
 	DBGINFO(("%s block_til_ready\n", tty->driver->name));
 
 	if (filp->f_flags & O_NONBLOCK || tty->flags & (1 << TTY_IO_ERROR)){
 		/* nonblock mode is set or port is not enabled */
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -3157,21 +3170,21 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 
 	/* Wait for carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->port.count is dropped by one, so that
+	 * this loop, port->count is dropped by one, so that
 	 * close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 
 	retval = 0;
-	add_wait_queue(&info->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 
 	spin_lock_irqsave(&info->lock, flags);
 	if (!tty_hung_up_p(filp)) {
 		extra_count = true;
-		info->port.count--;
+		port->count--;
 	}
 	spin_unlock_irqrestore(&info->lock, flags);
-	info->port.blocked_open++;
+	port->blocked_open++;
 
 	while (1) {
 		if ((tty->termios->c_cflag & CBAUD)) {
@@ -3183,20 +3196,16 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 
 		set_current_state(TASK_INTERRUPTIBLE);
 
-		if (tty_hung_up_p(filp) || !(info->port.flags & ASYNC_INITIALIZED)){
-			retval = (info->port.flags & ASYNC_HUP_NOTIFY) ?
+		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)){
+			retval = (port->flags & ASYNC_HUP_NOTIFY) ?
 					-EAGAIN : -ERESTARTSYS;
 			break;
 		}
 
-		spin_lock_irqsave(&info->lock,flags);
-	 	get_signals(info);
-		spin_unlock_irqrestore(&info->lock,flags);
+		cd = tty_port_carrier_raised(port);
 
- 		if (!(info->port.flags & ASYNC_CLOSING) &&
- 		    (do_clocal || (info->signals & SerialSignal_DCD)) ) {
+ 		if (!(port->flags & ASYNC_CLOSING) && (do_clocal || cd ))
  			break;
-		}
 
 		if (signal_pending(current)) {
 			retval = -ERESTARTSYS;
@@ -3208,14 +3217,14 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	}
 
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
 
 	if (extra_count)
-		info->port.count++;
-	info->port.blocked_open--;
+		port->count++;
+	port->blocked_open--;
 
 	if (!retval)
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 
 	DBGINFO(("%s block_til_ready ready, rc=%d\n", tty->driver->name, retval));
 	return retval;
@@ -3444,6 +3453,10 @@ static void add_device(struct slgt_info *info)
 #endif
 }
 
+static const struct tty_port_operations slgt_port_ops = {
+	.carrier_raised = carrier_raised,
+};
+
 /*
  *  allocate device instance structure, return NULL on failure
  */
@@ -3458,6 +3471,7 @@ static struct slgt_info *alloc_dev(int adapter_num, int port_num, struct pci_dev
 			driver_name, adapter_num, port_num));
 	} else {
 		tty_port_init(&info->port);
+		info->port.ops = &slgt_port_ops;
 		info->magic = MGSL_MAGIC;
 		INIT_WORK(&info->task, bh_handler);
 		info->max_frame_size = 4096;
diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c
index 6bdb44f7bec2..fcf1ec77450d 100644
--- a/drivers/char/synclinkmp.c
+++ b/drivers/char/synclinkmp.c
@@ -558,6 +558,7 @@ static void release_resources(SLMP_INFO *info);
 
 static int  startup(SLMP_INFO *info);
 static int  block_til_ready(struct tty_struct *tty, struct file * filp,SLMP_INFO *info);
+static int carrier_raised(struct tty_port *port);
 static void shutdown(SLMP_INFO *info);
 static void program_hw(SLMP_INFO *info);
 static void change_params(SLMP_INFO *info);
@@ -3318,7 +3319,17 @@ static int tiocmset(struct tty_struct *tty, struct file *file,
 	return 0;
 }
 
+static int carrier_raised(struct tty_port *port)
+{
+	SLMP_INFO *info = container_of(port, SLMP_INFO, port);
+	unsigned long flags;
 
+	spin_lock_irqsave(&info->lock,flags);
+ 	get_signals(info);
+	spin_unlock_irqrestore(&info->lock,flags);
+
+	return (info->serial_signals & SerialSignal_DCD) ? 1 : 0;
+}
 
 /* Block the current process until the specified port is ready to open.
  */
@@ -3330,6 +3341,8 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	bool		do_clocal = false;
 	bool		extra_count = false;
 	unsigned long	flags;
+	int		cd;
+	struct tty_port *port = &info->port;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):%s block_til_ready()\n",
@@ -3338,7 +3351,7 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	if (filp->f_flags & O_NONBLOCK || tty->flags & (1 << TTY_IO_ERROR)){
 		/* nonblock mode is set or port is not enabled */
 		/* just verify that callout device is not active */
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 		return 0;
 	}
 
@@ -3347,25 +3360,25 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 
 	/* Wait for carrier detect and the line to become
 	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->port.count is dropped by one, so that
+	 * this loop, port->count is dropped by one, so that
 	 * close() knows when to free things.  We restore it upon
 	 * exit, either normal or abnormal.
 	 */
 
 	retval = 0;
-	add_wait_queue(&info->port.open_wait, &wait);
+	add_wait_queue(&port->open_wait, &wait);
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):%s block_til_ready() before block, count=%d\n",
-			 __FILE__,__LINE__, tty->driver->name, info->port.count );
+			 __FILE__,__LINE__, tty->driver->name, port->count );
 
 	spin_lock_irqsave(&info->lock, flags);
 	if (!tty_hung_up_p(filp)) {
 		extra_count = true;
-		info->port.count--;
+		port->count--;
 	}
 	spin_unlock_irqrestore(&info->lock, flags);
-	info->port.blocked_open++;
+	port->blocked_open++;
 
 	while (1) {
 		if ((tty->termios->c_cflag & CBAUD)) {
@@ -3377,20 +3390,16 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 
 		set_current_state(TASK_INTERRUPTIBLE);
 
-		if (tty_hung_up_p(filp) || !(info->port.flags & ASYNC_INITIALIZED)){
-			retval = (info->port.flags & ASYNC_HUP_NOTIFY) ?
+		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)){
+			retval = (port->flags & ASYNC_HUP_NOTIFY) ?
 					-EAGAIN : -ERESTARTSYS;
 			break;
 		}
 
-		spin_lock_irqsave(&info->lock,flags);
-	 	get_signals(info);
-		spin_unlock_irqrestore(&info->lock,flags);
+		cd = tty_port_carrier_raised(port);
 
- 		if (!(info->port.flags & ASYNC_CLOSING) &&
- 		    (do_clocal || (info->serial_signals & SerialSignal_DCD)) ) {
+ 		if (!(port->flags & ASYNC_CLOSING) && (do_clocal || cd))
  			break;
-		}
 
 		if (signal_pending(current)) {
 			retval = -ERESTARTSYS;
@@ -3399,24 +3408,24 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 
 		if (debug_level >= DEBUG_LEVEL_INFO)
 			printk("%s(%d):%s block_til_ready() count=%d\n",
-				 __FILE__,__LINE__, tty->driver->name, info->port.count );
+				 __FILE__,__LINE__, tty->driver->name, port->count );
 
 		schedule();
 	}
 
 	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&info->port.open_wait, &wait);
+	remove_wait_queue(&port->open_wait, &wait);
 
 	if (extra_count)
-		info->port.count++;
-	info->port.blocked_open--;
+		port->count++;
+	port->blocked_open--;
 
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):%s block_til_ready() after, count=%d\n",
-			 __FILE__,__LINE__, tty->driver->name, info->port.count );
+			 __FILE__,__LINE__, tty->driver->name, port->count );
 
 	if (!retval)
-		info->port.flags |= ASYNC_NORMAL_ACTIVE;
+		port->flags |= ASYNC_NORMAL_ACTIVE;
 
 	return retval;
 }
@@ -3782,6 +3791,10 @@ static void add_device(SLMP_INFO *info)
 #endif
 }
 
+static const struct tty_port_operations port_ops = {
+	.carrier_raised = carrier_raised,
+};
+
 /* Allocate and initialize a device instance structure
  *
  * Return Value:	pointer to SLMP_INFO if success, otherwise NULL
@@ -3798,6 +3811,7 @@ static SLMP_INFO *alloc_dev(int adapter_num, int port_num, struct pci_dev *pdev)
 			__FILE__,__LINE__, adapter_num, port_num);
 	} else {
 		tty_port_init(&info->port);
+		info->port.ops = &port_ops;
 		info->magic = MGSL_MAGIC;
 		INIT_WORK(&info->task, bh_handler);
 		info->max_frame_size = 4096;
@@ -3940,6 +3954,7 @@ static const struct tty_operations ops = {
 	.tiocmset = tiocmset,
 };
 
+
 static void synclinkmp_cleanup(void)
 {
 	int rc;
diff --git a/drivers/char/tty_port.c b/drivers/char/tty_port.c
index c8f8024cb40e..f54e40cbf023 100644
--- a/drivers/char/tty_port.c
+++ b/drivers/char/tty_port.c
@@ -94,3 +94,20 @@ void tty_port_tty_set(struct tty_port *port, struct tty_struct *tty)
 	spin_unlock_irqrestore(&port->lock, flags);
 }
 EXPORT_SYMBOL(tty_port_tty_set);
+
+/**
+ *	tty_port_carrier_raised	-	carrier raised check
+ *	@port: tty port
+ *
+ *	Wrapper for the carrier detect logic. For the moment this is used
+ *	to hide some internal details. This will eventually become entirely
+ *	internal to the tty port.
+ */
+
+int tty_port_carrier_raised(struct tty_port *port)
+{
+	if (port->ops->carrier_raised == NULL)
+		return 1;
+	return port->ops->carrier_raised(port);
+}
+EXPORT_SYMBOL(tty_port_carrier_raised);
diff --git a/drivers/char/vme_scc.c b/drivers/char/vme_scc.c
index 1718b3c481db..d4e1534c0e03 100644
--- a/drivers/char/vme_scc.c
+++ b/drivers/char/vme_scc.c
@@ -69,7 +69,7 @@ static void scc_disable_tx_interrupts(void * ptr);
 static void scc_enable_tx_interrupts(void * ptr);
 static void scc_disable_rx_interrupts(void * ptr);
 static void scc_enable_rx_interrupts(void * ptr);
-static int  scc_get_CD(void * ptr);
+static int  scc_carrier_raised(struct tty_port *port);
 static void scc_shutdown_port(void * ptr);
 static int scc_set_real_termios(void  *ptr);
 static void scc_hungup(void  *ptr);
@@ -100,7 +100,6 @@ static struct real_driver scc_real_driver = {
         scc_enable_tx_interrupts,
         scc_disable_rx_interrupts,
         scc_enable_rx_interrupts,
-        scc_get_CD,
         scc_shutdown_port,
         scc_set_real_termios,
         scc_chars_in_buffer,
@@ -129,6 +128,10 @@ static const struct tty_operations scc_ops = {
 	.break_ctl = scc_break_ctl,
 };
 
+static const struct tty_port_operations scc_port_ops = {
+	.carrier_raised = scc_carrier_raised,
+};
+
 /*----------------------------------------------------------------------------
  * vme_scc_init() and support functions
  *---------------------------------------------------------------------------*/
@@ -176,6 +179,8 @@ static void scc_init_portstructs(void)
 
 	for (i = 0; i < 2; i++) {
 		port = scc_ports + i;
+		tty_port_init(&port->gs.port);
+		port->gs.port.ops = &scc_port_ops;
 		port->gs.magic = SCC_MAGIC;
 		port->gs.close_delay = HZ/2;
 		port->gs.closing_wait = 30 * HZ;
@@ -624,9 +629,9 @@ static void scc_enable_rx_interrupts(void *ptr)
 }
 
 
-static int scc_get_CD(void *ptr)
+static int scc_carrier_raised(struct tty_port *port)
 {
-	struct scc_port *port = ptr;
+	struct scc_port *scc = container_of(port, struct scc_port, gs.port);
 	unsigned channel = port->channel;
 
 	return !!(scc_last_status_reg[channel] & SR_DCD);
@@ -896,7 +901,7 @@ static int scc_open (struct tty_struct * tty, struct file * filp)
 		return retval;
 	}
 
-	port->c_dcd = scc_get_CD (port);
+	port->c_dcd = tty_port_carrier_raised(&port->gs.port);
 
 	scc_enable_rx_interrupts(port);
 
diff --git a/include/linux/generic_serial.h b/include/linux/generic_serial.h
index 4cc913939817..fadff28505bb 100644
--- a/include/linux/generic_serial.h
+++ b/include/linux/generic_serial.h
@@ -21,7 +21,6 @@ struct real_driver {
   void                    (*enable_tx_interrupts) (void *);
   void                    (*disable_rx_interrupts) (void *);
   void                    (*enable_rx_interrupts) (void *);
-  int                     (*get_CD) (void *);
   void                    (*shutdown_port) (void*);
   int                     (*set_real_termios) (void*);
   int                     (*chars_in_buffer) (void*);
diff --git a/include/linux/tty.h b/include/linux/tty.h
index bbbeaef99626..bc7bae78e22f 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -180,8 +180,16 @@ struct signal_struct;
  * until a hangup so don't use the wrong path.
  */
 
+struct tty_port;
+
+struct tty_port_operations {
+	/* Return 1 if the carrier is raised */
+	int (*carrier_raised)(struct tty_port *port);
+};
+	
 struct tty_port {
 	struct tty_struct	*tty;		/* Back pointer */
+	const struct tty_port_operations *ops;	/* Port operations */
 	spinlock_t		lock;		/* Lock protecting tty field */
 	int			blocked_open;	/* Waiting to open */
 	int			count;		/* Usage count */
@@ -427,6 +435,7 @@ extern int tty_port_alloc_xmit_buf(struct tty_port *port);
 extern void tty_port_free_xmit_buf(struct tty_port *port);
 extern struct tty_struct *tty_port_tty_get(struct tty_port *port);
 extern void tty_port_tty_set(struct tty_port *port, struct tty_struct *tty);
+extern int tty_port_carrier_raised(struct tty_port *port);
 
 extern int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc);
 extern int tty_unregister_ldisc(int disc);
-- 
cgit v1.2.3


From 5d951fb458f847e5485b5251597fbf326000bb3b Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Fri, 2 Jan 2009 13:45:19 +0000
Subject: tty: Pull the dtr raise into tty port

This moves another per device special out of what should be shared open
wait paths into private methods

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/isicom.c      | 13 ++++++++-----
 drivers/char/mxser.c       | 17 +++++++++++++----
 drivers/char/rocket.c      | 14 ++++++++++----
 drivers/char/synclink.c    | 21 +++++++++++++++------
 drivers/char/synclink_gt.c | 21 +++++++++++++++------
 drivers/char/tty_port.c    | 16 ++++++++++++++++
 include/linux/tty.h        |  2 ++
 7 files changed, 79 insertions(+), 25 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c
index b3da4858fd4a..a449449e301c 100644
--- a/drivers/char/isicom.c
+++ b/drivers/char/isicom.c
@@ -328,11 +328,13 @@ static inline void drop_rts(struct isi_port *port)
 }
 
 /* card->lock MUST NOT be held */
-static inline void raise_dtr_rts(struct isi_port *port)
+
+static void isicom_raise_dtr_rts(struct tty_port *port)
 {
-	struct isi_board *card = port->card;
+	struct isi_port *ip = container_of(port, struct isi_port, port);
+	struct isi_board *card = ip->card;
 	unsigned long base = card->base;
-	u16 channel = port->channel;
+	u16 channel = ip->channel;
 
 	if (!lock_card(card))
 		return;
@@ -340,7 +342,7 @@ static inline void raise_dtr_rts(struct isi_port *port)
 	outw(0x8000 | (channel << card->shift_count) | 0x02, base);
 	outw(0x0f04, base);
 	InterruptTheCard(base);
-	port->status |= (ISI_DTR | ISI_RTS);
+	ip->status |= (ISI_DTR | ISI_RTS);
 	unlock_card(card);
 }
 
@@ -881,7 +883,7 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	spin_unlock_irqrestore(&card->card_lock, flags);
 
 	while (1) {
-		raise_dtr_rts(ip);
+		tty_port_raise_dtr_rts(port);
 
 		set_current_state(TASK_INTERRUPTIBLE);
 		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)) {
@@ -1462,6 +1464,7 @@ static const struct tty_operations isicom_ops = {
 
 static const struct tty_port_operations isicom_port_ops = {
 	.carrier_raised		= isicom_carrier_raised,
+	.raise_dtr_rts		= isicom_raise_dtr_rts,
 };
 
 static int __devinit reset_card(struct pci_dev *pdev,
diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index eafbbcf355e7..ff5ff6188809 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -547,6 +547,17 @@ static int mxser_carrier_raised(struct tty_port *port)
 	return (inb(mp->ioaddr + UART_MSR) & UART_MSR_DCD)?1:0;
 }
 
+static void mxser_raise_dtr_rts(struct tty_port *port)
+{
+	struct mxser_port *mp = container_of(port, struct mxser_port, port);
+	unsigned long flags;
+
+	spin_lock_irqsave(&mp->slock, flags);
+	outb(inb(mp->ioaddr + UART_MCR) |
+		UART_MCR_DTR | UART_MCR_RTS, mp->ioaddr + UART_MCR);
+	spin_unlock_irqrestore(&mp->slock, flags);
+}
+
 static int mxser_block_til_ready(struct tty_struct *tty, struct file *filp,
 		struct mxser_port *mp)
 {
@@ -586,10 +597,7 @@ static int mxser_block_til_ready(struct tty_struct *tty, struct file *filp,
 	spin_unlock_irqrestore(&mp->slock, flags);
 	port->blocked_open++;
 	while (1) {
-		spin_lock_irqsave(&mp->slock, flags);
-		outb(inb(mp->ioaddr + UART_MCR) |
-			UART_MCR_DTR | UART_MCR_RTS, mp->ioaddr + UART_MCR);
-		spin_unlock_irqrestore(&mp->slock, flags);
+		tty_port_raise_dtr_rts(port);
 		set_current_state(TASK_INTERRUPTIBLE);
 		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)) {
 			if (port->flags & ASYNC_HUP_NOTIFY)
@@ -2458,6 +2466,7 @@ static const struct tty_operations mxser_ops = {
 
 struct tty_port_operations mxser_port_ops = {
 	.carrier_raised = mxser_carrier_raised,
+	.raise_dtr_rts = mxser_raise_dtr_rts,
 };
 
 /*
diff --git a/drivers/char/rocket.c b/drivers/char/rocket.c
index 4a4110e703a5..f4d9c3993488 100644
--- a/drivers/char/rocket.c
+++ b/drivers/char/rocket.c
@@ -870,6 +870,13 @@ static int carrier_raised(struct tty_port *port)
 	return (sGetChanStatusLo(&info->channel) & CD_ACT) ? 1 : 0;
 }
 
+static void raise_dtr_rts(struct tty_port *port)
+{
+	struct r_port *info = container_of(port, struct r_port, port);
+	sSetDTR(&info->channel);
+	sSetRTS(&info->channel);
+}
+
 /*  info->port.count is considered critical, protected by spinlocks.  */
 static int block_til_ready(struct tty_struct *tty, struct file *filp,
 			   struct r_port *info)
@@ -928,10 +935,8 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	spin_unlock_irqrestore(&info->slock, flags);
 
 	while (1) {
-		if (tty->termios->c_cflag & CBAUD) {
-			sSetDTR(&info->channel);
-			sSetRTS(&info->channel);
-		}
+		if (tty->termios->c_cflag & CBAUD)
+			tty_port_raise_dtr_rts(port);
 		set_current_state(TASK_INTERRUPTIBLE);
 		if (tty_hung_up_p(filp) || !(info->flags & ROCKET_INITIALIZED)) {
 			if (info->flags & ROCKET_HUP_NOTIFY)
@@ -2381,6 +2386,7 @@ static const struct tty_operations rocket_ops = {
 
 static const struct tty_port_operations rocket_port_ops = {
 	.carrier_raised = carrier_raised,
+	.raise_dtr_rts = raise_dtr_rts,
 };
 
 /*
diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c
index fb2e6b5e0ef1..ac9f21e18c3f 100644
--- a/drivers/char/synclink.c
+++ b/drivers/char/synclink.c
@@ -3298,6 +3298,18 @@ static int carrier_raised(struct tty_port *port)
 	return (info->serial_signals & SerialSignal_DCD) ? 1 : 0;
 }
 
+static void raise_dtr_rts(struct tty_port *port)
+{
+	struct mgsl_struct *info = container_of(port, struct mgsl_struct, port);
+	unsigned long flags;
+
+	spin_lock_irqsave(&info->irq_spinlock,flags);
+	info->serial_signals |= SerialSignal_RTS + SerialSignal_DTR;
+ 	usc_set_serial_signals(info);
+	spin_unlock_irqrestore(&info->irq_spinlock,flags);
+}
+
+
 /* block_til_ready()
  * 
  * 	Block the current process until the specified port
@@ -3358,12 +3370,8 @@ static int block_til_ready(struct tty_struct *tty, struct file * filp,
 	port->blocked_open++;
 	
 	while (1) {
-		if (tty->termios->c_cflag & CBAUD) {
-			spin_lock_irqsave(&info->irq_spinlock,flags);
-			info->serial_signals |= SerialSignal_RTS + SerialSignal_DTR;
-		 	usc_set_serial_signals(info);
-			spin_unlock_irqrestore(&info->irq_spinlock,flags);
-		}
+		if (tty->termios->c_cflag & CBAUD)
+			tty_port_raise_dtr_rts(port);
 		
 		set_current_state(TASK_INTERRUPTIBLE);
 		
@@ -4321,6 +4329,7 @@ static void mgsl_add_device( struct mgsl_struct *info )
 
 static const struct tty_port_operations mgsl_port_ops = {
 	.carrier_raised = carrier_raised,
+	.raise_dtr_rts = raise_dtr_rts,
 };
 
 
diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c
index 39ccaba8ca37..625c9bde3be8 100644
--- a/drivers/char/synclink_gt.c
+++ b/drivers/char/synclink_gt.c
@@ -3143,6 +3143,18 @@ static int carrier_raised(struct tty_port *port)
 	return (info->signals & SerialSignal_DCD) ? 1 : 0;
 }
 
+static void raise_dtr_rts(struct tty_port *port)
+{
+	unsigned long flags;
+	struct slgt_info *info = container_of(port, struct slgt_info, port);
+
+	spin_lock_irqsave(&info->lock,flags);
+	info->signals |= SerialSignal_RTS + SerialSignal_DTR;
+ 	set_signals(info);
+	spin_unlock_irqrestore(&info->lock,flags);
+}
+
+
 /*
  *  block current process until the device is ready to open
  */
@@ -3187,12 +3199,8 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	port->blocked_open++;
 
 	while (1) {
-		if ((tty->termios->c_cflag & CBAUD)) {
-			spin_lock_irqsave(&info->lock,flags);
-			info->signals |= SerialSignal_RTS + SerialSignal_DTR;
-		 	set_signals(info);
-			spin_unlock_irqrestore(&info->lock,flags);
-		}
+		if ((tty->termios->c_cflag & CBAUD))
+			tty_port_raise_dtr_rts(port);
 
 		set_current_state(TASK_INTERRUPTIBLE);
 
@@ -3455,6 +3463,7 @@ static void add_device(struct slgt_info *info)
 
 static const struct tty_port_operations slgt_port_ops = {
 	.carrier_raised = carrier_raised,
+	.raise_dtr_rts = raise_dtr_rts,
 };
 
 /*
diff --git a/drivers/char/tty_port.c b/drivers/char/tty_port.c
index f54e40cbf023..0557b6384747 100644
--- a/drivers/char/tty_port.c
+++ b/drivers/char/tty_port.c
@@ -111,3 +111,19 @@ int tty_port_carrier_raised(struct tty_port *port)
 	return port->ops->carrier_raised(port);
 }
 EXPORT_SYMBOL(tty_port_carrier_raised);
+
+/**
+ *	tty_port_raise_dtr_rts	-	Riase DTR/RTS
+ *	@port: tty port
+ *
+ *	Wrapper for the DTR/RTS raise logic. For the moment this is used
+ *	to hide some internal details. This will eventually become entirely
+ *	internal to the tty port.
+ */
+
+void tty_port_raise_dtr_rts(struct tty_port *port)
+{
+	if (port->ops->raise_dtr_rts)
+		port->ops->raise_dtr_rts(port);
+}
+EXPORT_SYMBOL(tty_port_raise_dtr_rts);
diff --git a/include/linux/tty.h b/include/linux/tty.h
index bc7bae78e22f..5001bbcacff6 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -185,6 +185,7 @@ struct tty_port;
 struct tty_port_operations {
 	/* Return 1 if the carrier is raised */
 	int (*carrier_raised)(struct tty_port *port);
+	void (*raise_dtr_rts)(struct tty_port *port);
 };
 	
 struct tty_port {
@@ -436,6 +437,7 @@ extern void tty_port_free_xmit_buf(struct tty_port *port);
 extern struct tty_struct *tty_port_tty_get(struct tty_port *port);
 extern void tty_port_tty_set(struct tty_port *port, struct tty_struct *tty);
 extern int tty_port_carrier_raised(struct tty_port *port);
+extern void tty_port_raise_dtr_rts(struct tty_port *port);
 
 extern int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc);
 extern int tty_unregister_ldisc(int disc);
-- 
cgit v1.2.3


From 3e61696bdc2103107674b06d0daf30b76193e922 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Fri, 2 Jan 2009 13:45:26 +0000
Subject: isicom: redo locking to use tty port locks

This helps set the basis for moving block_til_ready into common code. We also
introduce a tty_port_hangup helper as this will also be generally needed.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/isicom.c     | 35 +++++++++++++++--------------------
 drivers/char/synclinkmp.c | 20 ++++++++++++++------
 drivers/char/tty_port.c   | 24 ++++++++++++++++++++++++
 include/linux/tty.h       |  1 +
 4 files changed, 54 insertions(+), 26 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c
index a449449e301c..db53db91ae4a 100644
--- a/drivers/char/isicom.c
+++ b/drivers/char/isicom.c
@@ -841,7 +841,6 @@ static int isicom_carrier_raised(struct tty_port *port)
 static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	struct isi_port *ip)
 {
-	struct isi_board *card = ip->card;
 	struct tty_port *port = &ip->port;
 	int do_clocal = 0, retval;
 	unsigned long flags;
@@ -876,11 +875,11 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	retval = 0;
 	add_wait_queue(&port->open_wait, &wait);
 
-	spin_lock_irqsave(&card->card_lock, flags);
+	spin_lock_irqsave(&port->lock, flags);
 	if (!tty_hung_up_p(filp))
 		port->count--;
 	port->blocked_open++;
-	spin_unlock_irqrestore(&card->card_lock, flags);
+	spin_unlock_irqrestore(&port->lock, flags);
 
 	while (1) {
 		tty_port_raise_dtr_rts(port);
@@ -905,14 +904,13 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	}
 	set_current_state(TASK_RUNNING);
 	remove_wait_queue(&port->open_wait, &wait);
-	spin_lock_irqsave(&card->card_lock, flags);
+	spin_lock_irqsave(&port->lock, flags);
 	if (!tty_hung_up_p(filp))
 		port->count++;
 	port->blocked_open--;
-	spin_unlock_irqrestore(&card->card_lock, flags);
-	if (retval)
-		return retval;
-	port->flags |= ASYNC_NORMAL_ACTIVE;
+	if (retval == 0)
+		port->flags |= ASYNC_NORMAL_ACTIVE;
+	spin_unlock_irqrestore(&port->lock, flags);
 	return 0;
 }
 
@@ -1034,9 +1032,9 @@ static void isicom_close(struct tty_struct *tty, struct file *filp)
 
 	pr_dbg("Close start!!!.\n");
 
-	spin_lock_irqsave(&card->card_lock, flags);
+	spin_lock_irqsave(&port->port.lock, flags);
 	if (tty_hung_up_p(filp)) {
-		spin_unlock_irqrestore(&card->card_lock, flags);
+		spin_unlock_irqrestore(&port->port.lock, flags);
 		return;
 	}
 
@@ -1054,12 +1052,12 @@ static void isicom_close(struct tty_struct *tty, struct file *filp)
 	}
 
 	if (port->port.count) {
-		spin_unlock_irqrestore(&card->card_lock, flags);
+		spin_unlock_irqrestore(&port->port.lock, flags);
 		return;
 	}
 	port->port.flags |= ASYNC_CLOSING;
 	tty->closing = 1;
-	spin_unlock_irqrestore(&card->card_lock, flags);
+	spin_unlock_irqrestore(&port->port.lock, flags);
 
 	if (port->port.closing_wait != ASYNC_CLOSING_WAIT_NONE)
 		tty_wait_until_sent(tty, port->port.closing_wait);
@@ -1076,22 +1074,22 @@ static void isicom_close(struct tty_struct *tty, struct file *filp)
 	isicom_flush_buffer(tty);
 	tty_ldisc_flush(tty);
 
-	spin_lock_irqsave(&card->card_lock, flags);
+	spin_lock_irqsave(&port->port.lock, flags);
 	tty->closing = 0;
 
 	if (port->port.blocked_open) {
-		spin_unlock_irqrestore(&card->card_lock, flags);
+		spin_unlock_irqrestore(&port->port.lock, flags);
 		if (port->port.close_delay) {
 			pr_dbg("scheduling until time out.\n");
 			msleep_interruptible(
 				jiffies_to_msecs(port->port.close_delay));
 		}
-		spin_lock_irqsave(&card->card_lock, flags);
+		spin_lock_irqsave(&port->port.lock, flags);
 		wake_up_interruptible(&port->port.open_wait);
 	}
 	port->port.flags &= ~(ASYNC_NORMAL_ACTIVE | ASYNC_CLOSING);
 	wake_up_interruptible(&port->port.close_wait);
-	spin_unlock_irqrestore(&card->card_lock, flags);
+	spin_unlock_irqrestore(&port->port.lock, flags);
 }
 
 /* write et all */
@@ -1430,10 +1428,7 @@ static void isicom_hangup(struct tty_struct *tty)
 	isicom_shutdown_port(port);
 	spin_unlock_irqrestore(&port->card->card_lock, flags);
 
-	port->port.count = 0;
-	port->port.flags &= ~ASYNC_NORMAL_ACTIVE;
-	tty_port_tty_set(&port->port, NULL);
-	wake_up_interruptible(&port->port.open_wait);
+	tty_port_hangup(&port->port);
 }
 
 
diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c
index fcf1ec77450d..1f5c21ec4b14 100644
--- a/drivers/char/synclinkmp.c
+++ b/drivers/char/synclinkmp.c
@@ -3331,6 +3331,17 @@ static int carrier_raised(struct tty_port *port)
 	return (info->serial_signals & SerialSignal_DCD) ? 1 : 0;
 }
 
+static void raise_dtr_rts(struct tty_port *port)
+{
+	SLMP_INFO *info = container_of(port, SLMP_INFO, port);
+	unsigned long flags;
+
+	spin_lock_irqsave(&info->lock,flags);
+	info->serial_signals |= SerialSignal_RTS + SerialSignal_DTR;
+ 	set_signals(info);
+	spin_unlock_irqrestore(&info->lock,flags);
+}
+
 /* Block the current process until the specified port is ready to open.
  */
 static int block_til_ready(struct tty_struct *tty, struct file *filp,
@@ -3381,12 +3392,8 @@ static int block_til_ready(struct tty_struct *tty, struct file *filp,
 	port->blocked_open++;
 
 	while (1) {
-		if ((tty->termios->c_cflag & CBAUD)) {
-			spin_lock_irqsave(&info->lock,flags);
-			info->serial_signals |= SerialSignal_RTS + SerialSignal_DTR;
-		 	set_signals(info);
-			spin_unlock_irqrestore(&info->lock,flags);
-		}
+		if (tty->termios->c_cflag & CBAUD)
+			tty_port_raise_dtr_rts(port);
 
 		set_current_state(TASK_INTERRUPTIBLE);
 
@@ -3793,6 +3800,7 @@ static void add_device(SLMP_INFO *info)
 
 static const struct tty_port_operations port_ops = {
 	.carrier_raised = carrier_raised,
+	.raise_dtr_rts = raise_dtr_rts,
 };
 
 /* Allocate and initialize a device instance structure
diff --git a/drivers/char/tty_port.c b/drivers/char/tty_port.c
index 0557b6384747..9f418bca4a22 100644
--- a/drivers/char/tty_port.c
+++ b/drivers/char/tty_port.c
@@ -7,6 +7,7 @@
 #include <linux/tty.h>
 #include <linux/tty_driver.h>
 #include <linux/tty_flip.h>
+#include <linux/serial.h>
 #include <linux/timer.h>
 #include <linux/string.h>
 #include <linux/slab.h>
@@ -95,6 +96,29 @@ void tty_port_tty_set(struct tty_port *port, struct tty_struct *tty)
 }
 EXPORT_SYMBOL(tty_port_tty_set);
 
+/**
+ *	tty_port_hangup		-	hangup helper
+ *	@port: tty port
+ *
+ *	Perform port level tty hangup flag and count changes. Drop the tty
+ *	reference.
+ */
+
+void tty_port_hangup(struct tty_port *port)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&port->lock, flags);
+	port->count = 0;
+	port->flags &= ~ASYNC_NORMAL_ACTIVE;
+	if (port->tty)
+		tty_kref_put(port->tty);
+	port->tty = NULL;
+	spin_unlock_irqrestore(&port->lock, flags);
+	wake_up_interruptible(&port->open_wait);
+}
+EXPORT_SYMBOL(tty_port_hangup);
+
 /**
  *	tty_port_carrier_raised	-	carrier raised check
  *	@port: tty port
diff --git a/include/linux/tty.h b/include/linux/tty.h
index 5001bbcacff6..a1a93140e6e4 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -438,6 +438,7 @@ extern struct tty_struct *tty_port_tty_get(struct tty_port *port);
 extern void tty_port_tty_set(struct tty_port *port, struct tty_struct *tty);
 extern int tty_port_carrier_raised(struct tty_port *port);
 extern void tty_port_raise_dtr_rts(struct tty_port *port);
+extern void tty_port_hangup(struct tty_port *port);
 
 extern int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc);
 extern int tty_unregister_ldisc(int disc);
-- 
cgit v1.2.3


From 36c621d82b956ff6ff72273f848af53e6c581aba Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Fri, 2 Jan 2009 13:46:10 +0000
Subject: tty: Introduce a tty_port generic block_til_ready

Start sucking more commonality out of the drivers into a single piece of
core code.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/isicom.c   |  79 +-----------------------------------
 drivers/char/mxser.c    |  71 +-------------------------------
 drivers/char/riscom8.c  |  86 +--------------------------------------
 drivers/char/synclink.c |   1 +
 drivers/char/tty_port.c | 105 ++++++++++++++++++++++++++++++++++++++++++++++++
 drivers/char/vme_scc.c  |   6 +--
 include/linux/tty.h     |   2 +
 7 files changed, 115 insertions(+), 235 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c
index db53db91ae4a..bac55cf44243 100644
--- a/drivers/char/isicom.c
+++ b/drivers/char/isicom.c
@@ -838,82 +838,6 @@ static int isicom_carrier_raised(struct tty_port *port)
 	return (ip->status & ISI_DCD)?1 : 0;
 }
 
-static int block_til_ready(struct tty_struct *tty, struct file *filp,
-	struct isi_port *ip)
-{
-	struct tty_port *port = &ip->port;
-	int do_clocal = 0, retval;
-	unsigned long flags;
-	DECLARE_WAITQUEUE(wait, current);
-	int cd;
-
-	/* block if port is in the process of being closed */
-
-	if (tty_hung_up_p(filp) || port->flags & ASYNC_CLOSING) {
-		pr_dbg("block_til_ready: close in progress.\n");
-		interruptible_sleep_on(&port->close_wait);
-		if (port->flags & ASYNC_HUP_NOTIFY)
-			return -EAGAIN;
-		else
-			return -ERESTARTSYS;
-	}
-
-	/* if non-blocking mode is set ... */
-
-	if ((filp->f_flags & O_NONBLOCK) ||
-			(tty->flags & (1 << TTY_IO_ERROR))) {
-		pr_dbg("block_til_ready: non-block mode.\n");
-		port->flags |= ASYNC_NORMAL_ACTIVE;
-		return 0;
-	}
-
-	if (C_CLOCAL(tty))
-		do_clocal = 1;
-
-	/* block waiting for DCD to be asserted, and while
-						callout dev is busy */
-	retval = 0;
-	add_wait_queue(&port->open_wait, &wait);
-
-	spin_lock_irqsave(&port->lock, flags);
-	if (!tty_hung_up_p(filp))
-		port->count--;
-	port->blocked_open++;
-	spin_unlock_irqrestore(&port->lock, flags);
-
-	while (1) {
-		tty_port_raise_dtr_rts(port);
-
-		set_current_state(TASK_INTERRUPTIBLE);
-		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)) {
-			if (port->flags & ASYNC_HUP_NOTIFY)
-				retval = -EAGAIN;
-			else
-				retval = -ERESTARTSYS;
-			break;
-		}
-		cd = tty_port_carrier_raised(port);
-		if (!(port->flags & ASYNC_CLOSING) &&
-				(do_clocal || cd))
-			break;
-		if (signal_pending(current)) {
-			retval = -ERESTARTSYS;
-			break;
-		}
-		schedule();
-	}
-	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->open_wait, &wait);
-	spin_lock_irqsave(&port->lock, flags);
-	if (!tty_hung_up_p(filp))
-		port->count++;
-	port->blocked_open--;
-	if (retval == 0)
-		port->flags |= ASYNC_NORMAL_ACTIVE;
-	spin_unlock_irqrestore(&port->lock, flags);
-	return 0;
-}
-
 static int isicom_open(struct tty_struct *tty, struct file *filp)
 {
 	struct isi_port *port;
@@ -940,12 +864,13 @@ static int isicom_open(struct tty_struct *tty, struct file *filp)
 
 	isicom_setup_board(card);
 
+	/* FIXME: locking on port.count etc */
 	port->port.count++;
 	tty->driver_data = port;
 	tty_port_tty_set(&port->port, tty);
 	error = isicom_setup_port(tty);
 	if (error == 0)
-		error = block_til_ready(tty, filp, port);
+		error = tty_port_block_til_ready(&port->port, tty, filp);
 	return error;
 }
 
diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index e2471cf1ee95..08ba6eb1a380 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -558,75 +558,6 @@ static void mxser_raise_dtr_rts(struct tty_port *port)
 	spin_unlock_irqrestore(&mp->slock, flags);
 }
 
-static int mxser_block_til_ready(struct tty_struct *tty, struct file *filp,
-		struct mxser_port *mp)
-{
-	DECLARE_WAITQUEUE(wait, current);
-	int retval;
-	int do_clocal = 0;
-	unsigned long flags;
-	int cd;
-	struct tty_port *port = &mp->port;
-
-	/*
-	 * If non-blocking mode is set, or the port is not enabled,
-	 * then make the check up front and then exit.
-	 */
-	if ((filp->f_flags & O_NONBLOCK) ||
-			test_bit(TTY_IO_ERROR, &tty->flags)) {
-		port->flags |= ASYNC_NORMAL_ACTIVE;
-		return 0;
-	}
-
-	if (tty->termios->c_cflag & CLOCAL)
-		do_clocal = 1;
-
-	/*
-	 * Block waiting for the carrier detect and the line to become
-	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, port->count is dropped by one, so that
-	 * mxser_close() knows when to free things.  We restore it upon
-	 * exit, either normal or abnormal.
-	 */
-	retval = 0;
-	add_wait_queue(&port->open_wait, &wait);
-
-	spin_lock_irqsave(&port->lock, flags);
-	if (!tty_hung_up_p(filp))
-		port->count--;
-	port->blocked_open++;
-	spin_unlock_irqrestore(&port->lock, flags);
-	while (1) {
-		tty_port_raise_dtr_rts(port);
-		set_current_state(TASK_INTERRUPTIBLE);
-		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)) {
-			if (port->flags & ASYNC_HUP_NOTIFY)
-				retval = -EAGAIN;
-			else
-				retval = -ERESTARTSYS;
-			break;
-		}
-		cd = tty_port_carrier_raised(port);
-		if (!(port->flags & ASYNC_CLOSING) && (do_clocal || cd))
-			break;
-		if (signal_pending(current)) {
-			retval = -ERESTARTSYS;
-			break;
-		}
-		schedule();
-	}
-	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->open_wait, &wait);
-	spin_lock_irqsave(&port->lock, flags);
-	if (!tty_hung_up_p(filp))
-		port->count++;
-	port->blocked_open--;
-	if (retval == 0)
-		port->flags |= ASYNC_NORMAL_ACTIVE;
-	spin_unlock_irqrestore(&port->lock, flags);
-	return 0;
-}
-
 static int mxser_set_baud(struct tty_struct *tty, long newspd)
 {
 	struct mxser_port *info = tty->driver_data;
@@ -1110,7 +1041,7 @@ static int mxser_open(struct tty_struct *tty, struct file *filp)
 	if (retval)
 		return retval;
 
-	retval = mxser_block_til_ready(tty, filp, info);
+	retval = tty_port_block_til_ready(&info->port, tty, filp);
 	if (retval)
 		return retval;
 
diff --git a/drivers/char/riscom8.c b/drivers/char/riscom8.c
index 14662d7aa628..af34c2054a09 100644
--- a/drivers/char/riscom8.c
+++ b/drivers/char/riscom8.c
@@ -874,90 +874,6 @@ static int carrier_raised(struct tty_port *port)
 	return CD;
 }
 
-static int block_til_ready(struct tty_struct *tty, struct file *filp,
-			   struct riscom_port *rp)
-{
-	DECLARE_WAITQUEUE(wait, current);
-	int    retval;
-	int    do_clocal = 0;
-	int    CD;
-	unsigned long flags;
-	struct tty_port *port = &rp->port;
-
-	/*
-	 * If the device is in the middle of being closed, then block
-	 * until it's done, and then try again.
-	 */
-	if (tty_hung_up_p(filp) || port->flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&port->close_wait);
-		if (port->flags & ASYNC_HUP_NOTIFY)
-			return -EAGAIN;
-		else
-			return -ERESTARTSYS;
-	}
-
-	/*
-	 * If non-blocking mode is set, or the port is not enabled,
-	 * then make the check up front and then exit.
-	 */
-	if ((filp->f_flags & O_NONBLOCK) ||
-	    (tty->flags & (1 << TTY_IO_ERROR))) {
-		port->flags |= ASYNC_NORMAL_ACTIVE;
-		return 0;
-	}
-
-	if (C_CLOCAL(tty))
-		do_clocal = 1;
-
-	/*
-	 * Block waiting for the carrier detect and the line to become
-	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->count is dropped by one, so that
-	 * rs_close() knows when to free things.  We restore it upon
-	 * exit, either normal or abnormal.
-	 */
-	retval = 0;
-	add_wait_queue(&port->open_wait, &wait);
-
-	spin_lock_irqsave(&port->lock, flags);
-	if (!tty_hung_up_p(filp))
-		port->count--;
-	port->blocked_open++;
-	spin_unlock_irqrestore(&port->lock, flags);
-
-	while (1) {
-
-		CD = tty_port_carrier_raised(port);
-		set_current_state(TASK_INTERRUPTIBLE);
-		if (tty_hung_up_p(filp) ||
-		    !(port->flags & ASYNC_INITIALIZED)) {
-			if (port->flags & ASYNC_HUP_NOTIFY)
-				retval = -EAGAIN;
-			else
-				retval = -ERESTARTSYS;
-			break;
-		}
-		if (!(port->flags & ASYNC_CLOSING) &&
-		    (do_clocal || CD))
-			break;
-		if (signal_pending(current)) {
-			retval = -ERESTARTSYS;
-			break;
-		}
-		schedule();
-	}
-	__set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->open_wait, &wait);
-	spin_lock_irqsave(&port->lock, flags);
-	if (!tty_hung_up_p(filp))
-		port->count++;
-	port->blocked_open--;
-	if (retval == 0)
-		port->flags |= ASYNC_NORMAL_ACTIVE;
-	spin_unlock_irqrestore(&port->lock, flags);
-	return 0;
-}
-
 static int rc_open(struct tty_struct *tty, struct file *filp)
 {
 	int board;
@@ -984,7 +900,7 @@ static int rc_open(struct tty_struct *tty, struct file *filp)
 
 	error = rc_setup_port(bp, port);
 	if (error == 0)
-		error = block_til_ready(tty, filp, port);
+		error = tty_port_block_til_ready(&port->port, tty, filp);
 	return error;
 }
 
diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c
index ac9f21e18c3f..0ded4ed3da3c 100644
--- a/drivers/char/synclink.c
+++ b/drivers/char/synclink.c
@@ -3401,6 +3401,7 @@ static int block_til_ready(struct tty_struct *tty, struct file * filp,
 	set_current_state(TASK_RUNNING);
 	remove_wait_queue(&port->open_wait, &wait);
 	
+	/* FIXME: Racy on hangup during close wait */
 	if (extra_count)
 		port->count++;
 	port->blocked_open--;
diff --git a/drivers/char/tty_port.c b/drivers/char/tty_port.c
index 9f418bca4a22..ff94182b3813 100644
--- a/drivers/char/tty_port.c
+++ b/drivers/char/tty_port.c
@@ -151,3 +151,108 @@ void tty_port_raise_dtr_rts(struct tty_port *port)
 		port->ops->raise_dtr_rts(port);
 }
 EXPORT_SYMBOL(tty_port_raise_dtr_rts);
+
+/**
+ *	tty_port_block_til_ready	-	Waiting logic for tty open
+ *	@port: the tty port being opened
+ *	@tty: the tty device being bound
+ *	@filp: the file pointer of the opener
+ *
+ *	Implement the core POSIX/SuS tty behaviour when opening a tty device.
+ *	Handles:
+ *		- hangup (both before and during)
+ *		- non blocking open
+ *		- rts/dtr/dcd
+ *		- signals
+ *		- port flags and counts
+ *
+ *	The passed tty_port must implement the carrier_raised method if it can
+ *	do carrier detect and the raise_dtr_rts method if it supports software
+ *	management of these lines. Note that the dtr/rts raise is done each
+ *	iteration as a hangup may have previously dropped them while we wait.
+ */
+ 
+int tty_port_block_til_ready(struct tty_port *port,
+				struct tty_struct *tty, struct file *filp)
+{
+	int do_clocal = 0, retval;
+	unsigned long flags;
+	DECLARE_WAITQUEUE(wait, current);
+	int cd;
+
+	/* block if port is in the process of being closed */
+	if (tty_hung_up_p(filp) || port->flags & ASYNC_CLOSING) {
+		interruptible_sleep_on(&port->close_wait);
+		if (port->flags & ASYNC_HUP_NOTIFY)
+			return -EAGAIN;
+		else
+			return -ERESTARTSYS;
+	}
+
+	/* if non-blocking mode is set we can pass directly to open unless
+	   the port has just hung up or is in another error state */
+	if ((filp->f_flags & O_NONBLOCK) ||
+			(tty->flags & (1 << TTY_IO_ERROR))) {
+		port->flags |= ASYNC_NORMAL_ACTIVE;
+		return 0;
+	}
+
+	if (C_CLOCAL(tty))
+		do_clocal = 1;
+
+	/* Block waiting until we can proceed. We may need to wait for the
+	   carrier, but we must also wait for any close that is in progress
+	   before the next open may complete */
+
+	retval = 0;
+	add_wait_queue(&port->open_wait, &wait);
+
+	/* The port lock protects the port counts */
+	spin_lock_irqsave(&port->lock, flags);
+	if (!tty_hung_up_p(filp))
+		port->count--;
+	port->blocked_open++;
+	spin_unlock_irqrestore(&port->lock, flags);
+
+	while (1) {
+		/* Indicate we are open */
+		tty_port_raise_dtr_rts(port);
+
+		set_current_state(TASK_INTERRUPTIBLE);
+		/* Check for a hangup or uninitialised port. Return accordingly */
+		if (tty_hung_up_p(filp) || !(port->flags & ASYNC_INITIALIZED)) {
+			if (port->flags & ASYNC_HUP_NOTIFY)
+				retval = -EAGAIN;
+			else
+				retval = -ERESTARTSYS;
+			break;
+		}
+		/* Probe the carrier. For devices with no carrier detect this
+		   will always return true */
+		cd = tty_port_carrier_raised(port);
+		if (!(port->flags & ASYNC_CLOSING) &&
+				(do_clocal || cd))
+			break;
+		if (signal_pending(current)) {
+			retval = -ERESTARTSYS;
+			break;
+		}
+		schedule();
+	}
+	set_current_state(TASK_RUNNING);
+	remove_wait_queue(&port->open_wait, &wait);
+
+	/* Update counts. A parallel hangup will have set count to zero and
+	   we must not mess that up further */
+	spin_lock_irqsave(&port->lock, flags);
+	if (!tty_hung_up_p(filp))
+		port->count++;
+	port->blocked_open--;
+	if (retval == 0)
+		port->flags |= ASYNC_NORMAL_ACTIVE;
+	spin_unlock_irqrestore(&port->lock, flags);
+	return 0;
+	
+}
+EXPORT_SYMBOL(tty_port_block_til_ready);
+
diff --git a/drivers/char/vme_scc.c b/drivers/char/vme_scc.c
index d4e1534c0e03..2d9242a45a0d 100644
--- a/drivers/char/vme_scc.c
+++ b/drivers/char/vme_scc.c
@@ -631,8 +631,8 @@ static void scc_enable_rx_interrupts(void *ptr)
 
 static int scc_carrier_raised(struct tty_port *port)
 {
-	struct scc_port *scc = container_of(port, struct scc_port, gs.port);
-	unsigned channel = port->channel;
+	struct scc_port *sc = container_of(port, struct scc_port, gs.port);
+	unsigned channel = sc->channel;
 
 	return !!(scc_last_status_reg[channel] & SR_DCD);
 }
@@ -643,7 +643,7 @@ static void scc_shutdown_port(void *ptr)
 	struct scc_port *port = ptr;
 
 	port->gs.port.flags &= ~ GS_ACTIVE;
-	if (port->gs.port.tty && port->gs.port.tty->termios->c_cflag & HUPCL) {
+	if (port->gs.port.tty && (port->gs.port.tty->termios->c_cflag & HUPCL)) {
 		scc_setsignals (port, 0, 0);
 	}
 }
diff --git a/include/linux/tty.h b/include/linux/tty.h
index a1a93140e6e4..61a0ab32cf11 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -439,6 +439,8 @@ extern void tty_port_tty_set(struct tty_port *port, struct tty_struct *tty);
 extern int tty_port_carrier_raised(struct tty_port *port);
 extern void tty_port_raise_dtr_rts(struct tty_port *port);
 extern void tty_port_hangup(struct tty_port *port);
+extern int tty_port_block_til_ready(struct tty_port *port,
+				struct tty_struct *tty, struct file *filp);
 
 extern int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc);
 extern int tty_unregister_ldisc(int disc);
-- 
cgit v1.2.3


From 2a6eadbd5a2ae8f458e421f3614f1ad13c0f9a1c Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Fri, 2 Jan 2009 13:46:18 +0000
Subject: tty: Rework istallion to use the tty port changes

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/istallion.c  | 159 +++++++++++++---------------------------------
 include/linux/istallion.h |   1 -
 2 files changed, 43 insertions(+), 117 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c
index c4682f9e34bb..4c69ab97339a 100644
--- a/drivers/char/istallion.c
+++ b/drivers/char/istallion.c
@@ -626,8 +626,6 @@ static int	stli_hostcmd(struct stlibrd *brdp, struct stliport *portp);
 static int	stli_initopen(struct tty_struct *tty, struct stlibrd *brdp, struct stliport *portp);
 static int	stli_rawopen(struct stlibrd *brdp, struct stliport *portp, unsigned long arg, int wait);
 static int	stli_rawclose(struct stlibrd *brdp, struct stliport *portp, unsigned long arg, int wait);
-static int	stli_waitcarrier(struct tty_struct *tty, struct stlibrd *brdp,
-				struct stliport *portp, struct file *filp);
 static int	stli_setport(struct tty_struct *tty);
 static int	stli_cmdwait(struct stlibrd *brdp, struct stliport *portp, unsigned long cmd, void *arg, int size, int copyback);
 static void	stli_sendcmd(struct stlibrd *brdp, struct stliport *portp, unsigned long cmd, void *arg, int size, int copyback);
@@ -787,6 +785,7 @@ static int stli_open(struct tty_struct *tty, struct file *filp)
 {
 	struct stlibrd *brdp;
 	struct stliport *portp;
+	struct tty_port *port;
 	unsigned int minordev, brdnr, portnr;
 	int rc;
 
@@ -808,30 +807,19 @@ static int stli_open(struct tty_struct *tty, struct file *filp)
 		return -ENODEV;
 	if (portp->devnr < 1)
 		return -ENODEV;
-
-
-/*
- *	Check if this port is in the middle of closing. If so then wait
- *	until it is closed then return error status based on flag settings.
- *	The sleep here does not need interrupt protection since the wakeup
- *	for it is done with the same context.
- */
-	if (portp->port.flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&portp->port.close_wait);
-		if (portp->port.flags & ASYNC_HUP_NOTIFY)
-			return -EAGAIN;
-		return -ERESTARTSYS;
-	}
+	port = &portp->port;
 
 /*
  *	On the first open of the device setup the port hardware, and
  *	initialize the per port data structure. Since initializing the port
  *	requires several commands to the board we will need to wait for any
  *	other open that is already initializing the port.
+ *
+ *	Review - locking
  */
-	tty_port_tty_set(&portp->port, tty);
+	tty_port_tty_set(port, tty);
 	tty->driver_data = portp;
-	portp->port.count++;
+	port->count++;
 
 	wait_event_interruptible(portp->raw_wait,
 			!test_bit(ST_INITIALIZING, &portp->state));
@@ -841,7 +829,8 @@ static int stli_open(struct tty_struct *tty, struct file *filp)
 	if ((portp->port.flags & ASYNC_INITIALIZED) == 0) {
 		set_bit(ST_INITIALIZING, &portp->state);
 		if ((rc = stli_initopen(tty, brdp, portp)) >= 0) {
-			portp->port.flags |= ASYNC_INITIALIZED;
+			/* Locking */
+			port->flags |= ASYNC_INITIALIZED;
 			clear_bit(TTY_IO_ERROR, &tty->flags);
 		}
 		clear_bit(ST_INITIALIZING, &portp->state);
@@ -849,31 +838,7 @@ static int stli_open(struct tty_struct *tty, struct file *filp)
 		if (rc < 0)
 			return rc;
 	}
-
-/*
- *	Check if this port is in the middle of closing. If so then wait
- *	until it is closed then return error status, based on flag settings.
- *	The sleep here does not need interrupt protection since the wakeup
- *	for it is done with the same context.
- */
-	if (portp->port.flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&portp->port.close_wait);
-		if (portp->port.flags & ASYNC_HUP_NOTIFY)
-			return -EAGAIN;
-		return -ERESTARTSYS;
-	}
-
-/*
- *	Based on type of open being done check if it can overlap with any
- *	previous opens still in effect. If we are a normal serial device
- *	then also we might have to wait for carrier.
- */
-	if (!(filp->f_flags & O_NONBLOCK)) {
-		if ((rc = stli_waitcarrier(tty, brdp, portp, filp)) != 0)
-			return rc;
-	}
-	portp->port.flags |= ASYNC_NORMAL_ACTIVE;
-	return 0;
+	return tty_port_block_til_ready(&portp->port, tty, filp);
 }
 
 /*****************************************************************************/
@@ -882,25 +847,29 @@ static void stli_close(struct tty_struct *tty, struct file *filp)
 {
 	struct stlibrd *brdp;
 	struct stliport *portp;
+	struct tty_port *port;
 	unsigned long flags;
 
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
+	port = &portp->port;
 
-	spin_lock_irqsave(&stli_lock, flags);
+	spin_lock_irqsave(&port->lock, flags);
 	if (tty_hung_up_p(filp)) {
-		spin_unlock_irqrestore(&stli_lock, flags);
+		spin_unlock_irqrestore(&port->lock, flags);
 		return;
 	}
-	if ((tty->count == 1) && (portp->port.count != 1))
-		portp->port.count = 1;
-	if (portp->port.count-- > 1) {
-		spin_unlock_irqrestore(&stli_lock, flags);
+	if (tty->count == 1 && port->count != 1)
+		port->count = 1;
+	if (port->count-- > 1) {
+		spin_unlock_irqrestore(&port->lock, flags);
 		return;
 	}
 
-	portp->port.flags |= ASYNC_CLOSING;
+	port->flags |= ASYNC_CLOSING;
+	tty->closing = 1;
+	spin_unlock_irqrestore(&port->lock, flags);
 
 /*
  *	May want to wait for data to drain before closing. The BUSY flag
@@ -908,15 +877,17 @@ static void stli_close(struct tty_struct *tty, struct file *filp)
  *	updated by messages from the slave - indicating when all chars
  *	really have drained.
  */
+ 	spin_lock_irqsave(&stli_lock, flags);
 	if (tty == stli_txcooktty)
 		stli_flushchars(tty);
-	tty->closing = 1;
 	spin_unlock_irqrestore(&stli_lock, flags);
 
 	if (portp->closing_wait != ASYNC_CLOSING_WAIT_NONE)
 		tty_wait_until_sent(tty, portp->closing_wait);
 
-	portp->port.flags &= ~ASYNC_INITIALIZED;
+	/* FIXME: port locking here needs attending to */
+	port->flags &= ~ASYNC_INITIALIZED;
+
 	brdp = stli_brds[portp->brdnr];
 	stli_rawclose(brdp, portp, 0, 0);
 	if (tty->termios->c_cflag & HUPCL) {
@@ -937,14 +908,14 @@ static void stli_close(struct tty_struct *tty, struct file *filp)
 	tty->closing = 0;
 	tty_port_tty_set(&portp->port, NULL);
 
-	if (portp->openwaitcnt) {
+	if (port->blocked_open) {
 		if (portp->close_delay)
 			msleep_interruptible(jiffies_to_msecs(portp->close_delay));
-		wake_up_interruptible(&portp->port.open_wait);
+		wake_up_interruptible(&port->open_wait);
 	}
 
-	portp->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-	wake_up_interruptible(&portp->port.close_wait);
+	port->flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
+	wake_up_interruptible(&port->close_wait);
 }
 
 /*****************************************************************************/
@@ -1189,63 +1160,17 @@ static int stli_carrier_raised(struct tty_port *port)
 	return (portp->sigs & TIOCM_CD) ? 1 : 0;
 }
 
-/*
- *	Possibly need to wait for carrier (DCD signal) to come high. Say
- *	maybe because if we are clocal then we don't need to wait...
- */
-
-static int stli_waitcarrier(struct tty_struct *tty, struct stlibrd *brdp,
-				struct stliport *portp, struct file *filp)
+static void stli_raise_dtr_rts(struct tty_port *port)
 {
-	unsigned long flags;
-	int rc, doclocal;
-	struct tty_port *port = &portp->port;
-
-	rc = 0;
-	doclocal = 0;
-
-	if (tty->termios->c_cflag & CLOCAL)
-		doclocal++;
-
-	spin_lock_irqsave(&stli_lock, flags);
-	portp->openwaitcnt++;
-	if (! tty_hung_up_p(filp))
-		port->count--;
-	spin_unlock_irqrestore(&stli_lock, flags);
-
-	for (;;) {
-		stli_mkasysigs(&portp->asig, 1, 1);
-		if ((rc = stli_cmdwait(brdp, portp, A_SETSIGNALS,
-		    &portp->asig, sizeof(asysigs_t), 0)) < 0)
-			break;
-		if (tty_hung_up_p(filp) ||
-		    ((port->flags & ASYNC_INITIALIZED) == 0)) {
-			if (port->flags & ASYNC_HUP_NOTIFY)
-				rc = -EBUSY;
-			else
-				rc = -ERESTARTSYS;
-			break;
-		}
-		if (((port->flags & ASYNC_CLOSING) == 0) &&
-		    (doclocal || tty_port_carrier_raised(port))) {
-			break;
-		}
-		if (signal_pending(current)) {
-			rc = -ERESTARTSYS;
-			break;
-		}
-		interruptible_sleep_on(&port->open_wait);
-	}
-
-	spin_lock_irqsave(&stli_lock, flags);
-	if (! tty_hung_up_p(filp))
-		port->count++;
-	portp->openwaitcnt--;
-	spin_unlock_irqrestore(&stli_lock, flags);
-
-	return rc;
+	struct stliport *portp = container_of(port, struct stliport, port);
+	struct stlibrd *brdp = stli_brds[portp->brdnr];
+	stli_mkasysigs(&portp->asig, 1, 1);
+	if (stli_cmdwait(brdp, portp, A_SETSIGNALS, &portp->asig,
+		sizeof(asysigs_t), 0) < 0)
+			printk(KERN_WARNING "istallion: dtr raise failed.\n");
 }
 
+
 /*****************************************************************************/
 
 /*
@@ -1828,6 +1753,7 @@ static void stli_hangup(struct tty_struct *tty)
 {
 	struct stliport *portp;
 	struct stlibrd *brdp;
+	struct tty_port *port;
 	unsigned long flags;
 
 	portp = tty->driver_data;
@@ -1838,8 +1764,11 @@ static void stli_hangup(struct tty_struct *tty)
 	brdp = stli_brds[portp->brdnr];
 	if (brdp == NULL)
 		return;
+	port = &portp->port;
 
-	portp->port.flags &= ~ASYNC_INITIALIZED;
+	spin_lock_irqsave(&port->lock, flags);
+	port->flags &= ~ASYNC_INITIALIZED;
+	spin_unlock_irqrestore(&port->lock, flags);
 
 	if (!test_bit(ST_CLOSING, &portp->state))
 		stli_rawclose(brdp, portp, 0, 0);
@@ -1860,12 +1789,9 @@ static void stli_hangup(struct tty_struct *tty)
 	clear_bit(ST_TXBUSY, &portp->state);
 	clear_bit(ST_RXSTOP, &portp->state);
 	set_bit(TTY_IO_ERROR, &tty->flags);
-	tty_port_tty_set(&portp->port, NULL);
-	portp->port.flags &= ~ASYNC_NORMAL_ACTIVE;
-	portp->port.count = 0;
 	spin_unlock_irqrestore(&stli_lock, flags);
 
-	wake_up_interruptible(&portp->port.open_wait);
+	tty_port_hangup(port);
 }
 
 /*****************************************************************************/
@@ -4528,6 +4454,7 @@ static const struct tty_operations stli_ops = {
 
 static const struct tty_port_operations stli_port_ops = {
 	.carrier_raised = stli_carrier_raised,
+	.raise_dtr_rts = stli_raise_dtr_rts,
 };
 
 /*****************************************************************************/
diff --git a/include/linux/istallion.h b/include/linux/istallion.h
index 0d1840723249..053d5aea925c 100644
--- a/include/linux/istallion.h
+++ b/include/linux/istallion.h
@@ -61,7 +61,6 @@ struct stliport {
 	int			custom_divisor;
 	int			close_delay;
 	int			closing_wait;
-	int			openwaitcnt;
 	int			rc;
 	int			argsize;
 	void			*argp;
-- 
cgit v1.2.3


From a6614999e800cf3a134ce93ea46ef837e3c0e76e Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Fri, 2 Jan 2009 13:46:50 +0000
Subject: tty: Introduce some close helpers for ports

Again this is a lot of common code we can unify

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/isicom.c      | 66 ++++++---------------------------------
 drivers/char/istallion.c   | 78 +++++++++++++++++-----------------------------
 drivers/char/mxser.c       | 56 ++++++---------------------------
 drivers/char/riscom8.c     | 49 +++--------------------------
 drivers/char/stallion.c    | 39 +++--------------------
 drivers/char/synclink.c    | 58 ++--------------------------------
 drivers/char/synclink_gt.c | 51 ++----------------------------
 drivers/char/synclinkmp.c  | 58 ++--------------------------------
 drivers/char/tty_port.c    | 58 ++++++++++++++++++++++++++++++++++
 include/linux/istallion.h  |  1 -
 include/linux/tty.h        |  3 ++
 11 files changed, 126 insertions(+), 391 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c
index bac55cf44243..24aa6e88e223 100644
--- a/drivers/char/isicom.c
+++ b/drivers/char/isicom.c
@@ -945,76 +945,30 @@ static void isicom_flush_buffer(struct tty_struct *tty)
 
 static void isicom_close(struct tty_struct *tty, struct file *filp)
 {
-	struct isi_port *port = tty->driver_data;
+	struct isi_port *ip = tty->driver_data;
+	struct tty_port *port = &ip->port;
 	struct isi_board *card;
 	unsigned long flags;
 
-	if (!port)
-		return;
-	card = port->card;
-	if (isicom_paranoia_check(port, tty->name, "isicom_close"))
-		return;
-
-	pr_dbg("Close start!!!.\n");
-
-	spin_lock_irqsave(&port->port.lock, flags);
-	if (tty_hung_up_p(filp)) {
-		spin_unlock_irqrestore(&port->port.lock, flags);
-		return;
-	}
-
-	if (tty->count == 1 && port->port.count != 1) {
-		printk(KERN_WARNING "ISICOM:(0x%lx) isicom_close: bad port "
-			"count tty->count = 1 port count = %d.\n",
-			card->base, port->port.count);
-		port->port.count = 1;
-	}
-	if (--port->port.count < 0) {
-		printk(KERN_WARNING "ISICOM:(0x%lx) isicom_close: bad port "
-			"count for channel%d = %d", card->base, port->channel,
-			port->port.count);
-		port->port.count = 0;
-	}
+	BUG_ON(!ip);
 
-	if (port->port.count) {
-		spin_unlock_irqrestore(&port->port.lock, flags);
+	card = ip->card;
+	if (isicom_paranoia_check(ip, tty->name, "isicom_close"))
 		return;
-	}
-	port->port.flags |= ASYNC_CLOSING;
-	tty->closing = 1;
-	spin_unlock_irqrestore(&port->port.lock, flags);
 
-	if (port->port.closing_wait != ASYNC_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, port->port.closing_wait);
 	/* indicate to the card that no more data can be received
 	   on this port */
 	spin_lock_irqsave(&card->card_lock, flags);
-	if (port->port.flags & ASYNC_INITIALIZED) {
-		card->port_status &= ~(1 << port->channel);
+	if (port->flags & ASYNC_INITIALIZED) {
+		card->port_status &= ~(1 << ip->channel);
 		outw(card->port_status, card->base + 0x02);
 	}
-	isicom_shutdown_port(port);
+	isicom_shutdown_port(ip);
 	spin_unlock_irqrestore(&card->card_lock, flags);
 
 	isicom_flush_buffer(tty);
-	tty_ldisc_flush(tty);
-
-	spin_lock_irqsave(&port->port.lock, flags);
-	tty->closing = 0;
-
-	if (port->port.blocked_open) {
-		spin_unlock_irqrestore(&port->port.lock, flags);
-		if (port->port.close_delay) {
-			pr_dbg("scheduling until time out.\n");
-			msleep_interruptible(
-				jiffies_to_msecs(port->port.close_delay));
-		}
-		spin_lock_irqsave(&port->port.lock, flags);
-		wake_up_interruptible(&port->port.open_wait);
-	}
-	port->port.flags &= ~(ASYNC_NORMAL_ACTIVE | ASYNC_CLOSING);
-	wake_up_interruptible(&port->port.close_wait);
-	spin_unlock_irqrestore(&port->port.lock, flags);
+	
+	tty_port_close_end(port, tty);
 }
 
 /* write et all */
diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c
index 4c69ab97339a..5c3dc6b8411c 100644
--- a/drivers/char/istallion.c
+++ b/drivers/char/istallion.c
@@ -767,7 +767,7 @@ static int stli_parsebrd(struct stlconf *confp, char **argp)
 			break;
 	}
 	if (i == ARRAY_SIZE(stli_brdstr)) {
-		printk("STALLION: unknown board name, %s?\n", argp[0]);
+		printk(KERN_WARNING "istallion: unknown board name, %s?\n", argp[0]);
 		return 0;
 	}
 
@@ -855,21 +855,8 @@ static void stli_close(struct tty_struct *tty, struct file *filp)
 		return;
 	port = &portp->port;
 
-	spin_lock_irqsave(&port->lock, flags);
-	if (tty_hung_up_p(filp)) {
-		spin_unlock_irqrestore(&port->lock, flags);
-		return;
-	}
-	if (tty->count == 1 && port->count != 1)
-		port->count = 1;
-	if (port->count-- > 1) {
-		spin_unlock_irqrestore(&port->lock, flags);
+	if (tty_port_close_start(port, tty, filp) == 0)
 		return;
-	}
-
-	port->flags |= ASYNC_CLOSING;
-	tty->closing = 1;
-	spin_unlock_irqrestore(&port->lock, flags);
 
 /*
  *	May want to wait for data to drain before closing. The BUSY flag
@@ -882,6 +869,8 @@ static void stli_close(struct tty_struct *tty, struct file *filp)
 		stli_flushchars(tty);
 	spin_unlock_irqrestore(&stli_lock, flags);
 
+	/* We end up doing this twice for the moment. This needs looking at
+	   eventually. Note we still use portp->closing_wait as a result */
 	if (portp->closing_wait != ASYNC_CLOSING_WAIT_NONE)
 		tty_wait_until_sent(tty, portp->closing_wait);
 
@@ -905,17 +894,8 @@ static void stli_close(struct tty_struct *tty, struct file *filp)
 	set_bit(ST_DOFLUSHRX, &portp->state);
 	stli_flushbuffer(tty);
 
-	tty->closing = 0;
-	tty_port_tty_set(&portp->port, NULL);
-
-	if (port->blocked_open) {
-		if (portp->close_delay)
-			msleep_interruptible(jiffies_to_msecs(portp->close_delay));
-		wake_up_interruptible(&port->open_wait);
-	}
-
-	port->flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-	wake_up_interruptible(&port->close_wait);
+	tty_port_close_end(port, tty);
+	tty_port_tty_set(port, NULL);
 }
 
 /*****************************************************************************/
@@ -1482,7 +1462,7 @@ static int stli_getserial(struct stliport *portp, struct serial_struct __user *s
 	sio.irq = 0;
 	sio.flags = portp->port.flags;
 	sio.baud_base = portp->baud_base;
-	sio.close_delay = portp->close_delay;
+	sio.close_delay = portp->port.close_delay;
 	sio.closing_wait = portp->closing_wait;
 	sio.custom_divisor = portp->custom_divisor;
 	sio.xmit_fifo_size = 0;
@@ -1514,7 +1494,7 @@ static int stli_setserial(struct tty_struct *tty, struct serial_struct __user *s
 		return -EFAULT;
 	if (!capable(CAP_SYS_ADMIN)) {
 		if ((sio.baud_base != portp->baud_base) ||
-		    (sio.close_delay != portp->close_delay) ||
+		    (sio.close_delay != portp->port.close_delay) ||
 		    ((sio.flags & ~ASYNC_USR_MASK) !=
 		    (portp->port.flags & ~ASYNC_USR_MASK)))
 			return -EPERM;
@@ -1523,7 +1503,7 @@ static int stli_setserial(struct tty_struct *tty, struct serial_struct __user *s
 	portp->port.flags = (portp->port.flags & ~ASYNC_USR_MASK) |
 		(sio.flags & ASYNC_USR_MASK);
 	portp->baud_base = sio.baud_base;
-	portp->close_delay = sio.close_delay;
+	portp->port.close_delay = sio.close_delay;
 	portp->closing_wait = sio.closing_wait;
 	portp->custom_divisor = sio.custom_divisor;
 
@@ -2065,7 +2045,7 @@ static void __stli_sendcmd(struct stlibrd *brdp, struct stliport *portp, unsigne
 	unsigned char __iomem *bits;
 
 	if (test_bit(ST_CMDING, &portp->state)) {
-		printk(KERN_ERR "STALLION: command already busy, cmd=%x!\n",
+		printk(KERN_ERR "istallion: command already busy, cmd=%x!\n",
 				(int) cmd);
 		return;
 	}
@@ -2625,7 +2605,7 @@ static int stli_initports(struct stlibrd *brdp)
 	for (i = 0, panelnr = 0, panelport = 0; (i < brdp->nrports); i++) {
 		portp = kzalloc(sizeof(struct stliport), GFP_KERNEL);
 		if (!portp) {
-			printk("STALLION: failed to allocate port structure\n");
+			printk(KERN_WARNING "istallion: failed to allocate port structure\n");
 			continue;
 		}
 		tty_port_init(&portp->port);
@@ -2635,7 +2615,7 @@ static int stli_initports(struct stlibrd *brdp)
 		portp->brdnr = brdp->brdnr;
 		portp->panelnr = panelnr;
 		portp->baud_base = STL_BAUDBASE;
-		portp->close_delay = STL_CLOSEDELAY;
+		portp->port.close_delay = STL_CLOSEDELAY;
 		portp->closing_wait = 30 * HZ;
 		init_waitqueue_head(&portp->port.open_wait);
 		init_waitqueue_head(&portp->port.close_wait);
@@ -2692,7 +2672,7 @@ static void __iomem *stli_ecpgetmemptr(struct stlibrd *brdp, unsigned long offse
 	unsigned char val;
 
 	if (offset > brdp->memsize) {
-		printk(KERN_ERR "STALLION: shared memory pointer=%x out of "
+		printk(KERN_ERR "istallion: shared memory pointer=%x out of "
 				"range at line=%d(%d), brd=%d\n",
 			(int) offset, line, __LINE__, brdp->brdnr);
 		ptr = NULL;
@@ -2766,7 +2746,7 @@ static void __iomem *stli_ecpeigetmemptr(struct stlibrd *brdp, unsigned long off
 	unsigned char	val;
 
 	if (offset > brdp->memsize) {
-		printk(KERN_ERR "STALLION: shared memory pointer=%x out of "
+		printk(KERN_ERR "istallion: shared memory pointer=%x out of "
 				"range at line=%d(%d), brd=%d\n",
 			(int) offset, line, __LINE__, brdp->brdnr);
 		ptr = NULL;
@@ -2818,7 +2798,7 @@ static void __iomem *stli_ecpmcgetmemptr(struct stlibrd *brdp, unsigned long off
 	unsigned char val;
 
 	if (offset > brdp->memsize) {
-		printk(KERN_ERR "STALLION: shared memory pointer=%x out of "
+		printk(KERN_ERR "istallion: shared memory pointer=%x out of "
 				"range at line=%d(%d), brd=%d\n",
 			(int) offset, line, __LINE__, brdp->brdnr);
 		ptr = NULL;
@@ -2863,7 +2843,7 @@ static void __iomem *stli_ecppcigetmemptr(struct stlibrd *brdp, unsigned long of
 	unsigned char	val;
 
 	if (offset > brdp->memsize) {
-		printk(KERN_ERR "STALLION: shared memory pointer=%x out of "
+		printk(KERN_ERR "istallion: shared memory pointer=%x out of "
 				"range at line=%d(%d), board=%d\n",
 				(int) offset, line, __LINE__, brdp->brdnr);
 		ptr = NULL;
@@ -2928,7 +2908,7 @@ static void __iomem *stli_onbgetmemptr(struct stlibrd *brdp, unsigned long offse
 	void __iomem *ptr;
 
 	if (offset > brdp->memsize) {
-		printk(KERN_ERR "STALLION: shared memory pointer=%x out of "
+		printk(KERN_ERR "istallion: shared memory pointer=%x out of "
 				"range at line=%d(%d), brd=%d\n",
 				(int) offset, line, __LINE__, brdp->brdnr);
 		ptr = NULL;
@@ -2994,7 +2974,7 @@ static void __iomem *stli_onbegetmemptr(struct stlibrd *brdp, unsigned long offs
 	unsigned char val;
 
 	if (offset > brdp->memsize) {
-		printk(KERN_ERR "STALLION: shared memory pointer=%x out of "
+		printk(KERN_ERR "istallion: shared memory pointer=%x out of "
 				"range at line=%d(%d), brd=%d\n",
 			(int) offset, line, __LINE__, brdp->brdnr);
 		ptr = NULL;
@@ -3433,7 +3413,7 @@ static int stli_startbrd(struct stlibrd *brdp)
 #endif
 
 	if (nrdevs < (brdp->nrports + 1)) {
-		printk(KERN_ERR "STALLION: slave failed to allocate memory for "
+		printk(KERN_ERR "istallion: slave failed to allocate memory for "
 				"all devices, devices=%d\n", nrdevs);
 		brdp->nrports = nrdevs - 1;
 	}
@@ -3443,13 +3423,13 @@ static int stli_startbrd(struct stlibrd *brdp)
 	brdp->bitsize = (nrdevs + 7) / 8;
 	memoff = readl(&hdrp->memp);
 	if (memoff > brdp->memsize) {
-		printk(KERN_ERR "STALLION: corrupted shared memory region?\n");
+		printk(KERN_ERR "istallion: corrupted shared memory region?\n");
 		rc = -EIO;
 		goto stli_donestartup;
 	}
 	memp = (cdkmem_t __iomem *) EBRDGETMEMPTR(brdp, memoff);
 	if (readw(&memp->dtype) != TYP_ASYNCTRL) {
-		printk(KERN_ERR "STALLION: no slave control device found\n");
+		printk(KERN_ERR "istallion: no slave control device found\n");
 		goto stli_donestartup;
 	}
 	memp++;
@@ -3534,7 +3514,7 @@ static int __devinit stli_brdinit(struct stlibrd *brdp)
 		retval = stli_initonb(brdp);
 		break;
 	default:
-		printk(KERN_ERR "STALLION: board=%d is unknown board "
+		printk(KERN_ERR "istallion: board=%d is unknown board "
 				"type=%d\n", brdp->brdnr, brdp->brdtype);
 		retval = -ENODEV;
 	}
@@ -3543,7 +3523,7 @@ static int __devinit stli_brdinit(struct stlibrd *brdp)
 		return retval;
 
 	stli_initports(brdp);
-	printk(KERN_INFO "STALLION: %s found, board=%d io=%x mem=%x "
+	printk(KERN_INFO "istallion: %s found, board=%d io=%x mem=%x "
 		"nrpanels=%d nrports=%d\n", stli_brdnames[brdp->brdtype],
 		brdp->brdnr, brdp->iobase, (int) brdp->memaddr,
 		brdp->nrpanels, brdp->nrports);
@@ -3637,7 +3617,7 @@ static int stli_eisamemprobe(struct stlibrd *brdp)
 	if (! foundit) {
 		brdp->memaddr = 0;
 		brdp->membase = NULL;
-		printk(KERN_ERR "STALLION: failed to probe shared memory "
+		printk(KERN_ERR "istallion: failed to probe shared memory "
 				"region for %s in EISA slot=%d\n",
 			stli_brdnames[brdp->brdtype], (brdp->iobase >> 12));
 		return -ENODEV;
@@ -3782,7 +3762,7 @@ static int __devinit stli_pciprobe(struct pci_dev *pdev,
 	mutex_lock(&stli_brdslock);
 	brdnr = stli_getbrdnr();
 	if (brdnr < 0) {
-		printk(KERN_INFO "STALLION: too many boards found, "
+		printk(KERN_INFO "istallion: too many boards found, "
 			"maximum supported %d\n", STL_MAXBRDS);
 		mutex_unlock(&stli_brdslock);
 		retval = -EIO;
@@ -3854,7 +3834,7 @@ static struct stlibrd *stli_allocbrd(void)
 
 	brdp = kzalloc(sizeof(struct stlibrd), GFP_KERNEL);
 	if (!brdp) {
-		printk(KERN_ERR "STALLION: failed to allocate memory "
+		printk(KERN_ERR "istallion: failed to allocate memory "
 				"(size=%Zd)\n", sizeof(struct stlibrd));
 		return NULL;
 	}
@@ -4493,7 +4473,7 @@ static int __init istallion_module_init(void)
 
 	stli_txcookbuf = kmalloc(STLI_TXBUFSIZE, GFP_KERNEL);
 	if (!stli_txcookbuf) {
-		printk(KERN_ERR "STALLION: failed to allocate memory "
+		printk(KERN_ERR "istallion: failed to allocate memory "
 				"(size=%d)\n", STLI_TXBUFSIZE);
 		retval = -ENOMEM;
 		goto err;
@@ -4518,7 +4498,7 @@ static int __init istallion_module_init(void)
 
 	retval = tty_register_driver(stli_serial);
 	if (retval) {
-		printk(KERN_ERR "STALLION: failed to register serial driver\n");
+		printk(KERN_ERR "istallion: failed to register serial driver\n");
 		goto err_ttyput;
 	}
 
@@ -4532,7 +4512,7 @@ static int __init istallion_module_init(void)
  */
 	retval = register_chrdev(STL_SIOMEMMAJOR, "staliomem", &stli_fsiomem);
 	if (retval) {
-		printk(KERN_ERR "STALLION: failed to register serial memory "
+		printk(KERN_ERR "istallion: failed to register serial memory "
 				"device\n");
 		goto err_deinit;
 	}
diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index 08ba6eb1a380..402c9f217f83 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -1080,57 +1080,26 @@ static void mxser_flush_buffer(struct tty_struct *tty)
 static void mxser_close(struct tty_struct *tty, struct file *filp)
 {
 	struct mxser_port *info = tty->driver_data;
+	struct tty_port *port = &info->port;
 
 	unsigned long timeout;
-	unsigned long flags;
 
 	if (tty->index == MXSER_PORTS)
 		return;
 	if (!info)
 		return;
 
-	spin_lock_irqsave(&info->port.lock, flags);
-
-	if (tty_hung_up_p(filp)) {
-		spin_unlock_irqrestore(&info->port.lock, flags);
-		return;
-	}
-	if ((tty->count == 1) && (info->port.count != 1)) {
-		/*
-		 * Uh, oh.  tty->count is 1, which means that the tty
-		 * structure will be freed.  Info->port.count should always
-		 * be one in these conditions.  If it's greater than
-		 * one, we've got real problems, since it means the
-		 * serial port won't be shutdown.
-		 */
-		printk(KERN_ERR "mxser_close: bad serial port count; "
-			"tty->count is 1, info->port.count is %d\n", info->port.count);
-		info->port.count = 1;
-	}
-	if (--info->port.count < 0) {
-		printk(KERN_ERR "mxser_close: bad serial port count for "
-			"ttys%d: %d\n", tty->index, info->port.count);
-		info->port.count = 0;
-	}
-	if (info->port.count) {
-		spin_unlock_irqrestore(&info->port.lock, flags);
+	if (tty_port_close_start(port, tty, filp) == 0)
 		return;
-	}
-	info->port.flags |= ASYNC_CLOSING;
-	spin_unlock_irqrestore(&info->port.lock, flags);
+
 	/*
 	 * Save the termios structure, since this port may have
 	 * separate termios for callout and dialin.
+	 *
+	 * FIXME: Can this go ?
 	 */
 	if (info->port.flags & ASYNC_NORMAL_ACTIVE)
 		info->normal_termios = *tty->termios;
-	/*
-	 * Now we wait for the transmit buffer to clear; and we notify
-	 * the line discipline to only process XON/XOFF characters.
-	 */
-	tty->closing = 1;
-	if (info->port.closing_wait != ASYNC_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, info->port.closing_wait);
 	/*
 	 * At this point we stop accepting input.  To do this, we
 	 * disable the receive line status interrupts, and tell the
@@ -1156,19 +1125,12 @@ static void mxser_close(struct tty_struct *tty, struct file *filp)
 		}
 	}
 	mxser_shutdown(tty);
-
 	mxser_flush_buffer(tty);
-	tty_ldisc_flush(tty);
-
-	tty->closing = 0;
-	tty_port_tty_set(&info->port, NULL);
-	if (info->port.blocked_open) {
-		if (info->port.close_delay)
-			schedule_timeout_interruptible(info->port.close_delay);
-		wake_up_interruptible(&info->port.open_wait);
-	}
 
-	info->port.flags &= ~(ASYNC_NORMAL_ACTIVE | ASYNC_CLOSING);
+	/* Right now the tty_port set is done outside of the close_end helper
+	   as we don't yet have everyone using refcounts */	
+	tty_port_close_end(port, tty);
+	tty_port_tty_set(port, NULL);
 }
 
 static int mxser_write(struct tty_struct *tty, const unsigned char *buf, int count)
diff --git a/drivers/char/riscom8.c b/drivers/char/riscom8.c
index af34c2054a09..9ac5febd8abd 100644
--- a/drivers/char/riscom8.c
+++ b/drivers/char/riscom8.c
@@ -929,35 +929,11 @@ static void rc_close(struct tty_struct *tty, struct file *filp)
 	if (!port || rc_paranoia_check(port, tty->name, "close"))
 		return;
 
-	spin_lock_irqsave(&port->port.lock, flags);
-
-	if (tty_hung_up_p(filp))
-		goto out;
-
 	bp = port_Board(port);
-	if ((tty->count == 1) && (port->port.count != 1))  {
-		printk(KERN_INFO "rc%d: rc_close: bad port count;"
-		       " tty->count is 1, port count is %d\n",
-		       board_No(bp), port->port.count);
-		port->port.count = 1;
-	}
-	if (--port->port.count < 0)  {
-		printk(KERN_INFO "rc%d: rc_close: bad port count "
-				 "for tty%d: %d\n",
-		       board_No(bp), port_No(port), port->port.count);
-		port->port.count = 0;
-	}
-	if (port->port.count)
-		goto out;
-	port->port.flags |= ASYNC_CLOSING;
-	/*
-	 * Now we wait for the transmit buffer to clear; and we notify
-	 * the line discipline to only process XON/XOFF characters.
-	 */
-	tty->closing = 1;
-	spin_unlock_irqrestore(&port->port.lock, flags);
-	if (port->port.closing_wait != ASYNC_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, port->port.closing_wait);
+	
+	if (tty_port_close_start(&port->port, tty, filp) == 0)
+		return;
+	
 	/*
 	 * At this point we stop accepting input.  To do this, we
 	 * disable the receive line status interrupts, and tell the
@@ -989,23 +965,8 @@ static void rc_close(struct tty_struct *tty, struct file *filp)
 	rc_shutdown_port(tty, bp, port);
 	rc_flush_buffer(tty);
 	spin_unlock_irqrestore(&riscom_lock, flags);
-	tty_ldisc_flush(tty);
 
-	spin_lock_irqsave(&port->port.lock, flags);
-	tty->closing = 0;
-	port->port.tty = NULL;
-	if (port->port.blocked_open) {
-		spin_unlock_irqrestore(&port->port.lock, flags);
-		if (port->port.close_delay)
-			msleep_interruptible(jiffies_to_msecs(port->port.close_delay));
-		wake_up_interruptible(&port->port.open_wait);
-		spin_lock_irqsave(&port->port.lock, flags);
-	}
-	port->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-	wake_up_interruptible(&port->port.close_wait);
-
-out:
-	spin_unlock_irqrestore(&riscom_lock, flags);
+	tty_port_close_end(&port->port, tty);
 }
 
 static int rc_write(struct tty_struct *tty,
diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c
index 77eef61c46f3..e1e0dd89ac9a 100644
--- a/drivers/char/stallion.c
+++ b/drivers/char/stallion.c
@@ -833,40 +833,20 @@ static void stl_close(struct tty_struct *tty, struct file *filp)
 	pr_debug("stl_close(tty=%p,filp=%p)\n", tty, filp);
 
 	portp = tty->driver_data;
-	if (portp == NULL)
-		return;
+	BUG_ON(portp == NULL);
+
 	port = &portp->port;
 
-	spin_lock_irqsave(&port->lock, flags);
-	if (tty_hung_up_p(filp)) {
-		spin_unlock_irqrestore(&port->lock, flags);
+	if (tty_port_close_start(port, tty, filp) == 0)
 		return;
-	}
-	if (tty->count == 1 && port->count != 1)
-		port->count = 1;
-	if (port->count-- > 1) {
-		spin_unlock_irqrestore(&port->lock, flags);
-		return;
-	}
-
-	port->count = 0;
-	port->flags |= ASYNC_CLOSING;
-
 /*
  *	May want to wait for any data to drain before closing. The BUSY
  *	flag keeps track of whether we are still sending or not - it is
  *	very accurate for the cd1400, not quite so for the sc26198.
  *	(The sc26198 has no "end-of-data" interrupt only empty FIFO)
  */
-	tty->closing = 1;
-
-	spin_unlock_irqrestore(&port->lock, flags);
-
-	if (portp->closing_wait != ASYNC_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, portp->closing_wait);
 	stl_waituntilsent(tty, (HZ / 2));
 
-
 	spin_lock_irqsave(&port->lock, flags);
 	portp->port.flags &= ~ASYNC_INITIALIZED;
 	spin_unlock_irqrestore(&port->lock, flags);
@@ -883,20 +863,9 @@ static void stl_close(struct tty_struct *tty, struct file *filp)
 		portp->tx.head = NULL;
 		portp->tx.tail = NULL;
 	}
-	set_bit(TTY_IO_ERROR, &tty->flags);
-	tty_ldisc_flush(tty);
 
-	tty->closing = 0;
+	tty_port_close_end(port, tty);
 	tty_port_tty_set(port, NULL);
-
-	if (port->blocked_open) {
-		if (portp->close_delay)
-			msleep_interruptible(jiffies_to_msecs(portp->close_delay));
-		wake_up_interruptible(&portp->port.open_wait);
-	}
-
-	portp->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-	wake_up_interruptible(&port->close_wait);
 }
 
 /*****************************************************************************/
diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c
index 0ded4ed3da3c..fbd5a5ce2e1c 100644
--- a/drivers/char/synclink.c
+++ b/drivers/char/synclink.c
@@ -3104,70 +3104,18 @@ static void mgsl_close(struct tty_struct *tty, struct file * filp)
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):mgsl_close(%s) entry, count=%d\n",
 			 __FILE__,__LINE__, info->device_name, info->port.count);
-			 
-	if (!info->port.count)
-		return;
 
-	if (tty_hung_up_p(filp))
+	if (tty_port_close_start(&info->port, tty, filp) == 0)			 
 		goto cleanup;
 			
-	if ((tty->count == 1) && (info->port.count != 1)) {
-		/*
-		 * tty->count is 1 and the tty structure will be freed.
-		 * info->port.count should be one in this case.
-		 * if it's not, correct it so that the port is shutdown.
-		 */
-		printk("mgsl_close: bad refcount; tty->count is 1, "
-		       "info->port.count is %d\n", info->port.count);
-		info->port.count = 1;
-	}
-	
-	info->port.count--;
-	
-	/* if at least one open remaining, leave hardware active */
-	if (info->port.count)
-		goto cleanup;
-	
-	info->port.flags |= ASYNC_CLOSING;
-	
-	/* set tty->closing to notify line discipline to 
-	 * only process XON/XOFF characters. Only the N_TTY
-	 * discipline appears to use this (ppp does not).
-	 */
-	tty->closing = 1;
-	
-	/* wait for transmit data to clear all layers */
-	
-	if (info->port.closing_wait != ASYNC_CLOSING_WAIT_NONE) {
-		if (debug_level >= DEBUG_LEVEL_INFO)
-			printk("%s(%d):mgsl_close(%s) calling tty_wait_until_sent\n",
-				 __FILE__,__LINE__, info->device_name );
-		tty_wait_until_sent(tty, info->port.closing_wait);
-	}
-		
  	if (info->port.flags & ASYNC_INITIALIZED)
  		mgsl_wait_until_sent(tty, info->timeout);
-
 	mgsl_flush_buffer(tty);
-
 	tty_ldisc_flush(tty);
-		
 	shutdown(info);
-	
-	tty->closing = 0;
+
+	tty_port_close_end(&info->port, tty);	
 	info->port.tty = NULL;
-	
-	if (info->port.blocked_open) {
-		if (info->port.close_delay) {
-			msleep_interruptible(jiffies_to_msecs(info->port.close_delay));
-		}
-		wake_up_interruptible(&info->port.open_wait);
-	}
-	
-	info->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-			 
-	wake_up_interruptible(&info->port.close_wait);
-	
 cleanup:			
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):mgsl_close(%s) exit, count=%d\n", __FILE__,__LINE__,
diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c
index 625c9bde3be8..53544e21f191 100644
--- a/drivers/char/synclink_gt.c
+++ b/drivers/char/synclink_gt.c
@@ -720,44 +720,9 @@ static void close(struct tty_struct *tty, struct file *filp)
 		return;
 	DBGINFO(("%s close entry, count=%d\n", info->device_name, info->port.count));
 
-	if (!info->port.count)
-		return;
-
-	if (tty_hung_up_p(filp))
+	if (tty_port_close_start(&info->port, tty, filp) == 0)
 		goto cleanup;
 
-	if ((tty->count == 1) && (info->port.count != 1)) {
-		/*
-		 * tty->count is 1 and the tty structure will be freed.
-		 * info->port.count should be one in this case.
-		 * if it's not, correct it so that the port is shutdown.
-		 */
-		DBGERR(("%s close: bad refcount; tty->count=1, "
-		       "info->port.count=%d\n", info->device_name, info->port.count));
-		info->port.count = 1;
-	}
-
-	info->port.count--;
-
-	/* if at least one open remaining, leave hardware active */
-	if (info->port.count)
-		goto cleanup;
-
-	info->port.flags |= ASYNC_CLOSING;
-
-	/* set tty->closing to notify line discipline to
-	 * only process XON/XOFF characters. Only the N_TTY
-	 * discipline appears to use this (ppp does not).
-	 */
-	tty->closing = 1;
-
-	/* wait for transmit data to clear all layers */
-
-	if (info->port.closing_wait != ASYNC_CLOSING_WAIT_NONE) {
-		DBGINFO(("%s call tty_wait_until_sent\n", info->device_name));
-		tty_wait_until_sent(tty, info->port.closing_wait);
-	}
-
  	if (info->port.flags & ASYNC_INITIALIZED)
  		wait_until_sent(tty, info->timeout);
 	flush_buffer(tty);
@@ -765,20 +730,8 @@ static void close(struct tty_struct *tty, struct file *filp)
 
 	shutdown(info);
 
-	tty->closing = 0;
+	tty_port_close_end(&info->port, tty);
 	info->port.tty = NULL;
-
-	if (info->port.blocked_open) {
-		if (info->port.close_delay) {
-			msleep_interruptible(jiffies_to_msecs(info->port.close_delay));
-		}
-		wake_up_interruptible(&info->port.open_wait);
-	}
-
-	info->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-
-	wake_up_interruptible(&info->port.close_wait);
-
 cleanup:
 	DBGINFO(("%s close exit, count=%d\n", tty->driver->name, info->port.count));
 }
diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c
index 1f5c21ec4b14..2aac55bcf5fd 100644
--- a/drivers/char/synclinkmp.c
+++ b/drivers/char/synclinkmp.c
@@ -810,70 +810,18 @@ static void close(struct tty_struct *tty, struct file *filp)
 		printk("%s(%d):%s close() entry, count=%d\n",
 			 __FILE__,__LINE__, info->device_name, info->port.count);
 
-	if (!info->port.count)
-		return;
-
-	if (tty_hung_up_p(filp))
-		goto cleanup;
-
-	if ((tty->count == 1) && (info->port.count != 1)) {
-		/*
-		 * tty->count is 1 and the tty structure will be freed.
-		 * info->port.count should be one in this case.
-		 * if it's not, correct it so that the port is shutdown.
-		 */
-		printk("%s(%d):%s close: bad refcount; tty->count is 1, "
-		       "info->port.count is %d\n",
-			 __FILE__,__LINE__, info->device_name, info->port.count);
-		info->port.count = 1;
-	}
-
-	info->port.count--;
-
-	/* if at least one open remaining, leave hardware active */
-	if (info->port.count)
+	if (tty_port_close_start(&info->port, tty, filp) == 0)
 		goto cleanup;
-
-	info->port.flags |= ASYNC_CLOSING;
-
-	/* set tty->closing to notify line discipline to
-	 * only process XON/XOFF characters. Only the N_TTY
-	 * discipline appears to use this (ppp does not).
-	 */
-	tty->closing = 1;
-
-	/* wait for transmit data to clear all layers */
-
-	if (info->port.closing_wait != ASYNC_CLOSING_WAIT_NONE) {
-		if (debug_level >= DEBUG_LEVEL_INFO)
-			printk("%s(%d):%s close() calling tty_wait_until_sent\n",
-				 __FILE__,__LINE__, info->device_name );
-		tty_wait_until_sent(tty, info->port.closing_wait);
-	}
-
+		
  	if (info->port.flags & ASYNC_INITIALIZED)
  		wait_until_sent(tty, info->timeout);
 
 	flush_buffer(tty);
-
 	tty_ldisc_flush(tty);
-
 	shutdown(info);
 
-	tty->closing = 0;
+	tty_port_close_end(&info->port, tty);
 	info->port.tty = NULL;
-
-	if (info->port.blocked_open) {
-		if (info->port.close_delay) {
-			msleep_interruptible(jiffies_to_msecs(info->port.close_delay));
-		}
-		wake_up_interruptible(&info->port.open_wait);
-	}
-
-	info->port.flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-
-	wake_up_interruptible(&info->port.close_wait);
-
 cleanup:
 	if (debug_level >= DEBUG_LEVEL_INFO)
 		printk("%s(%d):%s close() exit, count=%d\n", __FILE__,__LINE__,
diff --git a/drivers/char/tty_port.c b/drivers/char/tty_port.c
index 0723664fe0ab..b3175f54fe05 100644
--- a/drivers/char/tty_port.c
+++ b/drivers/char/tty_port.c
@@ -257,3 +257,61 @@ int tty_port_block_til_ready(struct tty_port *port,
 }
 EXPORT_SYMBOL(tty_port_block_til_ready);
 
+int tty_port_close_start(struct tty_port *port, struct tty_struct *tty, struct file *filp)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&port->lock, flags);
+	if (tty_hung_up_p(filp)) {
+		spin_unlock_irqrestore(&port->lock, flags);
+		return 0;
+	}
+
+	if( tty->count == 1 && port->count != 1) {
+		printk(KERN_WARNING
+		    "tty_port_close_start: tty->count = 1 port count = %d.\n",
+								port->count);
+		port->count = 1;
+	}
+	if (--port->count < 0) {
+		printk(KERN_WARNING "tty_port_close_start: count = %d\n",
+								port->count);
+		port->count = 0;
+	}
+
+	if (port->count) {
+		spin_unlock_irqrestore(&port->lock, flags);
+		return 0;
+	}
+	port->flags |= ASYNC_CLOSING;
+	tty->closing = 1;
+	spin_unlock_irqrestore(&port->lock, flags);
+	if (port->closing_wait != ASYNC_CLOSING_WAIT_NONE)
+		tty_wait_until_sent(tty, port->closing_wait);
+	return 1;
+}
+EXPORT_SYMBOL(tty_port_close_start);
+
+void tty_port_close_end(struct tty_port *port, struct tty_struct *tty)
+{
+	unsigned long flags;
+
+	tty_ldisc_flush(tty);
+
+	spin_lock_irqsave(&port->lock, flags);
+	tty->closing = 0;
+
+	if (port->blocked_open) {
+		spin_unlock_irqrestore(&port->lock, flags);
+		if (port->close_delay) {
+			msleep_interruptible(
+				jiffies_to_msecs(port->close_delay));
+		}
+		spin_lock_irqsave(&port->lock, flags);
+		wake_up_interruptible(&port->open_wait);
+	}
+	port->flags &= ~(ASYNC_NORMAL_ACTIVE | ASYNC_CLOSING);
+	wake_up_interruptible(&port->close_wait);
+	spin_unlock_irqrestore(&port->lock, flags);
+}
+EXPORT_SYMBOL(tty_port_close_end);
diff --git a/include/linux/istallion.h b/include/linux/istallion.h
index 053d5aea925c..7faca98c7d14 100644
--- a/include/linux/istallion.h
+++ b/include/linux/istallion.h
@@ -59,7 +59,6 @@ struct stliport {
 	unsigned int		devnr;
 	int			baud_base;
 	int			custom_divisor;
-	int			close_delay;
 	int			closing_wait;
 	int			rc;
 	int			argsize;
diff --git a/include/linux/tty.h b/include/linux/tty.h
index 61a0ab32cf11..fc39db95499f 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -441,6 +441,9 @@ extern void tty_port_raise_dtr_rts(struct tty_port *port);
 extern void tty_port_hangup(struct tty_port *port);
 extern int tty_port_block_til_ready(struct tty_port *port,
 				struct tty_struct *tty, struct file *filp);
+extern int tty_port_close_start(struct tty_port *port,
+				struct tty_struct *tty, struct file *filp);
+extern void tty_port_close_end(struct tty_port *port, struct tty_struct *tty);
 
 extern int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc);
 extern int tty_unregister_ldisc(int disc);
-- 
cgit v1.2.3


From 39aced68d664291db3324d0fcf0985ab5626aac2 Mon Sep 17 00:00:00 2001
From: Niels de Vos <niels.devos@wincor-nixdorf.com>
Date: Fri, 2 Jan 2009 13:46:58 +0000
Subject: serial: set correct baud_base for Oxford Semiconductor Ltd EXSYS
 EX-41092 Dual 16950 Serial adapter

The PCI-card identified as "Oxford Semiconductor Ltd EXSYS EX-41092 Dual
16950 Serial adapter" is only usable with other devices (i.e. not the same
card) after doing a "setserial /dev/ttyS<n> baud_base 115200".  This
baud_base should be default for this card.

Signed-off-by: Niels de Vos <niels.devos@wincor-nixdorf.com>
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/8250_pci.c | 3 +++
 include/linux/pci_ids.h   | 1 +
 2 files changed, 4 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/serial/8250_pci.c b/drivers/serial/8250_pci.c
index 0b794138f686..2a2e1c717e8e 100644
--- a/drivers/serial/8250_pci.c
+++ b/drivers/serial/8250_pci.c
@@ -2387,6 +2387,9 @@ static struct pci_device_id serial_pci_tbl[] = {
 		 * www.ussg.iu.edu/hypermail/linux/kernel/0303.1/0516.html).
 		 * For now just used the hex ID 0x950a.
 		 */
+	{	PCI_VENDOR_ID_OXSEMI, 0x950a,
+		PCI_SUBVENDOR_ID_SIIG, PCI_SUBDEVICE_ID_SIIG_DUAL_SERIAL, 0, 0,
+		pbn_b0_2_115200 },
 	{	PCI_VENDOR_ID_OXSEMI, 0x950a,
 		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
 		pbn_b0_2_1130000 },
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index b6e694454280..fa83dfefc5e0 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1766,6 +1766,7 @@
 #define PCI_DEVICE_ID_SIIG_8S_20x_650	0x2081
 #define PCI_DEVICE_ID_SIIG_8S_20x_850	0x2082
 #define PCI_SUBDEVICE_ID_SIIG_QUARTET_SERIAL	0x2050
+#define PCI_SUBDEVICE_ID_SIIG_DUAL_SERIAL	0x2530
 
 #define PCI_VENDOR_ID_RADISYS		0x1331
 
-- 
cgit v1.2.3


From 60c20fb8c00a2b23308ae4517f145383bc66d291 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@canonical.com>
Date: Fri, 2 Jan 2009 13:49:04 +0000
Subject: serial: RS485 ioctl structure uses __u32 include linux/types.h
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In the commit below a new struct serial_rs485 was introduced for a new
ioctl:

    commit c26c56c0f40e200e61d1390629c806f6adaffbcc
    Author: Alan Cox <alan@redhat.com>
    Date:   Mon Oct 13 10:37:48 2008 +0100

	tty: Cris has a nice RS485 ioctl so we should steal it

This structure uses the __u32 types for some of its members, which leads
to the following compile error:

    $ cc -I.../include -c X.c
    In file included from X.c:2: .../include/linux/serial.h:185:
		error: expected specifier-qualifier-list before ‘__u32’
    $

It seems that these types are appropriate for this structure as it is
to be exposed to userspace.  These types are available via linux/types.h
so move the include of that outside the __KERNEL__ section.

Signed-off-by: Andy Whitcroft <apw@canonical.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/serial.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/serial.h b/include/linux/serial.h
index 1ea8d9265bf6..9136cc5608c3 100644
--- a/include/linux/serial.h
+++ b/include/linux/serial.h
@@ -10,8 +10,9 @@
 #ifndef _LINUX_SERIAL_H
 #define _LINUX_SERIAL_H
 
-#ifdef __KERNEL__
 #include <linux/types.h>
+
+#ifdef __KERNEL__
 #include <asm/page.h>
 
 /*
-- 
cgit v1.2.3


From f751928e0ddf54ea4fe5546f35e99efc5b5d9938 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Fri, 2 Jan 2009 13:49:21 +0000
Subject: tty: We want the port object to be persistent

Move the tty_port and uart_info bits around a little. By embedding the uart_info
into the uart_port we get rid of lots of corner case testing and also get the
ability to go port<->state<->info which is a bit more elegant than the current
data structures.

Downsides - we allocate a tiny bit more memory for unused ports, upside we've
removed as much code as it saved for most users..

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/jsm/jsm_tty.c |   2 +-
 drivers/serial/serial_core.c | 144 ++++++++++++++++++-------------------------
 include/linux/serial_core.h  |  62 ++++++++++---------
 3 files changed, 96 insertions(+), 112 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/serial/jsm/jsm_tty.c b/drivers/serial/jsm/jsm_tty.c
index a697914ae3d0..3547558d2caf 100644
--- a/drivers/serial/jsm/jsm_tty.c
+++ b/drivers/serial/jsm/jsm_tty.c
@@ -272,7 +272,7 @@ static void jsm_tty_close(struct uart_port *port)
 	jsm_printk(CLOSE, INFO, &channel->ch_bd->pci_dev, "start\n");
 
 	bd = channel->ch_bd;
-	ts = channel->uart_port.info->port.tty->termios;
+	ts = port->info->port.tty->termios;
 
 	channel->ch_flags &= ~(CH_STOPI);
 
diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c
index 874786a11fe9..daeba1c52c8e 100644
--- a/drivers/serial/serial_core.c
+++ b/drivers/serial/serial_core.c
@@ -50,7 +50,7 @@ static struct lock_class_key port_lock_key;
 
 #define HIGH_BITS_OFFSET	((sizeof(long)-sizeof(int))*8)
 
-#define uart_users(state)	((state)->count + ((state)->info ? (state)->info->port.blocked_open : 0))
+#define uart_users(state)	((state)->count + (state)->info.port.blocked_open)
 
 #ifdef CONFIG_SERIAL_CORE_CONSOLE
 #define uart_console(port)	((port)->cons && (port)->cons->index == (port)->line)
@@ -94,7 +94,7 @@ static void __uart_start(struct tty_struct *tty)
 	struct uart_state *state = tty->driver_data;
 	struct uart_port *port = state->port;
 
-	if (!uart_circ_empty(&state->info->xmit) && state->info->xmit.buf &&
+	if (!uart_circ_empty(&state->info.xmit) && state->info.xmit.buf &&
 	    !tty->stopped && !tty->hw_stopped)
 		port->ops->start_tx(port);
 }
@@ -113,7 +113,7 @@ static void uart_start(struct tty_struct *tty)
 static void uart_tasklet_action(unsigned long data)
 {
 	struct uart_state *state = (struct uart_state *)data;
-	tty_wakeup(state->info->port.tty);
+	tty_wakeup(state->info.port.tty);
 }
 
 static inline void
@@ -139,7 +139,7 @@ uart_update_mctrl(struct uart_port *port, unsigned int set, unsigned int clear)
  */
 static int uart_startup(struct uart_state *state, int init_hw)
 {
-	struct uart_info *info = state->info;
+	struct uart_info *info = &state->info;
 	struct uart_port *port = state->port;
 	unsigned long page;
 	int retval = 0;
@@ -212,14 +212,15 @@ static int uart_startup(struct uart_state *state, int init_hw)
  */
 static void uart_shutdown(struct uart_state *state)
 {
-	struct uart_info *info = state->info;
+	struct uart_info *info = &state->info;
 	struct uart_port *port = state->port;
+	struct tty_struct *tty = info->port.tty;
 
 	/*
 	 * Set the TTY IO error marker
 	 */
-	if (info->port.tty)
-		set_bit(TTY_IO_ERROR, &info->port.tty->flags);
+	if (tty)
+		set_bit(TTY_IO_ERROR, &tty->flags);
 
 	if (info->flags & UIF_INITIALIZED) {
 		info->flags &= ~UIF_INITIALIZED;
@@ -227,7 +228,7 @@ static void uart_shutdown(struct uart_state *state)
 		/*
 		 * Turn off DTR and RTS early.
 		 */
-		if (!info->port.tty || (info->port.tty->termios->c_cflag & HUPCL))
+		if (!tty || (tty->termios->c_cflag & HUPCL))
 			uart_clear_mctrl(port, TIOCM_DTR | TIOCM_RTS);
 
 		/*
@@ -427,7 +428,7 @@ EXPORT_SYMBOL(uart_get_divisor);
 static void
 uart_change_speed(struct uart_state *state, struct ktermios *old_termios)
 {
-	struct tty_struct *tty = state->info->port.tty;
+	struct tty_struct *tty = state->info.port.tty;
 	struct uart_port *port = state->port;
 	struct ktermios *termios;
 
@@ -444,14 +445,14 @@ uart_change_speed(struct uart_state *state, struct ktermios *old_termios)
 	 * Set flags based on termios cflag
 	 */
 	if (termios->c_cflag & CRTSCTS)
-		state->info->flags |= UIF_CTS_FLOW;
+		state->info.flags |= UIF_CTS_FLOW;
 	else
-		state->info->flags &= ~UIF_CTS_FLOW;
+		state->info.flags &= ~UIF_CTS_FLOW;
 
 	if (termios->c_cflag & CLOCAL)
-		state->info->flags &= ~UIF_CHECK_CD;
+		state->info.flags &= ~UIF_CHECK_CD;
 	else
-		state->info->flags |= UIF_CHECK_CD;
+		state->info.flags |= UIF_CHECK_CD;
 
 	port->ops->set_termios(port, termios, old_termios);
 }
@@ -479,7 +480,7 @@ static int uart_put_char(struct tty_struct *tty, unsigned char ch)
 {
 	struct uart_state *state = tty->driver_data;
 
-	return __uart_put_char(state->port, &state->info->xmit, ch);
+	return __uart_put_char(state->port, &state->info.xmit, ch);
 }
 
 static void uart_flush_chars(struct tty_struct *tty)
@@ -500,13 +501,13 @@ uart_write(struct tty_struct *tty, const unsigned char *buf, int count)
 	 * This means you called this function _after_ the port was
 	 * closed.  No cookie for you.
 	 */
-	if (!state || !state->info) {
+	if (!state) {
 		WARN_ON(1);
 		return -EL3HLT;
 	}
 
 	port = state->port;
-	circ = &state->info->xmit;
+	circ = &state->info.xmit;
 
 	if (!circ->buf)
 		return 0;
@@ -537,7 +538,7 @@ static int uart_write_room(struct tty_struct *tty)
 	int ret;
 
 	spin_lock_irqsave(&state->port->lock, flags);
-	ret = uart_circ_chars_free(&state->info->xmit);
+	ret = uart_circ_chars_free(&state->info.xmit);
 	spin_unlock_irqrestore(&state->port->lock, flags);
 	return ret;
 }
@@ -549,7 +550,7 @@ static int uart_chars_in_buffer(struct tty_struct *tty)
 	int ret;
 
 	spin_lock_irqsave(&state->port->lock, flags);
-	ret = uart_circ_chars_pending(&state->info->xmit);
+	ret = uart_circ_chars_pending(&state->info.xmit);
 	spin_unlock_irqrestore(&state->port->lock, flags);
 	return ret;
 }
@@ -564,7 +565,7 @@ static void uart_flush_buffer(struct tty_struct *tty)
 	 * This means you called this function _after_ the port was
 	 * closed.  No cookie for you.
 	 */
-	if (!state || !state->info) {
+	if (!state) {
 		WARN_ON(1);
 		return;
 	}
@@ -573,7 +574,7 @@ static void uart_flush_buffer(struct tty_struct *tty)
 	pr_debug("uart_flush_buffer(%d) called\n", tty->index);
 
 	spin_lock_irqsave(&port->lock, flags);
-	uart_circ_clear(&state->info->xmit);
+	uart_circ_clear(&state->info.xmit);
 	if (port->ops->flush_buffer)
 		port->ops->flush_buffer(port);
 	spin_unlock_irqrestore(&port->lock, flags);
@@ -837,15 +838,15 @@ static int uart_set_info(struct uart_state *state,
 	state->closing_wait    = closing_wait;
 	if (new_serial.xmit_fifo_size)
 		port->fifosize = new_serial.xmit_fifo_size;
-	if (state->info->port.tty)
-		state->info->port.tty->low_latency =
+	if (state->info.port.tty)
+		state->info.port.tty->low_latency =
 			(port->flags & UPF_LOW_LATENCY) ? 1 : 0;
 
  check_and_exit:
 	retval = 0;
 	if (port->type == PORT_UNKNOWN)
 		goto exit;
-	if (state->info->flags & UIF_INITIALIZED) {
+	if (state->info.flags & UIF_INITIALIZED) {
 		if (((old_flags ^ port->flags) & UPF_SPD_MASK) ||
 		    old_custom_divisor != port->custom_divisor) {
 			/*
@@ -858,7 +859,7 @@ static int uart_set_info(struct uart_state *state,
 				printk(KERN_NOTICE
 				       "%s sets custom speed on %s. This "
 				       "is deprecated.\n", current->comm,
-				       tty_name(state->info->port.tty, buf));
+				       tty_name(state->info.port.tty, buf));
 			}
 			uart_change_speed(state, NULL);
 		}
@@ -889,8 +890,8 @@ static int uart_get_lsr_info(struct uart_state *state,
 	 * interrupt happens).
 	 */
 	if (port->x_char ||
-	    ((uart_circ_chars_pending(&state->info->xmit) > 0) &&
-	     !state->info->port.tty->stopped && !state->info->port.tty->hw_stopped))
+	    ((uart_circ_chars_pending(&state->info.xmit) > 0) &&
+	     !state->info.port.tty->stopped && !state->info.port.tty->hw_stopped))
 		result &= ~TIOCSER_TEMT;
 
 	return put_user(result, value);
@@ -1017,7 +1018,7 @@ uart_wait_modem_status(struct uart_state *state, unsigned long arg)
 	port->ops->enable_ms(port);
 	spin_unlock_irq(&port->lock);
 
-	add_wait_queue(&state->info->delta_msr_wait, &wait);
+	add_wait_queue(&state->info.delta_msr_wait, &wait);
 	for (;;) {
 		spin_lock_irq(&port->lock);
 		memcpy(&cnow, &port->icount, sizeof(struct uart_icount));
@@ -1045,7 +1046,7 @@ uart_wait_modem_status(struct uart_state *state, unsigned long arg)
 	}
 
 	current->state = TASK_RUNNING;
-	remove_wait_queue(&state->info->delta_msr_wait, &wait);
+	remove_wait_queue(&state->info.delta_msr_wait, &wait);
 
 	return ret;
 }
@@ -1241,7 +1242,7 @@ static void uart_set_termios(struct tty_struct *tty,
 	 */
 	if (!(old_termios->c_cflag & CLOCAL) &&
 	    (tty->termios->c_cflag & CLOCAL))
-		wake_up_interruptible(&state->info->port.open_wait);
+		wake_up_interruptible(&info->port.open_wait);
 #endif
 }
 
@@ -1303,7 +1304,7 @@ static void uart_close(struct tty_struct *tty, struct file *filp)
 	 * At this point, we stop accepting input.  To do this, we
 	 * disable the receive line status interrupts.
 	 */
-	if (state->info->flags & UIF_INITIALIZED) {
+	if (state->info.flags & UIF_INITIALIZED) {
 		unsigned long flags;
 		spin_lock_irqsave(&port->lock, flags);
 		port->ops->stop_rx(port);
@@ -1322,9 +1323,9 @@ static void uart_close(struct tty_struct *tty, struct file *filp)
 	tty_ldisc_flush(tty);
 
 	tty->closing = 0;
-	state->info->port.tty = NULL;
+	state->info.port.tty = NULL;
 
-	if (state->info->port.blocked_open) {
+	if (state->info.port.blocked_open) {
 		if (state->close_delay)
 			msleep_interruptible(state->close_delay);
 	} else if (!uart_console(port)) {
@@ -1334,8 +1335,8 @@ static void uart_close(struct tty_struct *tty, struct file *filp)
 	/*
 	 * Wake up anyone trying to open this port.
 	 */
-	state->info->flags &= ~UIF_NORMAL_ACTIVE;
-	wake_up_interruptible(&state->info->port.open_wait);
+	state->info.flags &= ~UIF_NORMAL_ACTIVE;
+	wake_up_interruptible(&state->info.port.open_wait);
 
  done:
 	mutex_unlock(&state->mutex);
@@ -1409,19 +1410,20 @@ static void uart_wait_until_sent(struct tty_struct *tty, int timeout)
 static void uart_hangup(struct tty_struct *tty)
 {
 	struct uart_state *state = tty->driver_data;
+	struct uart_info *info = &state->info;
 
 	BUG_ON(!kernel_locked());
 	pr_debug("uart_hangup(%d)\n", state->port->line);
 
 	mutex_lock(&state->mutex);
-	if (state->info && state->info->flags & UIF_NORMAL_ACTIVE) {
+	if (info->flags & UIF_NORMAL_ACTIVE) {
 		uart_flush_buffer(tty);
 		uart_shutdown(state);
 		state->count = 0;
-		state->info->flags &= ~UIF_NORMAL_ACTIVE;
-		state->info->port.tty = NULL;
-		wake_up_interruptible(&state->info->port.open_wait);
-		wake_up_interruptible(&state->info->delta_msr_wait);
+		info->flags &= ~UIF_NORMAL_ACTIVE;
+		info->port.tty = NULL;
+		wake_up_interruptible(&info->port.open_wait);
+		wake_up_interruptible(&info->delta_msr_wait);
 	}
 	mutex_unlock(&state->mutex);
 }
@@ -1434,7 +1436,7 @@ static void uart_hangup(struct tty_struct *tty)
  */
 static void uart_update_termios(struct uart_state *state)
 {
-	struct tty_struct *tty = state->info->port.tty;
+	struct tty_struct *tty = state->info.port.tty;
 	struct uart_port *port = state->port;
 
 	if (uart_console(port) && port->cons->cflag) {
@@ -1469,7 +1471,7 @@ static int
 uart_block_til_ready(struct file *filp, struct uart_state *state)
 {
 	DECLARE_WAITQUEUE(wait, current);
-	struct uart_info *info = state->info;
+	struct uart_info *info = &state->info;
 	struct uart_port *port = state->port;
 	unsigned int mctrl;
 
@@ -1563,28 +1565,6 @@ static struct uart_state *uart_get(struct uart_driver *drv, int line)
 		ret = -ENXIO;
 		goto err_unlock;
 	}
-
-	/* BKL: RACE HERE - LEAK */
-	/* We should move this into the uart_state structure and kill off
-	   this whole complexity */
-	if (!state->info) {
-		state->info = kzalloc(sizeof(struct uart_info), GFP_KERNEL);
-		if (state->info) {
-			init_waitqueue_head(&state->info->port.open_wait);
-			init_waitqueue_head(&state->info->delta_msr_wait);
-
-			/*
-			 * Link the info into the other structures.
-			 */
-			state->port->info = state->info;
-
-			tasklet_init(&state->info->tlet, uart_tasklet_action,
-				     (unsigned long)state);
-		} else {
-			ret = -ENOMEM;
-			goto err_unlock;
-		}
-	}
 	return state;
 
  err_unlock:
@@ -1641,9 +1621,10 @@ static int uart_open(struct tty_struct *tty, struct file *filp)
 	 * Any failures from here onwards should not touch the count.
 	 */
 	tty->driver_data = state;
+	state->port->info = &state->info;
 	tty->low_latency = (state->port->flags & UPF_LOW_LATENCY) ? 1 : 0;
 	tty->alt_speed = 0;
-	state->info->port.tty = tty;
+	state->info.port.tty = tty;
 
 	/*
 	 * If the port is in the middle of closing, bail out now.
@@ -1676,8 +1657,8 @@ static int uart_open(struct tty_struct *tty, struct file *filp)
 	/*
 	 * If this is the first open to succeed, adjust things to suit.
 	 */
-	if (retval == 0 && !(state->info->flags & UIF_NORMAL_ACTIVE)) {
-		state->info->flags |= UIF_NORMAL_ACTIVE;
+	if (retval == 0 && !(state->info.flags & UIF_NORMAL_ACTIVE)) {
+		state->info.flags |= UIF_NORMAL_ACTIVE;
 
 		uart_update_termios(state);
 	}
@@ -2028,11 +2009,11 @@ int uart_suspend_port(struct uart_driver *drv, struct uart_port *port)
 	}
 	port->suspended = 1;
 
-	if (state->info && state->info->flags & UIF_INITIALIZED) {
+	if (state->info.flags & UIF_INITIALIZED) {
 		const struct uart_ops *ops = port->ops;
 		int tries;
 
-		state->info->flags = (state->info->flags & ~UIF_INITIALIZED)
+		state->info.flags = (state->info.flags & ~UIF_INITIALIZED)
 				     | UIF_SUSPENDED;
 
 		spin_lock_irq(&port->lock);
@@ -2107,15 +2088,15 @@ int uart_resume_port(struct uart_driver *drv, struct uart_port *port)
 		/*
 		 * If that's unset, use the tty termios setting.
 		 */
-		if (state->info && state->info->port.tty && termios.c_cflag == 0)
-			termios = *state->info->port.tty->termios;
+		if (state->info.port.tty && termios.c_cflag == 0)
+			termios = *state->info.port.tty->termios;
 
 		uart_change_pm(state, 0);
 		port->ops->set_termios(port, &termios, NULL);
 		console_start(port->cons);
 	}
 
-	if (state->info && state->info->flags & UIF_SUSPENDED) {
+	if (state->info.flags & UIF_SUSPENDED) {
 		const struct uart_ops *ops = port->ops;
 		int ret;
 
@@ -2130,7 +2111,7 @@ int uart_resume_port(struct uart_driver *drv, struct uart_port *port)
 			ops->set_mctrl(port, port->mctrl);
 			ops->start_tx(port);
 			spin_unlock_irq(&port->lock);
-			state->info->flags |= UIF_INITIALIZED;
+			state->info.flags |= UIF_INITIALIZED;
 		} else {
 			/*
 			 * Failed to resume - maybe hardware went away?
@@ -2140,7 +2121,7 @@ int uart_resume_port(struct uart_driver *drv, struct uart_port *port)
 			uart_shutdown(state);
 		}
 
-		state->info->flags &= ~UIF_SUSPENDED;
+		state->info.flags &= ~UIF_SUSPENDED;
 	}
 
 	mutex_unlock(&state->mutex);
@@ -2383,8 +2364,12 @@ int uart_register_driver(struct uart_driver *drv)
 
 		state->close_delay     = 500;	/* .5 seconds */
 		state->closing_wait    = 30000;	/* 30 seconds */
-
 		mutex_init(&state->mutex);
+
+		tty_port_init(&state->info.port);
+		init_waitqueue_head(&state->info.delta_msr_wait);
+		tasklet_init(&state->info.tlet, uart_tasklet_action,
+			     (unsigned long)state);
 	}
 
 	retval = tty_register_driver(normal);
@@ -2455,7 +2440,7 @@ int uart_add_one_port(struct uart_driver *drv, struct uart_port *port)
 	state->pm_state = -1;
 
 	port->cons = drv->cons;
-	port->info = state->info;
+	port->info = &state->info;
 
 	/*
 	 * If this port is a console, then the spinlock is already
@@ -2527,17 +2512,10 @@ int uart_remove_one_port(struct uart_driver *drv, struct uart_port *port)
 	 */
 	tty_unregister_device(drv->tty_driver, port->line);
 
-	info = state->info;
+	info = &state->info;
 	if (info && info->port.tty)
 		tty_vhangup(info->port.tty);
 
-	/*
-	 * All users of this port should now be disconnected from
-	 * this driver, and the port shut down.  We should be the
-	 * only thread fiddling with this port from now on.
-	 */
-	state->info = NULL;
-
 	/*
 	 * Free the port IO and memory resources, if any.
 	 */
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index feb3b939ec4b..2395969faa04 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -315,36 +315,14 @@ struct uart_port {
 	void			*private_data;		/* generic platform data pointer */
 };
 
-/*
- * This is the state information which is persistent across opens.
- * The low level driver must not to touch any elements contained
- * within.
- */
-struct uart_state {
-	unsigned int		close_delay;		/* msec */
-	unsigned int		closing_wait;		/* msec */
-
-#define USF_CLOSING_WAIT_INF	(0)
-#define USF_CLOSING_WAIT_NONE	(~0U)
-
-	int			count;
-	int			pm_state;
-	struct uart_info	*info;
-	struct uart_port	*port;
-
-	struct mutex		mutex;
-};
-
-#define UART_XMIT_SIZE	PAGE_SIZE
-
-typedef unsigned int __bitwise__ uif_t;
-
 /*
  * This is the state information which is only valid when the port
- * is open; it may be freed by the core driver once the device has
+ * is open; it may be cleared the core driver once the device has
  * been closed.  Either the low level driver or the core can modify
  * stuff here.
  */
+typedef unsigned int __bitwise__ uif_t;
+
 struct uart_info {
 	struct tty_port		port;
 	struct circ_buf		xmit;
@@ -366,6 +344,29 @@ struct uart_info {
 	wait_queue_head_t	delta_msr_wait;
 };
 
+/*
+ * This is the state information which is persistent across opens.
+ * The low level driver must not to touch any elements contained
+ * within.
+ */
+struct uart_state {
+	unsigned int		close_delay;		/* msec */
+	unsigned int		closing_wait;		/* msec */
+
+#define USF_CLOSING_WAIT_INF	(0)
+#define USF_CLOSING_WAIT_NONE	(~0U)
+
+	int			count;
+	int			pm_state;
+	struct uart_info	info;
+	struct uart_port	*port;
+
+	struct mutex		mutex;
+};
+
+#define UART_XMIT_SIZE	PAGE_SIZE
+
+
 /* number of characters left in xmit buffer before we ask for more */
 #define WAKEUP_CHARS		256
 
@@ -439,8 +440,13 @@ int uart_resume_port(struct uart_driver *reg, struct uart_port *port);
 #define uart_circ_chars_free(circ)	\
 	(CIRC_SPACE((circ)->head, (circ)->tail, UART_XMIT_SIZE))
 
-#define uart_tx_stopped(portp)		\
-	((portp)->info->port.tty->stopped || (portp)->info->port.tty->hw_stopped)
+static inline int uart_tx_stopped(struct uart_port *port)
+{
+	struct tty_struct *tty = port->info->port.tty;
+	if(tty->stopped || tty->hw_stopped)
+		return 1;
+	return 0;
+}
 
 /*
  * The following are helper functions for the low level drivers.
@@ -451,7 +457,7 @@ uart_handle_sysrq_char(struct uart_port *port, unsigned int ch)
 #ifdef SUPPORT_SYSRQ
 	if (port->sysrq) {
 		if (ch && time_before(jiffies, port->sysrq)) {
-			handle_sysrq(ch, port->info ? port->info->port.tty : NULL);
+			handle_sysrq(ch, port->info->port.tty);
 			port->sysrq = 0;
 			return 1;
 		}
-- 
cgit v1.2.3


From 7d6a07d123b62bf4fa71867420c23da3ca36c995 Mon Sep 17 00:00:00 2001
From: David Daney <ddaney@caviumnetworks.com>
Date: Fri, 2 Jan 2009 13:49:47 +0000
Subject: 8250: Serial driver changes to support future Cavium OCTEON serial
 patches.

In order to use Cavium OCTEON specific serial i/o drivers, we first
patch the 8250 driver to use replaceable I/O functions.  Compatible
I/O functions are added for existing iotypeS.

An added benefit of this change is that it makes it easy to factor
some of the existing special cases out to board/SOC specific support
code.

The alternative is to load up 8250.c with a bunch of OCTEON specific
iotype code and bug work-arounds.

Signed-off-by: David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Tomaso Paoletti <tpaoletti@caviumnetworks.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/8250.c       | 194 +++++++++++++++++++++++++++++++-------------
 include/linux/serial_8250.h |   2 +
 include/linux/serial_core.h |   2 +
 3 files changed, 140 insertions(+), 58 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index 8e28750a4058..849af9d21feb 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -303,16 +303,16 @@ static const u8 au_io_out_map[] = {
 };
 
 /* sane hardware needs no mapping */
-static inline int map_8250_in_reg(struct uart_8250_port *up, int offset)
+static inline int map_8250_in_reg(struct uart_port *p, int offset)
 {
-	if (up->port.iotype != UPIO_AU)
+	if (p->iotype != UPIO_AU)
 		return offset;
 	return au_io_in_map[offset];
 }
 
-static inline int map_8250_out_reg(struct uart_8250_port *up, int offset)
+static inline int map_8250_out_reg(struct uart_port *p, int offset)
 {
-	if (up->port.iotype != UPIO_AU)
+	if (p->iotype != UPIO_AU)
 		return offset;
 	return au_io_out_map[offset];
 }
@@ -341,16 +341,16 @@ static const u8
 		[UART_SCR]	= 0x2c
 	};
 
-static inline int map_8250_in_reg(struct uart_8250_port *up, int offset)
+static inline int map_8250_in_reg(struct uart_port *p, int offset)
 {
-	if (up->port.iotype != UPIO_RM9000)
+	if (p->iotype != UPIO_RM9000)
 		return offset;
 	return regmap_in[offset];
 }
 
-static inline int map_8250_out_reg(struct uart_8250_port *up, int offset)
+static inline int map_8250_out_reg(struct uart_port *p, int offset)
 {
-	if (up->port.iotype != UPIO_RM9000)
+	if (p->iotype != UPIO_RM9000)
 		return offset;
 	return regmap_out[offset];
 }
@@ -363,108 +363,170 @@ static inline int map_8250_out_reg(struct uart_8250_port *up, int offset)
 
 #endif
 
-static unsigned int serial_in(struct uart_8250_port *up, int offset)
+static unsigned int hub6_serial_in(struct uart_port *p, int offset)
 {
-	unsigned int tmp;
-	offset = map_8250_in_reg(up, offset) << up->port.regshift;
+	offset = map_8250_in_reg(p, offset) << p->regshift;
+	outb(p->hub6 - 1 + offset, p->iobase);
+	return inb(p->iobase + 1);
+}
 
-	switch (up->port.iotype) {
-	case UPIO_HUB6:
-		outb(up->port.hub6 - 1 + offset, up->port.iobase);
-		return inb(up->port.iobase + 1);
+static void hub6_serial_out(struct uart_port *p, int offset, int value)
+{
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	outb(p->hub6 - 1 + offset, p->iobase);
+	outb(value, p->iobase + 1);
+}
 
-	case UPIO_MEM:
-	case UPIO_DWAPB:
-		return readb(up->port.membase + offset);
+static unsigned int mem_serial_in(struct uart_port *p, int offset)
+{
+	offset = map_8250_in_reg(p, offset) << p->regshift;
+	return readb(p->membase + offset);
+}
 
-	case UPIO_RM9000:
-	case UPIO_MEM32:
-		return readl(up->port.membase + offset);
+static void mem_serial_out(struct uart_port *p, int offset, int value)
+{
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	writeb(value, p->membase + offset);
+}
+
+static void mem32_serial_out(struct uart_port *p, int offset, int value)
+{
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	writel(value, p->membase + offset);
+}
+
+static unsigned int mem32_serial_in(struct uart_port *p, int offset)
+{
+	offset = map_8250_in_reg(p, offset) << p->regshift;
+	return readl(p->membase + offset);
+}
 
 #ifdef CONFIG_SERIAL_8250_AU1X00
-	case UPIO_AU:
-		return __raw_readl(up->port.membase + offset);
+static unsigned int au_serial_in(struct uart_port *p, int offset)
+{
+	offset = map_8250_in_reg(p, offset) << p->regshift;
+	return __raw_readl(p->membase + offset);
+}
+
+static void au_serial_out(struct uart_port *p, int offset, int value)
+{
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	__raw_writel(value, p->membase + offset);
+}
 #endif
 
-	case UPIO_TSI:
-		if (offset == UART_IIR) {
-			tmp = readl(up->port.membase + (UART_IIR & ~3));
-			return (tmp >> 16) & 0xff; /* UART_IIR % 4 == 2 */
-		} else
-			return readb(up->port.membase + offset);
+static unsigned int tsi_serial_in(struct uart_port *p, int offset)
+{
+	unsigned int tmp;
+	offset = map_8250_in_reg(p, offset) << p->regshift;
+	if (offset == UART_IIR) {
+		tmp = readl(p->membase + (UART_IIR & ~3));
+		return (tmp >> 16) & 0xff; /* UART_IIR % 4 == 2 */
+	} else
+		return readb(p->membase + offset);
+}
 
-	default:
-		return inb(up->port.iobase + offset);
-	}
+static void tsi_serial_out(struct uart_port *p, int offset, int value)
+{
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	if (!((offset == UART_IER) && (value & UART_IER_UUE)))
+		writeb(value, p->membase + offset);
 }
 
-static void
-serial_out(struct uart_8250_port *up, int offset, int value)
+static void dwapb_serial_out(struct uart_port *p, int offset, int value)
 {
-	/* Save the offset before it's remapped */
 	int save_offset = offset;
-	offset = map_8250_out_reg(up, offset) << up->port.regshift;
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	/* Save the LCR value so it can be re-written when a
+	 * Busy Detect interrupt occurs. */
+	if (save_offset == UART_LCR) {
+		struct uart_8250_port *up = (struct uart_8250_port *)p;
+		up->lcr = value;
+	}
+	writeb(value, p->membase + offset);
+	/* Read the IER to ensure any interrupt is cleared before
+	 * returning from ISR. */
+	if (save_offset == UART_TX || save_offset == UART_IER)
+		value = p->serial_in(p, UART_IER);
+}
 
-	switch (up->port.iotype) {
+static unsigned int io_serial_in(struct uart_port *p, int offset)
+{
+	offset = map_8250_in_reg(p, offset) << p->regshift;
+	return inb(p->iobase + offset);
+}
+
+static void io_serial_out(struct uart_port *p, int offset, int value)
+{
+	offset = map_8250_out_reg(p, offset) << p->regshift;
+	outb(value, p->iobase + offset);
+}
+
+static void set_io_from_upio(struct uart_port *p)
+{
+	switch (p->iotype) {
 	case UPIO_HUB6:
-		outb(up->port.hub6 - 1 + offset, up->port.iobase);
-		outb(value, up->port.iobase + 1);
+		p->serial_in = hub6_serial_in;
+		p->serial_out = hub6_serial_out;
 		break;
 
 	case UPIO_MEM:
-		writeb(value, up->port.membase + offset);
+		p->serial_in = mem_serial_in;
+		p->serial_out = mem_serial_out;
 		break;
 
 	case UPIO_RM9000:
 	case UPIO_MEM32:
-		writel(value, up->port.membase + offset);
+		p->serial_in = mem32_serial_in;
+		p->serial_out = mem32_serial_out;
 		break;
 
 #ifdef CONFIG_SERIAL_8250_AU1X00
 	case UPIO_AU:
-		__raw_writel(value, up->port.membase + offset);
+		p->serial_in = au_serial_in;
+		p->serial_out = au_serial_out;
 		break;
 #endif
 	case UPIO_TSI:
-		if (!((offset == UART_IER) && (value & UART_IER_UUE)))
-			writeb(value, up->port.membase + offset);
+		p->serial_in = tsi_serial_in;
+		p->serial_out = tsi_serial_out;
 		break;
 
 	case UPIO_DWAPB:
-		/* Save the LCR value so it can be re-written when a
-		 * Busy Detect interrupt occurs. */
-		if (save_offset == UART_LCR)
-			up->lcr = value;
-		writeb(value, up->port.membase + offset);
-		/* Read the IER to ensure any interrupt is cleared before
-		 * returning from ISR. */
-		if (save_offset == UART_TX || save_offset == UART_IER)
-			value = serial_in(up, UART_IER);
+		p->serial_in = mem_serial_in;
+		p->serial_out = dwapb_serial_out;
 		break;
 
 	default:
-		outb(value, up->port.iobase + offset);
+		p->serial_in = io_serial_in;
+		p->serial_out = io_serial_out;
+		break;
 	}
 }
 
 static void
 serial_out_sync(struct uart_8250_port *up, int offset, int value)
 {
-	switch (up->port.iotype) {
+	struct uart_port *p = &up->port;
+	switch (p->iotype) {
 	case UPIO_MEM:
 	case UPIO_MEM32:
 #ifdef CONFIG_SERIAL_8250_AU1X00
 	case UPIO_AU:
 #endif
 	case UPIO_DWAPB:
-		serial_out(up, offset, value);
-		serial_in(up, UART_LCR);	/* safe, no side-effects */
+		p->serial_out(p, offset, value);
+		p->serial_in(p, UART_LCR);	/* safe, no side-effects */
 		break;
 	default:
-		serial_out(up, offset, value);
+		p->serial_out(p, offset, value);
 	}
 }
 
+#define serial_in(up, offset)		\
+	(up->port.serial_in(&(up)->port, (offset)))
+#define serial_out(up, offset, value)	\
+	(up->port.serial_out(&(up)->port, (offset), (value)))
 /*
  * We used to support using pause I/O for certain machines.  We
  * haven't supported this for a while, but just in case it's badly
@@ -2576,6 +2638,7 @@ static void __init serial8250_isa_init_ports(void)
 		up->port.membase  = old_serial_port[i].iomem_base;
 		up->port.iotype   = old_serial_port[i].io_type;
 		up->port.regshift = old_serial_port[i].iomem_reg_shift;
+		set_io_from_upio(&up->port);
 		if (share_irqs)
 			up->port.flags |= UPF_SHARE_IRQ;
 	}
@@ -2769,6 +2832,13 @@ int __init early_serial_setup(struct uart_port *port)
 	p->flags        = port->flags;
 	p->mapbase      = port->mapbase;
 	p->private_data = port->private_data;
+
+	set_io_from_upio(p);
+	if (port->serial_in)
+		p->serial_in = port->serial_in;
+	if (port->serial_out)
+		p->serial_out = port->serial_out;
+
 	return 0;
 }
 
@@ -2833,6 +2903,8 @@ static int __devinit serial8250_probe(struct platform_device *dev)
 		port.mapbase		= p->mapbase;
 		port.hub6		= p->hub6;
 		port.private_data	= p->private_data;
+		port.serial_in		= p->serial_in;
+		port.serial_out		= p->serial_out;
 		port.dev		= &dev->dev;
 		if (share_irqs)
 			port.flags |= UPF_SHARE_IRQ;
@@ -2986,6 +3058,12 @@ int serial8250_register_port(struct uart_port *port)
 		uart->port.private_data = port->private_data;
 		if (port->dev)
 			uart->port.dev = port->dev;
+		set_io_from_upio(&uart->port);
+		/* Possibly override default I/O functions.  */
+		if (port->serial_in)
+			uart->port.serial_in = port->serial_in;
+		if (port->serial_out)
+			uart->port.serial_out = port->serial_out;
 
 		ret = uart_add_one_port(&serial8250_reg, &uart->port);
 		if (ret == 0)
diff --git a/include/linux/serial_8250.h b/include/linux/serial_8250.h
index 3d37c94abbc8..77d83d929f2c 100644
--- a/include/linux/serial_8250.h
+++ b/include/linux/serial_8250.h
@@ -28,6 +28,8 @@ struct plat_serial8250_port {
 	unsigned char	iotype;		/* UPIO_* */
 	unsigned char	hub6;
 	upf_t		flags;		/* UPF_* flags */
+	unsigned int	(*serial_in)(struct uart_port *, int);
+	void		(*serial_out)(struct uart_port *, int, int);
 };
 
 /*
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index 2395969faa04..60061f44f3d8 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -248,6 +248,8 @@ struct uart_port {
 	spinlock_t		lock;			/* port lock */
 	unsigned long		iobase;			/* in/out[bwl] */
 	unsigned char __iomem	*membase;		/* read/write[bwl] */
+	unsigned int		(*serial_in)(struct uart_port *, int);
+	void			(*serial_out)(struct uart_port *, int, int);
 	unsigned int		irq;			/* irq number */
 	unsigned int		uartclk;		/* base uart clock */
 	unsigned int		fifosize;		/* tx fifo size */
-- 
cgit v1.2.3


From 8e23fcc89c8091790903927449f8efb9b4e23960 Mon Sep 17 00:00:00 2001
From: David Daney <ddaney@caviumnetworks.com>
Date: Fri, 2 Jan 2009 13:49:54 +0000
Subject: Serial: Allow port type to be specified when calling
 serial8250_register_port.

Add flag value UPF_FIXED_TYPE which specifies that the UART type is
known and should not be probed.  For this case the UARTs properties
are just copied out of the uart_config entry.

This allows us to keep SOC specific 8250 probe code out of 8250.c.  In
this case we know the serial hardware will not be changing as it is on
the same silicon as the CPU, and we can specify it with certainty in
the board/cpu setup code.

The alternative is to load up 8250.c with a bunch of OCTEON specific
special cases in the probing code.

Signed-off-by: David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/8250.c        | 9 +++++++++
 drivers/serial/serial_core.c | 7 +++++--
 include/linux/serial_8250.h  | 1 +
 include/linux/serial_core.h  | 2 ++
 4 files changed, 17 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index 849af9d21feb..3ae497422db5 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -2903,6 +2903,7 @@ static int __devinit serial8250_probe(struct platform_device *dev)
 		port.mapbase		= p->mapbase;
 		port.hub6		= p->hub6;
 		port.private_data	= p->private_data;
+		port.type		= p->type;
 		port.serial_in		= p->serial_in;
 		port.serial_out		= p->serial_out;
 		port.dev		= &dev->dev;
@@ -3058,6 +3059,14 @@ int serial8250_register_port(struct uart_port *port)
 		uart->port.private_data = port->private_data;
 		if (port->dev)
 			uart->port.dev = port->dev;
+
+		if (port->flags & UPF_FIXED_TYPE) {
+			uart->port.type = port->type;
+			uart->port.fifosize = uart_config[port->type].fifo_size;
+			uart->capabilities = uart_config[port->type].flags;
+			uart->tx_loadsz = uart_config[port->type].tx_loadsz;
+		}
+
 		set_io_from_upio(&uart->port);
 		/* Possibly override default I/O functions.  */
 		if (port->serial_in)
diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c
index 9425ed69e0f7..dc68b7e0c930 100644
--- a/drivers/serial/serial_core.c
+++ b/drivers/serial/serial_core.c
@@ -2179,11 +2179,14 @@ uart_configure_port(struct uart_driver *drv, struct uart_state *state,
 	 * Now do the auto configuration stuff.  Note that config_port
 	 * is expected to claim the resources and map the port for us.
 	 */
-	flags = UART_CONFIG_TYPE;
+	flags = 0;
 	if (port->flags & UPF_AUTO_IRQ)
 		flags |= UART_CONFIG_IRQ;
 	if (port->flags & UPF_BOOT_AUTOCONF) {
-		port->type = PORT_UNKNOWN;
+		if (!(port->flags & UPF_FIXED_TYPE)) {
+			port->type = PORT_UNKNOWN;
+			flags |= UART_CONFIG_TYPE;
+		}
 		port->ops->config_port(port, flags);
 	}
 
diff --git a/include/linux/serial_8250.h b/include/linux/serial_8250.h
index 77d83d929f2c..d4d2a78ad43e 100644
--- a/include/linux/serial_8250.h
+++ b/include/linux/serial_8250.h
@@ -28,6 +28,7 @@ struct plat_serial8250_port {
 	unsigned char	iotype;		/* UPIO_* */
 	unsigned char	hub6;
 	upf_t		flags;		/* UPF_* flags */
+	unsigned int	type;		/* If UPF_FIXED_TYPE */
 	unsigned int	(*serial_in)(struct uart_port *, int);
 	void		(*serial_out)(struct uart_port *, int, int);
 };
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index 60061f44f3d8..f155252f148c 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -295,6 +295,8 @@ struct uart_port {
 #define UPF_MAGIC_MULTIPLIER	((__force upf_t) (1 << 16))
 #define UPF_CONS_FLOW		((__force upf_t) (1 << 23))
 #define UPF_SHARE_IRQ		((__force upf_t) (1 << 24))
+/* The exact UART type is known and should not be probed.  */
+#define UPF_FIXED_TYPE		((__force upf_t) (1 << 27))
 #define UPF_BOOT_AUTOCONF	((__force upf_t) (1 << 28))
 #define UPF_FIXED_PORT		((__force upf_t) (1 << 29))
 #define UPF_DEAD		((__force upf_t) (1 << 30))
-- 
cgit v1.2.3


From 6b06f19151c335ee0c5b61839fa4e6838182ebb8 Mon Sep 17 00:00:00 2001
From: David Daney <ddaney@caviumnetworks.com>
Date: Fri, 2 Jan 2009 13:50:00 +0000
Subject: Serial: UART driver changes for Cavium OCTEON.

Cavium UART implementation is not covered by existing uart_configS.
Define a new uart_config (PORT_OCTEON) which is specified by OCTEON
platform device registration code.

Signed-off-by: Tomaso Paoletti <tpaoletti@caviumnetworks.com>
Signed-off-by: David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/8250.c       | 7 +++++++
 include/linux/serial_core.h | 3 ++-
 2 files changed, 9 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index 3ae497422db5..daa00567bc44 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -279,6 +279,13 @@ static const struct serial8250_config uart_config[] = {
 		.fcr		= UART_FCR_ENABLE_FIFO | UART_FCR_R_TRIG_10,
 		.flags		= UART_CAP_FIFO,
 	},
+	[PORT_OCTEON] = {
+		.name		= "OCTEON",
+		.fifo_size	= 64,
+		.tx_loadsz	= 64,
+		.fcr		= UART_FCR_ENABLE_FIFO | UART_FCR_R_TRIG_10,
+		.flags		= UART_CAP_FIFO,
+	},
 };
 
 #if defined (CONFIG_SERIAL_8250_AU1X00)
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index f155252f148c..b4199841f1fc 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -40,7 +40,8 @@
 #define PORT_NS16550A	14
 #define PORT_XSCALE	15
 #define PORT_RM9000	16	/* PMC-Sierra RM9xxx internal UART */
-#define PORT_MAX_8250	16	/* max port ID */
+#define PORT_OCTEON	17	/* Cavium OCTEON internal UART */
+#define PORT_MAX_8250	17	/* max port ID */
 
 /*
  * ARM specific type numbers.  These are not currently guaranteed
-- 
cgit v1.2.3


From e65f0f8271b1b0452334e5da37fd35413a000de4 Mon Sep 17 00:00:00 2001
From: Flavio Leitner <fleitner@redhat.com>
Date: Fri, 2 Jan 2009 13:50:43 +0000
Subject: serial_8250: support for Sealevel Systems Model 7803 COMM+8

Add support for Sealevel Systems Model 7803 COMM+8

Signed-off-by: Flavio Leitner <fleitner@redhat.com>
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/8250_pci.c | 3 +++
 include/linux/pci_ids.h   | 1 +
 2 files changed, 4 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/serial/8250_pci.c b/drivers/serial/8250_pci.c
index 2a2e1c717e8e..c088146b7513 100644
--- a/drivers/serial/8250_pci.c
+++ b/drivers/serial/8250_pci.c
@@ -2287,6 +2287,9 @@ static struct pci_device_id serial_pci_tbl[] = {
 	{	PCI_VENDOR_ID_SEALEVEL, PCI_DEVICE_ID_SEALEVEL_COMM8,
 		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
 		pbn_b2_8_115200 },
+	{	PCI_VENDOR_ID_SEALEVEL, PCI_DEVICE_ID_SEALEVEL_7803,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
+		pbn_b2_8_460800 },
 	{	PCI_VENDOR_ID_SEALEVEL, PCI_DEVICE_ID_SEALEVEL_UCOMM8,
 		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
 		pbn_b2_8_115200 },
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index fa83dfefc5e0..218c73b1e6d4 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1796,6 +1796,7 @@
 #define PCI_DEVICE_ID_SEALEVEL_UCOMM232	0x7202
 #define PCI_DEVICE_ID_SEALEVEL_COMM4	0x7401
 #define PCI_DEVICE_ID_SEALEVEL_COMM8	0x7801
+#define PCI_DEVICE_ID_SEALEVEL_7803	0x7803
 #define PCI_DEVICE_ID_SEALEVEL_UCOMM8	0x7804
 
 #define PCI_VENDOR_ID_HYPERCOPE		0x1365
-- 
cgit v1.2.3


From aecde8b53b8ee1330a5a8206200f0d6b8845a6e0 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Tue, 30 Dec 2008 07:14:19 -0300
Subject: V4L/DVB (10141): v4l2: debugging API changed to match against driver
 name instead of ID.

Since the i2c driver ID will be removed in the near future we have to
modify the v4l2 debugging API to use the driver name instead of driver ID.

Note that this API is not used in applications other than v4l2-dbg.cpp
as it is for debugging and testing only.

Should anyone use the old VIDIOC_G_CHIP_IDENT, then this will be logged
with a warning that it is deprecated and will be removed in 2.6.30.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 Documentation/video4linux/v4l2-framework.txt    |  2 +-
 drivers/media/video/bt8xx/bttv-driver.c         |  9 +++--
 drivers/media/video/cafe_ccic.c                 |  7 ++--
 drivers/media/video/cs5345.c                    | 13 +++----
 drivers/media/video/cs53l32a.c                  |  2 +-
 drivers/media/video/cx18/cx18-i2c.c             | 28 +-------------
 drivers/media/video/cx18/cx18-i2c.h             |  1 -
 drivers/media/video/cx18/cx18-ioctl.c           | 41 ++++++++------------
 drivers/media/video/cx23885/cx23885-video.c     |  8 ++--
 drivers/media/video/cx25840/cx25840-core.c      | 13 +++----
 drivers/media/video/cx88/cx88-video.c           | 13 ++++---
 drivers/media/video/em28xx/em28xx-video.c       | 28 +++++++-------
 drivers/media/video/ivtv/ivtv-driver.c          |  7 ++--
 drivers/media/video/ivtv/ivtv-ioctl.c           | 21 +++++-----
 drivers/media/video/m52790.c                    | 13 +++----
 drivers/media/video/msp3400-driver.c            |  2 +-
 drivers/media/video/mt9m001.c                   | 19 ++++-----
 drivers/media/video/mt9m111.c                   | 19 ++++-----
 drivers/media/video/mt9t031.c                   | 18 ++++-----
 drivers/media/video/mt9v022.c                   | 19 ++++-----
 drivers/media/video/ov7670.c                    |  2 +-
 drivers/media/video/ov772x.c                    |  7 ++--
 drivers/media/video/pvrusb2/pvrusb2-hdw.c       | 11 +++---
 drivers/media/video/pvrusb2/pvrusb2-hdw.h       |  4 +-
 drivers/media/video/pvrusb2/pvrusb2-v4l2.c      |  6 +--
 drivers/media/video/saa7115.c                   | 13 +++----
 drivers/media/video/saa7127.c                   | 13 +++----
 drivers/media/video/saa7134/saa6752hs.c         |  2 +-
 drivers/media/video/saa7134/saa7134-empress.c   | 14 +++----
 drivers/media/video/saa7134/saa7134-video.c     |  9 +++--
 drivers/media/video/saa717x.c                   |  9 +++--
 drivers/media/video/soc_camera.c                |  6 +--
 drivers/media/video/tvaudio.c                   |  2 +-
 drivers/media/video/tvp5150.c                   | 13 +++----
 drivers/media/video/tw9910.c                    |  6 +--
 drivers/media/video/upd64031a.c                 | 13 +++----
 drivers/media/video/upd64083.c                  | 13 +++----
 drivers/media/video/usbvision/usbvision-video.c |  9 +++--
 drivers/media/video/v4l2-common.c               | 29 +++++++++-----
 drivers/media/video/v4l2-compat-ioctl32.c       |  3 +-
 drivers/media/video/v4l2-ioctl.c                | 15 +++++---
 drivers/media/video/v4l2-subdev.c               |  2 +-
 drivers/media/video/vp27smpx.c                  |  2 +-
 drivers/media/video/wm8739.c                    |  2 +-
 drivers/media/video/wm8775.c                    |  2 +-
 include/linux/videodev2.h                       | 51 ++++++++++++++++++-------
 include/media/soc_camera.h                      |  6 +--
 include/media/v4l2-chip-ident.h                 |  4 +-
 include/media/v4l2-common.h                     |  6 +--
 include/media/v4l2-int-device.h                 |  2 +-
 include/media/v4l2-ioctl.h                      |  6 +--
 include/media/v4l2-subdev.h                     |  6 +--
 52 files changed, 290 insertions(+), 281 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/video4linux/v4l2-framework.txt b/Documentation/video4linux/v4l2-framework.txt
index 3b483c1e0124..ff124374e9ba 100644
--- a/Documentation/video4linux/v4l2-framework.txt
+++ b/Documentation/video4linux/v4l2-framework.txt
@@ -184,7 +184,7 @@ may be NULL if the subdev driver does not support anything from that category.
 It looks like this:
 
 struct v4l2_subdev_core_ops {
-	int (*g_chip_ident)(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip);
+	int (*g_chip_ident)(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip);
 	int (*log_status)(struct v4l2_subdev *sd);
 	int (*init)(struct v4l2_subdev *sd, u32 val);
 	...
diff --git a/drivers/media/video/bt8xx/bttv-driver.c b/drivers/media/video/bt8xx/bttv-driver.c
index ebcb8e5e9c4d..d2f43bd2f841 100644
--- a/drivers/media/video/bt8xx/bttv-driver.c
+++ b/drivers/media/video/bt8xx/bttv-driver.c
@@ -2039,7 +2039,7 @@ static int bttv_log_status(struct file *file, void *f)
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int bttv_g_register(struct file *file, void *f,
-					struct v4l2_register *reg)
+					struct v4l2_dbg_register *reg)
 {
 	struct bttv_fh *fh = f;
 	struct bttv *btv = fh->btv;
@@ -2047,18 +2047,19 @@ static int bttv_g_register(struct file *file, void *f,
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	if (!v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_host(&reg->match))
 		return -EINVAL;
 
 	/* bt848 has a 12-bit register space */
 	reg->reg &= 0xfff;
 	reg->val = btread(reg->reg);
+	reg->size = 1;
 
 	return 0;
 }
 
 static int bttv_s_register(struct file *file, void *f,
-					struct v4l2_register *reg)
+					struct v4l2_dbg_register *reg)
 {
 	struct bttv_fh *fh = f;
 	struct bttv *btv = fh->btv;
@@ -2066,7 +2067,7 @@ static int bttv_s_register(struct file *file, void *f,
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	if (!v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_host(&reg->match))
 		return -EINVAL;
 
 	/* bt848 has a 12-bit register space */
diff --git a/drivers/media/video/cafe_ccic.c b/drivers/media/video/cafe_ccic.c
index 476171cf5001..34a39d2e4703 100644
--- a/drivers/media/video/cafe_ccic.c
+++ b/drivers/media/video/cafe_ccic.c
@@ -859,7 +859,7 @@ static int __cafe_cam_reset(struct cafe_camera *cam)
  */
 static int cafe_cam_init(struct cafe_camera *cam)
 {
-	struct v4l2_chip_ident chip = { V4L2_CHIP_MATCH_I2C_ADDR, 0, 0, 0 };
+	struct v4l2_dbg_chip_ident chip;
 	int ret;
 
 	mutex_lock(&cam->s_mutex);
@@ -869,8 +869,9 @@ static int cafe_cam_init(struct cafe_camera *cam)
 	ret = __cafe_cam_reset(cam);
 	if (ret)
 		goto out;
-	chip.match_chip = cam->sensor->addr;
-	ret = __cafe_cam_cmd(cam, VIDIOC_G_CHIP_IDENT, &chip);
+	chip.match.type = V4L2_CHIP_MATCH_I2C_ADDR;
+	chip.match.addr = cam->sensor->addr;
+	ret = __cafe_cam_cmd(cam, VIDIOC_DBG_G_CHIP_IDENT, &chip);
 	if (ret)
 		goto out;
 	cam->sensor_type = chip.ident;
diff --git a/drivers/media/video/cs5345.c b/drivers/media/video/cs5345.c
index 70fcd0d5de13..14bebf8a116f 100644
--- a/drivers/media/video/cs5345.c
+++ b/drivers/media/video/cs5345.c
@@ -95,25 +95,24 @@ static int cs5345_s_ctrl(struct v4l2_subdev *sd, struct v4l2_control *ctrl)
 }
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
-static int cs5345_g_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int cs5345_g_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
+	reg->size = 1;
 	reg->val = cs5345_read(sd, reg->reg & 0x1f);
 	return 0;
 }
 
-static int cs5345_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int cs5345_s_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -122,7 +121,7 @@ static int cs5345_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
 }
 #endif
 
-static int cs5345_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip)
+static int cs5345_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
diff --git a/drivers/media/video/cs53l32a.c b/drivers/media/video/cs53l32a.c
index cb65d519cf78..7292a6316e63 100644
--- a/drivers/media/video/cs53l32a.c
+++ b/drivers/media/video/cs53l32a.c
@@ -102,7 +102,7 @@ static int cs53l32a_s_ctrl(struct v4l2_subdev *sd, struct v4l2_control *ctrl)
 	return 0;
 }
 
-static int cs53l32a_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip)
+static int cs53l32a_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
diff --git a/drivers/media/video/cx18/cx18-i2c.c b/drivers/media/video/cx18/cx18-i2c.c
index 8941f58bed7f..83e1c6333126 100644
--- a/drivers/media/video/cx18/cx18-i2c.c
+++ b/drivers/media/video/cx18/cx18-i2c.c
@@ -242,7 +242,7 @@ int cx18_call_i2c_client(struct cx18 *cx, int addr, unsigned cmd, void *arg)
 			return retval;
 		}
 	}
-	if (cmd != VIDIOC_G_CHIP_IDENT)
+	if (cmd != VIDIOC_DBG_G_CHIP_IDENT)
 		CX18_ERR("i2c addr 0x%02x not found for cmd 0x%x!\n",
 			       addr, cmd);
 	return -ENODEV;
@@ -268,17 +268,6 @@ static int cx18_i2c_id_addr(struct cx18 *cx, u32 id)
 	return retval;
 }
 
-/* Find the i2c device name matching the DRIVERID */
-static const char *cx18_i2c_id_name(u32 id)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(hw_driverids); i++)
-		if (hw_driverids[i] == id)
-			return hw_devicenames[i];
-	return "unknown device";
-}
-
 /* Find the i2c device name matching the CX18_HW_ flag */
 static const char *cx18_i2c_hw_name(u32 hw)
 {
@@ -326,21 +315,6 @@ int cx18_i2c_hw(struct cx18 *cx, u32 hw, unsigned int cmd, void *arg)
 	return cx18_call_i2c_client(cx, addr, cmd, arg);
 }
 
-/* Calls i2c device based on I2C driver ID. */
-int cx18_i2c_id(struct cx18 *cx, u32 id, unsigned int cmd, void *arg)
-{
-	int addr;
-
-	addr = cx18_i2c_id_addr(cx, id);
-	if (addr < 0) {
-		if (cmd != VIDIOC_G_CHIP_IDENT)
-			CX18_ERR("i2c ID 0x%08x (%s) not found for cmd 0x%x!\n",
-				id, cx18_i2c_id_name(id), cmd);
-		return addr;
-	}
-	return cx18_call_i2c_client(cx, addr, cmd, arg);
-}
-
 /* broadcast cmd for all I2C clients and for the gpio subsystem */
 void cx18_call_i2c_clients(struct cx18 *cx, unsigned int cmd, void *arg)
 {
diff --git a/drivers/media/video/cx18/cx18-i2c.h b/drivers/media/video/cx18/cx18-i2c.h
index 113c3f9a2cc0..4869739013bd 100644
--- a/drivers/media/video/cx18/cx18-i2c.h
+++ b/drivers/media/video/cx18/cx18-i2c.h
@@ -23,7 +23,6 @@
 
 int cx18_i2c_hw_addr(struct cx18 *cx, u32 hw);
 int cx18_i2c_hw(struct cx18 *cx, u32 hw, unsigned int cmd, void *arg);
-int cx18_i2c_id(struct cx18 *cx, u32 id, unsigned int cmd, void *arg);
 int cx18_call_i2c_client(struct cx18 *cx, int addr, unsigned cmd, void *arg);
 void cx18_call_i2c_clients(struct cx18 *cx, unsigned int cmd, void *arg);
 int cx18_i2c_register(struct cx18 *cx, unsigned idx);
diff --git a/drivers/media/video/cx18/cx18-ioctl.c b/drivers/media/video/cx18/cx18-ioctl.c
index 8aa152b39545..7086aaba77d6 100644
--- a/drivers/media/video/cx18/cx18-ioctl.c
+++ b/drivers/media/video/cx18/cx18-ioctl.c
@@ -254,30 +254,24 @@ static int cx18_s_fmt_sliced_vbi_cap(struct file *file, void *fh,
 }
 
 static int cx18_g_chip_ident(struct file *file, void *fh,
-				struct v4l2_chip_ident *chip)
+				struct v4l2_dbg_chip_ident *chip)
 {
 	struct cx18 *cx = ((struct cx18_open_id *)fh)->cx;
 
 	chip->ident = V4L2_IDENT_NONE;
 	chip->revision = 0;
-	if (chip->match_type == V4L2_CHIP_MATCH_HOST) {
-		if (v4l2_chip_match_host(chip->match_type, chip->match_chip))
-			chip->ident = V4L2_IDENT_CX23418;
+	if (v4l2_chip_match_host(&chip->match)) {
+		chip->ident = V4L2_IDENT_CX23418;
 		return 0;
 	}
-	if (chip->match_type == V4L2_CHIP_MATCH_I2C_DRIVER)
-		return cx18_i2c_id(cx, chip->match_chip, VIDIOC_G_CHIP_IDENT,
-					chip);
-	if (chip->match_type == V4L2_CHIP_MATCH_I2C_ADDR)
-		return cx18_call_i2c_client(cx, chip->match_chip,
-						VIDIOC_G_CHIP_IDENT, chip);
-	return -EINVAL;
+	cx18_call_i2c_clients(cx, VIDIOC_DBG_G_CHIP_IDENT, chip);
+	return 0;
 }
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int cx18_cxc(struct cx18 *cx, unsigned int cmd, void *arg)
 {
-	struct v4l2_register *regs = arg;
+	struct v4l2_dbg_register *regs = arg;
 	unsigned long flags;
 
 	if (!capable(CAP_SYS_ADMIN))
@@ -286,6 +280,7 @@ static int cx18_cxc(struct cx18 *cx, unsigned int cmd, void *arg)
 		return -EINVAL;
 
 	spin_lock_irqsave(&cx18_cards_lock, flags);
+	regs->size = 4;
 	if (cmd == VIDIOC_DBG_G_REGISTER)
 		regs->val = cx18_read_enc(cx, regs->reg);
 	else
@@ -295,31 +290,25 @@ static int cx18_cxc(struct cx18 *cx, unsigned int cmd, void *arg)
 }
 
 static int cx18_g_register(struct file *file, void *fh,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct cx18 *cx = ((struct cx18_open_id *)fh)->cx;
 
-	if (v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (v4l2_chip_match_host(&reg->match))
 		return cx18_cxc(cx, VIDIOC_DBG_G_REGISTER, reg);
-	if (reg->match_type == V4L2_CHIP_MATCH_I2C_DRIVER)
-		return cx18_i2c_id(cx, reg->match_chip, VIDIOC_DBG_G_REGISTER,
-					reg);
-	return cx18_call_i2c_client(cx, reg->match_chip, VIDIOC_DBG_G_REGISTER,
-					reg);
+	cx18_call_i2c_clients(cx, VIDIOC_DBG_G_REGISTER, reg);
+	return 0;
 }
 
 static int cx18_s_register(struct file *file, void *fh,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct cx18 *cx = ((struct cx18_open_id *)fh)->cx;
 
-	if (v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (v4l2_chip_match_host(&reg->match))
 		return cx18_cxc(cx, VIDIOC_DBG_S_REGISTER, reg);
-	if (reg->match_type == V4L2_CHIP_MATCH_I2C_DRIVER)
-		return cx18_i2c_id(cx, reg->match_chip, VIDIOC_DBG_S_REGISTER,
-					reg);
-	return cx18_call_i2c_client(cx, reg->match_chip, VIDIOC_DBG_S_REGISTER,
-					reg);
+	cx18_call_i2c_clients(cx, VIDIOC_DBG_S_REGISTER, reg);
+	return 0;
 }
 #endif
 
diff --git a/drivers/media/video/cx23885/cx23885-video.c b/drivers/media/video/cx23885/cx23885-video.c
index 637c4d008846..2d81c4d04340 100644
--- a/drivers/media/video/cx23885/cx23885-video.c
+++ b/drivers/media/video/cx23885/cx23885-video.c
@@ -1326,11 +1326,11 @@ static int vidioc_s_frequency(struct file *file, void *priv,
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int vidioc_g_register(struct file *file, void *fh,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct cx23885_dev *dev = ((struct cx23885_fh *)fh)->dev;
 
-	if (!v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_host(&reg->match))
 		return -EINVAL;
 
 	cx23885_call_i2c_clients(&dev->i2c_bus[2], VIDIOC_DBG_G_REGISTER, reg);
@@ -1339,11 +1339,11 @@ static int vidioc_g_register(struct file *file, void *fh,
 }
 
 static int vidioc_s_register(struct file *file, void *fh,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct cx23885_dev *dev = ((struct cx23885_fh *)fh)->dev;
 
-	if (!v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_host(&reg->match))
 		return -EINVAL;
 
 	cx23885_call_i2c_clients(&dev->i2c_bus[2], VIDIOC_DBG_S_REGISTER, reg);
diff --git a/drivers/media/video/cx25840/cx25840-core.c b/drivers/media/video/cx25840/cx25840-core.c
index 2ad277189da8..88f2fd32bfe3 100644
--- a/drivers/media/video/cx25840/cx25840-core.c
+++ b/drivers/media/video/cx25840/cx25840-core.c
@@ -1120,25 +1120,24 @@ static int cx25840_init(struct v4l2_subdev *sd, u32 val)
 }
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
-static int cx25840_g_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int cx25840_g_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
+	reg->size = 1;
 	reg->val = cx25840_read(client, reg->reg & 0x0fff);
 	return 0;
 }
 
-static int cx25840_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int cx25840_s_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -1362,7 +1361,7 @@ static int cx25840_reset(struct v4l2_subdev *sd, u32 val)
 	return 0;
 }
 
-static int cx25840_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip)
+static int cx25840_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip)
 {
 	struct cx25840_state *state = to_state(sd);
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
diff --git a/drivers/media/video/cx88/cx88-video.c b/drivers/media/video/cx88/cx88-video.c
index b93b7ab99d8c..791e69d804f9 100644
--- a/drivers/media/video/cx88/cx88-video.c
+++ b/drivers/media/video/cx88/cx88-video.c
@@ -1447,25 +1447,26 @@ static int vidioc_s_frequency (struct file *file, void *priv,
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int vidioc_g_register (struct file *file, void *fh,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct cx88_core *core = ((struct cx8800_fh*)fh)->dev->core;
 
-	if (!v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_host(&reg->match))
 		return -EINVAL;
 	/* cx2388x has a 24-bit register space */
-	reg->val = cx_read(reg->reg&0xffffff);
+	reg->val = cx_read(reg->reg & 0xffffff);
+	reg->size = 4;
 	return 0;
 }
 
 static int vidioc_s_register (struct file *file, void *fh,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct cx88_core *core = ((struct cx8800_fh*)fh)->dev->core;
 
-	if (!v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_host(&reg->match))
 		return -EINVAL;
-	cx_write(reg->reg&0xffffff, reg->val);
+	cx_write(reg->reg & 0xffffff, reg->val);
 	return 0;
 }
 #endif
diff --git a/drivers/media/video/em28xx/em28xx-video.c b/drivers/media/video/em28xx/em28xx-video.c
index 9cb7c64a88fa..416b691c33c1 100644
--- a/drivers/media/video/em28xx/em28xx-video.c
+++ b/drivers/media/video/em28xx/em28xx-video.c
@@ -1154,7 +1154,7 @@ static int em28xx_reg_len(int reg)
 }
 
 static int vidioc_g_chip_ident(struct file *file, void *priv,
-	       struct v4l2_chip_ident *chip)
+	       struct v4l2_dbg_chip_ident *chip)
 {
 	struct em28xx_fh      *fh  = priv;
 	struct em28xx         *dev = fh->dev;
@@ -1162,20 +1162,20 @@ static int vidioc_g_chip_ident(struct file *file, void *priv,
 	chip->ident = V4L2_IDENT_NONE;
 	chip->revision = 0;
 
-	em28xx_i2c_call_clients(dev, VIDIOC_G_CHIP_IDENT, chip);
+	em28xx_i2c_call_clients(dev, VIDIOC_DBG_G_CHIP_IDENT, chip);
 
 	return 0;
 }
 
 
 static int vidioc_g_register(struct file *file, void *priv,
-			     struct v4l2_register *reg)
+			     struct v4l2_dbg_register *reg)
 {
 	struct em28xx_fh      *fh  = priv;
 	struct em28xx         *dev = fh->dev;
 	int ret;
 
-	switch (reg->match_type) {
+	switch (reg->match.type) {
 	case V4L2_CHIP_MATCH_AC97:
 		mutex_lock(&dev->lock);
 		ret = em28xx_read_ac97(dev, reg->reg);
@@ -1184,6 +1184,7 @@ static int vidioc_g_register(struct file *file, void *priv,
 			return ret;
 
 		reg->val = ret;
+		reg->size = 1;
 		return 0;
 	case V4L2_CHIP_MATCH_I2C_DRIVER:
 		em28xx_i2c_call_clients(dev, VIDIOC_DBG_G_REGISTER, reg);
@@ -1192,12 +1193,13 @@ static int vidioc_g_register(struct file *file, void *priv,
 		/* Not supported yet */
 		return -EINVAL;
 	default:
-		if (!v4l2_chip_match_host(reg->match_type, reg->match_chip))
+		if (!v4l2_chip_match_host(&reg->match))
 			return -EINVAL;
 	}
 
 	/* Match host */
-	if (em28xx_reg_len(reg->reg) == 1) {
+	reg->size = em28xx_reg_len(reg->reg);
+	if (reg->size == 1) {
 		mutex_lock(&dev->lock);
 		ret = em28xx_read_reg(dev, reg->reg);
 		mutex_unlock(&dev->lock);
@@ -1207,7 +1209,7 @@ static int vidioc_g_register(struct file *file, void *priv,
 
 		reg->val = ret;
 	} else {
-		__le64 val = 0;
+		__le16 val = 0;
 		mutex_lock(&dev->lock);
 		ret = em28xx_read_reg_req_len(dev, USB_REQ_GET_STATUS,
 						   reg->reg, (char *)&val, 2);
@@ -1215,21 +1217,21 @@ static int vidioc_g_register(struct file *file, void *priv,
 		if (ret < 0)
 			return ret;
 
-		reg->val = le64_to_cpu(val);
+		reg->val = le16_to_cpu(val);
 	}
 
 	return 0;
 }
 
 static int vidioc_s_register(struct file *file, void *priv,
-			     struct v4l2_register *reg)
+			     struct v4l2_dbg_register *reg)
 {
 	struct em28xx_fh      *fh  = priv;
 	struct em28xx         *dev = fh->dev;
-	__le64 buf;
+	__le16 buf;
 	int    rc;
 
-	switch (reg->match_type) {
+	switch (reg->match.type) {
 	case V4L2_CHIP_MATCH_AC97:
 		mutex_lock(&dev->lock);
 		rc = em28xx_write_ac97(dev, reg->reg, reg->val);
@@ -1243,12 +1245,12 @@ static int vidioc_s_register(struct file *file, void *priv,
 		/* Not supported yet */
 		return -EINVAL;
 	default:
-		if (!v4l2_chip_match_host(reg->match_type, reg->match_chip))
+		if (!v4l2_chip_match_host(&reg->match))
 			return -EINVAL;
 	}
 
 	/* Match host */
-	buf = cpu_to_le64(reg->val);
+	buf = cpu_to_le16(reg->val);
 
 	mutex_lock(&dev->lock);
 	rc = em28xx_write_regs(dev, reg->reg, (char *)&buf,
diff --git a/drivers/media/video/ivtv/ivtv-driver.c b/drivers/media/video/ivtv/ivtv-driver.c
index 08b762951759..e8e5921cdc34 100644
--- a/drivers/media/video/ivtv/ivtv-driver.c
+++ b/drivers/media/video/ivtv/ivtv-driver.c
@@ -902,18 +902,19 @@ static void ivtv_load_and_init_modules(struct ivtv *itv)
 	}
 
 	if (hw & IVTV_HW_SAA711X) {
-		struct v4l2_chip_ident v = { V4L2_CHIP_MATCH_I2C_DRIVER, I2C_DRIVERID_SAA711X };
+		struct v4l2_dbg_chip_ident v;
 
 		/* determine the exact saa711x model */
 		itv->hw_flags &= ~IVTV_HW_SAA711X;
 
+		v.match.type = V4L2_CHIP_MATCH_I2C_DRIVER;
+		strlcpy(v.match.name, "saa7115", sizeof(v.match.name));
 		ivtv_call_hw(itv, IVTV_HW_SAA711X, core, g_chip_ident, &v);
 		if (v.ident == V4L2_IDENT_SAA7114) {
 			itv->hw_flags |= IVTV_HW_SAA7114;
 			/* VBI is not yet supported by the saa7114 driver. */
 			itv->v4l2_cap &= ~(V4L2_CAP_SLICED_VBI_CAPTURE|V4L2_CAP_VBI_CAPTURE);
-		}
-		else {
+		} else {
 			itv->hw_flags |= IVTV_HW_SAA7115;
 		}
 		itv->vbi.raw_decoder_line_size = 1443;
diff --git a/drivers/media/video/ivtv/ivtv-ioctl.c b/drivers/media/video/ivtv/ivtv-ioctl.c
index 1f6ca93b9840..f6b3ef6e691b 100644
--- a/drivers/media/video/ivtv/ivtv-ioctl.c
+++ b/drivers/media/video/ivtv/ivtv-ioctl.c
@@ -674,19 +674,19 @@ static int ivtv_s_fmt_vid_out_overlay(struct file *file, void *fh, struct v4l2_f
 	return ret;
 }
 
-static int ivtv_g_chip_ident(struct file *file, void *fh, struct v4l2_chip_ident *chip)
+static int ivtv_g_chip_ident(struct file *file, void *fh, struct v4l2_dbg_chip_ident *chip)
 {
 	struct ivtv *itv = ((struct ivtv_open_id *)fh)->itv;
 
 	chip->ident = V4L2_IDENT_NONE;
 	chip->revision = 0;
-	if (chip->match_type == V4L2_CHIP_MATCH_HOST) {
-		if (v4l2_chip_match_host(chip->match_type, chip->match_chip))
+	if (chip->match.type == V4L2_CHIP_MATCH_HOST) {
+		if (v4l2_chip_match_host(&chip->match))
 			chip->ident = itv->has_cx23415 ? V4L2_IDENT_CX23415 : V4L2_IDENT_CX23416;
 		return 0;
 	}
-	if (chip->match_type != V4L2_CHIP_MATCH_I2C_DRIVER &&
-	    chip->match_type != V4L2_CHIP_MATCH_I2C_ADDR)
+	if (chip->match.type != V4L2_CHIP_MATCH_I2C_DRIVER &&
+	    chip->match.type != V4L2_CHIP_MATCH_I2C_ADDR)
 		return -EINVAL;
 	/* TODO: is this correct? */
 	return ivtv_call_all_err(itv, core, g_chip_ident, chip);
@@ -695,7 +695,7 @@ static int ivtv_g_chip_ident(struct file *file, void *fh, struct v4l2_chip_ident
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int ivtv_itvc(struct ivtv *itv, unsigned int cmd, void *arg)
 {
-	struct v4l2_register *regs = arg;
+	struct v4l2_dbg_register *regs = arg;
 	volatile u8 __iomem *reg_start;
 
 	if (!capable(CAP_SYS_ADMIN))
@@ -710,6 +710,7 @@ static int ivtv_itvc(struct ivtv *itv, unsigned int cmd, void *arg)
 	else
 		return -EINVAL;
 
+	regs->size = 4;
 	if (cmd == VIDIOC_DBG_G_REGISTER)
 		regs->val = readl(regs->reg + reg_start);
 	else
@@ -717,11 +718,11 @@ static int ivtv_itvc(struct ivtv *itv, unsigned int cmd, void *arg)
 	return 0;
 }
 
-static int ivtv_g_register(struct file *file, void *fh, struct v4l2_register *reg)
+static int ivtv_g_register(struct file *file, void *fh, struct v4l2_dbg_register *reg)
 {
 	struct ivtv *itv = ((struct ivtv_open_id *)fh)->itv;
 
-	if (v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (v4l2_chip_match_host(&reg->match))
 		return ivtv_itvc(itv, VIDIOC_DBG_G_REGISTER, reg);
 	/* TODO: subdev errors should not be ignored, this should become a
 	   subdev helper function. */
@@ -729,11 +730,11 @@ static int ivtv_g_register(struct file *file, void *fh, struct v4l2_register *re
 	return 0;
 }
 
-static int ivtv_s_register(struct file *file, void *fh, struct v4l2_register *reg)
+static int ivtv_s_register(struct file *file, void *fh, struct v4l2_dbg_register *reg)
 {
 	struct ivtv *itv = ((struct ivtv_open_id *)fh)->itv;
 
-	if (v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (v4l2_chip_match_host(&reg->match))
 		return ivtv_itvc(itv, VIDIOC_DBG_S_REGISTER, reg);
 	/* TODO: subdev errors should not be ignored, this should become a
 	   subdev helper function. */
diff --git a/drivers/media/video/m52790.c b/drivers/media/video/m52790.c
index 07be14a9fe7b..de397ef57b44 100644
--- a/drivers/media/video/m52790.c
+++ b/drivers/media/video/m52790.c
@@ -80,29 +80,28 @@ static int m52790_s_routing(struct v4l2_subdev *sd, const struct v4l2_routing *r
 }
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
-static int m52790_g_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int m52790_g_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct m52790_state *state = to_state(sd);
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 	if (reg->reg != 0)
 		return -EINVAL;
+	reg->size = 1;
 	reg->val = state->input | state->output;
 	return 0;
 }
 
-static int m52790_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int m52790_s_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct m52790_state *state = to_state(sd);
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -115,7 +114,7 @@ static int m52790_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
 }
 #endif
 
-static int m52790_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip)
+static int m52790_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
diff --git a/drivers/media/video/msp3400-driver.c b/drivers/media/video/msp3400-driver.c
index b8577ade4050..4d7a91852117 100644
--- a/drivers/media/video/msp3400-driver.c
+++ b/drivers/media/video/msp3400-driver.c
@@ -733,7 +733,7 @@ static int msp_queryctrl(struct v4l2_subdev *sd, struct v4l2_queryctrl *qc)
 	return 0;
 }
 
-static int msp_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip)
+static int msp_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip)
 {
 	struct msp_state *state = to_state(sd);
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
diff --git a/drivers/media/video/mt9m001.c b/drivers/media/video/mt9m001.c
index 1a1a12453672..c1bf75ef2741 100644
--- a/drivers/media/video/mt9m001.c
+++ b/drivers/media/video/mt9m001.c
@@ -343,14 +343,14 @@ static int mt9m001_try_fmt(struct soc_camera_device *icd,
 }
 
 static int mt9m001_get_chip_id(struct soc_camera_device *icd,
-			       struct v4l2_chip_ident *id)
+			       struct v4l2_dbg_chip_ident *id)
 {
 	struct mt9m001 *mt9m001 = container_of(icd, struct mt9m001, icd);
 
-	if (id->match_type != V4L2_CHIP_MATCH_I2C_ADDR)
+	if (id->match.type != V4L2_CHIP_MATCH_I2C_ADDR)
 		return -EINVAL;
 
-	if (id->match_chip != mt9m001->client->addr)
+	if (id->match.addr != mt9m001->client->addr)
 		return -ENODEV;
 
 	id->ident	= mt9m001->model;
@@ -361,16 +361,17 @@ static int mt9m001_get_chip_id(struct soc_camera_device *icd,
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int mt9m001_get_register(struct soc_camera_device *icd,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct mt9m001 *mt9m001 = container_of(icd, struct mt9m001, icd);
 
-	if (reg->match_type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0xff)
+	if (reg->match.type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0xff)
 		return -EINVAL;
 
-	if (reg->match_chip != mt9m001->client->addr)
+	if (reg->match.addr != mt9m001->client->addr)
 		return -ENODEV;
 
+	reg->size = 2;
 	reg->val = reg_read(icd, reg->reg);
 
 	if (reg->val > 0xffff)
@@ -380,14 +381,14 @@ static int mt9m001_get_register(struct soc_camera_device *icd,
 }
 
 static int mt9m001_set_register(struct soc_camera_device *icd,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct mt9m001 *mt9m001 = container_of(icd, struct mt9m001, icd);
 
-	if (reg->match_type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0xff)
+	if (reg->match.type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0xff)
 		return -EINVAL;
 
-	if (reg->match_chip != mt9m001->client->addr)
+	if (reg->match.addr != mt9m001->client->addr)
 		return -ENODEV;
 
 	if (reg_write(icd, reg->reg, reg->val) < 0)
diff --git a/drivers/media/video/mt9m111.c b/drivers/media/video/mt9m111.c
index c89ea41fe259..5b8e20979cce 100644
--- a/drivers/media/video/mt9m111.c
+++ b/drivers/media/video/mt9m111.c
@@ -514,14 +514,14 @@ static int mt9m111_try_fmt(struct soc_camera_device *icd,
 }
 
 static int mt9m111_get_chip_id(struct soc_camera_device *icd,
-			       struct v4l2_chip_ident *id)
+			       struct v4l2_dbg_chip_ident *id)
 {
 	struct mt9m111 *mt9m111 = container_of(icd, struct mt9m111, icd);
 
-	if (id->match_type != V4L2_CHIP_MATCH_I2C_ADDR)
+	if (id->match.type != V4L2_CHIP_MATCH_I2C_ADDR)
 		return -EINVAL;
 
-	if (id->match_chip != mt9m111->client->addr)
+	if (id->match.addr != mt9m111->client->addr)
 		return -ENODEV;
 
 	id->ident	= mt9m111->model;
@@ -532,18 +532,19 @@ static int mt9m111_get_chip_id(struct soc_camera_device *icd,
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int mt9m111_get_register(struct soc_camera_device *icd,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	int val;
 
 	struct mt9m111 *mt9m111 = container_of(icd, struct mt9m111, icd);
 
-	if (reg->match_type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0x2ff)
+	if (reg->match.type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0x2ff)
 		return -EINVAL;
-	if (reg->match_chip != mt9m111->client->addr)
+	if (reg->match.addr != mt9m111->client->addr)
 		return -ENODEV;
 
 	val = mt9m111_reg_read(icd, reg->reg);
+	reg->size = 2;
 	reg->val = (u64)val;
 
 	if (reg->val > 0xffff)
@@ -553,14 +554,14 @@ static int mt9m111_get_register(struct soc_camera_device *icd,
 }
 
 static int mt9m111_set_register(struct soc_camera_device *icd,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct mt9m111 *mt9m111 = container_of(icd, struct mt9m111, icd);
 
-	if (reg->match_type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0x2ff)
+	if (reg->match.type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0x2ff)
 		return -EINVAL;
 
-	if (reg->match_chip != mt9m111->client->addr)
+	if (reg->match.addr != mt9m111->client->addr)
 		return -ENODEV;
 
 	if (mt9m111_reg_write(icd, reg->reg, reg->val) < 0)
diff --git a/drivers/media/video/mt9t031.c b/drivers/media/video/mt9t031.c
index 1a9d53966d06..349d8e365530 100644
--- a/drivers/media/video/mt9t031.c
+++ b/drivers/media/video/mt9t031.c
@@ -326,14 +326,14 @@ static int mt9t031_try_fmt(struct soc_camera_device *icd,
 }
 
 static int mt9t031_get_chip_id(struct soc_camera_device *icd,
-			       struct v4l2_chip_ident *id)
+			       struct v4l2_dbg_chip_ident *id)
 {
 	struct mt9t031 *mt9t031 = container_of(icd, struct mt9t031, icd);
 
-	if (id->match_type != V4L2_CHIP_MATCH_I2C_ADDR)
+	if (id->match.type != V4L2_CHIP_MATCH_I2C_ADDR)
 		return -EINVAL;
 
-	if (id->match_chip != mt9t031->client->addr)
+	if (id->match.addr != mt9t031->client->addr)
 		return -ENODEV;
 
 	id->ident	= mt9t031->model;
@@ -344,14 +344,14 @@ static int mt9t031_get_chip_id(struct soc_camera_device *icd,
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int mt9t031_get_register(struct soc_camera_device *icd,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct mt9t031 *mt9t031 = container_of(icd, struct mt9t031, icd);
 
-	if (reg->match_type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0xff)
+	if (reg->match.type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0xff)
 		return -EINVAL;
 
-	if (reg->match_chip != mt9t031->client->addr)
+	if (reg->match.addr != mt9t031->client->addr)
 		return -ENODEV;
 
 	reg->val = reg_read(icd, reg->reg);
@@ -363,14 +363,14 @@ static int mt9t031_get_register(struct soc_camera_device *icd,
 }
 
 static int mt9t031_set_register(struct soc_camera_device *icd,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct mt9t031 *mt9t031 = container_of(icd, struct mt9t031, icd);
 
-	if (reg->match_type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0xff)
+	if (reg->match.type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0xff)
 		return -EINVAL;
 
-	if (reg->match_chip != mt9t031->client->addr)
+	if (reg->match.addr != mt9t031->client->addr)
 		return -ENODEV;
 
 	if (reg_write(icd, reg->reg, reg->val) < 0)
diff --git a/drivers/media/video/mt9v022.c b/drivers/media/video/mt9v022.c
index 14a5f9c21ffa..b04c8cb1644d 100644
--- a/drivers/media/video/mt9v022.c
+++ b/drivers/media/video/mt9v022.c
@@ -422,14 +422,14 @@ static int mt9v022_try_fmt(struct soc_camera_device *icd,
 }
 
 static int mt9v022_get_chip_id(struct soc_camera_device *icd,
-			       struct v4l2_chip_ident *id)
+			       struct v4l2_dbg_chip_ident *id)
 {
 	struct mt9v022 *mt9v022 = container_of(icd, struct mt9v022, icd);
 
-	if (id->match_type != V4L2_CHIP_MATCH_I2C_ADDR)
+	if (id->match.type != V4L2_CHIP_MATCH_I2C_ADDR)
 		return -EINVAL;
 
-	if (id->match_chip != mt9v022->client->addr)
+	if (id->match.addr != mt9v022->client->addr)
 		return -ENODEV;
 
 	id->ident	= mt9v022->model;
@@ -440,16 +440,17 @@ static int mt9v022_get_chip_id(struct soc_camera_device *icd,
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int mt9v022_get_register(struct soc_camera_device *icd,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct mt9v022 *mt9v022 = container_of(icd, struct mt9v022, icd);
 
-	if (reg->match_type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0xff)
+	if (reg->match.type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0xff)
 		return -EINVAL;
 
-	if (reg->match_chip != mt9v022->client->addr)
+	if (reg->match.addr != mt9v022->client->addr)
 		return -ENODEV;
 
+	reg->size = 2;
 	reg->val = reg_read(icd, reg->reg);
 
 	if (reg->val > 0xffff)
@@ -459,14 +460,14 @@ static int mt9v022_get_register(struct soc_camera_device *icd,
 }
 
 static int mt9v022_set_register(struct soc_camera_device *icd,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct mt9v022 *mt9v022 = container_of(icd, struct mt9v022, icd);
 
-	if (reg->match_type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0xff)
+	if (reg->match.type != V4L2_CHIP_MATCH_I2C_ADDR || reg->reg > 0xff)
 		return -EINVAL;
 
-	if (reg->match_chip != mt9v022->client->addr)
+	if (reg->match.addr != mt9v022->client->addr)
 		return -ENODEV;
 
 	if (reg_write(icd, reg->reg, reg->val) < 0)
diff --git a/drivers/media/video/ov7670.c b/drivers/media/video/ov7670.c
index ea032f5f2f41..ca26b0c50cf2 100644
--- a/drivers/media/video/ov7670.c
+++ b/drivers/media/video/ov7670.c
@@ -1310,7 +1310,7 @@ static int ov7670_command(struct i2c_client *client, unsigned int cmd,
 		void *arg)
 {
 	switch (cmd) {
-	case VIDIOC_G_CHIP_IDENT:
+	case VIDIOC_DBG_G_CHIP_IDENT:
 		return v4l2_chip_ident_i2c_client(client, arg, V4L2_IDENT_OV7670, 0);
 
 	case VIDIOC_INT_RESET:
diff --git a/drivers/media/video/ov772x.c b/drivers/media/video/ov772x.c
index 54b736fcc07a..3c9e0ba974e9 100644
--- a/drivers/media/video/ov772x.c
+++ b/drivers/media/video/ov772x.c
@@ -724,7 +724,7 @@ static unsigned long ov772x_query_bus_param(struct soc_camera_device *icd)
 }
 
 static int ov772x_get_chip_id(struct soc_camera_device *icd,
-			      struct v4l2_chip_ident   *id)
+			      struct v4l2_dbg_chip_ident   *id)
 {
 	struct ov772x_priv *priv = container_of(icd, struct ov772x_priv, icd);
 
@@ -736,11 +736,12 @@ static int ov772x_get_chip_id(struct soc_camera_device *icd,
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int ov772x_get_register(struct soc_camera_device *icd,
-			       struct v4l2_register *reg)
+			       struct v4l2_dbg_register *reg)
 {
 	struct ov772x_priv *priv = container_of(icd, struct ov772x_priv, icd);
 	int                 ret;
 
+	reg->size = 1;
 	if (reg->reg > 0xff)
 		return -EINVAL;
 
@@ -754,7 +755,7 @@ static int ov772x_get_register(struct soc_camera_device *icd,
 }
 
 static int ov772x_set_register(struct soc_camera_device *icd,
-			       struct v4l2_register *reg)
+			       struct v4l2_dbg_register *reg)
 {
 	struct ov772x_priv *priv = container_of(icd, struct ov772x_priv, icd);
 
diff --git a/drivers/media/video/pvrusb2/pvrusb2-hdw.c b/drivers/media/video/pvrusb2/pvrusb2-hdw.c
index 4358079f1966..8fb92ac78c7b 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-hdw.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-hdw.c
@@ -4732,26 +4732,25 @@ static int pvr2_hdw_get_eeprom_addr(struct pvr2_hdw *hdw)
 
 
 int pvr2_hdw_register_access(struct pvr2_hdw *hdw,
-			     u32 match_type, u32 match_chip, u64 reg_id,
-			     int setFl,u64 *val_ptr)
+			     struct v4l2_dbg_match *match, u64 reg_id,
+			     int setFl, u64 *val_ptr)
 {
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 	struct pvr2_i2c_client *cp;
-	struct v4l2_register req;
+	struct v4l2_dbg_register req;
 	int stat = 0;
 	int okFl = 0;
 
 	if (!capable(CAP_SYS_ADMIN)) return -EPERM;
 
-	req.match_type = match_type;
-	req.match_chip = match_chip;
+	req.match = *match;
 	req.reg = reg_id;
 	if (setFl) req.val = *val_ptr;
 	mutex_lock(&hdw->i2c_list_lock); do {
 		list_for_each_entry(cp, &hdw->i2c_clients, list) {
 			if (!v4l2_chip_match_i2c_client(
 				    cp->client,
-				    req.match_type, req.match_chip)) {
+				    &req.match)) {
 				continue;
 			}
 			stat = pvr2_i2c_client_cmd(
diff --git a/drivers/media/video/pvrusb2/pvrusb2-hdw.h b/drivers/media/video/pvrusb2/pvrusb2-hdw.h
index 49482d1f2b28..1b4fec337c6b 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-hdw.h
+++ b/drivers/media/video/pvrusb2/pvrusb2-hdw.h
@@ -242,8 +242,8 @@ void pvr2_hdw_v4l_store_minor_number(struct pvr2_hdw *,
    setFl   - true to set the register, false to read it
    val_ptr - storage location for source / result. */
 int pvr2_hdw_register_access(struct pvr2_hdw *,
-			     u32 match_type, u32 match_chip,u64 reg_id,
-			     int setFl,u64 *val_ptr);
+			     struct v4l2_dbg_match *match, u64 reg_id,
+			     int setFl, u64 *val_ptr);
 
 /* The following entry points are all lower level things you normally don't
    want to worry about. */
diff --git a/drivers/media/video/pvrusb2/pvrusb2-v4l2.c b/drivers/media/video/pvrusb2/pvrusb2-v4l2.c
index b9aedceb2c44..878fd52a73b3 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-v4l2.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-v4l2.c
@@ -851,11 +851,11 @@ static long pvr2_v4l2_do_ioctl(struct file *file, unsigned int cmd, void *arg)
 	case VIDIOC_DBG_G_REGISTER:
 	{
 		u64 val;
-		struct v4l2_register *req = (struct v4l2_register *)arg;
+		struct v4l2_dbg_register *req = (struct v4l2_dbg_register *)arg;
 		if (cmd == VIDIOC_DBG_S_REGISTER) val = req->val;
 		ret = pvr2_hdw_register_access(
-			hdw,req->match_type,req->match_chip,req->reg,
-			cmd == VIDIOC_DBG_S_REGISTER,&val);
+			hdw, &req->match, req->reg,
+			cmd == VIDIOC_DBG_S_REGISTER, &val);
 		if (cmd == VIDIOC_DBG_G_REGISTER) req->val = val;
 		break;
 	}
diff --git a/drivers/media/video/saa7115.c b/drivers/media/video/saa7115.c
index 22708ecdf1bb..46c796c3fec8 100644
--- a/drivers/media/video/saa7115.c
+++ b/drivers/media/video/saa7115.c
@@ -1371,25 +1371,24 @@ static int saa711x_g_vbi_data(struct v4l2_subdev *sd, struct v4l2_sliced_vbi_dat
 }
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
-static int saa711x_g_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int saa711x_g_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 	reg->val = saa711x_read(sd, reg->reg & 0xff);
+	reg->size = 1;
 	return 0;
 }
 
-static int saa711x_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int saa711x_s_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -1398,7 +1397,7 @@ static int saa711x_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
 }
 #endif
 
-static int saa711x_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip)
+static int saa711x_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip)
 {
 	struct saa711x_state *state = to_state(sd);
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
diff --git a/drivers/media/video/saa7127.c b/drivers/media/video/saa7127.c
index bfc85654795e..d6848f7a503b 100644
--- a/drivers/media/video/saa7127.c
+++ b/drivers/media/video/saa7127.c
@@ -623,25 +623,24 @@ static int saa7127_s_vbi_data(struct v4l2_subdev *sd, const struct v4l2_sliced_v
 }
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
-static int saa7127_g_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int saa7127_g_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 	reg->val = saa7127_read(sd, reg->reg & 0xff);
+	reg->size = 1;
 	return 0;
 }
 
-static int saa7127_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int saa7127_s_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -650,7 +649,7 @@ static int saa7127_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
 }
 #endif
 
-static int saa7127_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip)
+static int saa7127_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip)
 {
 	struct saa7127_state *state = to_state(sd);
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
diff --git a/drivers/media/video/saa7134/saa6752hs.c b/drivers/media/video/saa7134/saa6752hs.c
index 1fb6eccdade3..1fee6e84a512 100644
--- a/drivers/media/video/saa7134/saa6752hs.c
+++ b/drivers/media/video/saa7134/saa6752hs.c
@@ -838,7 +838,7 @@ saa6752hs_command(struct i2c_client *client, unsigned int cmd, void *arg)
 		h->standard = *((v4l2_std_id *) arg);
 		break;
 
-	case VIDIOC_G_CHIP_IDENT:
+	case VIDIOC_DBG_G_CHIP_IDENT:
 		return v4l2_chip_ident_i2c_client(client,
 				arg, h->chip, h->revision);
 
diff --git a/drivers/media/video/saa7134/saa7134-empress.c b/drivers/media/video/saa7134/saa7134-empress.c
index 3beba480137f..c9d8beb87a60 100644
--- a/drivers/media/video/saa7134/saa7134-empress.c
+++ b/drivers/media/video/saa7134/saa7134-empress.c
@@ -405,7 +405,7 @@ static int empress_querymenu(struct file *file, void *priv,
 }
 
 static int empress_g_chip_ident(struct file *file, void *fh,
-	       struct v4l2_chip_ident *chip)
+	       struct v4l2_dbg_chip_ident *chip)
 {
 	struct saa7134_dev *dev = file->private_data;
 
@@ -413,12 +413,12 @@ static int empress_g_chip_ident(struct file *file, void *fh,
 	chip->revision = 0;
 	if (dev->mpeg_i2c_client == NULL)
 		return -EINVAL;
-	if (chip->match_type == V4L2_CHIP_MATCH_I2C_DRIVER &&
-	    chip->match_chip == I2C_DRIVERID_SAA6752HS)
-		return saa7134_i2c_call_saa6752(dev, VIDIOC_G_CHIP_IDENT, chip);
-	if (chip->match_type == V4L2_CHIP_MATCH_I2C_ADDR &&
-	    chip->match_chip == dev->mpeg_i2c_client->addr)
-		return saa7134_i2c_call_saa6752(dev, VIDIOC_G_CHIP_IDENT, chip);
+	if (chip->match.type == V4L2_CHIP_MATCH_I2C_DRIVER &&
+	    !strcmp(chip->match.name, "saa6752hs"))
+		return saa7134_i2c_call_saa6752(dev, VIDIOC_DBG_G_CHIP_IDENT, chip);
+	if (chip->match.type == V4L2_CHIP_MATCH_I2C_ADDR &&
+	    chip->match.addr == dev->mpeg_i2c_client->addr)
+		return saa7134_i2c_call_saa6752(dev, VIDIOC_DBG_G_CHIP_IDENT, chip);
 	return -EINVAL;
 }
 
diff --git a/drivers/media/video/saa7134/saa7134-video.c b/drivers/media/video/saa7134/saa7134-video.c
index 6b2ab57538ee..a1f7e351f572 100644
--- a/drivers/media/video/saa7134/saa7134-video.c
+++ b/drivers/media/video/saa7134/saa7134-video.c
@@ -2247,24 +2247,25 @@ static int saa7134_g_parm(struct file *file, void *fh,
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int vidioc_g_register (struct file *file, void *priv,
-			      struct v4l2_register *reg)
+			      struct v4l2_dbg_register *reg)
 {
 	struct saa7134_fh *fh = priv;
 	struct saa7134_dev *dev = fh->dev;
 
-	if (!v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_host(&reg->match))
 		return -EINVAL;
 	reg->val = saa_readb(reg->reg);
+	reg->size = 1;
 	return 0;
 }
 
 static int vidioc_s_register (struct file *file, void *priv,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct saa7134_fh *fh = priv;
 	struct saa7134_dev *dev = fh->dev;
 
-	if (!v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_host(&reg->match))
 		return -EINVAL;
 	saa_writeb(reg->reg&0xffffff, reg->val);
 	return 0;
diff --git a/drivers/media/video/saa717x.c b/drivers/media/video/saa717x.c
index 9befca65905e..454ad1dd7507 100644
--- a/drivers/media/video/saa717x.c
+++ b/drivers/media/video/saa717x.c
@@ -1171,25 +1171,26 @@ static int saa717x_queryctrl(struct v4l2_subdev *sd, struct v4l2_queryctrl *qc)
 }
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
-static int saa717x_g_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int saa717x_g_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client, reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 	reg->val = saa717x_read(sd, reg->reg);
+	reg->size = 1;
 	return 0;
 }
 
-static int saa717x_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int saa717x_s_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 	u16 addr = reg->reg & 0xffff;
 	u8 val = reg->val & 0xff;
 
-	if (!v4l2_chip_match_i2c_client(client, reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
diff --git a/drivers/media/video/soc_camera.c b/drivers/media/video/soc_camera.c
index 9986e02bcf1a..fcb05f06de8f 100644
--- a/drivers/media/video/soc_camera.c
+++ b/drivers/media/video/soc_camera.c
@@ -699,7 +699,7 @@ static int soc_camera_s_crop(struct file *file, void *fh,
 }
 
 static int soc_camera_g_chip_ident(struct file *file, void *fh,
-				   struct v4l2_chip_ident *id)
+				   struct v4l2_dbg_chip_ident *id)
 {
 	struct soc_camera_file *icf = file->private_data;
 	struct soc_camera_device *icd = icf->icd;
@@ -712,7 +712,7 @@ static int soc_camera_g_chip_ident(struct file *file, void *fh,
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int soc_camera_g_register(struct file *file, void *fh,
-				 struct v4l2_register *reg)
+				 struct v4l2_dbg_register *reg)
 {
 	struct soc_camera_file *icf = file->private_data;
 	struct soc_camera_device *icd = icf->icd;
@@ -724,7 +724,7 @@ static int soc_camera_g_register(struct file *file, void *fh,
 }
 
 static int soc_camera_s_register(struct file *file, void *fh,
-				 struct v4l2_register *reg)
+				 struct v4l2_dbg_register *reg)
 {
 	struct soc_camera_file *icf = file->private_data;
 	struct soc_camera_device *icd = icf->icd;
diff --git a/drivers/media/video/tvaudio.c b/drivers/media/video/tvaudio.c
index d0c794da735b..5aeccb301cea 100644
--- a/drivers/media/video/tvaudio.c
+++ b/drivers/media/video/tvaudio.c
@@ -1762,7 +1762,7 @@ static int tvaudio_s_frequency(struct v4l2_subdev *sd, struct v4l2_frequency *fr
 	return 0;
 }
 
-static int tvaudio_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip)
+static int tvaudio_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
diff --git a/drivers/media/video/tvp5150.c b/drivers/media/video/tvp5150.c
index a388a9f0cb18..2cd64ef27b95 100644
--- a/drivers/media/video/tvp5150.c
+++ b/drivers/media/video/tvp5150.c
@@ -963,7 +963,7 @@ static int tvp5150_g_fmt(struct v4l2_subdev *sd, struct v4l2_format *fmt)
 
 
 static int tvp5150_g_chip_ident(struct v4l2_subdev *sd,
-				struct v4l2_chip_ident *chip)
+				struct v4l2_dbg_chip_ident *chip)
 {
 	int rev;
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
@@ -977,25 +977,24 @@ static int tvp5150_g_chip_ident(struct v4l2_subdev *sd,
 
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
-static int tvp5150_g_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int tvp5150_g_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 	reg->val = tvp5150_read(sd, reg->reg & 0xff);
+	reg->size = 1;
 	return 0;
 }
 
-static int tvp5150_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int tvp5150_s_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
diff --git a/drivers/media/video/tw9910.c b/drivers/media/video/tw9910.c
index d5cdc4be1a35..52c0357faa5d 100644
--- a/drivers/media/video/tw9910.c
+++ b/drivers/media/video/tw9910.c
@@ -575,7 +575,7 @@ static unsigned long tw9910_query_bus_param(struct soc_camera_device *icd)
 }
 
 static int tw9910_get_chip_id(struct soc_camera_device *icd,
-			      struct v4l2_chip_ident *id)
+			      struct v4l2_dbg_chip_ident *id)
 {
 	id->ident = V4L2_IDENT_TW9910;
 	id->revision = 0;
@@ -606,7 +606,7 @@ static int tw9910_enum_input(struct soc_camera_device *icd,
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int tw9910_get_register(struct soc_camera_device *icd,
-			       struct v4l2_register *reg)
+			       struct v4l2_dbg_register *reg)
 {
 	struct tw9910_priv *priv = container_of(icd, struct tw9910_priv, icd);
 	int ret;
@@ -627,7 +627,7 @@ static int tw9910_get_register(struct soc_camera_device *icd,
 }
 
 static int tw9910_set_register(struct soc_camera_device *icd,
-			       struct v4l2_register *reg)
+			       struct v4l2_dbg_register *reg)
 {
 	struct tw9910_priv *priv = container_of(icd, struct tw9910_priv, icd);
 
diff --git a/drivers/media/video/upd64031a.c b/drivers/media/video/upd64031a.c
index 7a609a3a6dbe..4f16effb530f 100644
--- a/drivers/media/video/upd64031a.c
+++ b/drivers/media/video/upd64031a.c
@@ -147,7 +147,7 @@ static int upd64031a_s_routing(struct v4l2_subdev *sd, const struct v4l2_routing
 	return upd64031a_s_frequency(sd, NULL);
 }
 
-static int upd64031a_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip)
+static int upd64031a_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
@@ -162,25 +162,24 @@ static int upd64031a_log_status(struct v4l2_subdev *sd)
 }
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
-static int upd64031a_g_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int upd64031a_g_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 	reg->val = upd64031a_read(sd, reg->reg & 0xff);
+	reg->size = 1;
 	return 0;
 }
 
-static int upd64031a_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int upd64031a_s_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
diff --git a/drivers/media/video/upd64083.c b/drivers/media/video/upd64083.c
index 58412cb9c01a..4b712f69d1b7 100644
--- a/drivers/media/video/upd64083.c
+++ b/drivers/media/video/upd64083.c
@@ -120,25 +120,24 @@ static int upd64083_s_routing(struct v4l2_subdev *sd, const struct v4l2_routing
 }
 
 #ifdef CONFIG_VIDEO_ADV_DEBUG
-static int upd64083_g_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int upd64083_g_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 	reg->val = upd64083_read(sd, reg->reg & 0xff);
+	reg->size = 1;
 	return 0;
 }
 
-static int upd64083_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg)
+static int upd64083_s_register(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
-	if (!v4l2_chip_match_i2c_client(client,
-				reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_i2c_client(client, &reg->match))
 		return -EINVAL;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -147,7 +146,7 @@ static int upd64083_s_register(struct v4l2_subdev *sd, struct v4l2_register *reg
 }
 #endif
 
-static int upd64083_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip)
+static int upd64083_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
diff --git a/drivers/media/video/usbvision/usbvision-video.c b/drivers/media/video/usbvision/usbvision-video.c
index 7c61c6d5cede..2be5e47ed081 100644
--- a/drivers/media/video/usbvision/usbvision-video.c
+++ b/drivers/media/video/usbvision/usbvision-video.c
@@ -477,12 +477,12 @@ static int usbvision_v4l2_close(struct file *file)
  */
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 static int vidioc_g_register (struct file *file, void *priv,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct usb_usbvision *usbvision = video_drvdata(file);
 	int errCode;
 
-	if (!v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_host(&reg->match))
 		return -EINVAL;
 	/* NT100x has a 8-bit register space */
 	errCode = usbvision_read_reg(usbvision, reg->reg&0xff);
@@ -492,16 +492,17 @@ static int vidioc_g_register (struct file *file, void *priv,
 		return errCode;
 	}
 	reg->val = errCode;
+	reg->size = 1;
 	return 0;
 }
 
 static int vidioc_s_register (struct file *file, void *priv,
-				struct v4l2_register *reg)
+				struct v4l2_dbg_register *reg)
 {
 	struct usb_usbvision *usbvision = video_drvdata(file);
 	int errCode;
 
-	if (!v4l2_chip_match_host(reg->match_type, reg->match_chip))
+	if (!v4l2_chip_match_host(&reg->match))
 		return -EINVAL;
 	/* NT100x has a 8-bit register space */
 	errCode = usbvision_write_reg(usbvision, reg->reg&0xff, reg->val);
diff --git a/drivers/media/video/v4l2-common.c b/drivers/media/video/v4l2-common.c
index c676b0b0f708..b8f2be8d5c0e 100644
--- a/drivers/media/video/v4l2-common.c
+++ b/drivers/media/video/v4l2-common.c
@@ -797,11 +797,11 @@ u32 v4l2_ctrl_next(const u32 * const * ctrl_classes, u32 id)
 }
 EXPORT_SYMBOL(v4l2_ctrl_next);
 
-int v4l2_chip_match_host(u32 match_type, u32 match_chip)
+int v4l2_chip_match_host(const struct v4l2_dbg_match *match)
 {
-	switch (match_type) {
+	switch (match->type) {
 	case V4L2_CHIP_MATCH_HOST:
-		return match_chip == 0;
+		return match->addr == 0;
 	default:
 		return 0;
 	}
@@ -809,23 +809,34 @@ int v4l2_chip_match_host(u32 match_type, u32 match_chip)
 EXPORT_SYMBOL(v4l2_chip_match_host);
 
 #if defined(CONFIG_I2C) || (defined(CONFIG_I2C_MODULE) && defined(MODULE))
-int v4l2_chip_match_i2c_client(struct i2c_client *c, u32 match_type, u32 match_chip)
+int v4l2_chip_match_i2c_client(struct i2c_client *c, const struct v4l2_dbg_match *match)
 {
-	switch (match_type) {
+	int len;
+
+	if (c == NULL || match == NULL)
+		return 0;
+
+	switch (match->type) {
 	case V4L2_CHIP_MATCH_I2C_DRIVER:
-		return (c != NULL && c->driver != NULL && c->driver->id == match_chip);
+		if (c->driver == NULL || c->driver->driver.name == NULL)
+			return 0;
+		len = strlen(c->driver->driver.name);
+		/* legacy drivers have a ' suffix, don't try to match that */
+		if (len && c->driver->driver.name[len - 1] == '\'')
+			len--;
+		return len && !strncmp(c->driver->driver.name, match->name, len);
 	case V4L2_CHIP_MATCH_I2C_ADDR:
-		return (c != NULL && c->addr == match_chip);
+		return c->addr == match->addr;
 	default:
 		return 0;
 	}
 }
 EXPORT_SYMBOL(v4l2_chip_match_i2c_client);
 
-int v4l2_chip_ident_i2c_client(struct i2c_client *c, struct v4l2_chip_ident *chip,
+int v4l2_chip_ident_i2c_client(struct i2c_client *c, struct v4l2_dbg_chip_ident *chip,
 		u32 ident, u32 revision)
 {
-	if (!v4l2_chip_match_i2c_client(c, chip->match_type, chip->match_chip))
+	if (!v4l2_chip_match_i2c_client(c, &chip->match))
 		return 0;
 	if (chip->ident == V4L2_IDENT_NONE) {
 		chip->ident = ident;
diff --git a/drivers/media/video/v4l2-compat-ioctl32.c b/drivers/media/video/v4l2-compat-ioctl32.c
index ec81b9737bd7..110376be5d2b 100644
--- a/drivers/media/video/v4l2-compat-ioctl32.c
+++ b/drivers/media/video/v4l2-compat-ioctl32.c
@@ -1046,7 +1046,8 @@ long v4l2_compat_ioctl32(struct file *file, unsigned int cmd, unsigned long arg)
 	case VIDIOC_TRY_ENCODER_CMD:
 	case VIDIOC_DBG_S_REGISTER:
 	case VIDIOC_DBG_G_REGISTER:
-	case VIDIOC_G_CHIP_IDENT:
+	case VIDIOC_DBG_G_CHIP_IDENT:
+	case VIDIOC_G_CHIP_IDENT_OLD:
 	case VIDIOC_S_HW_FREQ_SEEK:
 		ret = do_video_ioctl(file, cmd, arg);
 		break;
diff --git a/drivers/media/video/v4l2-ioctl.c b/drivers/media/video/v4l2-ioctl.c
index 8f629ef5b9ec..52d687b165e0 100644
--- a/drivers/media/video/v4l2-ioctl.c
+++ b/drivers/media/video/v4l2-ioctl.c
@@ -266,7 +266,7 @@ static const char *v4l2_ioctls[] = {
 	[_IOC_NR(VIDIOC_DBG_S_REGISTER)]   = "VIDIOC_DBG_S_REGISTER",
 	[_IOC_NR(VIDIOC_DBG_G_REGISTER)]   = "VIDIOC_DBG_G_REGISTER",
 
-	[_IOC_NR(VIDIOC_G_CHIP_IDENT)]     = "VIDIOC_G_CHIP_IDENT",
+	[_IOC_NR(VIDIOC_DBG_G_CHIP_IDENT)] = "VIDIOC_DBG_G_CHIP_IDENT",
 	[_IOC_NR(VIDIOC_S_HW_FREQ_SEEK)]   = "VIDIOC_S_HW_FREQ_SEEK",
 #endif
 };
@@ -1720,7 +1720,7 @@ static long __video_do_ioctl(struct file *file,
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 	case VIDIOC_DBG_G_REGISTER:
 	{
-		struct v4l2_register *p = arg;
+		struct v4l2_dbg_register *p = arg;
 
 		if (!capable(CAP_SYS_ADMIN))
 			ret = -EPERM;
@@ -1730,7 +1730,7 @@ static long __video_do_ioctl(struct file *file,
 	}
 	case VIDIOC_DBG_S_REGISTER:
 	{
-		struct v4l2_register *p = arg;
+		struct v4l2_dbg_register *p = arg;
 
 		if (!capable(CAP_SYS_ADMIN))
 			ret = -EPERM;
@@ -1739,9 +1739,9 @@ static long __video_do_ioctl(struct file *file,
 		break;
 	}
 #endif
-	case VIDIOC_G_CHIP_IDENT:
+	case VIDIOC_DBG_G_CHIP_IDENT:
 	{
-		struct v4l2_chip_ident *p = arg;
+		struct v4l2_dbg_chip_ident *p = arg;
 
 		if (!ops->vidioc_g_chip_ident)
 			break;
@@ -1750,6 +1750,11 @@ static long __video_do_ioctl(struct file *file,
 			dbgarg(cmd, "chip_ident=%u, revision=0x%x\n", p->ident, p->revision);
 		break;
 	}
+	case VIDIOC_G_CHIP_IDENT_OLD:
+		printk(KERN_ERR "VIDIOC_G_CHIP_IDENT has been deprecated and will disappear in 2.6.30.\n");
+		printk(KERN_ERR "It is a debugging ioctl and must not be used in applications!\n");
+		return -EINVAL;
+
 	case VIDIOC_S_HW_FREQ_SEEK:
 	{
 		struct v4l2_hw_freq_seek *p = arg;
diff --git a/drivers/media/video/v4l2-subdev.c b/drivers/media/video/v4l2-subdev.c
index e3612f29d0df..fbe9cc0d433a 100644
--- a/drivers/media/video/v4l2-subdev.c
+++ b/drivers/media/video/v4l2-subdev.c
@@ -37,7 +37,7 @@ int v4l2_subdev_command(struct v4l2_subdev *sd, unsigned cmd, void *arg)
 		return v4l2_subdev_call(sd, core, queryctrl, arg);
 	case VIDIOC_LOG_STATUS:
 		return v4l2_subdev_call(sd, core, log_status);
-	case VIDIOC_G_CHIP_IDENT:
+	case VIDIOC_DBG_G_CHIP_IDENT:
 		return v4l2_subdev_call(sd, core, g_chip_ident, arg);
 	case VIDIOC_INT_S_STANDBY:
 		return v4l2_subdev_call(sd, core, s_standby, arg ? (*(u32 *)arg) : 0);
diff --git a/drivers/media/video/vp27smpx.c b/drivers/media/video/vp27smpx.c
index f72b859486ad..5d73f66d9f55 100644
--- a/drivers/media/video/vp27smpx.c
+++ b/drivers/media/video/vp27smpx.c
@@ -113,7 +113,7 @@ static int vp27smpx_g_tuner(struct v4l2_subdev *sd, struct v4l2_tuner *vt)
 	return 0;
 }
 
-static int vp27smpx_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip)
+static int vp27smpx_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
diff --git a/drivers/media/video/wm8739.c b/drivers/media/video/wm8739.c
index 12a31e7a5f6d..f2864d5cd180 100644
--- a/drivers/media/video/wm8739.c
+++ b/drivers/media/video/wm8739.c
@@ -233,7 +233,7 @@ static int wm8739_queryctrl(struct v4l2_subdev *sd, struct v4l2_queryctrl *qc)
 	return -EINVAL;
 }
 
-static int wm8739_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip)
+static int wm8739_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
diff --git a/drivers/media/video/wm8775.c b/drivers/media/video/wm8775.c
index d0220b0ec0bc..53fcd42843e0 100644
--- a/drivers/media/video/wm8775.c
+++ b/drivers/media/video/wm8775.c
@@ -130,7 +130,7 @@ static int wm8775_s_ctrl(struct v4l2_subdev *sd, struct v4l2_control *ctrl)
 	return 0;
 }
 
-static int wm8775_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip)
+static int wm8775_g_chip_ident(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip)
 {
 	struct i2c_client *client = v4l2_get_subdevdata(sd);
 
diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h
index 1f126e30766c..5571dbe1c0ad 100644
--- a/include/linux/videodev2.h
+++ b/include/linux/videodev2.h
@@ -1370,25 +1370,41 @@ struct v4l2_streamparm {
 /*
  *	A D V A N C E D   D E B U G G I N G
  *
- *	NOTE: EXPERIMENTAL API
+ *	NOTE: EXPERIMENTAL API, NEVER RELY ON THIS IN APPLICATIONS!
+ *	FOR DEBUGGING, TESTING AND INTERNAL USE ONLY!
  */
 
 /* VIDIOC_DBG_G_REGISTER and VIDIOC_DBG_S_REGISTER */
 
 #define V4L2_CHIP_MATCH_HOST       0  /* Match against chip ID on host (0 for the host) */
-#define V4L2_CHIP_MATCH_I2C_DRIVER 1  /* Match against I2C driver ID */
+#define V4L2_CHIP_MATCH_I2C_DRIVER 1  /* Match against I2C driver name */
 #define V4L2_CHIP_MATCH_I2C_ADDR   2  /* Match against I2C 7-bit address */
 #define V4L2_CHIP_MATCH_AC97       3  /* Match against anciliary AC97 chip */
 
-struct v4l2_register {
-	__u32 match_type; /* Match type */
-	__u32 match_chip; /* Match this chip, meaning determined by match_type */
+struct v4l2_dbg_match {
+	__u32 type; /* Match type */
+	union {     /* Match this chip, meaning determined by type */
+		__u32 addr;
+		char name[32];
+	};
+} __attribute__ ((packed));
+
+struct v4l2_dbg_register {
+	struct v4l2_dbg_match match;
+	__u32 size;	/* register size in bytes */
 	__u64 reg;
 	__u64 val;
-};
+} __attribute__ ((packed));
+
+/* VIDIOC_DBG_G_CHIP_IDENT */
+struct v4l2_dbg_chip_ident {
+	struct v4l2_dbg_match match;
+	__u32 ident;       /* chip identifier as specified in <media/v4l2-chip-ident.h> */
+	__u32 revision;    /* chip revision, chip specific */
+} __attribute__ ((packed));
 
-/* VIDIOC_G_CHIP_IDENT */
-struct v4l2_chip_ident {
+/* VIDIOC_G_CHIP_IDENT_OLD: Deprecated, do not use */
+struct v4l2_chip_ident_old {
 	__u32 match_type;  /* Match type */
 	__u32 match_chip;  /* Match this chip, meaning determined by match_type */
 	__u32 ident;       /* chip identifier as specified in <media/v4l2-chip-ident.h> */
@@ -1460,13 +1476,22 @@ struct v4l2_chip_ident {
 #define VIDIOC_G_ENC_INDEX       _IOR('V', 76, struct v4l2_enc_idx)
 #define VIDIOC_ENCODER_CMD      _IOWR('V', 77, struct v4l2_encoder_cmd)
 #define VIDIOC_TRY_ENCODER_CMD  _IOWR('V', 78, struct v4l2_encoder_cmd)
+#endif
 
-/* Experimental, only implemented if CONFIG_VIDEO_ADV_DEBUG is defined */
-#define	VIDIOC_DBG_S_REGISTER 	 _IOW('V', 79, struct v4l2_register)
-#define	VIDIOC_DBG_G_REGISTER 	_IOWR('V', 80, struct v4l2_register)
-
-#define VIDIOC_G_CHIP_IDENT     _IOWR('V', 81, struct v4l2_chip_ident)
+#if 1
+/* Experimental, meant for debugging, testing and internal use.
+   Only implemented if CONFIG_VIDEO_ADV_DEBUG is defined.
+   You must be root to use these ioctls. Never use these in applications! */
+#define	VIDIOC_DBG_S_REGISTER 	 _IOW('V', 79, struct v4l2_dbg_register)
+#define	VIDIOC_DBG_G_REGISTER 	_IOWR('V', 80, struct v4l2_dbg_register)
+
+/* Experimental, meant for debugging, testing and internal use.
+   Never use this ioctl in applications! */
+#define VIDIOC_DBG_G_CHIP_IDENT _IOWR('V', 81, struct v4l2_dbg_chip_ident)
+/* This is deprecated and will go away in 2.6.30 */
+#define VIDIOC_G_CHIP_IDENT_OLD _IOWR('V', 81, struct v4l2_chip_ident_old)
 #endif
+
 #define VIDIOC_S_HW_FREQ_SEEK	 _IOW('V', 82, struct v4l2_hw_freq_seek)
 /* Reminder: when adding new ioctls please add support for them to
    drivers/media/video/v4l2-compat-ioctl32.c as well! */
diff --git a/include/media/soc_camera.h b/include/media/soc_camera.h
index 425b6a98c95c..7440d9250665 100644
--- a/include/media/soc_camera.h
+++ b/include/media/soc_camera.h
@@ -164,12 +164,12 @@ struct soc_camera_ops {
 	unsigned long (*query_bus_param)(struct soc_camera_device *);
 	int (*set_bus_param)(struct soc_camera_device *, unsigned long);
 	int (*get_chip_id)(struct soc_camera_device *,
-			   struct v4l2_chip_ident *);
+			   struct v4l2_dbg_chip_ident *);
 	int (*set_std)(struct soc_camera_device *, v4l2_std_id *);
 	int (*enum_input)(struct soc_camera_device *, struct v4l2_input *);
 #ifdef CONFIG_VIDEO_ADV_DEBUG
-	int (*get_register)(struct soc_camera_device *, struct v4l2_register *);
-	int (*set_register)(struct soc_camera_device *, struct v4l2_register *);
+	int (*get_register)(struct soc_camera_device *, struct v4l2_dbg_register *);
+	int (*set_register)(struct soc_camera_device *, struct v4l2_dbg_register *);
 #endif
 	int (*get_control)(struct soc_camera_device *, struct v4l2_control *);
 	int (*set_control)(struct soc_camera_device *, struct v4l2_control *);
diff --git a/include/media/v4l2-chip-ident.h b/include/media/v4l2-chip-ident.h
index 43dbb659f1f5..9aaf652b20ef 100644
--- a/include/media/v4l2-chip-ident.h
+++ b/include/media/v4l2-chip-ident.h
@@ -2,7 +2,7 @@
     v4l2 chip identifiers header
 
     This header provides a list of chip identifiers that can be returned
-    through the VIDIOC_G_CHIP_IDENT ioctl.
+    through the VIDIOC_DBG_G_CHIP_IDENT ioctl.
 
     Copyright (C) 2007 Hans Verkuil <hverkuil@xs4all.nl>
 
@@ -24,7 +24,7 @@
 #ifndef V4L2_CHIP_IDENT_H_
 #define V4L2_CHIP_IDENT_H_
 
-/* VIDIOC_G_CHIP_IDENT: identifies the actual chip installed on the board */
+/* VIDIOC_DBG_G_CHIP_IDENT: identifies the actual chip installed on the board */
 enum {
 	/* general idents: reserved range 0-49 */
 	V4L2_IDENT_NONE      = 0,       /* No chip matched */
diff --git a/include/media/v4l2-common.h b/include/media/v4l2-common.h
index f99c866d8c37..95e74f1874e1 100644
--- a/include/media/v4l2-common.h
+++ b/include/media/v4l2-common.h
@@ -114,10 +114,10 @@ u32 v4l2_ctrl_next(const u32 * const *ctrl_classes, u32 id);
 /* Register/chip ident helper function */
 
 struct i2c_client; /* forward reference */
-int v4l2_chip_match_i2c_client(struct i2c_client *c, u32 id_type, u32 chip_id);
-int v4l2_chip_ident_i2c_client(struct i2c_client *c, struct v4l2_chip_ident *chip,
+int v4l2_chip_match_i2c_client(struct i2c_client *c, const struct v4l2_dbg_match *match);
+int v4l2_chip_ident_i2c_client(struct i2c_client *c, struct v4l2_dbg_chip_ident *chip,
 		u32 ident, u32 revision);
-int v4l2_chip_match_host(u32 id_type, u32 chip_id);
+int v4l2_chip_match_host(const struct v4l2_dbg_match *match);
 
 /* ------------------------------------------------------------------------- */
 
diff --git a/include/media/v4l2-int-device.h b/include/media/v4l2-int-device.h
index ecda3c725837..fbf585561570 100644
--- a/include/media/v4l2-int-device.h
+++ b/include/media/v4l2-int-device.h
@@ -219,7 +219,7 @@ enum v4l2_int_ioctl_num {
 	vidioc_int_reset_num,
 	/* VIDIOC_INT_INIT */
 	vidioc_int_init_num,
-	/* VIDIOC_INT_G_CHIP_IDENT */
+	/* VIDIOC_DBG_G_CHIP_IDENT */
 	vidioc_int_g_chip_ident_num,
 
 	/*
diff --git a/include/media/v4l2-ioctl.h b/include/media/v4l2-ioctl.h
index bf0e723a99c1..b01c044868d0 100644
--- a/include/media/v4l2-ioctl.h
+++ b/include/media/v4l2-ioctl.h
@@ -225,12 +225,12 @@ struct v4l2_ioctl_ops {
 	/* Debugging ioctls */
 #ifdef CONFIG_VIDEO_ADV_DEBUG
 	int (*vidioc_g_register)       (struct file *file, void *fh,
-					struct v4l2_register *reg);
+					struct v4l2_dbg_register *reg);
 	int (*vidioc_s_register)       (struct file *file, void *fh,
-					struct v4l2_register *reg);
+					struct v4l2_dbg_register *reg);
 #endif
 	int (*vidioc_g_chip_ident)     (struct file *file, void *fh,
-					struct v4l2_chip_ident *chip);
+					struct v4l2_dbg_chip_ident *chip);
 
 	int (*vidioc_enum_framesizes)   (struct file *file, void *fh,
 					 struct v4l2_frmsizeenum *fsize);
diff --git a/include/media/v4l2-subdev.h b/include/media/v4l2-subdev.h
index 2517344313b6..37b09e56e943 100644
--- a/include/media/v4l2-subdev.h
+++ b/include/media/v4l2-subdev.h
@@ -69,7 +69,7 @@ struct tuner_setup;
    not yet implemented) since ops provide proper type-checking.
  */
 struct v4l2_subdev_core_ops {
-	int (*g_chip_ident)(struct v4l2_subdev *sd, struct v4l2_chip_ident *chip);
+	int (*g_chip_ident)(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip);
 	int (*log_status)(struct v4l2_subdev *sd);
 	int (*init)(struct v4l2_subdev *sd, u32 val);
 	int (*s_standby)(struct v4l2_subdev *sd, u32 standby);
@@ -81,8 +81,8 @@ struct v4l2_subdev_core_ops {
 	int (*querymenu)(struct v4l2_subdev *sd, struct v4l2_querymenu *qm);
 	long (*ioctl)(struct v4l2_subdev *sd, unsigned int cmd, void *arg);
 #ifdef CONFIG_VIDEO_ADV_DEBUG
-	int (*g_register)(struct v4l2_subdev *sd, struct v4l2_register *reg);
-	int (*s_register)(struct v4l2_subdev *sd, struct v4l2_register *reg);
+	int (*g_register)(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg);
+	int (*s_register)(struct v4l2_subdev *sd, struct v4l2_dbg_register *reg);
 #endif
 };
 
-- 
cgit v1.2.3


From cb889a2f3515b140bef193cf6ffcdb099349b8aa Mon Sep 17 00:00:00 2001
From: Klaus Schmidinger <Klaus.Schmidinger@cadsoft.de>
Date: Wed, 31 Dec 2008 14:11:23 -0300
Subject: V4L/DVB (10164): Add missing S2 caps flag to S2API

The attached patch adds a capability flag that allows an application
to determine whether a particular device can handle "second generation
modulation" transponders. This is necessary in order for applications
to be able to decide which device to use for a given channel in
a multi device environment, where DVB-S and DVB-S2 devices are mixed.

It is assumed that a device capable of handling "second generation
modulation" can implicitly handle "first generation modulation".
The flag is not named anything with DVBS2 in order to allow its
use with future DVBT2 devices as well (should they ever come).

Signed-off by: Klaus Schmidinger <Klaus.Schmidinger@cadsoft.de>

Acked-by: Steven Toth <stoth@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 include/linux/dvb/frontend.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/dvb/frontend.h b/include/linux/dvb/frontend.h
index 79a8ed8e6a7d..926d28d526e7 100644
--- a/include/linux/dvb/frontend.h
+++ b/include/linux/dvb/frontend.h
@@ -63,6 +63,7 @@ typedef enum fe_caps {
 	FE_CAN_8VSB			= 0x200000,
 	FE_CAN_16VSB			= 0x400000,
 	FE_HAS_EXTENDED_CAPS		= 0x800000,   // We need more bitspace for newer APIs, indicate this.
+	FE_CAN_2G_MODULATION		= 0x10000000, // frontend supports "2nd generation modulation" (DVB-S2)
 	FE_NEEDS_BENDING		= 0x20000000, // not supported anymore, don't use (frontend requires frequency bending)
 	FE_CAN_RECOVER			= 0x40000000, // frontend can recover from a cable unplug automatically
 	FE_CAN_MUTE_TS			= 0x80000000  // frontend can stop spurious TS data output
-- 
cgit v1.2.3


From e4cda3e0728156c6be1d03e72ef20ea811da4ad5 Mon Sep 17 00:00:00 2001
From: Mauro Carvalho Chehab <mchehab@redhat.com>
Date: Wed, 31 Dec 2008 14:26:57 -0300
Subject: V4L/DVB (10166): dvb frontend: stop using non-C99 compliant comments

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 include/linux/dvb/frontend.h | 28 ++++++++++++++--------------
 1 file changed, 14 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/dvb/frontend.h b/include/linux/dvb/frontend.h
index 926d28d526e7..55026b1a40bd 100644
--- a/include/linux/dvb/frontend.h
+++ b/include/linux/dvb/frontend.h
@@ -62,11 +62,11 @@ typedef enum fe_caps {
 	FE_CAN_HIERARCHY_AUTO		= 0x100000,
 	FE_CAN_8VSB			= 0x200000,
 	FE_CAN_16VSB			= 0x400000,
-	FE_HAS_EXTENDED_CAPS		= 0x800000,   // We need more bitspace for newer APIs, indicate this.
-	FE_CAN_2G_MODULATION		= 0x10000000, // frontend supports "2nd generation modulation" (DVB-S2)
-	FE_NEEDS_BENDING		= 0x20000000, // not supported anymore, don't use (frontend requires frequency bending)
-	FE_CAN_RECOVER			= 0x40000000, // frontend can recover from a cable unplug automatically
-	FE_CAN_MUTE_TS			= 0x80000000  // frontend can stop spurious TS data output
+	FE_HAS_EXTENDED_CAPS		= 0x800000,   /* We need more bitspace for newer APIs, indicate this. */
+	FE_CAN_2G_MODULATION		= 0x10000000, /* frontend supports "2nd generation modulation" (DVB-S2) */
+	FE_NEEDS_BENDING		= 0x20000000, /* not supported anymore, don't use (frontend requires frequency bending) */
+	FE_CAN_RECOVER			= 0x40000000, /* frontend can recover from a cable unplug automatically */
+	FE_CAN_MUTE_TS			= 0x80000000  /* frontend can stop spurious TS data output */
 } fe_caps_t;
 
 
@@ -122,15 +122,15 @@ typedef enum fe_sec_mini_cmd {
 
 
 typedef enum fe_status {
-	FE_HAS_SIGNAL	= 0x01,   /*  found something above the noise level */
-	FE_HAS_CARRIER	= 0x02,   /*  found a DVB signal  */
-	FE_HAS_VITERBI	= 0x04,   /*  FEC is stable  */
-	FE_HAS_SYNC	= 0x08,   /*  found sync bytes  */
-	FE_HAS_LOCK	= 0x10,   /*  everything's working... */
-	FE_TIMEDOUT	= 0x20,   /*  no lock within the last ~2 seconds */
-	FE_REINIT	= 0x40    /*  frontend was reinitialized,  */
-} fe_status_t;			  /*  application is recommended to reset */
-				  /*  DiSEqC, tone and parameters */
+	FE_HAS_SIGNAL	= 0x01,   /* found something above the noise level */
+	FE_HAS_CARRIER	= 0x02,   /* found a DVB signal  */
+	FE_HAS_VITERBI	= 0x04,   /* FEC is stable  */
+	FE_HAS_SYNC	= 0x08,   /* found sync bytes  */
+	FE_HAS_LOCK	= 0x10,   /* everything's working... */
+	FE_TIMEDOUT	= 0x20,   /* no lock within the last ~2 seconds */
+	FE_REINIT	= 0x40    /* frontend was reinitialized,  */
+} fe_status_t;			  /* application is recommended to reset */
+				  /* DiSEqC, tone and parameters */
 
 typedef enum fe_spectral_inversion {
 	INVERSION_OFF,
-- 
cgit v1.2.3


From 6680598b44ed3c0052d155522eb21fc5a00de5f3 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 2 Jan 2009 18:53:14 +0100
Subject: Disallow gcc versions 3.{0,1}

GCC 3.0 and 3.1 are too old to build a working kernel.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
[ This check got dropped as obsolete when I simplified the gcc header
  inclusion mess in f153b82121b0366fe0e5f9553545cce237335175, but Willy
  Tarreau reports actually having those old versions still..  -Linus ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/compiler-gcc3.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/compiler-gcc3.h b/include/linux/compiler-gcc3.h
index 2befe6513ce4..8005effc04f1 100644
--- a/include/linux/compiler-gcc3.h
+++ b/include/linux/compiler-gcc3.h
@@ -2,6 +2,10 @@
 #error "Please don't include <linux/compiler-gcc3.h> directly, include <linux/compiler.h> instead."
 #endif
 
+#if __GNUC_MINOR__ < 2
+# error Sorry, your compiler is too old - please upgrade it.
+#endif
+
 #if __GNUC_MINOR__ >= 3
 # define __used			__attribute__((__used__))
 #else
-- 
cgit v1.2.3


From 015ab17dc2e9de805c26e74f498b12ee5e8de07e Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 20 Nov 2008 14:04:20 +0000
Subject: intel-iommu: remove some unused struct intel_iommu fields

The seg, saved_msg and sysdev fields appear to be unused since
before the code was first merged.

linux/msi.h is not needed in linux/intel-iommu.h anymore since
there is no longer a reference to struct msi_msg. The MSI code
in drivers/pci/intel-iommu.c still has linux/msi.h included
via linux/dmar.h.

linux/sysdev.h isn't needed because there is no reference to
struct sys_device.

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c   | 1 -
 include/linux/intel-iommu.h | 5 -----
 2 files changed, 6 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 5c8baa43ac9c..8e5a445ba93a 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -27,7 +27,6 @@
 #include <linux/slab.h>
 #include <linux/irq.h>
 #include <linux/interrupt.h>
-#include <linux/sysdev.h>
 #include <linux/spinlock.h>
 #include <linux/pci.h>
 #include <linux/dmar.h>
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index 3d017cfd245b..1bff7bf1bc2c 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -23,8 +23,6 @@
 #define _INTEL_IOMMU_H_
 
 #include <linux/types.h>
-#include <linux/msi.h>
-#include <linux/sysdev.h>
 #include <linux/iova.h>
 #include <linux/io.h>
 #include <linux/dma_remapping.h>
@@ -289,7 +287,6 @@ struct intel_iommu {
 	void __iomem	*reg; /* Pointer to hardware regs, virtual addr */
 	u64		cap;
 	u64		ecap;
-	int		seg;
 	u32		gcmd; /* Holds TE, EAFL. Don't need SRTP, SFL, WBF */
 	spinlock_t	register_lock; /* protect register handling */
 	int		seq_id;	/* sequence id of the iommu */
@@ -302,8 +299,6 @@ struct intel_iommu {
 
 	unsigned int irq;
 	unsigned char name[7];    /* Device Name */
-	struct msi_msg saved_msg;
-	struct sys_device sysdev;
 	struct iommu_flush flush;
 #endif
 	struct q_inval  *qi;            /* Queued invalidation info */
-- 
cgit v1.2.3


From 519a05491586dad04e687660e54c57882315b22b Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 20 Nov 2008 14:21:13 +0000
Subject: intel-iommu: make init_dmars() static

init_dmars() is not used outside of drivers/pci/intel-iommu.c

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c     | 2 +-
 include/linux/dma_remapping.h | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 8e5a445ba93a..95ae3a9aea8a 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1589,7 +1589,7 @@ static inline void iommu_prepare_isa(void)
 }
 #endif /* !CONFIG_DMAR_FLPY_WA */
 
-int __init init_dmars(void)
+static int __init init_dmars(void)
 {
 	struct dmar_drhd_unit *drhd;
 	struct dmar_rmrr_unit *rmrr;
diff --git a/include/linux/dma_remapping.h b/include/linux/dma_remapping.h
index 952df39c989d..cf92c4924b8c 100644
--- a/include/linux/dma_remapping.h
+++ b/include/linux/dma_remapping.h
@@ -141,7 +141,6 @@ struct device_domain_info {
 	struct dmar_domain *domain; /* pointer to domain */
 };
 
-extern int init_dmars(void);
 extern void free_dmar_iommu(struct intel_iommu *iommu);
 
 extern int dmar_disabled;
-- 
cgit v1.2.3


From f27be03b271851fd54529f292c0f25b4c1f1a553 Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 20 Nov 2008 15:49:43 +0000
Subject: intel-iommu: move DMA_32/64BIT_PFN into intel-iommu.c

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c     | 3 +++
 include/linux/dma_remapping.h | 5 -----
 2 files changed, 3 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 95ae3a9aea8a..6fadbb9bc180 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -53,6 +53,9 @@
 
 #define DOMAIN_MAX_ADDR(gaw) ((((u64)1) << gaw) - 1)
 
+#define IOVA_PFN(addr)		((addr) >> PAGE_SHIFT)
+#define DMA_32BIT_PFN		IOVA_PFN(DMA_32BIT_MASK)
+#define DMA_64BIT_PFN		IOVA_PFN(DMA_64BIT_MASK)
 
 static void flush_unmaps_timeout(unsigned long data);
 
diff --git a/include/linux/dma_remapping.h b/include/linux/dma_remapping.h
index cf92c4924b8c..2e5a5c0b6acd 100644
--- a/include/linux/dma_remapping.h
+++ b/include/linux/dma_remapping.h
@@ -9,11 +9,6 @@
 #define VTD_PAGE_MASK		(((u64)-1) << VTD_PAGE_SHIFT)
 #define VTD_PAGE_ALIGN(addr)	(((addr) + VTD_PAGE_SIZE - 1) & VTD_PAGE_MASK)
 
-#define IOVA_PFN(addr)		((addr) >> PAGE_SHIFT)
-#define DMA_32BIT_PFN		IOVA_PFN(DMA_32BIT_MASK)
-#define DMA_64BIT_PFN		IOVA_PFN(DMA_64BIT_MASK)
-
-
 /*
  * 0: Present
  * 1-11: Reserved
-- 
cgit v1.2.3


From 46b08e1a76b758193b0e7b889c6486a16eb1e9e2 Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 20 Nov 2008 15:49:44 +0000
Subject: intel-iommu: move root entry defs from dma_remapping.h

We keep the struct root_entry forward declaration for the
pointer in struct intel_iommu.

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c     | 33 +++++++++++++++++++++++++++++++++
 include/linux/dma_remapping.h | 34 +---------------------------------
 2 files changed, 34 insertions(+), 33 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 6fadbb9bc180..29bf2d8176e2 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -57,6 +57,39 @@
 #define DMA_32BIT_PFN		IOVA_PFN(DMA_32BIT_MASK)
 #define DMA_64BIT_PFN		IOVA_PFN(DMA_64BIT_MASK)
 
+/*
+ * 0: Present
+ * 1-11: Reserved
+ * 12-63: Context Ptr (12 - (haw-1))
+ * 64-127: Reserved
+ */
+struct root_entry {
+	u64	val;
+	u64	rsvd1;
+};
+#define ROOT_ENTRY_NR (VTD_PAGE_SIZE/sizeof(struct root_entry))
+static inline bool root_present(struct root_entry *root)
+{
+	return (root->val & 1);
+}
+static inline void set_root_present(struct root_entry *root)
+{
+	root->val |= 1;
+}
+static inline void set_root_value(struct root_entry *root, unsigned long value)
+{
+	root->val |= value & VTD_PAGE_MASK;
+}
+
+static inline struct context_entry *
+get_context_addr_from_root(struct root_entry *root)
+{
+	return (struct context_entry *)
+		(root_present(root)?phys_to_virt(
+		root->val & VTD_PAGE_MASK) :
+		NULL);
+}
+
 static void flush_unmaps_timeout(unsigned long data);
 
 DEFINE_TIMER(unmap_timer,  flush_unmaps_timeout, 0, 0);
diff --git a/include/linux/dma_remapping.h b/include/linux/dma_remapping.h
index 2e5a5c0b6acd..d8521662a495 100644
--- a/include/linux/dma_remapping.h
+++ b/include/linux/dma_remapping.h
@@ -9,39 +9,7 @@
 #define VTD_PAGE_MASK		(((u64)-1) << VTD_PAGE_SHIFT)
 #define VTD_PAGE_ALIGN(addr)	(((addr) + VTD_PAGE_SIZE - 1) & VTD_PAGE_MASK)
 
-/*
- * 0: Present
- * 1-11: Reserved
- * 12-63: Context Ptr (12 - (haw-1))
- * 64-127: Reserved
- */
-struct root_entry {
-	u64	val;
-	u64	rsvd1;
-};
-#define ROOT_ENTRY_NR (VTD_PAGE_SIZE/sizeof(struct root_entry))
-static inline bool root_present(struct root_entry *root)
-{
-	return (root->val & 1);
-}
-static inline void set_root_present(struct root_entry *root)
-{
-	root->val |= 1;
-}
-static inline void set_root_value(struct root_entry *root, unsigned long value)
-{
-	root->val |= value & VTD_PAGE_MASK;
-}
-
-struct context_entry;
-static inline struct context_entry *
-get_context_addr_from_root(struct root_entry *root)
-{
-	return (struct context_entry *)
-		(root_present(root)?phys_to_virt(
-		root->val & VTD_PAGE_MASK) :
-		NULL);
-}
+struct root_entry;
 
 /*
  * low 64 bits:
-- 
cgit v1.2.3


From 7a8fc25e0cc6e75fa6fdb0a856490e324218550b Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 20 Nov 2008 15:49:45 +0000
Subject: intel-iommu: move context entry defs out from dma_remapping.h

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c     | 38 ++++++++++++++++++++++++++++++++++++++
 include/linux/dma_remapping.h | 38 --------------------------------------
 2 files changed, 38 insertions(+), 38 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 29bf2d8176e2..9d06f4bb6b5e 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -90,6 +90,44 @@ get_context_addr_from_root(struct root_entry *root)
 		NULL);
 }
 
+/*
+ * low 64 bits:
+ * 0: present
+ * 1: fault processing disable
+ * 2-3: translation type
+ * 12-63: address space root
+ * high 64 bits:
+ * 0-2: address width
+ * 3-6: aval
+ * 8-23: domain id
+ */
+struct context_entry {
+	u64 lo;
+	u64 hi;
+};
+#define context_present(c) ((c).lo & 1)
+#define context_fault_disable(c) (((c).lo >> 1) & 1)
+#define context_translation_type(c) (((c).lo >> 2) & 3)
+#define context_address_root(c) ((c).lo & VTD_PAGE_MASK)
+#define context_address_width(c) ((c).hi &  7)
+#define context_domain_id(c) (((c).hi >> 8) & ((1 << 16) - 1))
+
+#define context_set_present(c) do {(c).lo |= 1;} while (0)
+#define context_set_fault_enable(c) \
+	do {(c).lo &= (((u64)-1) << 2) | 1;} while (0)
+#define context_set_translation_type(c, val) \
+	do { \
+		(c).lo &= (((u64)-1) << 4) | 3; \
+		(c).lo |= ((val) & 3) << 2; \
+	} while (0)
+#define CONTEXT_TT_MULTI_LEVEL 0
+#define context_set_address_root(c, val) \
+	do {(c).lo |= (val) & VTD_PAGE_MASK; } while (0)
+#define context_set_address_width(c, val) do {(c).hi |= (val) & 7;} while (0)
+#define context_set_domain_id(c, val) \
+	do {(c).hi |= ((val) & ((1 << 16) - 1)) << 8;} while (0)
+#define context_clear_entry(c) do {(c).lo = 0; (c).hi = 0;} while (0)
+
 static void flush_unmaps_timeout(unsigned long data);
 
 DEFINE_TIMER(unmap_timer,  flush_unmaps_timeout, 0, 0);
diff --git a/include/linux/dma_remapping.h b/include/linux/dma_remapping.h
index d8521662a495..9a88f7d0262f 100644
--- a/include/linux/dma_remapping.h
+++ b/include/linux/dma_remapping.h
@@ -11,44 +11,6 @@
 
 struct root_entry;
 
-/*
- * low 64 bits:
- * 0: present
- * 1: fault processing disable
- * 2-3: translation type
- * 12-63: address space root
- * high 64 bits:
- * 0-2: address width
- * 3-6: aval
- * 8-23: domain id
- */
-struct context_entry {
-	u64 lo;
-	u64 hi;
-};
-#define context_present(c) ((c).lo & 1)
-#define context_fault_disable(c) (((c).lo >> 1) & 1)
-#define context_translation_type(c) (((c).lo >> 2) & 3)
-#define context_address_root(c) ((c).lo & VTD_PAGE_MASK)
-#define context_address_width(c) ((c).hi &  7)
-#define context_domain_id(c) (((c).hi >> 8) & ((1 << 16) - 1))
-
-#define context_set_present(c) do {(c).lo |= 1;} while (0)
-#define context_set_fault_enable(c) \
-	do {(c).lo &= (((u64)-1) << 2) | 1;} while (0)
-#define context_set_translation_type(c, val) \
-	do { \
-		(c).lo &= (((u64)-1) << 4) | 3; \
-		(c).lo |= ((val) & 3) << 2; \
-	} while (0)
-#define CONTEXT_TT_MULTI_LEVEL 0
-#define context_set_address_root(c, val) \
-	do {(c).lo |= (val) & VTD_PAGE_MASK; } while (0)
-#define context_set_address_width(c, val) do {(c).hi |= (val) & 7;} while (0)
-#define context_set_domain_id(c, val) \
-	do {(c).hi |= ((val) & ((1 << 16) - 1)) << 8;} while (0)
-#define context_clear_entry(c) do {(c).lo = 0; (c).hi = 0;} while (0)
-
 /*
  * 0: readable
  * 1: writable
-- 
cgit v1.2.3


From 622ba12a4c2148999bda9b891bfd0c6ddcb6c57e Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 20 Nov 2008 15:49:46 +0000
Subject: intel-iommu: move DMA PTE defs out of dma_remapping.h

DMA_PTE_READ/WRITE are needed by kvm.

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c     | 22 ++++++++++++++++++++++
 include/linux/dma_remapping.h | 22 ----------------------
 2 files changed, 22 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 9d06f4bb6b5e..26c5402b6f7c 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -128,6 +128,28 @@ struct context_entry {
 	do {(c).hi |= ((val) & ((1 << 16) - 1)) << 8;} while (0)
 #define context_clear_entry(c) do {(c).lo = 0; (c).hi = 0;} while (0)
 
+/*
+ * 0: readable
+ * 1: writable
+ * 2-6: reserved
+ * 7: super page
+ * 8-11: available
+ * 12-63: Host physcial address
+ */
+struct dma_pte {
+	u64 val;
+};
+#define dma_clear_pte(p)	do {(p).val = 0;} while (0)
+
+#define dma_set_pte_readable(p) do {(p).val |= DMA_PTE_READ;} while (0)
+#define dma_set_pte_writable(p) do {(p).val |= DMA_PTE_WRITE;} while (0)
+#define dma_set_pte_prot(p, prot) \
+		do {(p).val = ((p).val & ~3) | ((prot) & 3); } while (0)
+#define dma_pte_addr(p) ((p).val & VTD_PAGE_MASK)
+#define dma_set_pte_addr(p, addr) do {\
+		(p).val |= ((addr) & VTD_PAGE_MASK); } while (0)
+#define dma_pte_present(p) (((p).val & 3) != 0)
+
 static void flush_unmaps_timeout(unsigned long data);
 
 DEFINE_TIMER(unmap_timer,  flush_unmaps_timeout, 0, 0);
diff --git a/include/linux/dma_remapping.h b/include/linux/dma_remapping.h
index 9a88f7d0262f..9d5874e3bec9 100644
--- a/include/linux/dma_remapping.h
+++ b/include/linux/dma_remapping.h
@@ -11,31 +11,9 @@
 
 struct root_entry;
 
-/*
- * 0: readable
- * 1: writable
- * 2-6: reserved
- * 7: super page
- * 8-11: available
- * 12-63: Host physcial address
- */
-struct dma_pte {
-	u64 val;
-};
-#define dma_clear_pte(p)	do {(p).val = 0;} while (0)
-
 #define DMA_PTE_READ (1)
 #define DMA_PTE_WRITE (2)
 
-#define dma_set_pte_readable(p) do {(p).val |= DMA_PTE_READ;} while (0)
-#define dma_set_pte_writable(p) do {(p).val |= DMA_PTE_WRITE;} while (0)
-#define dma_set_pte_prot(p, prot) \
-		do {(p).val = ((p).val & ~3) | ((prot) & 3); } while (0)
-#define dma_pte_addr(p) ((p).val & VTD_PAGE_MASK)
-#define dma_set_pte_addr(p, addr) do {\
-		(p).val |= ((addr) & VTD_PAGE_MASK); } while (0)
-#define dma_pte_present(p) (((p).val & 3) != 0)
-
 struct intel_iommu;
 
 struct dmar_domain {
-- 
cgit v1.2.3


From 99126f7ce14aff5f9371b2fa81fddb82be815794 Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 20 Nov 2008 15:49:47 +0000
Subject: intel-iommu: move struct dmar_domain def out dma_remapping.h

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c     | 18 ++++++++++++++++++
 include/linux/dma_remapping.h | 22 ++--------------------
 2 files changed, 20 insertions(+), 20 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 26c5402b6f7c..97c36b2ee611 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -150,6 +150,24 @@ struct dma_pte {
 		(p).val |= ((addr) & VTD_PAGE_MASK); } while (0)
 #define dma_pte_present(p) (((p).val & 3) != 0)
 
+struct dmar_domain {
+	int	id;			/* domain id */
+	struct intel_iommu *iommu;	/* back pointer to owning iommu */
+
+	struct list_head devices; 	/* all devices' list */
+	struct iova_domain iovad;	/* iova's that belong to this domain */
+
+	struct dma_pte	*pgd;		/* virtual address */
+	spinlock_t	mapping_lock;	/* page table lock */
+	int		gaw;		/* max guest address width */
+
+	/* adjusted guest address width, 0 is level 2 30-bit */
+	int		agaw;
+
+#define DOMAIN_FLAG_MULTIPLE_DEVICES 1
+	int		flags;
+};
+
 static void flush_unmaps_timeout(unsigned long data);
 
 DEFINE_TIMER(unmap_timer,  flush_unmaps_timeout, 0, 0);
diff --git a/include/linux/dma_remapping.h b/include/linux/dma_remapping.h
index 9d5874e3bec9..333014468f17 100644
--- a/include/linux/dma_remapping.h
+++ b/include/linux/dma_remapping.h
@@ -9,30 +9,12 @@
 #define VTD_PAGE_MASK		(((u64)-1) << VTD_PAGE_SHIFT)
 #define VTD_PAGE_ALIGN(addr)	(((addr) + VTD_PAGE_SIZE - 1) & VTD_PAGE_MASK)
 
-struct root_entry;
-
 #define DMA_PTE_READ (1)
 #define DMA_PTE_WRITE (2)
 
 struct intel_iommu;
-
-struct dmar_domain {
-	int	id;			/* domain id */
-	struct intel_iommu *iommu;	/* back pointer to owning iommu */
-
-	struct list_head devices; 	/* all devices' list */
-	struct iova_domain iovad;	/* iova's that belong to this domain */
-
-	struct dma_pte	*pgd;		/* virtual address */
-	spinlock_t	mapping_lock;	/* page table lock */
-	int		gaw;		/* max guest address width */
-
-	/* adjusted guest address width, 0 is level 2 30-bit */
-	int		agaw;
-
-#define DOMAIN_FLAG_MULTIPLE_DEVICES 1
-	int		flags;
-};
+struct dmar_domain;
+struct root_entry;
 
 /* PCI domain-device relationship */
 struct device_domain_info {
-- 
cgit v1.2.3


From a647dacbb1389aa6a5fa631766c1eaea35905890 Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 20 Nov 2008 15:49:48 +0000
Subject: intel-iommu: move struct device_domain_info out of dma_remapping.h

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c     | 10 ++++++++++
 include/linux/dma_remapping.h | 10 ----------
 2 files changed, 10 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 97c36b2ee611..f23a02054bf7 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -168,6 +168,16 @@ struct dmar_domain {
 	int		flags;
 };
 
+/* PCI domain-device relationship */
+struct device_domain_info {
+	struct list_head link;	/* link to domain siblings */
+	struct list_head global; /* link to global list */
+	u8 bus;			/* PCI bus numer */
+	u8 devfn;		/* PCI devfn number */
+	struct pci_dev *dev; /* it's NULL for PCIE-to-PCI bridge */
+	struct dmar_domain *domain; /* pointer to domain */
+};
+
 static void flush_unmaps_timeout(unsigned long data);
 
 DEFINE_TIMER(unmap_timer,  flush_unmaps_timeout, 0, 0);
diff --git a/include/linux/dma_remapping.h b/include/linux/dma_remapping.h
index 333014468f17..4ef5f6bc0d68 100644
--- a/include/linux/dma_remapping.h
+++ b/include/linux/dma_remapping.h
@@ -16,16 +16,6 @@ struct intel_iommu;
 struct dmar_domain;
 struct root_entry;
 
-/* PCI domain-device relationship */
-struct device_domain_info {
-	struct list_head link;	/* link to domain siblings */
-	struct list_head global; /* link to global list */
-	u8 bus;			/* PCI bus numer */
-	u8 devfn;		/* PCI devfn number */
-	struct pci_dev *dev; /* it's NULL for PCIE-to-PCI bridge */
-	struct dmar_domain *domain; /* pointer to domain */
-};
-
 extern void free_dmar_iommu(struct intel_iommu *iommu);
 
 extern int dmar_disabled;
-- 
cgit v1.2.3


From 58fa7304a2c2bfd46e505c293ef779aa1d9715c2 Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 20 Nov 2008 15:49:49 +0000
Subject: intel-iommu: kill off duplicate def of dmar_disabled

This is only used in dmar.c and intel-iommu.h, so dma_remapping.h
seems like the appropriate place for it.

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 include/linux/dmar.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index f1984fc3e06d..f28440784cf0 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -144,7 +144,6 @@ struct dmar_rmrr_unit {
 	list_for_each_entry(rmrr, &dmar_rmrr_units, list)
 /* Intel DMAR  initialization functions */
 extern int intel_iommu_init(void);
-extern int dmar_disabled;
 #else
 static inline int intel_iommu_init(void)
 {
-- 
cgit v1.2.3


From 2abd7e167c1b281f99bb58d302225872bfae9123 Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Thu, 20 Nov 2008 15:49:50 +0000
Subject: intel-iommu: move iommu_prepare_gfx_mapping() out of dma_remapping.h

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c     | 5 +++++
 include/linux/dma_remapping.h | 7 -------
 2 files changed, 5 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index f23a02054bf7..c1c59a619650 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1686,6 +1686,11 @@ static void __init iommu_prepare_gfx_mapping(void)
 			printk(KERN_ERR "IOMMU: mapping reserved region failed\n");
 	}
 }
+#else /* !CONFIG_DMAR_GFX_WA */
+static inline void iommu_prepare_gfx_mapping(void)
+{
+	return;
+}
 #endif
 
 #ifdef CONFIG_DMAR_FLOPPY_WA
diff --git a/include/linux/dma_remapping.h b/include/linux/dma_remapping.h
index 4ef5f6bc0d68..7799a85614c1 100644
--- a/include/linux/dma_remapping.h
+++ b/include/linux/dma_remapping.h
@@ -20,11 +20,4 @@ extern void free_dmar_iommu(struct intel_iommu *iommu);
 
 extern int dmar_disabled;
 
-#ifndef CONFIG_DMAR_GFX_WA
-static inline void iommu_prepare_gfx_mapping(void)
-{
-	return;
-}
-#endif /* !CONFIG_DMAR_GFX_WA */
-
 #endif
-- 
cgit v1.2.3


From 1b5736839ae13dadc5947940144f95dd0f4a4a8c Mon Sep 17 00:00:00 2001
From: Weidong Han <weidong.han@intel.com>
Date: Mon, 8 Dec 2008 15:34:06 +0800
Subject: calculate agaw for each iommu

"SAGAW" capability may be different across iommus. Use a default agaw, but if default agaw is not supported in some iommus, choose a less supported agaw.

Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 drivers/pci/dmar.c            | 10 ++++++++++
 drivers/pci/intel-iommu.c     | 22 ++++++++++++++++++++++
 include/linux/dma_remapping.h |  1 +
 include/linux/intel-iommu.h   |  1 +
 4 files changed, 34 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index 5f164ff3026e..f5a662a50acb 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -491,6 +491,7 @@ int alloc_iommu(struct dmar_drhd_unit *drhd)
 	int map_size;
 	u32 ver;
 	static int iommu_allocated = 0;
+	int agaw;
 
 	iommu = kzalloc(sizeof(*iommu), GFP_KERNEL);
 	if (!iommu)
@@ -506,6 +507,15 @@ int alloc_iommu(struct dmar_drhd_unit *drhd)
 	iommu->cap = dmar_readq(iommu->reg + DMAR_CAP_REG);
 	iommu->ecap = dmar_readq(iommu->reg + DMAR_ECAP_REG);
 
+	agaw = iommu_calculate_agaw(iommu);
+	if (agaw < 0) {
+		printk(KERN_ERR
+			"Cannot get a valid agaw for iommu (seq_id = %d)\n",
+			iommu->seq_id);
+		goto error;
+	}
+	iommu->agaw = agaw;
+
 	/* the registers might be more than one page */
 	map_size = max_t(int, ecap_max_iotlb_offset(iommu->ecap),
 		cap_max_fault_reg_offset(iommu->cap));
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 9dca689215eb..3ecfa2304c2c 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -362,6 +362,28 @@ void free_iova_mem(struct iova *iova)
 	kmem_cache_free(iommu_iova_cache, iova);
 }
 
+
+static inline int width_to_agaw(int width);
+
+/* calculate agaw for each iommu.
+ * "SAGAW" may be different across iommus, use a default agaw, and
+ * get a supported less agaw for iommus that don't support the default agaw.
+ */
+int iommu_calculate_agaw(struct intel_iommu *iommu)
+{
+	unsigned long sagaw;
+	int agaw = -1;
+
+	sagaw = cap_sagaw(iommu->cap);
+	for (agaw = width_to_agaw(DEFAULT_DOMAIN_ADDRESS_WIDTH);
+	     agaw >= 0; agaw--) {
+		if (test_bit(agaw, &sagaw))
+			break;
+	}
+
+	return agaw;
+}
+
 /* in native case, each domain is related to only one iommu */
 static struct intel_iommu *domain_get_iommu(struct dmar_domain *domain)
 {
diff --git a/include/linux/dma_remapping.h b/include/linux/dma_remapping.h
index 7799a85614c1..136f170cecc2 100644
--- a/include/linux/dma_remapping.h
+++ b/include/linux/dma_remapping.h
@@ -17,6 +17,7 @@ struct dmar_domain;
 struct root_entry;
 
 extern void free_dmar_iommu(struct intel_iommu *iommu);
+extern int iommu_calculate_agaw(struct intel_iommu *iommu);
 
 extern int dmar_disabled;
 
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index 1bff7bf1bc2c..06349fd5871b 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -290,6 +290,7 @@ struct intel_iommu {
 	u32		gcmd; /* Holds TE, EAFL. Don't need SRTP, SFL, WBF */
 	spinlock_t	register_lock; /* protect register handling */
 	int		seq_id;	/* sequence id of the iommu */
+	int		agaw; /* agaw of this iommu */
 
 #ifdef CONFIG_DMAR
 	unsigned long 	*domain_ids; /* bitmap of domains */
-- 
cgit v1.2.3


From faa3d6f5ffe7bf60ebfd0d36513fbcda0eb0ea1a Mon Sep 17 00:00:00 2001
From: Weidong Han <weidong.han@intel.com>
Date: Mon, 8 Dec 2008 23:09:29 +0800
Subject: Change intel iommu APIs of virtual machine domain

These APIs are used by KVM to use VT-d

Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 drivers/pci/intel-iommu.c   | 129 ++++++++++++++++++++------------------------
 include/linux/intel-iommu.h |  20 +++----
 2 files changed, 70 insertions(+), 79 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 8a204d5bb427..f1380269cabd 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -2944,96 +2944,87 @@ static void vm_domain_exit(struct dmar_domain *domain)
 	free_domain_mem(domain);
 }
 
-void intel_iommu_domain_exit(struct dmar_domain *domain)
+struct dmar_domain *intel_iommu_alloc_domain(void)
 {
-	u64 end;
-
-	/* Domain 0 is reserved, so dont process it */
-	if (!domain)
-		return;
-
-	end = DOMAIN_MAX_ADDR(domain->gaw);
-	end = end & (~VTD_PAGE_MASK);
-
-	/* clear ptes */
-	dma_pte_clear_range(domain, 0, end);
-
-	/* free page tables */
-	dma_pte_free_pagetable(domain, 0, end);
-
-	iommu_free_domain(domain);
-	free_domain_mem(domain);
-}
-EXPORT_SYMBOL_GPL(intel_iommu_domain_exit);
-
-struct dmar_domain *intel_iommu_domain_alloc(struct pci_dev *pdev)
-{
-	struct dmar_drhd_unit *drhd;
 	struct dmar_domain *domain;
-	struct intel_iommu *iommu;
 
-	drhd = dmar_find_matched_drhd_unit(pdev);
-	if (!drhd) {
-		printk(KERN_ERR "intel_iommu_domain_alloc: drhd == NULL\n");
-		return NULL;
-	}
-
-	iommu = drhd->iommu;
-	if (!iommu) {
-		printk(KERN_ERR
-			"intel_iommu_domain_alloc: iommu == NULL\n");
-		return NULL;
-	}
-	domain = iommu_alloc_domain(iommu);
+	domain = iommu_alloc_vm_domain();
 	if (!domain) {
 		printk(KERN_ERR
 			"intel_iommu_domain_alloc: domain == NULL\n");
 		return NULL;
 	}
-	if (domain_init(domain, DEFAULT_DOMAIN_ADDRESS_WIDTH)) {
+	if (vm_domain_init(domain, DEFAULT_DOMAIN_ADDRESS_WIDTH)) {
 		printk(KERN_ERR
 			"intel_iommu_domain_alloc: domain_init() failed\n");
-		intel_iommu_domain_exit(domain);
+		vm_domain_exit(domain);
 		return NULL;
 	}
+
 	return domain;
 }
-EXPORT_SYMBOL_GPL(intel_iommu_domain_alloc);
+EXPORT_SYMBOL_GPL(intel_iommu_alloc_domain);
 
-int intel_iommu_context_mapping(
-	struct dmar_domain *domain, struct pci_dev *pdev)
+void intel_iommu_free_domain(struct dmar_domain *domain)
 {
-	int rc;
-	rc = domain_context_mapping(domain, pdev);
-	return rc;
+	vm_domain_exit(domain);
 }
-EXPORT_SYMBOL_GPL(intel_iommu_context_mapping);
+EXPORT_SYMBOL_GPL(intel_iommu_free_domain);
 
-int intel_iommu_page_mapping(
-	struct dmar_domain *domain, dma_addr_t iova,
-	u64 hpa, size_t size, int prot)
+int intel_iommu_attach_device(struct dmar_domain *domain,
+			      struct pci_dev *pdev)
 {
-	int rc;
-	rc = domain_page_mapping(domain, iova, hpa, size, prot);
-	return rc;
+	int ret;
+
+	/* normally pdev is not mapped */
+	if (unlikely(domain_context_mapped(pdev))) {
+		struct dmar_domain *old_domain;
+
+		old_domain = find_domain(pdev);
+		if (old_domain) {
+			if (domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE)
+				vm_domain_remove_one_dev_info(old_domain, pdev);
+			else
+				domain_remove_dev_info(old_domain);
+		}
+	}
+
+	ret = domain_context_mapping(domain, pdev);
+	if (ret)
+		return ret;
+
+	ret = vm_domain_add_dev_info(domain, pdev);
+	return ret;
 }
-EXPORT_SYMBOL_GPL(intel_iommu_page_mapping);
+EXPORT_SYMBOL_GPL(intel_iommu_attach_device);
 
-void intel_iommu_detach_dev(struct dmar_domain *domain, u8 bus, u8 devfn)
+void intel_iommu_detach_device(struct dmar_domain *domain,
+			       struct pci_dev *pdev)
 {
-	struct intel_iommu *iommu;
+	vm_domain_remove_one_dev_info(domain, pdev);
+}
+EXPORT_SYMBOL_GPL(intel_iommu_detach_device);
 
-	iommu = device_to_iommu(bus, devfn);
-	iommu_detach_dev(iommu, bus, devfn);
+int intel_iommu_map_address(struct dmar_domain *domain, dma_addr_t iova,
+			    u64 hpa, size_t size, int prot)
+{
+	int ret;
+	ret = domain_page_mapping(domain, iova, hpa, size, prot);
+	return ret;
 }
-EXPORT_SYMBOL_GPL(intel_iommu_detach_dev);
+EXPORT_SYMBOL_GPL(intel_iommu_map_address);
 
-struct dmar_domain *
-intel_iommu_find_domain(struct pci_dev *pdev)
+void intel_iommu_unmap_address(struct dmar_domain *domain,
+			       dma_addr_t iova, size_t size)
 {
-	return find_domain(pdev);
+	dma_addr_t base;
+
+	/* The address might not be aligned */
+	base = iova & VTD_PAGE_MASK;
+	size = VTD_PAGE_ALIGN(size);
+	dma_pte_clear_range(domain, base, base + size);
 }
-EXPORT_SYMBOL_GPL(intel_iommu_find_domain);
+EXPORT_SYMBOL_GPL(intel_iommu_unmap_address);
 
 int intel_iommu_found(void)
 {
@@ -3041,17 +3032,15 @@ int intel_iommu_found(void)
 }
 EXPORT_SYMBOL_GPL(intel_iommu_found);
 
-u64 intel_iommu_iova_to_pfn(struct dmar_domain *domain, u64 iova)
+u64 intel_iommu_iova_to_phys(struct dmar_domain *domain, u64 iova)
 {
 	struct dma_pte *pte;
-	u64 pfn;
+	u64 phys = 0;
 
-	pfn = 0;
 	pte = addr_to_dma_pte(domain, iova);
-
 	if (pte)
-		pfn = dma_pte_addr(pte);
+		phys = dma_pte_addr(pte);
 
-	return pfn >> VTD_PAGE_SHIFT;
+	return phys;
 }
-EXPORT_SYMBOL_GPL(intel_iommu_iova_to_pfn);
+EXPORT_SYMBOL_GPL(intel_iommu_iova_to_phys);
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index 06349fd5871b..07973c4e4acc 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -330,15 +330,17 @@ extern int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
 
 extern void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu);
 
-void intel_iommu_domain_exit(struct dmar_domain *domain);
-struct dmar_domain *intel_iommu_domain_alloc(struct pci_dev *pdev);
-int intel_iommu_context_mapping(struct dmar_domain *domain,
-				struct pci_dev *pdev);
-int intel_iommu_page_mapping(struct dmar_domain *domain, dma_addr_t iova,
-			     u64 hpa, size_t size, int prot);
-void intel_iommu_detach_dev(struct dmar_domain *domain, u8 bus, u8 devfn);
-struct dmar_domain *intel_iommu_find_domain(struct pci_dev *pdev);
-u64 intel_iommu_iova_to_pfn(struct dmar_domain *domain, u64 iova);
+struct dmar_domain *intel_iommu_alloc_domain(void);
+void intel_iommu_free_domain(struct dmar_domain *domain);
+int intel_iommu_attach_device(struct dmar_domain *domain,
+			      struct pci_dev *pdev);
+void intel_iommu_detach_device(struct dmar_domain *domain,
+			       struct pci_dev *pdev);
+int intel_iommu_map_address(struct dmar_domain *domain, dma_addr_t iova,
+			    u64 hpa, size_t size, int prot);
+void intel_iommu_unmap_address(struct dmar_domain *domain,
+			       dma_addr_t iova, size_t size);
+u64 intel_iommu_iova_to_phys(struct dmar_domain *domain, u64 iova);
 
 #ifdef CONFIG_DMAR
 int intel_iommu_found(void);
-- 
cgit v1.2.3


From 260782bcfdaaa7850f29d6bb2ec6603019168c57 Mon Sep 17 00:00:00 2001
From: Weidong Han <weidong.han@intel.com>
Date: Tue, 2 Dec 2008 21:03:39 +0800
Subject: KVM: use the new intel iommu APIs

intel iommu APIs are updated, use the new APIs.

In addition, change kvm_iommu_map_guest() to just create the domain, let kvm_iommu_assign_device() assign device.

Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 include/linux/kvm_host.h | 15 +++++---
 virt/kvm/kvm_main.c      |  7 +++-
 virt/kvm/vtd.c           | 98 +++++++++++++++++++++++++++---------------------
 3 files changed, 71 insertions(+), 49 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index eafabd5c66b2..c96739b4b7a3 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -330,9 +330,10 @@ void kvm_free_irq_source_id(struct kvm *kvm, int irq_source_id);
 #ifdef CONFIG_DMAR
 int kvm_iommu_map_pages(struct kvm *kvm, gfn_t base_gfn,
 			unsigned long npages);
-int kvm_iommu_map_guest(struct kvm *kvm,
-			struct kvm_assigned_dev_kernel *assigned_dev);
+int kvm_iommu_map_guest(struct kvm *kvm);
 int kvm_iommu_unmap_guest(struct kvm *kvm);
+int kvm_assign_device(struct kvm *kvm,
+		      struct kvm_assigned_dev_kernel *assigned_dev);
 #else /* CONFIG_DMAR */
 static inline int kvm_iommu_map_pages(struct kvm *kvm,
 				      gfn_t base_gfn,
@@ -341,9 +342,7 @@ static inline int kvm_iommu_map_pages(struct kvm *kvm,
 	return 0;
 }
 
-static inline int kvm_iommu_map_guest(struct kvm *kvm,
-				      struct kvm_assigned_dev_kernel
-				      *assigned_dev)
+static inline int kvm_iommu_map_guest(struct kvm *kvm)
 {
 	return -ENODEV;
 }
@@ -352,6 +351,12 @@ static inline int kvm_iommu_unmap_guest(struct kvm *kvm)
 {
 	return 0;
 }
+
+static inline int kvm_assign_device(struct kvm *kvm,
+		struct kvm_assigned_dev_kernel *assigned_dev)
+{
+	return 0;
+}
 #endif /* CONFIG_DMAR */
 
 static inline void kvm_guest_enter(void)
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index fc6127cbea1f..c92b63462b79 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -503,7 +503,12 @@ static int kvm_vm_ioctl_assign_device(struct kvm *kvm,
 	list_add(&match->list, &kvm->arch.assigned_dev_head);
 
 	if (assigned_dev->flags & KVM_DEV_ASSIGN_ENABLE_IOMMU) {
-		r = kvm_iommu_map_guest(kvm, match);
+		if (!kvm->arch.intel_iommu_domain) {
+			r = kvm_iommu_map_guest(kvm);
+			if (r)
+				goto out_list_del;
+		}
+		r = kvm_assign_device(kvm, match);
 		if (r)
 			goto out_list_del;
 	}
diff --git a/virt/kvm/vtd.c b/virt/kvm/vtd.c
index a770874f3a3a..44bb58a395a5 100644
--- a/virt/kvm/vtd.c
+++ b/virt/kvm/vtd.c
@@ -45,20 +45,18 @@ int kvm_iommu_map_pages(struct kvm *kvm,
 
 	for (i = 0; i < npages; i++) {
 		/* check if already mapped */
-		pfn = (pfn_t)intel_iommu_iova_to_pfn(domain,
-						     gfn_to_gpa(gfn));
-		if (pfn)
+		if (intel_iommu_iova_to_phys(domain,
+					     gfn_to_gpa(gfn)))
 			continue;
 
 		pfn = gfn_to_pfn(kvm, gfn);
-		r = intel_iommu_page_mapping(domain,
-					     gfn_to_gpa(gfn),
-					     pfn_to_hpa(pfn),
-					     PAGE_SIZE,
-					     DMA_PTE_READ |
-					     DMA_PTE_WRITE);
+		r = intel_iommu_map_address(domain,
+					    gfn_to_gpa(gfn),
+					    pfn_to_hpa(pfn),
+					    PAGE_SIZE,
+					    DMA_PTE_READ | DMA_PTE_WRITE);
 		if (r) {
-			printk(KERN_ERR "kvm_iommu_map_pages:"
+			printk(KERN_ERR "kvm_iommu_map_address:"
 			       "iommu failed to map pfn=%lx\n", pfn);
 			goto unmap_pages;
 		}
@@ -86,50 +84,55 @@ static int kvm_iommu_map_memslots(struct kvm *kvm)
 	return r;
 }
 
-int kvm_iommu_map_guest(struct kvm *kvm,
-			struct kvm_assigned_dev_kernel *assigned_dev)
+int kvm_assign_device(struct kvm *kvm,
+		      struct kvm_assigned_dev_kernel *assigned_dev)
 {
 	struct pci_dev *pdev = NULL;
+	struct dmar_domain *domain = kvm->arch.intel_iommu_domain;
 	int r;
 
-	if (!intel_iommu_found()) {
-		printk(KERN_ERR "%s: intel iommu not found\n", __func__);
+	/* check if iommu exists and in use */
+	if (!domain)
+		return 0;
+
+	pdev = assigned_dev->dev;
+	if (pdev == NULL)
 		return -ENODEV;
+
+	r = intel_iommu_attach_device(domain, pdev);
+	if (r) {
+		printk(KERN_ERR "assign device %x:%x.%x failed",
+			pdev->bus->number,
+			PCI_SLOT(pdev->devfn),
+			PCI_FUNC(pdev->devfn));
+		return r;
 	}
 
-	printk(KERN_DEBUG "VT-d direct map: host bdf = %x:%x:%x\n",
-	       assigned_dev->host_busnr,
-	       PCI_SLOT(assigned_dev->host_devfn),
-	       PCI_FUNC(assigned_dev->host_devfn));
+	printk(KERN_DEBUG "assign device: host bdf = %x:%x:%x\n",
+		assigned_dev->host_busnr,
+		PCI_SLOT(assigned_dev->host_devfn),
+		PCI_FUNC(assigned_dev->host_devfn));
 
-	pdev = assigned_dev->dev;
+	return 0;
+}
 
-	if (pdev == NULL) {
-		if (kvm->arch.intel_iommu_domain) {
-			intel_iommu_domain_exit(kvm->arch.intel_iommu_domain);
-			kvm->arch.intel_iommu_domain = NULL;
-		}
+int kvm_iommu_map_guest(struct kvm *kvm)
+{
+	int r;
+
+	if (!intel_iommu_found()) {
+		printk(KERN_ERR "%s: intel iommu not found\n", __func__);
 		return -ENODEV;
 	}
 
-	kvm->arch.intel_iommu_domain = intel_iommu_domain_alloc(pdev);
+	kvm->arch.intel_iommu_domain = intel_iommu_alloc_domain();
 	if (!kvm->arch.intel_iommu_domain)
-		return -ENODEV;
+		return -ENOMEM;
 
 	r = kvm_iommu_map_memslots(kvm);
 	if (r)
 		goto out_unmap;
 
-	intel_iommu_detach_dev(kvm->arch.intel_iommu_domain,
-			       pdev->bus->number, pdev->devfn);
-
-	r = intel_iommu_context_mapping(kvm->arch.intel_iommu_domain,
-					pdev);
-	if (r) {
-		printk(KERN_ERR "Domain context map for %s failed",
-		       pci_name(pdev));
-		goto out_unmap;
-	}
 	return 0;
 
 out_unmap:
@@ -138,19 +141,29 @@ out_unmap:
 }
 
 static void kvm_iommu_put_pages(struct kvm *kvm,
-			       gfn_t base_gfn, unsigned long npages)
+				gfn_t base_gfn, unsigned long npages)
 {
 	gfn_t gfn = base_gfn;
 	pfn_t pfn;
 	struct dmar_domain *domain = kvm->arch.intel_iommu_domain;
-	int i;
+	unsigned long i;
+	u64 phys;
+
+	/* check if iommu exists and in use */
+	if (!domain)
+		return;
 
 	for (i = 0; i < npages; i++) {
-		pfn = (pfn_t)intel_iommu_iova_to_pfn(domain,
-						     gfn_to_gpa(gfn));
+		phys = intel_iommu_iova_to_phys(domain,
+						gfn_to_gpa(gfn));
+		pfn = phys >> PAGE_SHIFT;
 		kvm_release_pfn_clean(pfn);
 		gfn++;
 	}
+
+	intel_iommu_unmap_address(domain,
+				  gfn_to_gpa(base_gfn),
+				  PAGE_SIZE * npages);
 }
 
 static int kvm_iommu_unmap_memslots(struct kvm *kvm)
@@ -182,10 +195,9 @@ int kvm_iommu_unmap_guest(struct kvm *kvm)
 		       PCI_FUNC(entry->host_devfn));
 
 		/* detach kvm dmar domain */
-		intel_iommu_detach_dev(domain, entry->host_busnr,
-				       entry->host_devfn);
+		intel_iommu_detach_device(domain, entry->dev);
 	}
 	kvm_iommu_unmap_memslots(kvm);
-	intel_iommu_domain_exit(domain);
+	intel_iommu_free_domain(domain);
 	return 0;
 }
-- 
cgit v1.2.3


From 0a920356748df4fb06e86c21c23d2ed6d31d37ad Mon Sep 17 00:00:00 2001
From: Weidong Han <weidong.han@intel.com>
Date: Tue, 2 Dec 2008 21:24:23 +0800
Subject: KVM: support device deassignment

Support device deassignment, it can be used in device hotplug.

Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 include/linux/kvm_host.h |  8 ++++++++
 virt/kvm/kvm_main.c      | 42 ++++++++++++++++++++++++++++++++++++++++++
 virt/kvm/vtd.c           | 24 ++++++++++++++++++++++++
 3 files changed, 74 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index c96739b4b7a3..ce5d1c17ce26 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -334,6 +334,8 @@ int kvm_iommu_map_guest(struct kvm *kvm);
 int kvm_iommu_unmap_guest(struct kvm *kvm);
 int kvm_assign_device(struct kvm *kvm,
 		      struct kvm_assigned_dev_kernel *assigned_dev);
+int kvm_deassign_device(struct kvm *kvm,
+			struct kvm_assigned_dev_kernel *assigned_dev);
 #else /* CONFIG_DMAR */
 static inline int kvm_iommu_map_pages(struct kvm *kvm,
 				      gfn_t base_gfn,
@@ -357,6 +359,12 @@ static inline int kvm_assign_device(struct kvm *kvm,
 {
 	return 0;
 }
+
+static inline int kvm_deassign_device(struct kvm *kvm,
+		struct kvm_assigned_dev_kernel *assigned_dev)
+{
+	return 0;
+}
 #endif /* CONFIG_DMAR */
 
 static inline void kvm_guest_enter(void)
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index c92b63462b79..3238e08e4651 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -530,6 +530,35 @@ out_free:
 }
 #endif
 
+#ifdef KVM_CAP_DEVICE_DEASSIGNMENT
+static int kvm_vm_ioctl_deassign_device(struct kvm *kvm,
+		struct kvm_assigned_pci_dev *assigned_dev)
+{
+	int r = 0;
+	struct kvm_assigned_dev_kernel *match;
+
+	mutex_lock(&kvm->lock);
+
+	match = kvm_find_assigned_dev(&kvm->arch.assigned_dev_head,
+				      assigned_dev->assigned_dev_id);
+	if (!match) {
+		printk(KERN_INFO "%s: device hasn't been assigned before, "
+		  "so cannot be deassigned\n", __func__);
+		r = -EINVAL;
+		goto out;
+	}
+
+	if (assigned_dev->flags & KVM_DEV_ASSIGN_ENABLE_IOMMU)
+		kvm_deassign_device(kvm, match);
+
+	kvm_free_assigned_device(kvm, match);
+
+out:
+	mutex_unlock(&kvm->lock);
+	return r;
+}
+#endif
+
 static inline int valid_vcpu(int n)
 {
 	return likely(n >= 0 && n < KVM_MAX_VCPUS);
@@ -1862,6 +1891,19 @@ static long kvm_vm_ioctl(struct file *filp,
 			goto out;
 		break;
 	}
+#endif
+#ifdef KVM_CAP_DEVICE_DEASSIGNMENT
+	case KVM_DEASSIGN_PCI_DEVICE: {
+		struct kvm_assigned_pci_dev assigned_dev;
+
+		r = -EFAULT;
+		if (copy_from_user(&assigned_dev, argp, sizeof assigned_dev))
+			goto out;
+		r = kvm_vm_ioctl_deassign_device(kvm, &assigned_dev);
+		if (r)
+			goto out;
+		break;
+	}
 #endif
 	default:
 		r = kvm_arch_vm_ioctl(filp, ioctl, arg);
diff --git a/virt/kvm/vtd.c b/virt/kvm/vtd.c
index 44bb58a395a5..174ea1f8cee5 100644
--- a/virt/kvm/vtd.c
+++ b/virt/kvm/vtd.c
@@ -116,6 +116,30 @@ int kvm_assign_device(struct kvm *kvm,
 	return 0;
 }
 
+int kvm_deassign_device(struct kvm *kvm,
+			struct kvm_assigned_dev_kernel *assigned_dev)
+{
+	struct dmar_domain *domain = kvm->arch.intel_iommu_domain;
+	struct pci_dev *pdev = NULL;
+
+	/* check if iommu exists and in use */
+	if (!domain)
+		return 0;
+
+	pdev = assigned_dev->dev;
+	if (pdev == NULL)
+		return -ENODEV;
+
+	intel_iommu_detach_device(domain, pdev);
+
+	printk(KERN_DEBUG "deassign device: host bdf = %x:%x:%x\n",
+		assigned_dev->host_busnr,
+		PCI_SLOT(assigned_dev->host_devfn),
+		PCI_FUNC(assigned_dev->host_devfn));
+
+	return 0;
+}
+
 int kvm_iommu_map_guest(struct kvm *kvm)
 {
 	int r;
-- 
cgit v1.2.3


From b653574a7d14b663cc812cb20be6a114939ba186 Mon Sep 17 00:00:00 2001
From: Weidong Han <weidong.han@intel.com>
Date: Mon, 8 Dec 2008 23:29:53 +0800
Subject: Deassign device in kvm_free_assgined_device

In kvm_iommu_unmap_memslots(), assigned_dev_head is already empty.

Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 include/linux/kvm_host.h |  1 +
 virt/kvm/kvm_main.c      |  1 +
 virt/kvm/vtd.c           | 10 ----------
 3 files changed, 2 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index ce5d1c17ce26..e62a4629e51c 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -316,6 +316,7 @@ struct kvm_assigned_dev_kernel {
 #define KVM_ASSIGNED_DEV_HOST_MSI	(1 << 9)
 	unsigned long irq_requested_type;
 	int irq_source_id;
+	int flags;
 	struct pci_dev *dev;
 	struct kvm *kvm;
 };
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 3238e08e4651..4ef0fb43d1f9 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -496,6 +496,7 @@ static int kvm_vm_ioctl_assign_device(struct kvm *kvm,
 	match->assigned_dev_id = assigned_dev->assigned_dev_id;
 	match->host_busnr = assigned_dev->busnr;
 	match->host_devfn = assigned_dev->devfn;
+	match->flags = assigned_dev->flags;
 	match->dev = dev;
 	match->irq_source_id = -1;
 	match->kvm = kvm;
diff --git a/virt/kvm/vtd.c b/virt/kvm/vtd.c
index 174ea1f8cee5..d46de9af838b 100644
--- a/virt/kvm/vtd.c
+++ b/virt/kvm/vtd.c
@@ -205,22 +205,12 @@ static int kvm_iommu_unmap_memslots(struct kvm *kvm)
 
 int kvm_iommu_unmap_guest(struct kvm *kvm)
 {
-	struct kvm_assigned_dev_kernel *entry;
 	struct dmar_domain *domain = kvm->arch.intel_iommu_domain;
 
 	/* check if iommu exists and in use */
 	if (!domain)
 		return 0;
 
-	list_for_each_entry(entry, &kvm->arch.assigned_dev_head, list) {
-		printk(KERN_DEBUG "VT-d unmap: host bdf = %x:%x:%x\n",
-		       entry->host_busnr,
-		       PCI_SLOT(entry->host_devfn),
-		       PCI_FUNC(entry->host_devfn));
-
-		/* detach kvm dmar domain */
-		intel_iommu_detach_device(domain, entry->dev);
-	}
 	kvm_iommu_unmap_memslots(kvm);
 	intel_iommu_free_domain(domain);
 	return 0;
-- 
cgit v1.2.3


From 4a77a6cf6d9bf9f5c74b27f62bd2bfe6dcc88392 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 26 Nov 2008 17:02:33 +0100
Subject: introcude linux/iommu.h for an iommu api

This patch introduces the API to abstract the exported VT-d functions
for KVM into a generic API. This way the AMD IOMMU implementation can
plug into this API later.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 include/linux/iommu.h | 112 ++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 112 insertions(+)
 create mode 100644 include/linux/iommu.h

(limited to 'include/linux')

diff --git a/include/linux/iommu.h b/include/linux/iommu.h
new file mode 100644
index 000000000000..8a7bfb1b6ca0
--- /dev/null
+++ b/include/linux/iommu.h
@@ -0,0 +1,112 @@
+/*
+ * Copyright (C) 2007-2008 Advanced Micro Devices, Inc.
+ * Author: Joerg Roedel <joerg.roedel@amd.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
+ */
+
+#ifndef __LINUX_IOMMU_H
+#define __LINUX_IOMMU_H
+
+#define IOMMU_READ	(1)
+#define IOMMU_WRITE	(2)
+
+struct device;
+
+struct iommu_domain {
+	void *priv;
+};
+
+struct iommu_ops {
+	int (*domain_init)(struct iommu_domain *domain);
+	void (*domain_destroy)(struct iommu_domain *domain);
+	int (*attach_dev)(struct iommu_domain *domain, struct device *dev);
+	void (*detach_dev)(struct iommu_domain *domain, struct device *dev);
+	int (*map)(struct iommu_domain *domain, unsigned long iova,
+		   phys_addr_t paddr, size_t size, int prot);
+	void (*unmap)(struct iommu_domain *domain, unsigned long iova,
+		      size_t size);
+	phys_addr_t (*iova_to_phys)(struct iommu_domain *domain,
+				    unsigned long iova);
+};
+
+#ifdef CONFIG_IOMMU_API
+
+extern void register_iommu(struct iommu_ops *ops);
+extern bool iommu_found(void);
+extern struct iommu_domain *iommu_domain_alloc(void);
+extern void iommu_domain_free(struct iommu_domain *domain);
+extern int iommu_attach_device(struct iommu_domain *domain,
+			       struct device *dev);
+extern void iommu_detach_device(struct iommu_domain *domain,
+				struct device *dev);
+extern int iommu_map_range(struct iommu_domain *domain, unsigned long iova,
+			   phys_addr_t paddr, size_t size, int prot);
+extern void iommu_unmap_range(struct iommu_domain *domain, unsigned long iova,
+			      size_t size);
+extern phys_addr_t iommu_iova_to_phys(struct iommu_domain *domain,
+				      unsigned long iova);
+
+#else /* CONFIG_IOMMU_API */
+
+static inline void register_iommu(struct iommu_ops *ops)
+{
+}
+
+static inline bool iommu_found(void)
+{
+	return false;
+}
+
+static inline struct iommu_domain *iommu_domain_alloc(void)
+{
+	return NULL;
+}
+
+static inline void iommu_domain_free(struct iommu_domain *domain)
+{
+}
+
+static inline int iommu_attach_device(struct iommu_domain *domain,
+				      struct device *dev)
+{
+	return -ENODEV;
+}
+
+static inline void iommu_detach_device(struct iommu_domain *domain,
+				       struct device *dev)
+{
+}
+
+static inline int iommu_map_range(struct iommu_domain *domain,
+				  unsigned long iova, phys_addr_t paddr,
+				  size_t size, int prot)
+{
+	return -ENODEV;
+}
+
+static inline void iommu_unmap_range(struct iommu_domain *domain,
+				     unsigned long iova, size_t size)
+{
+}
+
+static inline phys_addr_t iommu_iova_to_phys(struct iommu_domain *domain,
+					     unsigned long iova)
+{
+	return 0;
+}
+
+#endif /* CONFIG_IOMMU_API */
+
+#endif /* __LINUX_IOMMU_H */
-- 
cgit v1.2.3


From 19de40a8472fa64693eab844911eec277d489f6c Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 3 Dec 2008 14:43:34 +0100
Subject: KVM: change KVM to use IOMMU API

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/ia64/include/asm/kvm_host.h |  2 +-
 arch/ia64/kvm/Makefile           |  2 +-
 arch/ia64/kvm/kvm-ia64.c         |  3 ++-
 arch/x86/include/asm/kvm_host.h  |  2 +-
 arch/x86/kvm/Makefile            |  2 +-
 arch/x86/kvm/x86.c               |  3 ++-
 include/linux/kvm_host.h         |  6 +++---
 virt/kvm/iommu.c                 | 45 +++++++++++++++++++---------------------
 virt/kvm/kvm_main.c              |  2 +-
 9 files changed, 33 insertions(+), 34 deletions(-)

(limited to 'include/linux')

diff --git a/arch/ia64/include/asm/kvm_host.h b/arch/ia64/include/asm/kvm_host.h
index 0560f3fae538..348663661659 100644
--- a/arch/ia64/include/asm/kvm_host.h
+++ b/arch/ia64/include/asm/kvm_host.h
@@ -467,7 +467,7 @@ struct kvm_arch {
 	struct kvm_sal_data rdv_sal_data;
 
 	struct list_head assigned_dev_head;
-	struct dmar_domain *intel_iommu_domain;
+	struct iommu_domain *iommu_domain;
 	struct hlist_head irq_ack_notifier_list;
 
 	unsigned long irq_sources_bitmap;
diff --git a/arch/ia64/kvm/Makefile b/arch/ia64/kvm/Makefile
index cb69dfcfb1a6..0bb99b732908 100644
--- a/arch/ia64/kvm/Makefile
+++ b/arch/ia64/kvm/Makefile
@@ -51,7 +51,7 @@ EXTRA_AFLAGS += -Ivirt/kvm -Iarch/ia64/kvm/
 common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o \
 		coalesced_mmio.o irq_comm.o)
 
-ifeq ($(CONFIG_DMAR),y)
+ifeq ($(CONFIG_IOMMU_API),y)
 common-objs += $(addprefix ../../../virt/kvm/, iommu.o)
 endif
 
diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c
index 0f5ebd948437..4e586f6110aa 100644
--- a/arch/ia64/kvm/kvm-ia64.c
+++ b/arch/ia64/kvm/kvm-ia64.c
@@ -31,6 +31,7 @@
 #include <linux/bitops.h>
 #include <linux/hrtimer.h>
 #include <linux/uaccess.h>
+#include <linux/iommu.h>
 #include <linux/intel-iommu.h>
 
 #include <asm/pgtable.h>
@@ -188,7 +189,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 		r = KVM_COALESCED_MMIO_PAGE_OFFSET;
 		break;
 	case KVM_CAP_IOMMU:
-		r = intel_iommu_found();
+		r = iommu_found();
 		break;
 	default:
 		r = 0;
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 97215a458e5f..730843d1d2fb 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -360,7 +360,7 @@ struct kvm_arch{
 	struct list_head active_mmu_pages;
 	struct list_head assigned_dev_head;
 	struct list_head oos_global_pages;
-	struct dmar_domain *intel_iommu_domain;
+	struct iommu_domain *iommu_domain;
 	struct kvm_pic *vpic;
 	struct kvm_ioapic *vioapic;
 	struct kvm_pit *vpit;
diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile
index 00f46c2d14bd..d3ec292f00f2 100644
--- a/arch/x86/kvm/Makefile
+++ b/arch/x86/kvm/Makefile
@@ -7,7 +7,7 @@ common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o \
 ifeq ($(CONFIG_KVM_TRACE),y)
 common-objs += $(addprefix ../../../virt/kvm/, kvm_trace.o)
 endif
-ifeq ($(CONFIG_DMAR),y)
+ifeq ($(CONFIG_IOMMU_API),y)
 common-objs += $(addprefix ../../../virt/kvm/, iommu.o)
 endif
 
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 0e6aa8141dcd..cc17546a2406 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -34,6 +34,7 @@
 #include <linux/module.h>
 #include <linux/mman.h>
 #include <linux/highmem.h>
+#include <linux/iommu.h>
 #include <linux/intel-iommu.h>
 
 #include <asm/uaccess.h>
@@ -989,7 +990,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 		r = !tdp_enabled;
 		break;
 	case KVM_CAP_IOMMU:
-		r = intel_iommu_found();
+		r = iommu_found();
 		break;
 	default:
 		r = 0;
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index e62a4629e51c..ec49d0be7f52 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -328,7 +328,7 @@ void kvm_unregister_irq_ack_notifier(struct kvm_irq_ack_notifier *kian);
 int kvm_request_irq_source_id(struct kvm *kvm);
 void kvm_free_irq_source_id(struct kvm *kvm, int irq_source_id);
 
-#ifdef CONFIG_DMAR
+#ifdef CONFIG_IOMMU_API
 int kvm_iommu_map_pages(struct kvm *kvm, gfn_t base_gfn,
 			unsigned long npages);
 int kvm_iommu_map_guest(struct kvm *kvm);
@@ -337,7 +337,7 @@ int kvm_assign_device(struct kvm *kvm,
 		      struct kvm_assigned_dev_kernel *assigned_dev);
 int kvm_deassign_device(struct kvm *kvm,
 			struct kvm_assigned_dev_kernel *assigned_dev);
-#else /* CONFIG_DMAR */
+#else /* CONFIG_IOMMU_API */
 static inline int kvm_iommu_map_pages(struct kvm *kvm,
 				      gfn_t base_gfn,
 				      unsigned long npages)
@@ -366,7 +366,7 @@ static inline int kvm_deassign_device(struct kvm *kvm,
 {
 	return 0;
 }
-#endif /* CONFIG_DMAR */
+#endif /* CONFIG_IOMMU_API */
 
 static inline void kvm_guest_enter(void)
 {
diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c
index d46de9af838b..d0bebaa5bf0a 100644
--- a/virt/kvm/iommu.c
+++ b/virt/kvm/iommu.c
@@ -25,6 +25,7 @@
 #include <linux/kvm_host.h>
 #include <linux/pci.h>
 #include <linux/dmar.h>
+#include <linux/iommu.h>
 #include <linux/intel-iommu.h>
 
 static int kvm_iommu_unmap_memslots(struct kvm *kvm);
@@ -37,7 +38,7 @@ int kvm_iommu_map_pages(struct kvm *kvm,
 	gfn_t gfn = base_gfn;
 	pfn_t pfn;
 	int i, r = 0;
-	struct dmar_domain *domain = kvm->arch.intel_iommu_domain;
+	struct iommu_domain *domain = kvm->arch.iommu_domain;
 
 	/* check if iommu exists and in use */
 	if (!domain)
@@ -45,16 +46,15 @@ int kvm_iommu_map_pages(struct kvm *kvm,
 
 	for (i = 0; i < npages; i++) {
 		/* check if already mapped */
-		if (intel_iommu_iova_to_phys(domain,
-					     gfn_to_gpa(gfn)))
+		if (iommu_iova_to_phys(domain, gfn_to_gpa(gfn)))
 			continue;
 
 		pfn = gfn_to_pfn(kvm, gfn);
-		r = intel_iommu_map_address(domain,
-					    gfn_to_gpa(gfn),
-					    pfn_to_hpa(pfn),
-					    PAGE_SIZE,
-					    DMA_PTE_READ | DMA_PTE_WRITE);
+		r = iommu_map_range(domain,
+				    gfn_to_gpa(gfn),
+				    pfn_to_hpa(pfn),
+				    PAGE_SIZE,
+				    IOMMU_READ | IOMMU_WRITE);
 		if (r) {
 			printk(KERN_ERR "kvm_iommu_map_address:"
 			       "iommu failed to map pfn=%lx\n", pfn);
@@ -88,7 +88,7 @@ int kvm_assign_device(struct kvm *kvm,
 		      struct kvm_assigned_dev_kernel *assigned_dev)
 {
 	struct pci_dev *pdev = NULL;
-	struct dmar_domain *domain = kvm->arch.intel_iommu_domain;
+	struct iommu_domain *domain = kvm->arch.iommu_domain;
 	int r;
 
 	/* check if iommu exists and in use */
@@ -99,7 +99,7 @@ int kvm_assign_device(struct kvm *kvm,
 	if (pdev == NULL)
 		return -ENODEV;
 
-	r = intel_iommu_attach_device(domain, pdev);
+	r = iommu_attach_device(domain, &pdev->dev);
 	if (r) {
 		printk(KERN_ERR "assign device %x:%x.%x failed",
 			pdev->bus->number,
@@ -119,7 +119,7 @@ int kvm_assign_device(struct kvm *kvm,
 int kvm_deassign_device(struct kvm *kvm,
 			struct kvm_assigned_dev_kernel *assigned_dev)
 {
-	struct dmar_domain *domain = kvm->arch.intel_iommu_domain;
+	struct iommu_domain *domain = kvm->arch.iommu_domain;
 	struct pci_dev *pdev = NULL;
 
 	/* check if iommu exists and in use */
@@ -130,7 +130,7 @@ int kvm_deassign_device(struct kvm *kvm,
 	if (pdev == NULL)
 		return -ENODEV;
 
-	intel_iommu_detach_device(domain, pdev);
+	iommu_detach_device(domain, &pdev->dev);
 
 	printk(KERN_DEBUG "deassign device: host bdf = %x:%x:%x\n",
 		assigned_dev->host_busnr,
@@ -144,13 +144,13 @@ int kvm_iommu_map_guest(struct kvm *kvm)
 {
 	int r;
 
-	if (!intel_iommu_found()) {
-		printk(KERN_ERR "%s: intel iommu not found\n", __func__);
+	if (!iommu_found()) {
+		printk(KERN_ERR "%s: iommu not found\n", __func__);
 		return -ENODEV;
 	}
 
-	kvm->arch.intel_iommu_domain = intel_iommu_alloc_domain();
-	if (!kvm->arch.intel_iommu_domain)
+	kvm->arch.iommu_domain = iommu_domain_alloc();
+	if (!kvm->arch.iommu_domain)
 		return -ENOMEM;
 
 	r = kvm_iommu_map_memslots(kvm);
@@ -169,7 +169,7 @@ static void kvm_iommu_put_pages(struct kvm *kvm,
 {
 	gfn_t gfn = base_gfn;
 	pfn_t pfn;
-	struct dmar_domain *domain = kvm->arch.intel_iommu_domain;
+	struct iommu_domain *domain = kvm->arch.iommu_domain;
 	unsigned long i;
 	u64 phys;
 
@@ -178,16 +178,13 @@ static void kvm_iommu_put_pages(struct kvm *kvm,
 		return;
 
 	for (i = 0; i < npages; i++) {
-		phys = intel_iommu_iova_to_phys(domain,
-						gfn_to_gpa(gfn));
+		phys = iommu_iova_to_phys(domain, gfn_to_gpa(gfn));
 		pfn = phys >> PAGE_SHIFT;
 		kvm_release_pfn_clean(pfn);
 		gfn++;
 	}
 
-	intel_iommu_unmap_address(domain,
-				  gfn_to_gpa(base_gfn),
-				  PAGE_SIZE * npages);
+	iommu_unmap_range(domain, gfn_to_gpa(base_gfn), PAGE_SIZE * npages);
 }
 
 static int kvm_iommu_unmap_memslots(struct kvm *kvm)
@@ -205,13 +202,13 @@ static int kvm_iommu_unmap_memslots(struct kvm *kvm)
 
 int kvm_iommu_unmap_guest(struct kvm *kvm)
 {
-	struct dmar_domain *domain = kvm->arch.intel_iommu_domain;
+	struct iommu_domain *domain = kvm->arch.iommu_domain;
 
 	/* check if iommu exists and in use */
 	if (!domain)
 		return 0;
 
 	kvm_iommu_unmap_memslots(kvm);
-	intel_iommu_free_domain(domain);
+	iommu_domain_free(domain);
 	return 0;
 }
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 4ef0fb43d1f9..3a5a08298aab 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -504,7 +504,7 @@ static int kvm_vm_ioctl_assign_device(struct kvm *kvm,
 	list_add(&match->list, &kvm->arch.assigned_dev_head);
 
 	if (assigned_dev->flags & KVM_DEV_ASSIGN_ENABLE_IOMMU) {
-		if (!kvm->arch.intel_iommu_domain) {
+		if (!kvm->arch.iommu_domain) {
 			r = kvm_iommu_map_guest(kvm);
 			if (r)
 				goto out_list_del;
-- 
cgit v1.2.3


From 5d450806eb0e569c5846a5825e7f535980b0da32 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 3 Dec 2008 14:52:32 +0100
Subject: VT-d: adapt domain init and destroy functions for IOMMU API

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 drivers/pci/intel-iommu.c   | 33 ++++++++++++++++++---------------
 include/linux/intel-iommu.h |  2 --
 2 files changed, 18 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 772fb22e1be0..5c95a5a65440 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -34,6 +34,7 @@
 #include <linux/mempool.h>
 #include <linux/timer.h>
 #include <linux/iova.h>
+#include <linux/iommu.h>
 #include <linux/intel-iommu.h>
 #include <asm/cacheflush.h>
 #include <asm/iommu.h>
@@ -2962,32 +2963,34 @@ static void vm_domain_exit(struct dmar_domain *domain)
 	free_domain_mem(domain);
 }
 
-struct dmar_domain *intel_iommu_alloc_domain(void)
+static int intel_iommu_domain_init(struct iommu_domain *domain)
 {
-	struct dmar_domain *domain;
+	struct dmar_domain *dmar_domain;
 
-	domain = iommu_alloc_vm_domain();
-	if (!domain) {
+	dmar_domain = iommu_alloc_vm_domain();
+	if (!dmar_domain) {
 		printk(KERN_ERR
-			"intel_iommu_domain_alloc: domain == NULL\n");
-		return NULL;
+			"intel_iommu_domain_init: dmar_domain == NULL\n");
+		return -ENOMEM;
 	}
-	if (vm_domain_init(domain, DEFAULT_DOMAIN_ADDRESS_WIDTH)) {
+	if (vm_domain_init(dmar_domain, DEFAULT_DOMAIN_ADDRESS_WIDTH)) {
 		printk(KERN_ERR
-			"intel_iommu_domain_alloc: domain_init() failed\n");
-		vm_domain_exit(domain);
-		return NULL;
+			"intel_iommu_domain_init() failed\n");
+		vm_domain_exit(dmar_domain);
+		return -ENOMEM;
 	}
+	domain->priv = dmar_domain;
 
-	return domain;
+	return 0;
 }
-EXPORT_SYMBOL_GPL(intel_iommu_alloc_domain);
 
-void intel_iommu_free_domain(struct dmar_domain *domain)
+static void intel_iommu_domain_destroy(struct iommu_domain *domain)
 {
-	vm_domain_exit(domain);
+	struct dmar_domain *dmar_domain = domain->priv;
+
+	domain->priv = NULL;
+	vm_domain_exit(dmar_domain);
 }
-EXPORT_SYMBOL_GPL(intel_iommu_free_domain);
 
 int intel_iommu_attach_device(struct dmar_domain *domain,
 			      struct pci_dev *pdev)
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index 07973c4e4acc..0a7ba0cefc74 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -330,8 +330,6 @@ extern int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
 
 extern void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu);
 
-struct dmar_domain *intel_iommu_alloc_domain(void);
-void intel_iommu_free_domain(struct dmar_domain *domain);
 int intel_iommu_attach_device(struct dmar_domain *domain,
 			      struct pci_dev *pdev);
 void intel_iommu_detach_device(struct dmar_domain *domain,
-- 
cgit v1.2.3


From 4c5478c94eb29e6101f1f13175f7455bc8b5d953 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 3 Dec 2008 14:58:24 +0100
Subject: VT-d: adapt device attach and detach functions for IOMMU API

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 drivers/pci/intel-iommu.c   | 27 +++++++++++++++------------
 include/linux/intel-iommu.h |  4 ----
 2 files changed, 15 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 5c95a5a65440..db9a26cfeb8f 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -2992,9 +2992,11 @@ static void intel_iommu_domain_destroy(struct iommu_domain *domain)
 	vm_domain_exit(dmar_domain);
 }
 
-int intel_iommu_attach_device(struct dmar_domain *domain,
-			      struct pci_dev *pdev)
+static int intel_iommu_attach_device(struct iommu_domain *domain,
+				     struct device *dev)
 {
+	struct dmar_domain *dmar_domain = domain->priv;
+	struct pci_dev *pdev = to_pci_dev(dev);
 	struct intel_iommu *iommu;
 	int addr_width;
 	u64 end;
@@ -3006,7 +3008,7 @@ int intel_iommu_attach_device(struct dmar_domain *domain,
 
 		old_domain = find_domain(pdev);
 		if (old_domain) {
-			if (domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE)
+			if (dmar_domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE)
 				vm_domain_remove_one_dev_info(old_domain, pdev);
 			else
 				domain_remove_dev_info(old_domain);
@@ -3021,28 +3023,29 @@ int intel_iommu_attach_device(struct dmar_domain *domain,
 	addr_width = agaw_to_width(iommu->agaw);
 	end = DOMAIN_MAX_ADDR(addr_width);
 	end = end & VTD_PAGE_MASK;
-	if (end < domain->max_addr) {
+	if (end < dmar_domain->max_addr) {
 		printk(KERN_ERR "%s: iommu agaw (%d) is not "
 		       "sufficient for the mapped address (%llx)\n",
-		       __func__, iommu->agaw, domain->max_addr);
+		       __func__, iommu->agaw, dmar_domain->max_addr);
 		return -EFAULT;
 	}
 
-	ret = domain_context_mapping(domain, pdev);
+	ret = domain_context_mapping(dmar_domain, pdev);
 	if (ret)
 		return ret;
 
-	ret = vm_domain_add_dev_info(domain, pdev);
+	ret = vm_domain_add_dev_info(dmar_domain, pdev);
 	return ret;
 }
-EXPORT_SYMBOL_GPL(intel_iommu_attach_device);
 
-void intel_iommu_detach_device(struct dmar_domain *domain,
-			       struct pci_dev *pdev)
+static void intel_iommu_detach_device(struct iommu_domain *domain,
+				      struct device *dev)
 {
-	vm_domain_remove_one_dev_info(domain, pdev);
+	struct dmar_domain *dmar_domain = domain->priv;
+	struct pci_dev *pdev = to_pci_dev(dev);
+
+	vm_domain_remove_one_dev_info(dmar_domain, pdev);
 }
-EXPORT_SYMBOL_GPL(intel_iommu_detach_device);
 
 int intel_iommu_map_address(struct dmar_domain *domain, dma_addr_t iova,
 			    u64 hpa, size_t size, int prot)
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index 0a7ba0cefc74..9909c5a1b20f 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -330,10 +330,6 @@ extern int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
 
 extern void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu);
 
-int intel_iommu_attach_device(struct dmar_domain *domain,
-			      struct pci_dev *pdev);
-void intel_iommu_detach_device(struct dmar_domain *domain,
-			       struct pci_dev *pdev);
 int intel_iommu_map_address(struct dmar_domain *domain, dma_addr_t iova,
 			    u64 hpa, size_t size, int prot);
 void intel_iommu_unmap_address(struct dmar_domain *domain,
-- 
cgit v1.2.3


From dde57a210dcdce85e2813bab8f88687761d9f6a6 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 3 Dec 2008 15:04:09 +0100
Subject: VT-d: adapt domain map and unmap functions for IOMMU API

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 drivers/pci/intel-iommu.c   | 33 ++++++++++++++++++++-------------
 include/linux/intel-iommu.h |  4 ----
 2 files changed, 20 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index db9a26cfeb8f..8af6c96f31b3 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -3047,20 +3047,28 @@ static void intel_iommu_detach_device(struct iommu_domain *domain,
 	vm_domain_remove_one_dev_info(dmar_domain, pdev);
 }
 
-int intel_iommu_map_address(struct dmar_domain *domain, dma_addr_t iova,
-			    u64 hpa, size_t size, int prot)
+static int intel_iommu_map_range(struct iommu_domain *domain,
+				 unsigned long iova, phys_addr_t hpa,
+				 size_t size, int iommu_prot)
 {
+	struct dmar_domain *dmar_domain = domain->priv;
 	u64 max_addr;
 	int addr_width;
+	int prot = 0;
 	int ret;
 
+	if (iommu_prot & IOMMU_READ)
+		prot |= DMA_PTE_READ;
+	if (iommu_prot & IOMMU_WRITE)
+		prot |= DMA_PTE_WRITE;
+
 	max_addr = (iova & VTD_PAGE_MASK) + VTD_PAGE_ALIGN(size);
-	if (domain->max_addr < max_addr) {
+	if (dmar_domain->max_addr < max_addr) {
 		int min_agaw;
 		u64 end;
 
 		/* check if minimum agaw is sufficient for mapped address */
-		min_agaw = vm_domain_min_agaw(domain);
+		min_agaw = vm_domain_min_agaw(dmar_domain);
 		addr_width = agaw_to_width(min_agaw);
 		end = DOMAIN_MAX_ADDR(addr_width);
 		end = end & VTD_PAGE_MASK;
@@ -3070,28 +3078,27 @@ int intel_iommu_map_address(struct dmar_domain *domain, dma_addr_t iova,
 			       __func__, min_agaw, max_addr);
 			return -EFAULT;
 		}
-		domain->max_addr = max_addr;
+		dmar_domain->max_addr = max_addr;
 	}
 
-	ret = domain_page_mapping(domain, iova, hpa, size, prot);
+	ret = domain_page_mapping(dmar_domain, iova, hpa, size, prot);
 	return ret;
 }
-EXPORT_SYMBOL_GPL(intel_iommu_map_address);
 
-void intel_iommu_unmap_address(struct dmar_domain *domain,
-			       dma_addr_t iova, size_t size)
+static void intel_iommu_unmap_range(struct iommu_domain *domain,
+				    unsigned long iova, size_t size)
 {
+	struct dmar_domain *dmar_domain = domain->priv;
 	dma_addr_t base;
 
 	/* The address might not be aligned */
 	base = iova & VTD_PAGE_MASK;
 	size = VTD_PAGE_ALIGN(size);
-	dma_pte_clear_range(domain, base, base + size);
+	dma_pte_clear_range(dmar_domain, base, base + size);
 
-	if (domain->max_addr == base + size)
-		domain->max_addr = base;
+	if (dmar_domain->max_addr == base + size)
+		dmar_domain->max_addr = base;
 }
-EXPORT_SYMBOL_GPL(intel_iommu_unmap_address);
 
 int intel_iommu_found(void)
 {
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index 9909c5a1b20f..6bc26e03858c 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -330,10 +330,6 @@ extern int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
 
 extern void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu);
 
-int intel_iommu_map_address(struct dmar_domain *domain, dma_addr_t iova,
-			    u64 hpa, size_t size, int prot);
-void intel_iommu_unmap_address(struct dmar_domain *domain,
-			       dma_addr_t iova, size_t size);
 u64 intel_iommu_iova_to_phys(struct dmar_domain *domain, u64 iova);
 
 #ifdef CONFIG_DMAR
-- 
cgit v1.2.3


From d14d65777c2491dd5baf1e17f444b8f653f3cbb1 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 3 Dec 2008 15:06:57 +0100
Subject: VT-d: adapt domain iova_to_phys function for IOMMU API

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 drivers/pci/intel-iommu.c   | 7 ++++---
 include/linux/intel-iommu.h | 2 --
 2 files changed, 4 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 8af6c96f31b3..712810598a2e 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -3106,15 +3106,16 @@ int intel_iommu_found(void)
 }
 EXPORT_SYMBOL_GPL(intel_iommu_found);
 
-u64 intel_iommu_iova_to_phys(struct dmar_domain *domain, u64 iova)
+static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain,
+					    unsigned long iova)
 {
+	struct dmar_domain *dmar_domain = domain->priv;
 	struct dma_pte *pte;
 	u64 phys = 0;
 
-	pte = addr_to_dma_pte(domain, iova);
+	pte = addr_to_dma_pte(dmar_domain, iova);
 	if (pte)
 		phys = dma_pte_addr(pte);
 
 	return phys;
 }
-EXPORT_SYMBOL_GPL(intel_iommu_iova_to_phys);
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index 6bc26e03858c..26ccc0294567 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -330,8 +330,6 @@ extern int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
 
 extern void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu);
 
-u64 intel_iommu_iova_to_phys(struct dmar_domain *domain, u64 iova);
-
 #ifdef CONFIG_DMAR
 int intel_iommu_found(void);
 #else /* CONFIG_DMAR */
-- 
cgit v1.2.3


From e4754c96cf8b82a754dc5ba791d6c0bf1fbe8e8e Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Wed, 3 Dec 2008 15:26:42 +0100
Subject: VT-d: remove now unused intel_iommu_found function

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 drivers/pci/intel-iommu.c   | 6 ------
 include/linux/intel-iommu.h | 9 ---------
 2 files changed, 15 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 81e04ec85d97..ecb5fd3b71f7 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -3105,12 +3105,6 @@ static void intel_iommu_unmap_range(struct iommu_domain *domain,
 		dmar_domain->max_addr = base;
 }
 
-int intel_iommu_found(void)
-{
-	return g_num_of_iommus;
-}
-EXPORT_SYMBOL_GPL(intel_iommu_found);
-
 static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain,
 					    unsigned long iova)
 {
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index 26ccc0294567..c4f6c101dbcd 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -330,15 +330,6 @@ extern int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
 
 extern void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu);
 
-#ifdef CONFIG_DMAR
-int intel_iommu_found(void);
-#else /* CONFIG_DMAR */
-static inline int intel_iommu_found(void)
-{
-	return 0;
-}
-#endif /* CONFIG_DMAR */
-
 extern void *intel_alloc_coherent(struct device *, size_t, dma_addr_t *, gfp_t);
 extern void intel_free_coherent(struct device *, size_t, void *, dma_addr_t);
 extern dma_addr_t intel_map_single(struct device *, phys_addr_t, size_t, int);
-- 
cgit v1.2.3


From 87d8fe1ee6b8d2f95076142d58c440dba4e7bdc2 Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Sat, 3 Jan 2009 09:47:09 -0500
Subject: add releasepage hooks to block devices which can be used by file
 systems

Implement blkdev_releasepage() to release the buffer_heads and pages
after we release private data belonging to a mounted filesystem.

Cc: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 fs/block_dev.c     | 15 +++++++++++++++
 fs/super.c         |  2 ++
 include/linux/fs.h |  2 ++
 3 files changed, 19 insertions(+)

(limited to 'include/linux')

diff --git a/fs/block_dev.c b/fs/block_dev.c
index 349a26c10001..1dd07e66e98a 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1220,6 +1220,20 @@ static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 	return blkdev_ioctl(bdev, mode, cmd, arg);
 }
 
+/*
+ * Try to release a page associated with block device when the system
+ * is under memory pressure.
+ */
+static int blkdev_releasepage(struct page *page, gfp_t wait)
+{
+	struct super_block *super = BDEV_I(page->mapping->host)->bdev.bd_super;
+
+	if (super && super->s_op->bdev_try_to_free_page)
+		return super->s_op->bdev_try_to_free_page(super, page, wait);
+
+	return try_to_free_buffers(page);
+}
+
 static const struct address_space_operations def_blk_aops = {
 	.readpage	= blkdev_readpage,
 	.writepage	= blkdev_writepage,
@@ -1227,6 +1241,7 @@ static const struct address_space_operations def_blk_aops = {
 	.write_begin	= blkdev_write_begin,
 	.write_end	= blkdev_write_end,
 	.writepages	= generic_writepages,
+	.releasepage	= blkdev_releasepage,
 	.direct_IO	= blkdev_direct_IO,
 };
 
diff --git a/fs/super.c b/fs/super.c
index ddba069d7a99..d5fd4498548a 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -800,6 +800,7 @@ int get_sb_bdev(struct file_system_type *fs_type,
 		}
 
 		s->s_flags |= MS_ACTIVE;
+		bdev->bd_super = s;
 	}
 
 	return simple_set_mnt(mnt, s);
@@ -819,6 +820,7 @@ void kill_block_super(struct super_block *sb)
 	struct block_device *bdev = sb->s_bdev;
 	fmode_t mode = sb->s_mode;
 
+	bdev->bd_super = 0;
 	generic_shutdown_super(sb);
 	sync_blockdev(bdev);
 	close_bdev_exclusive(bdev, mode);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index f2a3010140e3..0f54ae0f0ccd 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -565,6 +565,7 @@ struct address_space {
 struct block_device {
 	dev_t			bd_dev;  /* not a kdev_t - it's a search key */
 	struct inode *		bd_inode;	/* will die */
+	struct super_block *	bd_super;
 	int			bd_openers;
 	struct mutex		bd_mutex;	/* open/close mutex */
 	struct semaphore	bd_mount_sem;
@@ -1385,6 +1386,7 @@ struct super_operations {
 	ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t);
 	ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t);
 #endif
+	int (*bdev_try_to_free_page)(struct super_block*, struct page*, gfp_t);
 };
 
 /*
-- 
cgit v1.2.3


From 2f983570010a0dcb26d988da02d7ccfad00c807c Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Sat, 3 Jan 2009 00:06:34 -0800
Subject: sparseirq: move set/get_timer_rand_state back to .c

those two functions only used in that C file

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/random.c  | 40 ++++++++++++++++++++++++++++++++++------
 include/linux/random.h | 50 --------------------------------------------------
 2 files changed, 34 insertions(+), 56 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/random.c b/drivers/char/random.c
index d26891bfcd41..c7afc068c28d 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -559,7 +559,40 @@ struct timer_rand_state {
 };
 
 #ifndef CONFIG_SPARSE_IRQ
-struct timer_rand_state *irq_timer_state[NR_IRQS];
+
+static struct timer_rand_state *irq_timer_state[NR_IRQS];
+
+static struct timer_rand_state *get_timer_rand_state(unsigned int irq)
+{
+	return irq_timer_state[irq];
+}
+
+static void set_timer_rand_state(unsigned int irq,
+				 struct timer_rand_state *state)
+{
+	irq_timer_state[irq] = state;
+}
+
+#else
+
+static struct timer_rand_state *get_timer_rand_state(unsigned int irq)
+{
+	struct irq_desc *desc;
+
+	desc = irq_to_desc(irq);
+
+	return desc->timer_rand_state;
+}
+
+static void set_timer_rand_state(unsigned int irq,
+				 struct timer_rand_state *state)
+{
+	struct irq_desc *desc;
+
+	desc = irq_to_desc(irq);
+
+	desc->timer_rand_state = state;
+}
 #endif
 
 static struct timer_rand_state input_timer_state;
@@ -919,11 +952,6 @@ void rand_initialize_irq(int irq)
 {
 	struct timer_rand_state *state;
 
-#ifndef CONFIG_SPARSE_IRQ
-	if (irq >= nr_irqs)
-		return;
-#endif
-
 	state = get_timer_rand_state(irq);
 
 	if (state)
diff --git a/include/linux/random.h b/include/linux/random.h
index adbf3bd3c6b3..407ea3646f8f 100644
--- a/include/linux/random.h
+++ b/include/linux/random.h
@@ -45,56 +45,6 @@ struct rand_pool_info {
 
 extern void rand_initialize_irq(int irq);
 
-struct timer_rand_state;
-#ifndef CONFIG_SPARSE_IRQ
-
-extern struct timer_rand_state *irq_timer_state[];
-
-static inline struct timer_rand_state *get_timer_rand_state(unsigned int irq)
-{
-	if (irq >= nr_irqs)
-		return NULL;
-
-	return irq_timer_state[irq];
-}
-
-static inline void set_timer_rand_state(unsigned int irq, struct timer_rand_state *state)
-{
-	if (irq >= nr_irqs)
-		return;
-
-	irq_timer_state[irq] = state;
-}
-
-#else
-
-#include <linux/irq.h>
-static inline struct timer_rand_state *get_timer_rand_state(unsigned int irq)
-{
-	struct irq_desc *desc;
-
-	desc = irq_to_desc(irq);
-
-	if (!desc)
-		return NULL;
-
-	return desc->timer_rand_state;
-}
-
-static inline void set_timer_rand_state(unsigned int irq, struct timer_rand_state *state)
-{
-	struct irq_desc *desc;
-
-	desc = irq_to_desc(irq);
-
-	if (!desc)
-		return;
-
-	desc->timer_rand_state = state;
-}
-#endif
-
-
 extern void add_input_randomness(unsigned int type, unsigned int code,
 				 unsigned int value);
 extern void add_interrupt_randomness(int irq);
-- 
cgit v1.2.3


From 9188e79ec3fd43a0a605274324aecfb731baa88b Mon Sep 17 00:00:00 2001
From: Jiri Kosina <jkosina@suse.cz>
Date: Wed, 12 Nov 2008 16:14:08 +0100
Subject: HID: add phys and name ioctls to hidraw

The hiddev interface provides ioctl() calls which can be used
to obtain phys and raw name of the underlying device.

Add the corresponding support also into hidraw.

Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 drivers/hid/hidraw.c   | 28 ++++++++++++++++++++++++++++
 include/linux/hidraw.h |  2 ++
 2 files changed, 30 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/hid/hidraw.c b/drivers/hid/hidraw.c
index 7685ae6808c4..975edd88a3db 100644
--- a/drivers/hid/hidraw.c
+++ b/drivers/hid/hidraw.c
@@ -265,6 +265,34 @@ static long hidraw_ioctl(struct file *file, unsigned int cmd,
 				break;
 			}
 		default:
+			{
+				struct hid_device *hid = dev->hid;
+				if (_IOC_TYPE(cmd) != 'H' || _IOC_DIR(cmd) != _IOC_READ)
+					return -EINVAL;
+
+				if (_IOC_NR(cmd) == _IOC_NR(HIDIOCGRAWNAME(0))) {
+					int len;
+					if (!hid->name)
+						return 0;
+					len = strlen(hid->name) + 1;
+					if (len > _IOC_SIZE(cmd))
+						len = _IOC_SIZE(cmd);
+					return copy_to_user(user_arg, hid->name, len) ?
+						-EFAULT : len;
+				}
+
+				if (_IOC_NR(cmd) == _IOC_NR(HIDIOCGRAWPHYS(0))) {
+					int len;
+					if (!hid->phys)
+						return 0;
+					len = strlen(hid->phys) + 1;
+					if (len > _IOC_SIZE(cmd))
+						len = _IOC_SIZE(cmd);
+					return copy_to_user(user_arg, hid->phys, len) ?
+						-EFAULT : len;
+				}
+                }
+
 			ret = -ENOTTY;
 	}
 	unlock_kernel();
diff --git a/include/linux/hidraw.h b/include/linux/hidraw.h
index dbb5c8c374f0..dd8d69269176 100644
--- a/include/linux/hidraw.h
+++ b/include/linux/hidraw.h
@@ -33,6 +33,8 @@ struct hidraw_devinfo {
 #define HIDIOCGRDESCSIZE	_IOR('H', 0x01, int)
 #define HIDIOCGRDESC		_IOR('H', 0x02, struct hidraw_report_descriptor)
 #define HIDIOCGRAWINFO		_IOR('H', 0x03, struct hidraw_devinfo)
+#define HIDIOCGRAWNAME(len)     _IOC(_IOC_READ, 'H', 0x04, len)
+#define HIDIOCGRAWPHYS(len)     _IOC(_IOC_READ, 'H', 0x05, len)
 
 #define HIDRAW_FIRST_MINOR 0
 #define HIDRAW_MAX_DEVICES 64
-- 
cgit v1.2.3


From 0ed94b334265b6ee3e3336b4fedacfa9cb2ccaba Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Mon, 24 Nov 2008 16:20:07 +0100
Subject: HID: move usbhid flags to usbhid.h

Move usbhid specific flags from global hid.h into local usbhid.h.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 drivers/hid/usbhid/usbhid.h | 10 ++++++++++
 include/linux/hid.h         |  9 ---------
 2 files changed, 10 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/hid/usbhid/usbhid.h b/drivers/hid/usbhid/usbhid.h
index 332abcdf9956..9eb30564be9c 100644
--- a/drivers/hid/usbhid/usbhid.h
+++ b/drivers/hid/usbhid/usbhid.h
@@ -40,6 +40,16 @@ int usbhid_open(struct hid_device *hid);
 void usbhid_init_reports(struct hid_device *hid);
 void usbhid_submit_report(struct hid_device *hid, struct hid_report *report, unsigned char dir);
 
+/* iofl flags */
+#define HID_CTRL_RUNNING	1
+#define HID_OUT_RUNNING		2
+#define HID_IN_RUNNING		3
+#define HID_RESET_PENDING	4
+#define HID_SUSPENDED		5
+#define HID_CLEAR_HALT		6
+#define HID_DISCONNECTED	7
+#define HID_STARTED		8
+
 /*
  * USB-specific HID struct, to be pointed to
  * from struct hid_device->driver_data
diff --git a/include/linux/hid.h b/include/linux/hid.h
index e5780f8c934a..2c20f20283b2 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -403,15 +403,6 @@ struct hid_output_fifo {
 #define HID_STAT_ADDED		1
 #define HID_STAT_PARSED		2
 
-#define HID_CTRL_RUNNING	1
-#define HID_OUT_RUNNING		2
-#define HID_IN_RUNNING		3
-#define HID_RESET_PENDING	4
-#define HID_SUSPENDED		5
-#define HID_CLEAR_HALT		6
-#define HID_DISCONNECTED	7
-#define HID_STARTED		8
-
 struct hid_input {
 	struct list_head list;
 	struct hid_report *report;
-- 
cgit v1.2.3


From 3a6f82f7a22cf19687f556997c6978b31c109360 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Mon, 24 Nov 2008 16:20:09 +0100
Subject: HID: add dynids facility

Allow adding new devices to the hid drivers on the fly without
a need of kernel recompilation.

Now, one can test a driver e.g. by:
echo 0003:045E:00F0.0003 > ../generic-usb/unbind
echo 0003 045E 00F0 > new_id
from some driver subdir.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 drivers/hid/hid-core.c | 101 +++++++++++++++++++++++++++++++++++++++++++++++--
 include/linux/hid.h    |   5 +++
 2 files changed, 103 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c
index 344f8fdb2824..34cc3b0d01f6 100644
--- a/drivers/hid/hid-core.c
+++ b/drivers/hid/hid-core.c
@@ -1304,12 +1304,92 @@ static const struct hid_device_id hid_blacklist[] = {
 	{ }
 };
 
+struct hid_dynid {
+	struct list_head list;
+	struct hid_device_id id;
+};
+
+/**
+ * store_new_id - add a new HID device ID to this driver and re-probe devices
+ * @driver: target device driver
+ * @buf: buffer for scanning device ID data
+ * @count: input size
+ *
+ * Adds a new dynamic hid device ID to this driver,
+ * and causes the driver to probe for all devices again.
+ */
+static ssize_t store_new_id(struct device_driver *drv, const char *buf,
+		size_t count)
+{
+	struct hid_driver *hdrv = container_of(drv, struct hid_driver, driver);
+	struct hid_dynid *dynid;
+	__u32 bus, vendor, product;
+	unsigned long driver_data = 0;
+	int ret;
+
+	ret = sscanf(buf, "%x %x %x %lx",
+			&bus, &vendor, &product, &driver_data);
+	if (ret < 3)
+		return -EINVAL;
+
+	dynid = kzalloc(sizeof(*dynid), GFP_KERNEL);
+	if (!dynid)
+		return -ENOMEM;
+
+	dynid->id.bus = bus;
+	dynid->id.vendor = vendor;
+	dynid->id.product = product;
+	dynid->id.driver_data = driver_data;
+
+	spin_lock(&hdrv->dyn_lock);
+	list_add_tail(&dynid->list, &hdrv->dyn_list);
+	spin_unlock(&hdrv->dyn_lock);
+
+	ret = 0;
+	if (get_driver(&hdrv->driver)) {
+		ret = driver_attach(&hdrv->driver);
+		put_driver(&hdrv->driver);
+	}
+
+	return ret ? : count;
+}
+static DRIVER_ATTR(new_id, S_IWUSR, NULL, store_new_id);
+
+static void hid_free_dynids(struct hid_driver *hdrv)
+{
+	struct hid_dynid *dynid, *n;
+
+	spin_lock(&hdrv->dyn_lock);
+	list_for_each_entry_safe(dynid, n, &hdrv->dyn_list, list) {
+		list_del(&dynid->list);
+		kfree(dynid);
+	}
+	spin_unlock(&hdrv->dyn_lock);
+}
+
+static const struct hid_device_id *hid_match_device(struct hid_device *hdev,
+		struct hid_driver *hdrv)
+{
+	struct hid_dynid *dynid;
+
+	spin_lock(&hdrv->dyn_lock);
+	list_for_each_entry(dynid, &hdrv->dyn_list, list) {
+		if (hid_match_one_id(hdev, &dynid->id)) {
+			spin_unlock(&hdrv->dyn_lock);
+			return &dynid->id;
+		}
+	}
+	spin_unlock(&hdrv->dyn_lock);
+
+	return hid_match_id(hdev, hdrv->id_table);
+}
+
 static int hid_bus_match(struct device *dev, struct device_driver *drv)
 {
 	struct hid_driver *hdrv = container_of(drv, struct hid_driver, driver);
 	struct hid_device *hdev = container_of(dev, struct hid_device, dev);
 
-	if (!hid_match_id(hdev, hdrv->id_table))
+	if (!hid_match_device(hdev, hdrv))
 		return 0;
 
 	/* generic wants all non-blacklisted */
@@ -1328,7 +1408,7 @@ static int hid_device_probe(struct device *dev)
 	int ret = 0;
 
 	if (!hdev->driver) {
-		id = hid_match_id(hdev, hdrv->id_table);
+		id = hid_match_device(hdev, hdrv);
 		if (id == NULL)
 			return -ENODEV;
 
@@ -1695,18 +1775,33 @@ EXPORT_SYMBOL_GPL(hid_destroy_device);
 int __hid_register_driver(struct hid_driver *hdrv, struct module *owner,
 		const char *mod_name)
 {
+	int ret;
+
 	hdrv->driver.name = hdrv->name;
 	hdrv->driver.bus = &hid_bus_type;
 	hdrv->driver.owner = owner;
 	hdrv->driver.mod_name = mod_name;
 
-	return driver_register(&hdrv->driver);
+	INIT_LIST_HEAD(&hdrv->dyn_list);
+	spin_lock_init(&hdrv->dyn_lock);
+
+	ret = driver_register(&hdrv->driver);
+	if (ret)
+		return ret;
+
+	ret = driver_create_file(&hdrv->driver, &driver_attr_new_id);
+	if (ret)
+		driver_unregister(&hdrv->driver);
+
+	return ret;
 }
 EXPORT_SYMBOL_GPL(__hid_register_driver);
 
 void hid_unregister_driver(struct hid_driver *hdrv)
 {
+	driver_remove_file(&hdrv->driver, &driver_attr_new_id);
 	driver_unregister(&hdrv->driver);
+	hid_free_dynids(hdrv);
 }
 EXPORT_SYMBOL_GPL(hid_unregister_driver);
 
diff --git a/include/linux/hid.h b/include/linux/hid.h
index 2c20f20283b2..215035bbb288 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -531,6 +531,8 @@ struct hid_usage_id {
  * @name: driver name (e.g. "Footech_bar-wheel")
  * @id_table: which devices is this driver for (must be non-NULL for probe
  * 	      to be called)
+ * @dyn_list: list of dynamically added device ids
+ * @dyn_lock: lock protecting @dyn_list
  * @probe: new device inserted
  * @remove: device removed (NULL if not a hot-plug capable driver)
  * @report_table: on which reports to call raw_event (NULL means all)
@@ -558,6 +560,9 @@ struct hid_driver {
 	char *name;
 	const struct hid_device_id *id_table;
 
+	struct list_head dyn_list;
+	spinlock_t dyn_lock;
+
 	int (*probe)(struct hid_device *dev, const struct hid_device_id *id);
 	void (*remove)(struct hid_device *dev);
 
-- 
cgit v1.2.3


From 725cf0f47dbb02e0482f081828cff73f55479b79 Mon Sep 17 00:00:00 2001
From: Hannes Eder <hannes@hanneseder.net>
Date: Tue, 16 Dec 2008 14:20:23 +0100
Subject: HID: avoid sparse warning in HID_COMPAT_LOAD_DRIVER

Impact: include a prototype for the exported function in the macro

Fix about 20 of this warnings:

  drivers/hid/hid-a4tech.c:162:1: warning: symbol 'hid_compat_a4tech' was not declared. Should it be static?

Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 include/linux/hid.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/hid.h b/include/linux/hid.h
index 215035bbb288..81aa84d60c6b 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -793,6 +793,8 @@ dbg_hid(const char *fmt, ...)
 
 #ifdef CONFIG_HID_COMPAT
 #define HID_COMPAT_LOAD_DRIVER(name)	\
+/* prototype to avoid sparse warning */	\
+extern void hid_compat_##name(void);	\
 void hid_compat_##name(void) { }	\
 EXPORT_SYMBOL(hid_compat_##name)
 #else
-- 
cgit v1.2.3


From c31910672376dfb8d020e32afa7249763bcd924a Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Tue, 6 Jan 2009 11:14:25 -0500
Subject: ext4: Remove code to create the journal inode

This code has been obsolete in quite some time, since the supported
method for adding a journal inode is to use tune2fs (or to creating
new filesystem with a journal via mke2fs or mkfs.ext4).

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 Documentation/filesystems/ext4.txt |  4 ---
 fs/ext4/super.c                    | 68 +++--------------------------------
 fs/jbd2/journal.c                  | 72 --------------------------------------
 include/linux/jbd2.h               |  1 -
 4 files changed, 4 insertions(+), 141 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt
index e3fcbea3ec8c..9ec29d86ff8b 100644
--- a/Documentation/filesystems/ext4.txt
+++ b/Documentation/filesystems/ext4.txt
@@ -149,10 +149,6 @@ journal_async_commit	Commit block can be written to disk without waiting
 journal=update		Update the ext4 file system's journal to the current
 			format.
 
-journal=inum		When a journal already exists, this option is ignored.
-			Otherwise, it specifies the number of the inode which
-			will represent the ext4 file system's journal file.
-
 journal_dev=devnum	When the external journal device's major/minor numbers
 			have changed, this option allows the user to specify
 			the new journal location.  The journal device is
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index e5ab520724da..8036392b2121 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -51,8 +51,6 @@ struct proc_dir_entry *ext4_proc_root;
 
 static int ext4_load_journal(struct super_block *, struct ext4_super_block *,
 			     unsigned long journal_devnum);
-static int ext4_create_journal(struct super_block *, struct ext4_super_block *,
-			       unsigned int);
 static void ext4_commit_super(struct super_block *sb,
 			      struct ext4_super_block *es, int sync);
 static void ext4_mark_recovery_complete(struct super_block *sb,
@@ -1006,7 +1004,7 @@ enum {
 	Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl,
 	Opt_reservation, Opt_noreservation, Opt_noload, Opt_nobh, Opt_bh,
 	Opt_commit, Opt_min_batch_time, Opt_max_batch_time,
-	Opt_journal_update, Opt_journal_inum, Opt_journal_dev,
+	Opt_journal_update, Opt_journal_dev,
 	Opt_journal_checksum, Opt_journal_async_commit,
 	Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
 	Opt_data_err_abort, Opt_data_err_ignore,
@@ -1048,7 +1046,6 @@ static const match_table_t tokens = {
 	{Opt_min_batch_time, "min_batch_time=%u"},
 	{Opt_max_batch_time, "max_batch_time=%u"},
 	{Opt_journal_update, "journal=update"},
-	{Opt_journal_inum, "journal=%u"},
 	{Opt_journal_dev, "journal_dev=%u"},
 	{Opt_journal_checksum, "journal_checksum"},
 	{Opt_journal_async_commit, "journal_async_commit"},
@@ -1102,7 +1099,7 @@ static ext4_fsblk_t get_sb_block(void **data)
 }
 
 static int parse_options(char *options, struct super_block *sb,
-			 unsigned int *inum, unsigned long *journal_devnum,
+			 unsigned long *journal_devnum,
 			 ext4_fsblk_t *n_blocks_count, int is_remount)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
@@ -1226,16 +1223,6 @@ static int parse_options(char *options, struct super_block *sb,
 			}
 			set_opt(sbi->s_mount_opt, UPDATE_JOURNAL);
 			break;
-		case Opt_journal_inum:
-			if (is_remount) {
-				printk(KERN_ERR "EXT4-fs: cannot specify "
-				       "journal on remount\n");
-				return 0;
-			}
-			if (match_int(&args[0], &option))
-				return 0;
-			*inum = option;
-			break;
 		case Opt_journal_dev:
 			if (is_remount) {
 				printk(KERN_ERR "EXT4-fs: cannot specify "
@@ -2035,7 +2022,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	ext4_fsblk_t sb_block = get_sb_block(&data);
 	ext4_fsblk_t logical_sb_block;
 	unsigned long offset = 0;
-	unsigned int journal_inum = 0;
 	unsigned long journal_devnum = 0;
 	unsigned long def_mount_opts;
 	struct inode *root;
@@ -2155,8 +2141,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	set_opt(sbi->s_mount_opt, DELALLOC);
 
 
-	if (!parse_options((char *) data, sb, &journal_inum, &journal_devnum,
-			   NULL, 0))
+	if (!parse_options((char *) data, sb, &journal_devnum, NULL, 0))
 		goto failed_mount;
 
 	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
@@ -2460,9 +2445,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 				goto failed_mount4;
 			}
 		}
-	} else if (journal_inum) {
-		if (ext4_create_journal(sb, es, journal_inum))
-			goto failed_mount3;
 	} else if (test_opt(sb, NOLOAD) && !(sb->s_flags & MS_RDONLY) &&
 	      EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)) {
 		printk(KERN_ERR "EXT4-fs: required journal recovery "
@@ -2926,48 +2908,6 @@ static int ext4_load_journal(struct super_block *sb,
 	return 0;
 }
 
-static int ext4_create_journal(struct super_block *sb,
-			       struct ext4_super_block *es,
-			       unsigned int journal_inum)
-{
-	journal_t *journal;
-	int err;
-
-	if (sb->s_flags & MS_RDONLY) {
-		printk(KERN_ERR "EXT4-fs: readonly filesystem when trying to "
-				"create journal.\n");
-		return -EROFS;
-	}
-
-	journal = ext4_get_journal(sb, journal_inum);
-	if (!journal)
-		return -EINVAL;
-
-	printk(KERN_INFO "EXT4-fs: creating new journal on inode %u\n",
-	       journal_inum);
-
-	err = jbd2_journal_create(journal);
-	if (err) {
-		printk(KERN_ERR "EXT4-fs: error creating journal.\n");
-		jbd2_journal_destroy(journal);
-		return -EIO;
-	}
-
-	EXT4_SB(sb)->s_journal = journal;
-
-	ext4_update_dynamic_rev(sb);
-	EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
-	EXT4_SET_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL);
-
-	es->s_journal_inum = cpu_to_le32(journal_inum);
-	sb->s_dirt = 1;
-
-	/* Make sure we flush the recovery flag to disk. */
-	ext4_commit_super(sb, es, 1);
-
-	return 0;
-}
-
 static void ext4_commit_super(struct super_block *sb,
 			      struct ext4_super_block *es, int sync)
 {
@@ -3209,7 +3149,7 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data)
 	/*
 	 * Allow the "check" option to be passed as a remount option.
 	 */
-	if (!parse_options(data, sb, NULL, NULL, &n_blocks_count, 1)) {
+	if (!parse_options(data, sb, NULL, &n_blocks_count, 1)) {
 		err = -EINVAL;
 		goto restore_opts;
 	}
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 34ef98057202..b10d7283ba5b 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -66,7 +66,6 @@ EXPORT_SYMBOL(jbd2_journal_update_format);
 EXPORT_SYMBOL(jbd2_journal_check_used_features);
 EXPORT_SYMBOL(jbd2_journal_check_available_features);
 EXPORT_SYMBOL(jbd2_journal_set_features);
-EXPORT_SYMBOL(jbd2_journal_create);
 EXPORT_SYMBOL(jbd2_journal_load);
 EXPORT_SYMBOL(jbd2_journal_destroy);
 EXPORT_SYMBOL(jbd2_journal_abort);
@@ -1162,77 +1161,6 @@ static int journal_reset(journal_t *journal)
 	return jbd2_journal_start_thread(journal);
 }
 
-/**
- * int jbd2_journal_create() - Initialise the new journal file
- * @journal: Journal to create. This structure must have been initialised
- *
- * Given a journal_t structure which tells us which disk blocks we can
- * use, create a new journal superblock and initialise all of the
- * journal fields from scratch.
- **/
-int jbd2_journal_create(journal_t *journal)
-{
-	unsigned long long blocknr;
-	struct buffer_head *bh;
-	journal_superblock_t *sb;
-	int i, err;
-
-	if (journal->j_maxlen < JBD2_MIN_JOURNAL_BLOCKS) {
-		printk (KERN_ERR "Journal length (%d blocks) too short.\n",
-			journal->j_maxlen);
-		journal_fail_superblock(journal);
-		return -EINVAL;
-	}
-
-	if (journal->j_inode == NULL) {
-		/*
-		 * We don't know what block to start at!
-		 */
-		printk(KERN_EMERG
-		       "%s: creation of journal on external device!\n",
-		       __func__);
-		BUG();
-	}
-
-	/* Zero out the entire journal on disk.  We cannot afford to
-	   have any blocks on disk beginning with JBD2_MAGIC_NUMBER. */
-	jbd_debug(1, "JBD: Zeroing out journal blocks...\n");
-	for (i = 0; i < journal->j_maxlen; i++) {
-		err = jbd2_journal_bmap(journal, i, &blocknr);
-		if (err)
-			return err;
-		bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize);
-		lock_buffer(bh);
-		memset (bh->b_data, 0, journal->j_blocksize);
-		BUFFER_TRACE(bh, "marking dirty");
-		mark_buffer_dirty(bh);
-		BUFFER_TRACE(bh, "marking uptodate");
-		set_buffer_uptodate(bh);
-		unlock_buffer(bh);
-		__brelse(bh);
-	}
-
-	sync_blockdev(journal->j_dev);
-	jbd_debug(1, "JBD: journal cleared.\n");
-
-	/* OK, fill in the initial static fields in the new superblock */
-	sb = journal->j_superblock;
-
-	sb->s_header.h_magic	 = cpu_to_be32(JBD2_MAGIC_NUMBER);
-	sb->s_header.h_blocktype = cpu_to_be32(JBD2_SUPERBLOCK_V2);
-
-	sb->s_blocksize	= cpu_to_be32(journal->j_blocksize);
-	sb->s_maxlen	= cpu_to_be32(journal->j_maxlen);
-	sb->s_first	= cpu_to_be32(1);
-
-	journal->j_transaction_sequence = 1;
-
-	journal->j_flags &= ~JBD2_ABORT;
-	journal->j_format_version = 2;
-
-	return journal_reset(journal);
-}
-
 /**
  * void jbd2_journal_update_superblock() - Update journal sb on disk.
  * @journal: The journal to update.
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index 9d82084a1605..adef1c9940d3 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -1104,7 +1104,6 @@ extern int	   jbd2_journal_set_features
 		   (journal_t *, unsigned long, unsigned long, unsigned long);
 extern void	   jbd2_journal_clear_features
 		   (journal_t *, unsigned long, unsigned long, unsigned long);
-extern int	   jbd2_journal_create     (journal_t *);
 extern int	   jbd2_journal_load       (journal_t *journal);
 extern int	   jbd2_journal_destroy    (journal_t *);
 extern int	   jbd2_journal_recover    (journal_t *journal);
-- 
cgit v1.2.3


From c66b9906f863696159e05890bb7123269bb9a9de Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sun, 4 Jan 2009 10:55:02 +0100
Subject: intel-iommu: fix build error with INTR_REMAP=y and DMAR=n

dmar.o can be built in the CONFIG_INTR_REMAP=y case but
iommu_calculate_agaw() is only available if VT-d is built as well.

So create an inline version of iommu_calculate_agaw() for the
!CONFIG_DMAR case. The iommu->agaw value wont be used in this
case, but the code is cleaner (has less #ifdefs) if we have it around
unconditionally.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/dma_remapping.h | 8 ++++++++
 1 file changed, 8 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/dma_remapping.h b/include/linux/dma_remapping.h
index 136f170cecc2..af1dab41674b 100644
--- a/include/linux/dma_remapping.h
+++ b/include/linux/dma_remapping.h
@@ -17,7 +17,15 @@ struct dmar_domain;
 struct root_entry;
 
 extern void free_dmar_iommu(struct intel_iommu *iommu);
+
+#ifdef CONFIG_DMAR
 extern int iommu_calculate_agaw(struct intel_iommu *iommu);
+#else
+static inline int iommu_calculate_agaw(struct intel_iommu *iommu)
+{
+	return 0;
+}
+#endif
 
 extern int dmar_disabled;
 
-- 
cgit v1.2.3


From 0c8a601678960fbcc1c1185a283d6d107575810b Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Sat, 8 Nov 2008 01:10:16 +0100
Subject: mfd: Add WM8350 revision H support

No other software changes are required.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/wm8350-core.c       | 7 ++++++-
 include/linux/mfd/wm8350/core.h | 1 +
 2 files changed, 7 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/wm8350-core.c b/drivers/mfd/wm8350-core.c
index d63a530c4fe8..c013afde260d 100644
--- a/drivers/mfd/wm8350-core.c
+++ b/drivers/mfd/wm8350-core.c
@@ -1202,9 +1202,14 @@ int wm8350_device_init(struct wm8350 *wm8350, int irq,
 			dev_info(wm8350->dev, "Found Rev G device\n");
 			wm8350->rev = WM8350_REV_G;
 			break;
+		case WM8350_REV_H:
+			dev_info(wm8350->dev, "Found Rev H device\n");
+			wm8350->rev = WM8350_REV_H;
+			break;
 		default:
 			/* For safety we refuse to run on unknown hardware */
-			dev_info(wm8350->dev, "Found unknown rev\n");
+			dev_info(wm8350->dev, "Found unknown rev %x\n",
+				 (id2 & WM8350_CHIP_REV_MASK) >> 12);
 			ret = -ENODEV;
 			goto err;
 		}
diff --git a/include/linux/mfd/wm8350/core.h b/include/linux/mfd/wm8350/core.h
index 6ebf97f2a475..9490ec175d5a 100644
--- a/include/linux/mfd/wm8350/core.h
+++ b/include/linux/mfd/wm8350/core.h
@@ -536,6 +536,7 @@
 #define WM8350_REV_E				0x4
 #define WM8350_REV_F				0x5
 #define WM8350_REV_G				0x6
+#define WM8350_REV_H				0x7
 
 #define WM8350_NUM_IRQ				63
 
-- 
cgit v1.2.3


From 67488526349d043372d141c054f4dc6313780b3c Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Sat, 8 Nov 2008 01:10:21 +0100
Subject: mfd: Add AUXADC support for WM8350

The auxiliary ADC in the WM8350 is shared between several subdevices
so access to it needs to be arbitrated by the core driver.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/wm8350-core.c             | 51 ++++++++++++++++++++++++++++++++++-
 include/linux/mfd/wm8350/comparator.h |  8 ++++++
 include/linux/mfd/wm8350/core.h       |  2 ++
 3 files changed, 60 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/wm8350-core.c b/drivers/mfd/wm8350-core.c
index c013afde260d..60439bd3984d 100644
--- a/drivers/mfd/wm8350-core.c
+++ b/drivers/mfd/wm8350-core.c
@@ -63,7 +63,6 @@
  */
 static DEFINE_MUTEX(io_mutex);
 static DEFINE_MUTEX(reg_lock_mutex);
-static DEFINE_MUTEX(auxadc_mutex);
 
 /* Perform a physical read from the device.
  */
@@ -1082,6 +1081,55 @@ int wm8350_unmask_irq(struct wm8350 *wm8350, int irq)
 }
 EXPORT_SYMBOL_GPL(wm8350_unmask_irq);
 
+int wm8350_read_auxadc(struct wm8350 *wm8350, int channel, int scale, int vref)
+{
+	u16 reg, result = 0;
+	int tries = 5;
+
+	if (channel < WM8350_AUXADC_AUX1 || channel > WM8350_AUXADC_TEMP)
+		return -EINVAL;
+	if (channel >= WM8350_AUXADC_USB && channel <= WM8350_AUXADC_TEMP
+	    && (scale != 0 || vref != 0))
+		return -EINVAL;
+
+	mutex_lock(&wm8350->auxadc_mutex);
+
+	/* Turn on the ADC */
+	reg = wm8350_reg_read(wm8350, WM8350_POWER_MGMT_5);
+	wm8350_reg_write(wm8350, WM8350_POWER_MGMT_5, reg | WM8350_AUXADC_ENA);
+
+	if (scale || vref) {
+		reg = scale << 13;
+		reg |= vref << 12;
+		wm8350_reg_write(wm8350, WM8350_AUX1_READBACK + channel, reg);
+	}
+
+	reg = wm8350_reg_read(wm8350, WM8350_DIGITISER_CONTROL_1);
+	reg |= 1 << channel | WM8350_AUXADC_POLL;
+	wm8350_reg_write(wm8350, WM8350_DIGITISER_CONTROL_1, reg);
+
+	do {
+		schedule_timeout_interruptible(1);
+		reg = wm8350_reg_read(wm8350, WM8350_DIGITISER_CONTROL_1);
+	} while (tries-- && (reg & WM8350_AUXADC_POLL));
+
+	if (!tries)
+		dev_err(wm8350->dev, "adc chn %d read timeout\n", channel);
+	else
+		result = wm8350_reg_read(wm8350,
+					 WM8350_AUX1_READBACK + channel);
+
+	/* Turn off the ADC */
+	reg = wm8350_reg_read(wm8350, WM8350_POWER_MGMT_5);
+	wm8350_reg_write(wm8350, WM8350_POWER_MGMT_5,
+			 reg & ~WM8350_AUXADC_ENA);
+
+	mutex_unlock(&wm8350->auxadc_mutex);
+
+	return result & WM8350_AUXADC_DATA1_MASK;
+}
+EXPORT_SYMBOL_GPL(wm8350_read_auxadc);
+
 /*
  * Cache is always host endian.
  */
@@ -1239,6 +1287,7 @@ int wm8350_device_init(struct wm8350 *wm8350, int irq,
 		}
 	}
 
+	mutex_init(&wm8350->auxadc_mutex);
 	mutex_init(&wm8350->irq_mutex);
 	INIT_WORK(&wm8350->irq_work, wm8350_irq_worker);
 	if (irq) {
diff --git a/include/linux/mfd/wm8350/comparator.h b/include/linux/mfd/wm8350/comparator.h
index 053788649452..54bc5d0fd502 100644
--- a/include/linux/mfd/wm8350/comparator.h
+++ b/include/linux/mfd/wm8350/comparator.h
@@ -164,4 +164,12 @@
 #define WM8350_AUXADC_BATT			6
 #define WM8350_AUXADC_TEMP			7
 
+struct wm8350;
+
+/*
+ * AUX ADC Readback
+ */
+int wm8350_read_auxadc(struct wm8350 *wm8350, int channel, int scale,
+		       int vref);
+
 #endif
diff --git a/include/linux/mfd/wm8350/core.h b/include/linux/mfd/wm8350/core.h
index 9490ec175d5a..cc190055b9c4 100644
--- a/include/linux/mfd/wm8350/core.h
+++ b/include/linux/mfd/wm8350/core.h
@@ -573,6 +573,8 @@ struct wm8350 {
 			 void *src);
 	u16 *reg_cache;
 
+	struct mutex auxadc_mutex;
+
 	/* Interrupt handling */
 	struct work_struct irq_work;
 	struct mutex irq_mutex; /* IRQ table mutex */
-- 
cgit v1.2.3


From 3fba19ec1ae5b460c73a7f32efed8d3b3300b246 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Sat, 8 Nov 2008 01:13:16 +0100
Subject: mfd: allow reading entire register banks on twl4030

Minor change to the TWL4030 utility interface:  support reads
of all 256 bytes in each register bank (vs just 255).  This
can help when debugging, but is otherwise a NOP.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/twl4030-core.c  | 4 ++--
 include/linux/i2c/twl4030.h | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/twl4030-core.c b/drivers/mfd/twl4030-core.c
index dd843c4fbcc7..ef9a971e3ead 100644
--- a/drivers/mfd/twl4030-core.c
+++ b/drivers/mfd/twl4030-core.c
@@ -225,7 +225,7 @@ static struct twl4030mapping twl4030_map[TWL4030_MODULE_LAST + 1] = {
  *
  * Returns the result of operation - 0 is success
  */
-int twl4030_i2c_write(u8 mod_no, u8 *value, u8 reg, u8 num_bytes)
+int twl4030_i2c_write(u8 mod_no, u8 *value, u8 reg, unsigned num_bytes)
 {
 	int ret;
 	int sid;
@@ -274,7 +274,7 @@ EXPORT_SYMBOL(twl4030_i2c_write);
  *
  * Returns result of operation - num_bytes is success else failure.
  */
-int twl4030_i2c_read(u8 mod_no, u8 *value, u8 reg, u8 num_bytes)
+int twl4030_i2c_read(u8 mod_no, u8 *value, u8 reg, unsigned num_bytes)
 {
 	int ret;
 	u8 val;
diff --git a/include/linux/i2c/twl4030.h b/include/linux/i2c/twl4030.h
index fb604dcd38f1..ae25c907b7cf 100644
--- a/include/linux/i2c/twl4030.h
+++ b/include/linux/i2c/twl4030.h
@@ -78,8 +78,8 @@ int twl4030_i2c_read_u8(u8 mod_no, u8 *val, u8 reg);
  * IMPORTANT:  For twl4030_i2c_write(), allocate num_bytes + 1
  * for the value, and populate your data starting at offset 1.
  */
-int twl4030_i2c_write(u8 mod_no, u8 *value, u8 reg, u8 num_bytes);
-int twl4030_i2c_read(u8 mod_no, u8 *value, u8 reg, u8 num_bytes);
+int twl4030_i2c_write(u8 mod_no, u8 *value, u8 reg, unsigned num_bytes);
+int twl4030_i2c_read(u8 mod_no, u8 *value, u8 reg, unsigned num_bytes);
 
 /*----------------------------------------------------------------------*/
 
-- 
cgit v1.2.3


From 14431aa0c5a443d13d24e6f865a8838f97dab973 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Sun, 16 Nov 2008 20:16:47 +0100
Subject: power_supply: Add support for WM8350 PMU

This patch adds support for the PMU provided by the WM8350 which
implements battery, line and USB supplies including a battery charger.
The hardware functions largely autonomously, with minimal software
control required to initiate fast charging.

Support for configuration of the USB supply is not yet implemented.
This means that the hardware will remain in the mode configured at
startup, by default limiting the current drawn from USB to 100mA.

This driver was originally written by Liam Girdwood with subsequent
updates for submission by Mark Brown.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Anton Vorontsov <cbouatmailru@gmail.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/power/Kconfig             |   7 +
 drivers/power/Makefile            |   1 +
 drivers/power/wm8350_power.c      | 515 ++++++++++++++++++++++++++++++++++++++
 include/linux/mfd/wm8350/core.h   |  26 ++
 include/linux/mfd/wm8350/supply.h |  23 +-
 5 files changed, 571 insertions(+), 1 deletion(-)
 create mode 100644 drivers/power/wm8350_power.c

(limited to 'include/linux')

diff --git a/drivers/power/Kconfig b/drivers/power/Kconfig
index 8e0c2b47803c..52f86767f72e 100644
--- a/drivers/power/Kconfig
+++ b/drivers/power/Kconfig
@@ -29,6 +29,13 @@ config APM_POWER
 	  Say Y here to enable support APM status emulation using
 	  battery class devices.
 
+config WM8350_POWER
+        tristate "WM8350 PMU support"
+        depends on MFD_WM8350
+        help
+          Say Y here to enable support for the power management unit
+	  provided by the Wolfson Microelectronics WM8350 PMIC.
+
 config BATTERY_DS2760
 	tristate "DS2760 battery driver (HP iPAQ & others)"
 	select W1
diff --git a/drivers/power/Makefile b/drivers/power/Makefile
index e8f1ecec5d8f..e6f68655d9e8 100644
--- a/drivers/power/Makefile
+++ b/drivers/power/Makefile
@@ -16,6 +16,7 @@ obj-$(CONFIG_POWER_SUPPLY)	+= power_supply.o
 
 obj-$(CONFIG_PDA_POWER)		+= pda_power.o
 obj-$(CONFIG_APM_POWER)		+= apm_power.o
+obj-$(CONFIG_WM8350_POWER)	+= wm8350_power.o
 
 obj-$(CONFIG_BATTERY_DS2760)	+= ds2760_battery.o
 obj-$(CONFIG_BATTERY_PMU)	+= pmu_battery.o
diff --git a/drivers/power/wm8350_power.c b/drivers/power/wm8350_power.c
new file mode 100644
index 000000000000..9c0a847a1139
--- /dev/null
+++ b/drivers/power/wm8350_power.c
@@ -0,0 +1,515 @@
+/*
+ * Battery driver for wm8350 PMIC
+ *
+ * Copyright 2007, 2008 Wolfson Microelectronics PLC.
+ *
+ * Based on OLPC Battery Driver
+ *
+ * Copyright 2006  David Woodhouse <dwmw2@infradead.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/err.h>
+#include <linux/platform_device.h>
+#include <linux/power_supply.h>
+#include <linux/mfd/wm8350/supply.h>
+#include <linux/mfd/wm8350/core.h>
+#include <linux/mfd/wm8350/comparator.h>
+
+static int wm8350_read_battery_uvolts(struct wm8350 *wm8350)
+{
+	return wm8350_read_auxadc(wm8350, WM8350_AUXADC_BATT, 0, 0)
+		* WM8350_AUX_COEFF;
+}
+
+static int wm8350_read_line_uvolts(struct wm8350 *wm8350)
+{
+	return wm8350_read_auxadc(wm8350, WM8350_AUXADC_LINE, 0, 0)
+		* WM8350_AUX_COEFF;
+}
+
+static int wm8350_read_usb_uvolts(struct wm8350 *wm8350)
+{
+	return wm8350_read_auxadc(wm8350, WM8350_AUXADC_USB, 0, 0)
+		* WM8350_AUX_COEFF;
+}
+
+#define WM8350_BATT_SUPPLY	1
+#define WM8350_USB_SUPPLY	2
+#define WM8350_LINE_SUPPLY	4
+
+static inline int wm8350_charge_time_min(struct wm8350 *wm8350, int min)
+{
+	if (wm8350->rev < WM8350_REV_G)
+		return (((min - 30) / 15) & 0xf) << 8;
+	else
+		return (((min - 30) / 30) & 0xf) << 8;
+}
+
+static int wm8350_get_supplies(struct wm8350 *wm8350)
+{
+	u16 sm, ov, co, chrg;
+	int supplies = 0;
+
+	sm = wm8350_reg_read(wm8350, WM8350_STATE_MACHINE_STATUS);
+	ov = wm8350_reg_read(wm8350, WM8350_MISC_OVERRIDES);
+	co = wm8350_reg_read(wm8350, WM8350_COMPARATOR_OVERRIDES);
+	chrg = wm8350_reg_read(wm8350, WM8350_BATTERY_CHARGER_CONTROL_2);
+
+	/* USB_SM */
+	sm = (sm & WM8350_USB_SM_MASK) >> WM8350_USB_SM_SHIFT;
+
+	/* CHG_ISEL */
+	chrg &= WM8350_CHG_ISEL_MASK;
+
+	/* If the USB state machine is active then we're using that with or
+	 * without battery, otherwise check for wall supply */
+	if (((sm == WM8350_USB_SM_100_SLV) ||
+	     (sm == WM8350_USB_SM_500_SLV) ||
+	     (sm == WM8350_USB_SM_STDBY_SLV))
+	    && !(ov & WM8350_USB_LIMIT_OVRDE))
+		supplies = WM8350_USB_SUPPLY;
+	else if (((sm == WM8350_USB_SM_100_SLV) ||
+		  (sm == WM8350_USB_SM_500_SLV) ||
+		  (sm == WM8350_USB_SM_STDBY_SLV))
+		 && (ov & WM8350_USB_LIMIT_OVRDE) && (chrg == 0))
+		supplies = WM8350_USB_SUPPLY | WM8350_BATT_SUPPLY;
+	else if (co & WM8350_WALL_FB_OVRDE)
+		supplies = WM8350_LINE_SUPPLY;
+	else
+		supplies = WM8350_BATT_SUPPLY;
+
+	return supplies;
+}
+
+static int wm8350_charger_config(struct wm8350 *wm8350,
+				 struct wm8350_charger_policy *policy)
+{
+	u16 reg, eoc_mA, fast_limit_mA;
+
+	if (!policy) {
+		dev_warn(wm8350->dev,
+			 "No charger policy, charger not configured.\n");
+		return -EINVAL;
+	}
+
+	/* make sure USB fast charge current is not > 500mA */
+	if (policy->fast_limit_USB_mA > 500) {
+		dev_err(wm8350->dev, "USB fast charge > 500mA\n");
+		return -EINVAL;
+	}
+
+	eoc_mA = WM8350_CHG_EOC_mA(policy->eoc_mA);
+
+	wm8350_reg_unlock(wm8350);
+
+	reg = wm8350_reg_read(wm8350, WM8350_BATTERY_CHARGER_CONTROL_1)
+		& WM8350_CHG_ENA_R168;
+	wm8350_reg_write(wm8350, WM8350_BATTERY_CHARGER_CONTROL_1,
+			 reg | eoc_mA | policy->trickle_start_mV |
+			 WM8350_CHG_TRICKLE_TEMP_CHOKE |
+			 WM8350_CHG_TRICKLE_USB_CHOKE |
+			 WM8350_CHG_FAST_USB_THROTTLE);
+
+	if (wm8350_get_supplies(wm8350) & WM8350_USB_SUPPLY) {
+		fast_limit_mA =
+			WM8350_CHG_FAST_LIMIT_mA(policy->fast_limit_USB_mA);
+		wm8350_reg_write(wm8350, WM8350_BATTERY_CHARGER_CONTROL_2,
+			    policy->charge_mV | policy->trickle_charge_USB_mA |
+			    fast_limit_mA | wm8350_charge_time_min(wm8350,
+						policy->charge_timeout));
+
+	} else {
+		fast_limit_mA =
+			WM8350_CHG_FAST_LIMIT_mA(policy->fast_limit_mA);
+		wm8350_reg_write(wm8350, WM8350_BATTERY_CHARGER_CONTROL_2,
+			    policy->charge_mV | policy->trickle_charge_mA |
+			    fast_limit_mA | wm8350_charge_time_min(wm8350,
+						policy->charge_timeout));
+	}
+
+	wm8350_reg_lock(wm8350);
+	return 0;
+}
+
+static int wm8350_batt_status(struct wm8350 *wm8350)
+{
+	u16 state;
+
+	state = wm8350_reg_read(wm8350, WM8350_BATTERY_CHARGER_CONTROL_2);
+	state &= WM8350_CHG_STS_MASK;
+
+	switch (state) {
+	case WM8350_CHG_STS_OFF:
+		return POWER_SUPPLY_STATUS_DISCHARGING;
+
+	case WM8350_CHG_STS_TRICKLE:
+	case WM8350_CHG_STS_FAST:
+		return POWER_SUPPLY_STATUS_CHARGING;
+
+	default:
+		return POWER_SUPPLY_STATUS_UNKNOWN;
+	}
+}
+
+static ssize_t charger_state_show(struct device *dev,
+				 struct device_attribute *attr, char *buf)
+{
+	struct wm8350 *wm8350 = dev_get_drvdata(dev);
+	char *charge;
+	int state;
+
+	state = wm8350_reg_read(wm8350, WM8350_BATTERY_CHARGER_CONTROL_2) &
+	    WM8350_CHG_STS_MASK;
+	switch (state) {
+	case WM8350_CHG_STS_OFF:
+		charge = "Charger Off";
+		break;
+	case WM8350_CHG_STS_TRICKLE:
+		charge = "Trickle Charging";
+		break;
+	case WM8350_CHG_STS_FAST:
+		charge = "Fast Charging";
+		break;
+	default:
+		return 0;
+	}
+
+	return sprintf(buf, "%s\n", charge);
+}
+
+static DEVICE_ATTR(charger_state, 0444, charger_state_show, NULL);
+
+static void wm8350_charger_handler(struct wm8350 *wm8350, int irq, void *data)
+{
+	struct wm8350_power *power = &wm8350->power;
+	struct wm8350_charger_policy *policy = power->policy;
+
+	switch (irq) {
+	case WM8350_IRQ_CHG_BAT_HOT:
+		dev_err(wm8350->dev, "battery too hot\n");
+		break;
+	case WM8350_IRQ_CHG_BAT_COLD:
+		dev_err(wm8350->dev, "battery too cold\n");
+		break;
+	case WM8350_IRQ_CHG_BAT_FAIL:
+		dev_err(wm8350->dev, "battery failed\n");
+		break;
+	case WM8350_IRQ_CHG_TO:
+		dev_err(wm8350->dev, "charger timeout\n");
+		break;
+	case WM8350_IRQ_CHG_END:
+		power_supply_changed(&power->battery);
+		break;
+	case WM8350_IRQ_CHG_START:
+		power_supply_changed(&power->battery);
+		break;
+
+	case WM8350_IRQ_CHG_FAST_RDY:
+		dev_dbg(wm8350->dev, "fast charger ready\n");
+		wm8350_charger_config(wm8350, policy);
+		wm8350_reg_unlock(wm8350);
+		wm8350_set_bits(wm8350, WM8350_BATTERY_CHARGER_CONTROL_1,
+				WM8350_CHG_FAST);
+		wm8350_reg_lock(wm8350);
+		break;
+
+	case WM8350_IRQ_CHG_VBATT_LT_3P9:
+		dev_warn(wm8350->dev, "battery < 3.9V\n");
+		break;
+	case WM8350_IRQ_CHG_VBATT_LT_3P1:
+		dev_warn(wm8350->dev, "battery < 3.1V\n");
+		break;
+	case WM8350_IRQ_CHG_VBATT_LT_2P85:
+		dev_warn(wm8350->dev, "battery < 2.85V\n");
+		break;
+
+		/* Supply change.  We will overnotify but it should do
+		 * no harm. */
+	case WM8350_IRQ_EXT_USB_FB:
+	case WM8350_IRQ_EXT_WALL_FB:
+		wm8350_charger_config(wm8350, policy);
+	case WM8350_IRQ_EXT_BAT_FB:   /* Fall through */
+		power_supply_changed(&power->battery);
+		power_supply_changed(&power->usb);
+		power_supply_changed(&power->ac);
+		break;
+
+	default:
+		dev_err(wm8350->dev, "Unknown interrupt %d\n", irq);
+	}
+}
+
+/*********************************************************************
+ *		AC Power
+ *********************************************************************/
+static int wm8350_ac_get_prop(struct power_supply *psy,
+			      enum power_supply_property psp,
+			      union power_supply_propval *val)
+{
+	struct wm8350 *wm8350 = dev_get_drvdata(psy->dev->parent);
+	int ret = 0;
+
+	switch (psp) {
+	case POWER_SUPPLY_PROP_ONLINE:
+		val->intval = !!(wm8350_get_supplies(wm8350) &
+				 WM8350_LINE_SUPPLY);
+		break;
+	case POWER_SUPPLY_PROP_VOLTAGE_NOW:
+		val->intval = wm8350_read_line_uvolts(wm8350);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+	return ret;
+}
+
+static enum power_supply_property wm8350_ac_props[] = {
+	POWER_SUPPLY_PROP_ONLINE,
+	POWER_SUPPLY_PROP_VOLTAGE_NOW,
+};
+
+/*********************************************************************
+ *		USB Power
+ *********************************************************************/
+static int wm8350_usb_get_prop(struct power_supply *psy,
+			       enum power_supply_property psp,
+			       union power_supply_propval *val)
+{
+	struct wm8350 *wm8350 = dev_get_drvdata(psy->dev->parent);
+	int ret = 0;
+
+	switch (psp) {
+	case POWER_SUPPLY_PROP_ONLINE:
+		val->intval = !!(wm8350_get_supplies(wm8350) &
+				 WM8350_USB_SUPPLY);
+		break;
+	case POWER_SUPPLY_PROP_VOLTAGE_NOW:
+		val->intval = wm8350_read_usb_uvolts(wm8350);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+	return ret;
+}
+
+static enum power_supply_property wm8350_usb_props[] = {
+	POWER_SUPPLY_PROP_ONLINE,
+	POWER_SUPPLY_PROP_VOLTAGE_NOW,
+};
+
+/*********************************************************************
+ *		Battery properties
+ *********************************************************************/
+
+static int wm8350_bat_get_property(struct power_supply *psy,
+				   enum power_supply_property psp,
+				   union power_supply_propval *val)
+{
+	struct wm8350 *wm8350 = dev_get_drvdata(psy->dev->parent);
+	int ret = 0;
+
+	switch (psp) {
+	case POWER_SUPPLY_PROP_STATUS:
+		val->intval = wm8350_batt_status(wm8350);
+		break;
+	case POWER_SUPPLY_PROP_ONLINE:
+		val->intval = !!(wm8350_get_supplies(wm8350) &
+				 WM8350_BATT_SUPPLY);
+		break;
+	case POWER_SUPPLY_PROP_VOLTAGE_NOW:
+		val->intval = wm8350_read_battery_uvolts(wm8350);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	return ret;
+}
+
+static enum power_supply_property wm8350_bat_props[] = {
+	POWER_SUPPLY_PROP_STATUS,
+	POWER_SUPPLY_PROP_ONLINE,
+	POWER_SUPPLY_PROP_VOLTAGE_NOW,
+};
+
+/*********************************************************************
+ *		Initialisation
+ *********************************************************************/
+
+static void wm8350_init_charger(struct wm8350 *wm8350)
+{
+	/* register our interest in charger events */
+	wm8350_register_irq(wm8350, WM8350_IRQ_CHG_BAT_HOT,
+			    wm8350_charger_handler, NULL);
+	wm8350_unmask_irq(wm8350, WM8350_IRQ_CHG_BAT_HOT);
+	wm8350_register_irq(wm8350, WM8350_IRQ_CHG_BAT_COLD,
+			    wm8350_charger_handler, NULL);
+	wm8350_unmask_irq(wm8350, WM8350_IRQ_CHG_BAT_COLD);
+	wm8350_register_irq(wm8350, WM8350_IRQ_CHG_BAT_FAIL,
+			    wm8350_charger_handler, NULL);
+	wm8350_unmask_irq(wm8350, WM8350_IRQ_CHG_BAT_FAIL);
+	wm8350_register_irq(wm8350, WM8350_IRQ_CHG_TO,
+			    wm8350_charger_handler, NULL);
+	wm8350_unmask_irq(wm8350, WM8350_IRQ_CHG_TO);
+	wm8350_register_irq(wm8350, WM8350_IRQ_CHG_END,
+			    wm8350_charger_handler, NULL);
+	wm8350_unmask_irq(wm8350, WM8350_IRQ_CHG_END);
+	wm8350_register_irq(wm8350, WM8350_IRQ_CHG_START,
+			    wm8350_charger_handler, NULL);
+	wm8350_unmask_irq(wm8350, WM8350_IRQ_CHG_START);
+	wm8350_register_irq(wm8350, WM8350_IRQ_CHG_FAST_RDY,
+			    wm8350_charger_handler, NULL);
+	wm8350_unmask_irq(wm8350, WM8350_IRQ_CHG_FAST_RDY);
+	wm8350_register_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_3P9,
+			    wm8350_charger_handler, NULL);
+	wm8350_unmask_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_3P9);
+	wm8350_register_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_3P1,
+			    wm8350_charger_handler, NULL);
+	wm8350_unmask_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_3P1);
+	wm8350_register_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_2P85,
+			    wm8350_charger_handler, NULL);
+	wm8350_unmask_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_2P85);
+
+	/* and supply change events */
+	wm8350_register_irq(wm8350, WM8350_IRQ_EXT_USB_FB,
+			    wm8350_charger_handler, NULL);
+	wm8350_unmask_irq(wm8350, WM8350_IRQ_EXT_USB_FB);
+	wm8350_register_irq(wm8350, WM8350_IRQ_EXT_WALL_FB,
+			    wm8350_charger_handler, NULL);
+	wm8350_unmask_irq(wm8350, WM8350_IRQ_EXT_WALL_FB);
+	wm8350_register_irq(wm8350, WM8350_IRQ_EXT_BAT_FB,
+			    wm8350_charger_handler, NULL);
+	wm8350_unmask_irq(wm8350, WM8350_IRQ_EXT_BAT_FB);
+}
+
+static void free_charger_irq(struct wm8350 *wm8350)
+{
+	wm8350_mask_irq(wm8350, WM8350_IRQ_CHG_BAT_HOT);
+	wm8350_free_irq(wm8350, WM8350_IRQ_CHG_BAT_HOT);
+	wm8350_mask_irq(wm8350, WM8350_IRQ_CHG_BAT_COLD);
+	wm8350_free_irq(wm8350, WM8350_IRQ_CHG_BAT_COLD);
+	wm8350_mask_irq(wm8350, WM8350_IRQ_CHG_BAT_FAIL);
+	wm8350_free_irq(wm8350, WM8350_IRQ_CHG_BAT_FAIL);
+	wm8350_mask_irq(wm8350, WM8350_IRQ_CHG_TO);
+	wm8350_free_irq(wm8350, WM8350_IRQ_CHG_TO);
+	wm8350_mask_irq(wm8350, WM8350_IRQ_CHG_END);
+	wm8350_free_irq(wm8350, WM8350_IRQ_CHG_END);
+	wm8350_mask_irq(wm8350, WM8350_IRQ_CHG_START);
+	wm8350_free_irq(wm8350, WM8350_IRQ_CHG_START);
+	wm8350_mask_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_3P9);
+	wm8350_free_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_3P9);
+	wm8350_mask_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_3P1);
+	wm8350_free_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_3P1);
+	wm8350_mask_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_2P85);
+	wm8350_free_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_2P85);
+	wm8350_mask_irq(wm8350, WM8350_IRQ_EXT_USB_FB);
+	wm8350_free_irq(wm8350, WM8350_IRQ_EXT_USB_FB);
+	wm8350_mask_irq(wm8350, WM8350_IRQ_EXT_WALL_FB);
+	wm8350_free_irq(wm8350, WM8350_IRQ_EXT_WALL_FB);
+	wm8350_mask_irq(wm8350, WM8350_IRQ_EXT_BAT_FB);
+	wm8350_free_irq(wm8350, WM8350_IRQ_EXT_BAT_FB);
+}
+
+static __devinit int wm8350_power_probe(struct platform_device *pdev)
+{
+	struct wm8350 *wm8350 = platform_get_drvdata(pdev);
+	struct wm8350_power *power = &wm8350->power;
+	struct wm8350_charger_policy *policy = power->policy;
+	struct power_supply *usb = &power->usb;
+	struct power_supply *battery = &power->battery;
+	struct power_supply *ac = &power->ac;
+	int ret;
+
+	ac->name = "wm8350-ac";
+	ac->type = POWER_SUPPLY_TYPE_MAINS;
+	ac->properties = wm8350_ac_props;
+	ac->num_properties = ARRAY_SIZE(wm8350_ac_props);
+	ac->get_property = wm8350_ac_get_prop;
+	ret = power_supply_register(&pdev->dev, ac);
+	if (ret)
+		return ret;
+
+	battery->name = "wm8350-battery";
+	battery->properties = wm8350_bat_props;
+	battery->num_properties = ARRAY_SIZE(wm8350_bat_props);
+	battery->get_property = wm8350_bat_get_property;
+	battery->use_for_apm = 1;
+	ret = power_supply_register(&pdev->dev, battery);
+	if (ret)
+		goto battery_failed;
+
+	usb->name = "wm8350-usb",
+	usb->type = POWER_SUPPLY_TYPE_USB;
+	usb->properties = wm8350_usb_props;
+	usb->num_properties = ARRAY_SIZE(wm8350_usb_props);
+	usb->get_property = wm8350_usb_get_prop;
+	ret = power_supply_register(&pdev->dev, usb);
+	if (ret)
+		goto usb_failed;
+
+	ret = device_create_file(&pdev->dev, &dev_attr_charger_state);
+	if (ret < 0)
+		dev_warn(wm8350->dev, "failed to add charge sysfs: %d\n", ret);
+	ret = 0;
+
+	wm8350_init_charger(wm8350);
+	if (wm8350_charger_config(wm8350, policy) == 0) {
+		wm8350_reg_unlock(wm8350);
+		wm8350_set_bits(wm8350, WM8350_POWER_MGMT_5, WM8350_CHG_ENA);
+		wm8350_reg_lock(wm8350);
+	}
+
+	return ret;
+
+usb_failed:
+	power_supply_unregister(battery);
+battery_failed:
+	power_supply_unregister(ac);
+
+	return ret;
+}
+
+static __devexit int wm8350_power_remove(struct platform_device *pdev)
+{
+	struct wm8350 *wm8350 = platform_get_drvdata(pdev);
+	struct wm8350_power *power = &wm8350->power;
+
+	free_charger_irq(wm8350);
+	device_remove_file(&pdev->dev, &dev_attr_charger_state);
+	power_supply_unregister(&power->battery);
+	power_supply_unregister(&power->ac);
+	power_supply_unregister(&power->usb);
+	return 0;
+}
+
+static struct platform_driver wm8350_power_driver = {
+	.probe = wm8350_power_probe,
+	.remove = __devexit_p(wm8350_power_remove),
+	.driver = {
+		.name = "wm8350-power",
+	},
+};
+
+static int __init wm8350_power_init(void)
+{
+	return platform_driver_register(&wm8350_power_driver);
+}
+module_init(wm8350_power_init);
+
+static void __exit wm8350_power_exit(void)
+{
+	platform_driver_unregister(&wm8350_power_driver);
+}
+module_exit(wm8350_power_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Power supply driver for WM8350");
+MODULE_ALIAS("platform:wm8350-power");
diff --git a/include/linux/mfd/wm8350/core.h b/include/linux/mfd/wm8350/core.h
index cc190055b9c4..d2614dfc9397 100644
--- a/include/linux/mfd/wm8350/core.h
+++ b/include/linux/mfd/wm8350/core.h
@@ -57,6 +57,9 @@
 #define WM8350_OVER_CURRENT_INT_STATUS_MASK     0x25
 #define WM8350_GPIO_INT_STATUS_MASK             0x26
 #define WM8350_COMPARATOR_INT_STATUS_MASK       0x27
+#define WM8350_MISC_OVERRIDES			0xE3
+#define WM8350_COMPARATOR_OVERRIDES		0xE7
+#define WM8350_STATE_MACHINE_STATUS		0xE9
 
 #define WM8350_MAX_REGISTER                     0xFF
 
@@ -523,6 +526,29 @@
 #define WM8350_DC2_STS                          0x0002
 #define WM8350_DC1_STS                          0x0001
 
+/*
+ * R227 (0xE3) - Misc Overrides
+ */
+#define WM8350_USB_LIMIT_OVRDE			0x0400
+
+/*
+ * R227 (0xE7) - Comparator Overrides
+ */
+#define WM8350_USB_FB_OVRDE			0x8000
+#define WM8350_WALL_FB_OVRDE			0x4000
+#define WM8350_BATT_FB_OVRDE			0x2000
+
+
+/*
+ * R233 (0xE9) - State Machinine Status
+ */
+#define WM8350_USB_SM_MASK			0x0700
+#define WM8350_USB_SM_SHIFT			8
+
+#define WM8350_USB_SM_100_SLV   1
+#define WM8350_USB_SM_500_SLV   5
+#define WM8350_USB_SM_STDBY_SLV 7
+
 /* WM8350 wake up conditions */
 #define WM8350_IRQ_WKUP_OFF_STATE		43
 #define WM8350_IRQ_WKUP_HIB_STATE		44
diff --git a/include/linux/mfd/wm8350/supply.h b/include/linux/mfd/wm8350/supply.h
index 1c8f3cde79b0..79721513fa9f 100644
--- a/include/linux/mfd/wm8350/supply.h
+++ b/include/linux/mfd/wm8350/supply.h
@@ -13,7 +13,8 @@
 #ifndef __LINUX_MFD_WM8350_SUPPLY_H_
 #define __LINUX_MFD_WM8350_SUPPLY_H_
 
-#include <linux/platform_device.h>
+#include <linux/mutex.h>
+#include <linux/power_supply.h>
 
 /*
  * Charger registers
@@ -104,8 +105,28 @@
 #define WM8350_IRQ_EXT_WALL_FB			37
 #define WM8350_IRQ_EXT_BAT_FB			38
 
+/*
+ * Policy to control charger state machine.
+ */
+struct wm8350_charger_policy {
+
+	/* charger state machine policy  - set in machine driver */
+	int eoc_mA;		/* end of charge current (mA)  */
+	int charge_mV;		/* charge voltage */
+	int fast_limit_mA;	/* fast charge current limit */
+	int fast_limit_USB_mA;	/* USB fast charge current limit */
+	int charge_timeout;	/* charge timeout (mins) */
+	int trickle_start_mV;	/* trickle charge starts at mV */
+	int trickle_charge_mA;	/* trickle charge current */
+	int trickle_charge_USB_mA;	/* USB trickle charge current */
+};
+
 struct wm8350_power {
 	struct platform_device *pdev;
+	struct power_supply battery;
+	struct power_supply usb;
+	struct power_supply ac;
+	struct wm8350_charger_policy *policy;
 };
 
 #endif
-- 
cgit v1.2.3


From d756f4a4446227ca9626087939a6769ca55ab036 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Mon, 24 Nov 2008 20:20:30 +0100
Subject: mfd: Switch WM8350 revision detection to a feature based model

Rather than check for chip revisions in the WM8350 drivers have the core
code set flags for relevant differences.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/wm8350-core.c         | 6 ++----
 drivers/power/wm8350_power.c      | 2 +-
 include/linux/mfd/wm8350/core.h   | 2 --
 include/linux/mfd/wm8350/supply.h | 2 ++
 4 files changed, 5 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/wm8350-core.c b/drivers/mfd/wm8350-core.c
index 60439bd3984d..764bf15ea684 100644
--- a/drivers/mfd/wm8350-core.c
+++ b/drivers/mfd/wm8350-core.c
@@ -1240,19 +1240,17 @@ int wm8350_device_init(struct wm8350 *wm8350, int irq,
 		switch ((id2 & WM8350_CHIP_REV_MASK) >> 12) {
 		case WM8350_REV_E:
 			dev_info(wm8350->dev, "Found Rev E device\n");
-			wm8350->rev = WM8350_REV_E;
 			break;
 		case WM8350_REV_F:
 			dev_info(wm8350->dev, "Found Rev F device\n");
-			wm8350->rev = WM8350_REV_F;
 			break;
 		case WM8350_REV_G:
 			dev_info(wm8350->dev, "Found Rev G device\n");
-			wm8350->rev = WM8350_REV_G;
+			wm8350->power.rev_g_coeff = 1;
 			break;
 		case WM8350_REV_H:
 			dev_info(wm8350->dev, "Found Rev H device\n");
-			wm8350->rev = WM8350_REV_H;
+			wm8350->power.rev_g_coeff = 1;
 			break;
 		default:
 			/* For safety we refuse to run on unknown hardware */
diff --git a/drivers/power/wm8350_power.c b/drivers/power/wm8350_power.c
index 9c0a847a1139..74e7593beffb 100644
--- a/drivers/power/wm8350_power.c
+++ b/drivers/power/wm8350_power.c
@@ -44,7 +44,7 @@ static int wm8350_read_usb_uvolts(struct wm8350 *wm8350)
 
 static inline int wm8350_charge_time_min(struct wm8350 *wm8350, int min)
 {
-	if (wm8350->rev < WM8350_REV_G)
+	if (!wm8350->power.rev_g_coeff)
 		return (((min - 30) / 15) & 0xf) << 8;
 	else
 		return (((min - 30) / 30) & 0xf) << 8;
diff --git a/include/linux/mfd/wm8350/core.h b/include/linux/mfd/wm8350/core.h
index d2614dfc9397..3c9735663f36 100644
--- a/include/linux/mfd/wm8350/core.h
+++ b/include/linux/mfd/wm8350/core.h
@@ -585,8 +585,6 @@ struct wm8350_irq {
 };
 
 struct wm8350 {
-	int rev;		/* chip revision */
-
 	struct device *dev;
 
 	/* device IO */
diff --git a/include/linux/mfd/wm8350/supply.h b/include/linux/mfd/wm8350/supply.h
index 79721513fa9f..2b9479310bbd 100644
--- a/include/linux/mfd/wm8350/supply.h
+++ b/include/linux/mfd/wm8350/supply.h
@@ -127,6 +127,8 @@ struct wm8350_power {
 	struct power_supply usb;
 	struct power_supply ac;
 	struct wm8350_charger_policy *policy;
+
+	int rev_g_coeff;
 };
 
 #endif
-- 
cgit v1.2.3


From b797a5551979da22b0a35632198ffc8a330d9537 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Mon, 24 Nov 2008 20:22:58 +0100
Subject: mfd: Refactor WM8350 chip identification

Since the WM8350 driver was originally written the semantics for the
identification registers of the chip have been clarified, allowing
us to do an exact match on all the fields. This avoids mistakenly
running on unsupported hardware.

Also change to using the datasheet names more consistently for
legibility and fix a printk() that should be dev_err().

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/wm8350-core.c       | 54 ++++++++++++++++++++++++++++-------------
 include/linux/mfd/wm8350/core.h |  6 +++++
 2 files changed, 43 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/wm8350-core.c b/drivers/mfd/wm8350-core.c
index 764bf15ea684..2188d759cbde 100644
--- a/drivers/mfd/wm8350-core.c
+++ b/drivers/mfd/wm8350-core.c
@@ -1227,52 +1227,72 @@ int wm8350_device_init(struct wm8350 *wm8350, int irq,
 		       struct wm8350_platform_data *pdata)
 {
 	int ret = -EINVAL;
-	u16 id1, id2, mask, mode;
+	u16 id1, id2, mask_rev;
+	u16 cust_id, mode, chip_rev;
 
 	/* get WM8350 revision and config mode */
 	wm8350->read_dev(wm8350, WM8350_RESET_ID, sizeof(id1), &id1);
 	wm8350->read_dev(wm8350, WM8350_ID, sizeof(id2), &id2);
+	wm8350->read_dev(wm8350, WM8350_REVISION, sizeof(mask_rev), &mask_rev);
 
 	id1 = be16_to_cpu(id1);
 	id2 = be16_to_cpu(id2);
+	mask_rev = be16_to_cpu(mask_rev);
 
-	if (id1 == 0x6143) {
-		switch ((id2 & WM8350_CHIP_REV_MASK) >> 12) {
+	if (id1 != 0x6143) {
+		dev_err(wm8350->dev,
+			"Device with ID %x is not a WM8350\n", id1);
+		ret = -ENODEV;
+		goto err;
+	}
+
+	mode = id2 & WM8350_CONF_STS_MASK >> 10;
+	cust_id = id2 & WM8350_CUST_ID_MASK;
+	chip_rev = (id2 & WM8350_CHIP_REV_MASK) >> 12;
+	dev_info(wm8350->dev,
+		 "CONF_STS %d, CUST_ID %d, MASK_REV %d, CHIP_REV %d\n",
+		 mode, cust_id, mask_rev, chip_rev);
+
+	if (cust_id != 0) {
+		dev_err(wm8350->dev, "Unsupported CUST_ID\n");
+		ret = -ENODEV;
+		goto err;
+	}
+
+	switch (mask_rev) {
+	case 0:
+		switch (chip_rev) {
 		case WM8350_REV_E:
-			dev_info(wm8350->dev, "Found Rev E device\n");
+			dev_info(wm8350->dev, "WM8350 Rev E\n");
 			break;
 		case WM8350_REV_F:
-			dev_info(wm8350->dev, "Found Rev F device\n");
+			dev_info(wm8350->dev, "WM8350 Rev F\n");
 			break;
 		case WM8350_REV_G:
-			dev_info(wm8350->dev, "Found Rev G device\n");
+			dev_info(wm8350->dev, "WM8350 Rev G\n");
 			wm8350->power.rev_g_coeff = 1;
 			break;
 		case WM8350_REV_H:
-			dev_info(wm8350->dev, "Found Rev H device\n");
+			dev_info(wm8350->dev, "WM8350 Rev H\n");
 			wm8350->power.rev_g_coeff = 1;
 			break;
 		default:
 			/* For safety we refuse to run on unknown hardware */
-			dev_info(wm8350->dev, "Found unknown rev %x\n",
-				 (id2 & WM8350_CHIP_REV_MASK) >> 12);
+			dev_err(wm8350->dev, "Unknown WM8350 CHIP_REV\n");
 			ret = -ENODEV;
 			goto err;
 		}
-	} else {
-		dev_info(wm8350->dev, "Device with ID %x is not a WM8350\n",
-			 id1);
+		break;
+
+	default:
+		dev_err(wm8350->dev, "Unknown MASK_REV\n");
 		ret = -ENODEV;
 		goto err;
 	}
 
-	mode = id2 & WM8350_CONF_STS_MASK >> 10;
-	mask = id2 & WM8350_CUST_ID_MASK;
-	dev_info(wm8350->dev, "Config mode %d, ROM mask %d\n", mode, mask);
-
 	ret = wm8350_create_cache(wm8350, mode);
 	if (ret < 0) {
-		printk(KERN_ERR "wm8350: failed to create register cache\n");
+		dev_err(wm8350->dev, "Failed to create register cache\n");
 		return ret;
 	}
 
diff --git a/include/linux/mfd/wm8350/core.h b/include/linux/mfd/wm8350/core.h
index 3c9735663f36..2a7abeebe777 100644
--- a/include/linux/mfd/wm8350/core.h
+++ b/include/linux/mfd/wm8350/core.h
@@ -29,6 +29,7 @@
  */
 #define WM8350_RESET_ID                         0x00
 #define WM8350_ID                               0x01
+#define WM8350_REVISION				0x02
 #define WM8350_SYSTEM_CONTROL_1                 0x03
 #define WM8350_SYSTEM_CONTROL_2                 0x04
 #define WM8350_SYSTEM_HIBERNATE                 0x05
@@ -79,6 +80,11 @@
 #define WM8350_CONF_STS_MASK                    0x0C00
 #define WM8350_CUST_ID_MASK                     0x00FF
 
+/*
+ * R2 (0x02) - Revision
+ */
+#define WM8350_MASK_REV_MASK			0x00FF
+
 /*
  * R3 (0x03) - System Control 1
  */
-- 
cgit v1.2.3


From 7e386e6e0e4f34f0545e8923e22fe4dd61ef9d48 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Sun, 30 Nov 2008 22:43:21 +0100
Subject: power_supply: Add cold to the POWER_SUPPLY_HEALTH report values

Some systems are able to report problems with batteries being under
temperature.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Anton Vorontsov <cbouatmailru@gmail.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/power/power_supply_sysfs.c | 2 +-
 include/linux/power_supply.h       | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/power/power_supply_sysfs.c b/drivers/power/power_supply_sysfs.c
index 23ae8460f5c1..ac01e06817fb 100644
--- a/drivers/power/power_supply_sysfs.c
+++ b/drivers/power/power_supply_sysfs.c
@@ -45,7 +45,7 @@ static ssize_t power_supply_show_property(struct device *dev,
 	};
 	static char *health_text[] = {
 		"Unknown", "Good", "Overheat", "Dead", "Over voltage",
-		"Unspecified failure"
+		"Unspecified failure", "Cold",
 	};
 	static char *technology_text[] = {
 		"Unknown", "NiMH", "Li-ion", "Li-poly", "LiFe", "NiCd",
diff --git a/include/linux/power_supply.h b/include/linux/power_supply.h
index f9348cba6dc1..8ff25e0e7f7a 100644
--- a/include/linux/power_supply.h
+++ b/include/linux/power_supply.h
@@ -45,6 +45,7 @@ enum {
 	POWER_SUPPLY_HEALTH_DEAD,
 	POWER_SUPPLY_HEALTH_OVERVOLTAGE,
 	POWER_SUPPLY_HEALTH_UNSPEC_FAILURE,
+	POWER_SUPPLY_HEALTH_COLD,
 };
 
 enum {
-- 
cgit v1.2.3


From 4008e879e1325c29362aa2c3fa4b527273ae15a8 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Sun, 30 Nov 2008 22:45:14 +0100
Subject: power_supply: Add battery health reporting for WM8350

Implement support for reporting battery health in the WM8350 battery
interface. Since we are now able to report this via the classs remove
the diagnostics from the interrupt handler.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Anton Vorontsov <cbouatmailru@gmail.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/wm8350-regmap.c     |  2 +-
 drivers/power/wm8350_power.c    | 33 +++++++++++++++++++++++++--------
 include/linux/mfd/wm8350/core.h |  7 +++++++
 3 files changed, 33 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/wm8350-regmap.c b/drivers/mfd/wm8350-regmap.c
index aaf394aa9c2d..b43d64c2b920 100644
--- a/drivers/mfd/wm8350-regmap.c
+++ b/drivers/mfd/wm8350-regmap.c
@@ -1314,7 +1314,7 @@ const struct wm8350_reg_access wm8350_reg_io_map[] = {
 	{ 0xFFFF, 0xFFFF, 0xFFFF }, /* R223 */
 	{ 0x0000, 0x0000, 0x0000 }, /* R224 */
 	{ 0x8F3F, 0x0000, 0xFFFF }, /* R225 - DCDC/LDO status */
-	{ 0x0000, 0x0000, 0x0000 }, /* R226 */
+	{ 0x0000, 0x0000, 0xFFFF }, /* R226 - Charger status */
 	{ 0x0000, 0x0000, 0xFFFF }, /* R227 */
 	{ 0x0000, 0x0000, 0x0000 }, /* R228 */
 	{ 0x0000, 0x0000, 0x0000 }, /* R229 */
diff --git a/drivers/power/wm8350_power.c b/drivers/power/wm8350_power.c
index 74e7593beffb..1b16bf343f2f 100644
--- a/drivers/power/wm8350_power.c
+++ b/drivers/power/wm8350_power.c
@@ -190,22 +190,18 @@ static void wm8350_charger_handler(struct wm8350 *wm8350, int irq, void *data)
 	struct wm8350_charger_policy *policy = power->policy;
 
 	switch (irq) {
-	case WM8350_IRQ_CHG_BAT_HOT:
-		dev_err(wm8350->dev, "battery too hot\n");
-		break;
-	case WM8350_IRQ_CHG_BAT_COLD:
-		dev_err(wm8350->dev, "battery too cold\n");
-		break;
 	case WM8350_IRQ_CHG_BAT_FAIL:
 		dev_err(wm8350->dev, "battery failed\n");
 		break;
 	case WM8350_IRQ_CHG_TO:
 		dev_err(wm8350->dev, "charger timeout\n");
-		break;
-	case WM8350_IRQ_CHG_END:
 		power_supply_changed(&power->battery);
 		break;
+
+	case WM8350_IRQ_CHG_BAT_HOT:
+	case WM8350_IRQ_CHG_BAT_COLD:
 	case WM8350_IRQ_CHG_START:
+	case WM8350_IRQ_CHG_END:
 		power_supply_changed(&power->battery);
 		break;
 
@@ -308,6 +304,23 @@ static enum power_supply_property wm8350_usb_props[] = {
  *		Battery properties
  *********************************************************************/
 
+static int wm8350_bat_check_health(struct wm8350 *wm8350)
+{
+	u16 reg;
+
+	if (wm8350_read_battery_uvolts(wm8350) < 2850000)
+		return POWER_SUPPLY_HEALTH_UNSPEC_FAILURE;
+
+	reg = wm8350_reg_read(wm8350, WM8350_CHARGER_OVERRIDES);
+	if (reg & WM8350_CHG_BATT_HOT_OVRDE)
+		return POWER_SUPPLY_HEALTH_OVERHEAT;
+
+	if (reg & WM8350_CHG_BATT_COLD_OVRDE)
+		return POWER_SUPPLY_HEALTH_COLD;
+
+	return POWER_SUPPLY_HEALTH_GOOD;
+}
+
 static int wm8350_bat_get_property(struct power_supply *psy,
 				   enum power_supply_property psp,
 				   union power_supply_propval *val)
@@ -326,6 +339,9 @@ static int wm8350_bat_get_property(struct power_supply *psy,
 	case POWER_SUPPLY_PROP_VOLTAGE_NOW:
 		val->intval = wm8350_read_battery_uvolts(wm8350);
 		break;
+	case POWER_SUPPLY_PROP_HEALTH:
+		val->intval = wm8350_bat_check_health(wm8350);
+		break;
 	default:
 		ret = -EINVAL;
 		break;
@@ -338,6 +354,7 @@ static enum power_supply_property wm8350_bat_props[] = {
 	POWER_SUPPLY_PROP_STATUS,
 	POWER_SUPPLY_PROP_ONLINE,
 	POWER_SUPPLY_PROP_VOLTAGE_NOW,
+	POWER_SUPPLY_PROP_HEALTH,
 };
 
 /*********************************************************************
diff --git a/include/linux/mfd/wm8350/core.h b/include/linux/mfd/wm8350/core.h
index 2a7abeebe777..afeff6f1316c 100644
--- a/include/linux/mfd/wm8350/core.h
+++ b/include/linux/mfd/wm8350/core.h
@@ -58,6 +58,7 @@
 #define WM8350_OVER_CURRENT_INT_STATUS_MASK     0x25
 #define WM8350_GPIO_INT_STATUS_MASK             0x26
 #define WM8350_COMPARATOR_INT_STATUS_MASK       0x27
+#define WM8350_CHARGER_OVERRIDES		0xE2
 #define WM8350_MISC_OVERRIDES			0xE3
 #define WM8350_COMPARATOR_OVERRIDES		0xE7
 #define WM8350_STATE_MACHINE_STATUS		0xE9
@@ -532,6 +533,12 @@
 #define WM8350_DC2_STS                          0x0002
 #define WM8350_DC1_STS                          0x0001
 
+/*
+ * R226 (0xE2) - Charger status
+ */
+#define WM8350_CHG_BATT_HOT_OVRDE		0x8000
+#define WM8350_CHG_BATT_COLD_OVRDE		0x4000
+
 /*
  * R227 (0xE3) - Misc Overrides
  */
-- 
cgit v1.2.3


From 67460a7c26271fd7a32e5d51b2c806a84ce78a62 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Mon, 1 Dec 2008 00:35:33 +0100
Subject: mfd: twl4030: cleanup symbols and OMAP dependency

Finish removing dependency of TWL driver stack on platform-specific
IRQ definitions ... and remove the build dependency on OMAP.

This lets the TWL4030 code be included in test builds for most
platforms, and will make it easier for non-OMAP folk to update
most of this code for new APIs etc.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/Kconfig         |  2 +-
 include/linux/i2c/twl4030.h | 31 -------------------------------
 2 files changed, 1 insertion(+), 32 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index 257277394f8c..8cd3dd9a69b2 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -63,7 +63,7 @@ config UCB1400_CORE
 
 config TWL4030_CORE
 	bool "Texas Instruments TWL4030/TPS659x0 Support"
-	depends on I2C=y && GENERIC_HARDIRQS && (ARCH_OMAP2 || ARCH_OMAP3)
+	depends on I2C=y && GENERIC_HARDIRQS
 	help
 	  Say yes here if you have TWL4030 family chip on your board.
 	  This core driver provides register access and IRQ handling
diff --git a/include/linux/i2c/twl4030.h b/include/linux/i2c/twl4030.h
index ae25c907b7cf..d4846695bcd0 100644
--- a/include/linux/i2c/twl4030.h
+++ b/include/linux/i2c/twl4030.h
@@ -285,33 +285,6 @@ struct twl4030_platform_data {
 
 int twl4030_sih_setup(int module);
 
-/*
- * FIXME completely stop using TWL4030_IRQ_BASE ... instead, pass the
- * IRQ data to subsidiary devices using platform device resources.
- */
-
-/* IRQ information-need base */
-#include <mach/irqs.h>
-/* TWL4030 interrupts */
-
-/* #define TWL4030_MODIRQ_GPIO		(TWL4030_IRQ_BASE + 0) */
-#define TWL4030_MODIRQ_KEYPAD		(TWL4030_IRQ_BASE + 1)
-#define TWL4030_MODIRQ_BCI		(TWL4030_IRQ_BASE + 2)
-#define TWL4030_MODIRQ_MADC		(TWL4030_IRQ_BASE + 3)
-/* #define TWL4030_MODIRQ_USB		(TWL4030_IRQ_BASE + 4) */
-/* #define TWL4030_MODIRQ_PWR		(TWL4030_IRQ_BASE + 5) */
-
-#define TWL4030_PWRIRQ_PWRBTN		(TWL4030_PWR_IRQ_BASE + 0)
-/* #define TWL4030_PWRIRQ_CHG_PRES		(TWL4030_PWR_IRQ_BASE + 1) */
-/* #define TWL4030_PWRIRQ_USB_PRES		(TWL4030_PWR_IRQ_BASE + 2) */
-/* #define TWL4030_PWRIRQ_RTC		(TWL4030_PWR_IRQ_BASE + 3) */
-/* #define TWL4030_PWRIRQ_HOT_DIE		(TWL4030_PWR_IRQ_BASE + 4) */
-/* #define TWL4030_PWRIRQ_PWROK_TIMEOUT	(TWL4030_PWR_IRQ_BASE + 5) */
-/* #define TWL4030_PWRIRQ_MBCHG		(TWL4030_PWR_IRQ_BASE + 6) */
-/* #define TWL4030_PWRIRQ_SC_DETECT	(TWL4030_PWR_IRQ_BASE + 7) */
-
-/* Rest are unsued currently*/
-
 /* Offsets to Power Registers */
 #define TWL4030_VDAC_DEV_GRP		0x3B
 #define TWL4030_VDAC_DEDICATED		0x3E
@@ -322,10 +295,6 @@ int twl4030_sih_setup(int module);
 #define TWL4030_VAUX3_DEV_GRP		0x1F
 #define TWL4030_VAUX3_DEDICATED		0x22
 
-/* TWL4030 GPIO interrupt definitions */
-
-#define TWL4030_GPIO_IRQ_NO(n)		(TWL4030_GPIO_IRQ_BASE + (n))
-
 /*
  * Exported TWL4030 GPIO APIs
  *
-- 
cgit v1.2.3


From dad759ff8ba79927766e3f0159bfc5fb6de0f982 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Mon, 1 Dec 2008 00:43:58 +0100
Subject: mfd: twl4030: create some regulator devices

Initial code to create twl4030 voltage regulator devices, using
the new regulator framework.  Note that this now starts to care
what name is used to declare the TWL chip:

 - TWL4030 is the "old" chip; newer ones have a bigger variety
   of VAUX2 voltages.

 - TWL5030 is the core "new" chip; TPS65950 is its catalog version.

 - The TPS65930 and TPS65920 are cost-reduced catalog versions of
   TWL5030 parts ... fewer regulators, no battery charger, etc.

Board-specific regulator configuration should be provided, listing
which regulators are used and their constraints (e.g. 1.8V only).

Code that could ("should"?) leverage the regulator stuff includes
TWL4030 USB transceiver support and MMC glue, LCD support for the
3430SDP and Labrador boards, and S-Video output.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/twl4030-core.c  | 174 +++++++++++++++++++++++++++++++++++++++++---
 include/linux/i2c/twl4030.h |  47 ++++++++++++
 2 files changed, 211 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/twl4030-core.c b/drivers/mfd/twl4030-core.c
index f5486cce86f4..8ab9ee8543a6 100644
--- a/drivers/mfd/twl4030-core.c
+++ b/drivers/mfd/twl4030-core.c
@@ -33,6 +33,8 @@
 #include <linux/clk.h>
 #include <linux/err.h>
 
+#include <linux/regulator/machine.h>
+
 #include <linux/i2c.h>
 #include <linux/i2c/twl4030.h>
 
@@ -71,6 +73,13 @@
 #define twl_has_gpio()	false
 #endif
 
+#if defined(CONFIG_REGULATOR_TWL4030) \
+	|| defined(CONFIG_REGULATOR_TWL4030_MODULE)
+#define twl_has_regulator()	true
+#else
+#define twl_has_regulator()	false
+#endif
+
 #if defined(CONFIG_TWL4030_MADC) || defined(CONFIG_TWL4030_MADC_MODULE)
 #define twl_has_madc()	true
 #else
@@ -149,6 +158,10 @@
 #define HIGH_PERF_SQ			(1 << 3)
 
 
+/* chip-specific feature flags, for i2c_device_id.driver_data */
+#define TWL4030_VAUX2		BIT(0)	/* pre-5030 voltage ranges */
+#define TPS_SUBSET		BIT(1)	/* tps659[23]0 have fewer LDOs */
+
 /*----------------------------------------------------------------------*/
 
 /* is driver active, bound to a chip? */
@@ -352,7 +365,8 @@ EXPORT_SYMBOL(twl4030_i2c_read_u8);
 
 /*----------------------------------------------------------------------*/
 
-static struct device *add_child(unsigned chip, const char *name,
+static struct device *
+add_numbered_child(unsigned chip, const char *name, int num,
 		void *pdata, unsigned pdata_len,
 		bool can_wakeup, int irq0, int irq1)
 {
@@ -360,7 +374,7 @@ static struct device *add_child(unsigned chip, const char *name,
 	struct twl4030_client	*twl = &twl4030_modules[chip];
 	int			status;
 
-	pdev = platform_device_alloc(name, -1);
+	pdev = platform_device_alloc(name, num);
 	if (!pdev) {
 		dev_dbg(&twl->client->dev, "can't alloc dev\n");
 		status = -ENOMEM;
@@ -402,17 +416,52 @@ err:
 	return &pdev->dev;
 }
 
+static inline struct device *add_child(unsigned chip, const char *name,
+		void *pdata, unsigned pdata_len,
+		bool can_wakeup, int irq0, int irq1)
+{
+	return add_numbered_child(chip, name, -1, pdata, pdata_len,
+		can_wakeup, irq0, irq1);
+}
+
+static struct device *
+add_regulator_linked(int num, struct regulator_init_data *pdata,
+		struct regulator_consumer_supply *consumers,
+		unsigned num_consumers)
+{
+	/* regulator framework demands init_data ... */
+	if (!pdata)
+		return NULL;
+
+	if (consumers && !pdata->consumer_supplies) {
+		pdata->consumer_supplies = consumers;
+		pdata->num_consumer_supplies = num_consumers;
+	}
+
+	/* NOTE:  we currently ignore regulator IRQs, e.g. for short circuits */
+	return add_numbered_child(3, "twl4030_reg", num,
+		pdata, sizeof(*pdata), false, 0, 0);
+}
+
+static struct device *
+add_regulator(int num, struct regulator_init_data *pdata)
+{
+	return add_regulator_linked(num, pdata, NULL, 0);
+}
+
 /*
  * NOTE:  We know the first 8 IRQs after pdata->base_irq are
  * for the PIH, and the next are for the PWR_INT SIH, since
  * that's how twl_init_irq() sets things up.
  */
 
-static int add_children(struct twl4030_platform_data *pdata)
+static int
+add_children(struct twl4030_platform_data *pdata, unsigned long features)
 {
 	struct device	*child;
+	struct device	*usb_transceiver = NULL;
 
-	if (twl_has_bci() && pdata->bci) {
+	if (twl_has_bci() && pdata->bci && !(features & TPS_SUBSET)) {
 		child = add_child(3, "twl4030_bci",
 				pdata->bci, sizeof(*pdata->bci),
 				false,
@@ -469,6 +518,111 @@ static int add_children(struct twl4030_platform_data *pdata)
 				pdata->irq_base + 8 + 2, pdata->irq_base + 4);
 		if (IS_ERR(child))
 			return PTR_ERR(child);
+
+		/* we need to connect regulators to this transceiver */
+		usb_transceiver = child;
+	}
+
+	if (twl_has_regulator()) {
+		/*
+		child = add_regulator(TWL4030_REG_VPLL1, pdata->vpll1);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+		*/
+
+		child = add_regulator(TWL4030_REG_VMMC1, pdata->vmmc1);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+
+		child = add_regulator(TWL4030_REG_VDAC, pdata->vdac);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+
+		child = add_regulator((features & TWL4030_VAUX2)
+					? TWL4030_REG_VAUX2_4030
+					: TWL4030_REG_VAUX2,
+				pdata->vaux2);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+	}
+
+	if (twl_has_regulator() && usb_transceiver) {
+		static struct regulator_consumer_supply usb1v5 = {
+			.supply =	"usb1v5",
+		};
+		static struct regulator_consumer_supply usb1v8 = {
+			.supply =	"usb1v8",
+		};
+		static struct regulator_consumer_supply usb3v1 = {
+			.supply =	"usb3v1",
+		};
+		static struct regulator_consumer_supply usbcp = {
+			.supply =	"usbcp",
+		};
+
+		/* this is a template that gets copied */
+		struct regulator_init_data usb_fixed = {
+			.constraints.valid_modes_mask =
+				  REGULATOR_MODE_NORMAL
+				| REGULATOR_MODE_STANDBY,
+			.constraints.valid_ops_mask =
+				  REGULATOR_CHANGE_MODE
+				| REGULATOR_CHANGE_STATUS,
+		};
+
+		usb1v5.dev = usb_transceiver;
+		usb1v8.dev = usb_transceiver;
+		usb3v1.dev = usb_transceiver;
+		usbcp.dev = usb_transceiver;
+
+		child = add_regulator_linked(TWL4030_REG_VUSB1V5, &usb_fixed,
+				&usb1v5, 1);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+
+		child = add_regulator_linked(TWL4030_REG_VUSB1V8, &usb_fixed,
+				&usb1v8, 1);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+
+		child = add_regulator_linked(TWL4030_REG_VUSB3V1, &usb_fixed,
+				&usb3v1, 1);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+
+		child = add_regulator_linked(TWL4030_REG_VUSBCP, &usb_fixed,
+				&usbcp, 1);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+	}
+
+	/* maybe add LDOs that are omitted on cost-reduced parts */
+	if (twl_has_regulator() && !(features & TPS_SUBSET)) {
+		/*
+		child = add_regulator(TWL4030_REG_VPLL2, pdata->vpll2);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+		*/
+
+		child = add_regulator(TWL4030_REG_VMMC2, pdata->vmmc2);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+
+		child = add_regulator(TWL4030_REG_VSIM, pdata->vsim);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+
+		child = add_regulator(TWL4030_REG_VAUX1, pdata->vaux1);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+
+		child = add_regulator(TWL4030_REG_VAUX3, pdata->vaux3);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+
+		child = add_regulator(TWL4030_REG_VAUX4, pdata->vaux4);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
 	}
 
 	return 0;
@@ -632,7 +786,7 @@ twl4030_probe(struct i2c_client *client, const struct i2c_device_id *id)
 			goto fail;
 	}
 
-	status = add_children(pdata);
+	status = add_children(pdata, id->driver_data);
 fail:
 	if (status < 0)
 		twl4030_remove(client);
@@ -640,11 +794,11 @@ fail:
 }
 
 static const struct i2c_device_id twl4030_ids[] = {
-	{ "twl4030", 0 },	/* "Triton 2" */
-	{ "tps65950", 0 },	/* catalog version of twl4030 */
-	{ "tps65930", 0 },	/* fewer LDOs and DACs; no charger */
-	{ "tps65920", 0 },	/* fewer LDOs; no codec or charger */
-	{ "twl5030", 0 },	/* T2 updated */
+	{ "twl4030", TWL4030_VAUX2 },	/* "Triton 2" */
+	{ "twl5030", 0 },		/* T2 updated */
+	{ "tps65950", 0 },		/* catalog version of twl5030 */
+	{ "tps65930", TPS_SUBSET },	/* fewer LDOs and DACs; no charger */
+	{ "tps65920", TPS_SUBSET },	/* fewer LDOs; no codec or charger */
 	{ /* end of list */ },
 };
 MODULE_DEVICE_TABLE(i2c, twl4030_ids);
diff --git a/include/linux/i2c/twl4030.h b/include/linux/i2c/twl4030.h
index d4846695bcd0..e06555d40d35 100644
--- a/include/linux/i2c/twl4030.h
+++ b/include/linux/i2c/twl4030.h
@@ -278,6 +278,18 @@ struct twl4030_platform_data {
 	struct twl4030_keypad_data		*keypad;
 	struct twl4030_usb_data			*usb;
 
+	/* LDO regulators */
+	struct regulator_init_data		*vdac;
+	struct regulator_init_data		*vpll1;
+	struct regulator_init_data		*vpll2;
+	struct regulator_init_data		*vmmc1;
+	struct regulator_init_data		*vmmc2;
+	struct regulator_init_data		*vsim;
+	struct regulator_init_data		*vaux1;
+	struct regulator_init_data		*vaux2;
+	struct regulator_init_data		*vaux3;
+	struct regulator_init_data		*vaux4;
+
 	/* REVISIT more to come ... _nothing_ should be hard-wired */
 };
 
@@ -309,4 +321,39 @@ int twl4030_set_gpio_debounce(int gpio, int enable);
 	static inline int twl4030charger_usb_en(int enable) { return 0; }
 #endif
 
+/*----------------------------------------------------------------------*/
+
+/* Linux-specific regulator identifiers ... for now, we only support
+ * the LDOs, and leave the three buck converters alone.  VDD1 and VDD2
+ * need to tie into hardware based voltage scaling (cpufreq etc), while
+ * VIO is generally fixed.
+ */
+
+/* EXTERNAL dc-to-dc buck converters */
+#define TWL4030_REG_VDD1	0
+#define TWL4030_REG_VDD2	1
+#define TWL4030_REG_VIO		2
+
+/* EXTERNAL LDOs */
+#define TWL4030_REG_VDAC	3
+#define TWL4030_REG_VPLL1	4
+#define TWL4030_REG_VPLL2	5	/* not on all chips */
+#define TWL4030_REG_VMMC1	6
+#define TWL4030_REG_VMMC2	7	/* not on all chips */
+#define TWL4030_REG_VSIM	8	/* not on all chips */
+#define TWL4030_REG_VAUX1	9	/* not on all chips */
+#define TWL4030_REG_VAUX2_4030	10	/* (twl4030-specific) */
+#define TWL4030_REG_VAUX2	11	/* (twl5030 and newer) */
+#define TWL4030_REG_VAUX3	12	/* not on all chips */
+#define TWL4030_REG_VAUX4	13	/* not on all chips */
+
+/* INTERNAL LDOs */
+#define TWL4030_REG_VINTANA1	14
+#define TWL4030_REG_VINTANA2	15
+#define TWL4030_REG_VINTDIG	16
+#define TWL4030_REG_VUSB1V5	17
+#define TWL4030_REG_VUSB1V8	18
+#define TWL4030_REG_VUSB3V1	19
+#define TWL4030_REG_VUSBCP	20
+
 #endif /* End of __TWL4030_H */
-- 
cgit v1.2.3


From b73eac7871d002835be17d4602cced2c15c0db4b Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Sun, 7 Dec 2008 19:10:58 +0100
Subject: mfd: twl4030 regulator bug fixes

This contains two bugfixes to the initial twl4030 regulator
support patch related to USB:

 (a) always overwrite the old list of consumers ... else
     the regulator handles all use the same "usb1v5" name;
 (b) don't set up the "usbcp" regulator, which turns out
     to be managed through separate controls, usually ULPI
     directly from the OTG controller.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/twl4030-core.c  | 11 +----------
 include/linux/i2c/twl4030.h |  1 -
 2 files changed, 1 insertion(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/twl4030-core.c b/drivers/mfd/twl4030-core.c
index 8ab9ee8543a6..fdfbd313ae09 100644
--- a/drivers/mfd/twl4030-core.c
+++ b/drivers/mfd/twl4030-core.c
@@ -433,7 +433,7 @@ add_regulator_linked(int num, struct regulator_init_data *pdata,
 	if (!pdata)
 		return NULL;
 
-	if (consumers && !pdata->consumer_supplies) {
+	if (consumers) {
 		pdata->consumer_supplies = consumers;
 		pdata->num_consumer_supplies = num_consumers;
 	}
@@ -556,9 +556,6 @@ add_children(struct twl4030_platform_data *pdata, unsigned long features)
 		static struct regulator_consumer_supply usb3v1 = {
 			.supply =	"usb3v1",
 		};
-		static struct regulator_consumer_supply usbcp = {
-			.supply =	"usbcp",
-		};
 
 		/* this is a template that gets copied */
 		struct regulator_init_data usb_fixed = {
@@ -573,7 +570,6 @@ add_children(struct twl4030_platform_data *pdata, unsigned long features)
 		usb1v5.dev = usb_transceiver;
 		usb1v8.dev = usb_transceiver;
 		usb3v1.dev = usb_transceiver;
-		usbcp.dev = usb_transceiver;
 
 		child = add_regulator_linked(TWL4030_REG_VUSB1V5, &usb_fixed,
 				&usb1v5, 1);
@@ -589,11 +585,6 @@ add_children(struct twl4030_platform_data *pdata, unsigned long features)
 				&usb3v1, 1);
 		if (IS_ERR(child))
 			return PTR_ERR(child);
-
-		child = add_regulator_linked(TWL4030_REG_VUSBCP, &usb_fixed,
-				&usbcp, 1);
-		if (IS_ERR(child))
-			return PTR_ERR(child);
 	}
 
 	/* maybe add LDOs that are omitted on cost-reduced parts */
diff --git a/include/linux/i2c/twl4030.h b/include/linux/i2c/twl4030.h
index e06555d40d35..a8f84c01f82e 100644
--- a/include/linux/i2c/twl4030.h
+++ b/include/linux/i2c/twl4030.h
@@ -354,6 +354,5 @@ int twl4030_set_gpio_debounce(int gpio, int enable);
 #define TWL4030_REG_VUSB1V5	17
 #define TWL4030_REG_VUSB1V8	18
 #define TWL4030_REG_VUSB3V1	19
-#define TWL4030_REG_VUSBCP	20
 
 #endif /* End of __TWL4030_H */
-- 
cgit v1.2.3


From 856f6fd119411d5701d5db96e1aae1dd69923887 Mon Sep 17 00:00:00 2001
From: Mike Rapoport <mike@compulab.co.il>
Date: Thu, 18 Dec 2008 10:54:27 +0100
Subject: mfd: Dialog DA9030 battery charger MFD driver

This patch amends DA903x MFD driver with definitions and methods
needed for battery charger driver.

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/da903x.c       | 12 ++++++++++++
 include/linux/mfd/da903x.h | 44 ++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 54 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/da903x.c b/drivers/mfd/da903x.c
index 0b5bd85dfcec..fcaf1f6028dd 100644
--- a/drivers/mfd/da903x.c
+++ b/drivers/mfd/da903x.c
@@ -151,12 +151,24 @@ int da903x_write(struct device *dev, int reg, uint8_t val)
 }
 EXPORT_SYMBOL_GPL(da903x_write);
 
+int da903x_writes(struct device *dev, int reg, int len, uint8_t *val)
+{
+	return __da903x_writes(to_i2c_client(dev), reg, len, val);
+}
+EXPORT_SYMBOL_GPL(da903x_writes);
+
 int da903x_read(struct device *dev, int reg, uint8_t *val)
 {
 	return __da903x_read(to_i2c_client(dev), reg, val);
 }
 EXPORT_SYMBOL_GPL(da903x_read);
 
+int da903x_reads(struct device *dev, int reg, int len, uint8_t *val)
+{
+	return __da903x_reads(to_i2c_client(dev), reg, len, val);
+}
+EXPORT_SYMBOL_GPL(da903x_reads);
+
 int da903x_set_bits(struct device *dev, int reg, uint8_t bit_mask)
 {
 	struct da903x_chip *chip = dev_get_drvdata(dev);
diff --git a/include/linux/mfd/da903x.h b/include/linux/mfd/da903x.h
index cad314c12439..115dbe965082 100644
--- a/include/linux/mfd/da903x.h
+++ b/include/linux/mfd/da903x.h
@@ -32,6 +32,7 @@ enum {
 	DA9030_ID_LDO18,
 	DA9030_ID_LDO19,
 	DA9030_ID_LDO_INT,	/* LDO Internal */
+	DA9030_ID_BAT,		/* battery charger */
 
 	DA9034_ID_LED_1,
 	DA9034_ID_LED_2,
@@ -93,6 +94,43 @@ struct da9034_touch_pdata {
 	int	y_inverted;
 };
 
+/* DA9030 battery charger data */
+struct power_supply_info;
+
+struct da9030_battery_info {
+	/* battery parameters */
+	struct power_supply_info *battery_info;
+
+	/* current and voltage to use for battery charging */
+	unsigned int charge_milliamp;
+	unsigned int charge_millivolt;
+
+	/* voltage thresholds (in millivolts) */
+	int vbat_low;
+	int vbat_crit;
+	int vbat_charge_start;
+	int vbat_charge_stop;
+	int vbat_charge_restart;
+
+	/* battery nominal minimal and maximal voltages in millivolts */
+	int vcharge_min;
+	int vcharge_max;
+
+	/* Temperature thresholds. These are DA9030 register values
+	   "as is" and should be measured for each battery type */
+	int tbat_low;
+	int tbat_high;
+	int tbat_restart;
+
+
+	/* battery monitor interval (seconds) */
+	unsigned int batmon_interval;
+
+	/* platform callbacks for battery low and critical events */
+	void (*battery_low)(void);
+	void (*battery_critical)(void);
+};
+
 struct da903x_subdev_info {
 	int		id;
 	const char	*name;
@@ -190,11 +228,13 @@ extern int da903x_unregister_notifier(struct device *dev,
 extern int da903x_query_status(struct device *dev, unsigned int status);
 
 
-/* NOTE: the two functions below are not intended for use outside
- * of the DA9034 sub-device drivers
+/* NOTE: the functions below are not intended for use outside
+ * of the DA903x sub-device drivers
  */
 extern int da903x_write(struct device *dev, int reg, uint8_t val);
+extern int da903x_writes(struct device *dev, int reg, int len, uint8_t *val);
 extern int da903x_read(struct device *dev, int reg, uint8_t *val);
+extern int da903x_reads(struct device *dev, int reg, int len, uint8_t *val);
 extern int da903x_update(struct device *dev, int reg, uint8_t val, uint8_t mask);
 extern int da903x_set_bits(struct device *dev, int reg, uint8_t bit_mask);
 extern int da903x_clr_bits(struct device *dev, int reg, uint8_t bit_mask);
-- 
cgit v1.2.3


From 96920630624868add3f63f596523e70dbb64549a Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Thu, 18 Dec 2008 23:09:50 +0100
Subject: mfd: Add WM8352 support

The WM8352 is a variant of the WM8350. Aside from the register defaults
there are no software visible differences to the WM8350.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/Kconfig             |   16 +
 drivers/mfd/wm8350-core.c       |   86 +++-
 drivers/mfd/wm8350-i2c.c        |    1 +
 drivers/mfd/wm8350-regmap.c     | 1052 +++++++++++++++++++++++++++++++++++++++
 include/linux/mfd/wm8350/core.h |    4 +
 5 files changed, 1143 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index ddfb12b52f54..76a482dfc177 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -143,6 +143,22 @@ config MFD_WM8350_CONFIG_MODE_3
 	bool
 	depends on MFD_WM8350
 
+config MFD_WM8352_CONFIG_MODE_0
+	bool
+	depends on MFD_WM8350
+
+config MFD_WM8352_CONFIG_MODE_1
+	bool
+	depends on MFD_WM8350
+
+config MFD_WM8352_CONFIG_MODE_2
+	bool
+	depends on MFD_WM8350
+
+config MFD_WM8352_CONFIG_MODE_3
+	bool
+	depends on MFD_WM8350
+
 config MFD_WM8350_I2C
 	tristate "Support Wolfson Microelectronics WM8350 with I2C"
 	select MFD_WM8350
diff --git a/drivers/mfd/wm8350-core.c b/drivers/mfd/wm8350-core.c
index 2188d759cbde..fa505ac76c84 100644
--- a/drivers/mfd/wm8350-core.c
+++ b/drivers/mfd/wm8350-core.c
@@ -1133,35 +1133,75 @@ EXPORT_SYMBOL_GPL(wm8350_read_auxadc);
 /*
  * Cache is always host endian.
  */
-static int wm8350_create_cache(struct wm8350 *wm8350, int mode)
+static int wm8350_create_cache(struct wm8350 *wm8350, int type, int mode)
 {
 	int i, ret = 0;
 	u16 value;
 	const u16 *reg_map;
 
-	switch (mode) {
-#ifdef CONFIG_MFD_WM8350_CONFIG_MODE_0
+	switch (type) {
 	case 0:
-		reg_map = wm8350_mode0_defaults;
-		break;
+		switch (mode) {
+#ifdef CONFIG_MFD_WM8350_CONFIG_MODE_0
+		case 0:
+			reg_map = wm8350_mode0_defaults;
+			break;
 #endif
 #ifdef CONFIG_MFD_WM8350_CONFIG_MODE_1
-	case 1:
-		reg_map = wm8350_mode1_defaults;
-		break;
+		case 1:
+			reg_map = wm8350_mode1_defaults;
+			break;
 #endif
 #ifdef CONFIG_MFD_WM8350_CONFIG_MODE_2
-	case 2:
-		reg_map = wm8350_mode2_defaults;
-		break;
+		case 2:
+			reg_map = wm8350_mode2_defaults;
+			break;
 #endif
 #ifdef CONFIG_MFD_WM8350_CONFIG_MODE_3
-	case 3:
-		reg_map = wm8350_mode3_defaults;
-		break;
+		case 3:
+			reg_map = wm8350_mode3_defaults;
+			break;
 #endif
+		default:
+			dev_err(wm8350->dev,
+				"WM8350 configuration mode %d not supported\n",
+				mode);
+			return -EINVAL;
+		}
+
+	case 2:
+		switch (mode) {
+#ifdef CONFIG_MFD_WM8352_CONFIG_MODE_0
+		case 0:
+			reg_map = wm8352_mode0_defaults;
+			break;
+#endif
+#ifdef CONFIG_MFD_WM8352_CONFIG_MODE_1
+		case 1:
+			reg_map = wm8352_mode1_defaults;
+			break;
+#endif
+#ifdef CONFIG_MFD_WM8352_CONFIG_MODE_2
+		case 2:
+			reg_map = wm8352_mode2_defaults;
+			break;
+#endif
+#ifdef CONFIG_MFD_WM8352_CONFIG_MODE_3
+		case 3:
+			reg_map = wm8352_mode3_defaults;
+			break;
+#endif
+		default:
+			dev_err(wm8350->dev,
+				"WM8352 configuration mode %d not supported\n",
+				mode);
+			return -EINVAL;
+		}
+		break;
+
 	default:
-		dev_err(wm8350->dev, "Configuration mode %d not supported\n",
+		dev_err(wm8350->dev,
+			"WM835x configuration mode %d not supported\n",
 			mode);
 		return -EINVAL;
 	}
@@ -1284,13 +1324,27 @@ int wm8350_device_init(struct wm8350 *wm8350, int irq,
 		}
 		break;
 
+	case 2:
+		switch (chip_rev) {
+		case 0:
+			dev_info(wm8350->dev, "WM8352 Rev A\n");
+			wm8350->power.rev_g_coeff = 1;
+			break;
+
+		default:
+			dev_err(wm8350->dev, "Unknown WM8352 CHIP_REV\n");
+			ret = -ENODEV;
+			goto err;
+		}
+		break;
+
 	default:
 		dev_err(wm8350->dev, "Unknown MASK_REV\n");
 		ret = -ENODEV;
 		goto err;
 	}
 
-	ret = wm8350_create_cache(wm8350, mode);
+	ret = wm8350_create_cache(wm8350, mask_rev, mode);
 	if (ret < 0) {
 		dev_err(wm8350->dev, "Failed to create register cache\n");
 		return ret;
diff --git a/drivers/mfd/wm8350-i2c.c b/drivers/mfd/wm8350-i2c.c
index 876e693582bd..878051253084 100644
--- a/drivers/mfd/wm8350-i2c.c
+++ b/drivers/mfd/wm8350-i2c.c
@@ -97,6 +97,7 @@ static int wm8350_i2c_remove(struct i2c_client *i2c)
 
 static const struct i2c_device_id wm8350_i2c_id[] = {
        { "wm8350", 0 },
+       { "wm8352", 0 },
        { }
 };
 MODULE_DEVICE_TABLE(i2c, wm8350_i2c_id);
diff --git a/drivers/mfd/wm8350-regmap.c b/drivers/mfd/wm8350-regmap.c
index b43d64c2b920..3e2cc37961f3 100644
--- a/drivers/mfd/wm8350-regmap.c
+++ b/drivers/mfd/wm8350-regmap.c
@@ -1074,6 +1074,1058 @@ const u16 wm8350_mode3_defaults[] = {
 };
 #endif
 
+#ifdef CONFIG_MFD_WM8352_CONFIG_MODE_0
+
+#undef WM8350_HAVE_CONFIG_MODE
+#define WM8350_HAVE_CONFIG_MODE
+
+const u16 wm8352_mode0_defaults[] = {
+	0x6143,     /* R0   - Reset/ID */
+	0x0000,     /* R1   - ID */
+	0x0002,     /* R2   - Revision */
+	0x1C02,     /* R3   - System Control 1 */
+	0x0004,     /* R4   - System Control 2 */
+	0x0000,     /* R5   - System Hibernate */
+	0x8A00,     /* R6   - Interface Control */
+	0x0000,     /* R7 */
+	0x8000,     /* R8   - Power mgmt (1) */
+	0x0000,     /* R9   - Power mgmt (2) */
+	0x0000,     /* R10  - Power mgmt (3) */
+	0x2000,     /* R11  - Power mgmt (4) */
+	0x0E00,     /* R12  - Power mgmt (5) */
+	0x0000,     /* R13  - Power mgmt (6) */
+	0x0000,     /* R14  - Power mgmt (7) */
+	0x0000,     /* R15 */
+	0x0000,     /* R16  - RTC Seconds/Minutes */
+	0x0100,     /* R17  - RTC Hours/Day */
+	0x0101,     /* R18  - RTC Date/Month */
+	0x1400,     /* R19  - RTC Year */
+	0x0000,     /* R20  - Alarm Seconds/Minutes */
+	0x0000,     /* R21  - Alarm Hours/Day */
+	0x0000,     /* R22  - Alarm Date/Month */
+	0x0320,     /* R23  - RTC Time Control */
+	0x0000,     /* R24  - System Interrupts */
+	0x0000,     /* R25  - Interrupt Status 1 */
+	0x0000,     /* R26  - Interrupt Status 2 */
+	0x0000,     /* R27 */
+	0x0000,     /* R28  - Under Voltage Interrupt status */
+	0x0000,     /* R29  - Over Current Interrupt status */
+	0x0000,     /* R30  - GPIO Interrupt Status */
+	0x0000,     /* R31  - Comparator Interrupt Status */
+	0x3FFF,     /* R32  - System Interrupts Mask */
+	0x0000,     /* R33  - Interrupt Status 1 Mask */
+	0x0000,     /* R34  - Interrupt Status 2 Mask */
+	0x0000,     /* R35 */
+	0x0000,     /* R36  - Under Voltage Interrupt status Mask */
+	0x0000,     /* R37  - Over Current Interrupt status Mask */
+	0x0000,     /* R38  - GPIO Interrupt Status Mask */
+	0x0000,     /* R39  - Comparator Interrupt Status Mask */
+	0x0040,     /* R40  - Clock Control 1 */
+	0x0000,     /* R41  - Clock Control 2 */
+	0x3A00,     /* R42  - FLL Control 1 */
+	0x7086,     /* R43  - FLL Control 2 */
+	0xC226,     /* R44  - FLL Control 3 */
+	0x0000,     /* R45  - FLL Control 4 */
+	0x0000,     /* R46 */
+	0x0000,     /* R47 */
+	0x0000,     /* R48  - DAC Control */
+	0x0000,     /* R49 */
+	0x00C0,     /* R50  - DAC Digital Volume L */
+	0x00C0,     /* R51  - DAC Digital Volume R */
+	0x0000,     /* R52 */
+	0x0040,     /* R53  - DAC LR Rate */
+	0x0000,     /* R54  - DAC Clock Control */
+	0x0000,     /* R55 */
+	0x0000,     /* R56 */
+	0x0000,     /* R57 */
+	0x4000,     /* R58  - DAC Mute */
+	0x0000,     /* R59  - DAC Mute Volume */
+	0x0000,     /* R60  - DAC Side */
+	0x0000,     /* R61 */
+	0x0000,     /* R62 */
+	0x0000,     /* R63 */
+	0x8000,     /* R64  - ADC Control */
+	0x0000,     /* R65 */
+	0x00C0,     /* R66  - ADC Digital Volume L */
+	0x00C0,     /* R67  - ADC Digital Volume R */
+	0x0000,     /* R68  - ADC Divider */
+	0x0000,     /* R69 */
+	0x0040,     /* R70  - ADC LR Rate */
+	0x0000,     /* R71 */
+	0x0303,     /* R72  - Input Control */
+	0x0000,     /* R73  - IN3 Input Control */
+	0x0000,     /* R74  - Mic Bias Control */
+	0x0000,     /* R75 */
+	0x0000,     /* R76  - Output Control */
+	0x0000,     /* R77  - Jack Detect */
+	0x0000,     /* R78  - Anti Pop Control */
+	0x0000,     /* R79 */
+	0x0040,     /* R80  - Left Input Volume */
+	0x0040,     /* R81  - Right Input Volume */
+	0x0000,     /* R82 */
+	0x0000,     /* R83 */
+	0x0000,     /* R84 */
+	0x0000,     /* R85 */
+	0x0000,     /* R86 */
+	0x0000,     /* R87 */
+	0x0800,     /* R88  - Left Mixer Control */
+	0x1000,     /* R89  - Right Mixer Control */
+	0x0000,     /* R90 */
+	0x0000,     /* R91 */
+	0x0000,     /* R92  - OUT3 Mixer Control */
+	0x0000,     /* R93  - OUT4 Mixer Control */
+	0x0000,     /* R94 */
+	0x0000,     /* R95 */
+	0x0000,     /* R96  - Output Left Mixer Volume */
+	0x0000,     /* R97  - Output Right Mixer Volume */
+	0x0000,     /* R98  - Input Mixer Volume L */
+	0x0000,     /* R99  - Input Mixer Volume R */
+	0x0000,     /* R100 - Input Mixer Volume */
+	0x0000,     /* R101 */
+	0x0000,     /* R102 */
+	0x0000,     /* R103 */
+	0x00E4,     /* R104 - OUT1L Volume */
+	0x00E4,     /* R105 - OUT1R Volume */
+	0x00E4,     /* R106 - OUT2L Volume */
+	0x02E4,     /* R107 - OUT2R Volume */
+	0x0000,     /* R108 */
+	0x0000,     /* R109 */
+	0x0000,     /* R110 */
+	0x0000,     /* R111 - BEEP Volume */
+	0x0A00,     /* R112 - AI Formating */
+	0x0000,     /* R113 - ADC DAC COMP */
+	0x0020,     /* R114 - AI ADC Control */
+	0x0020,     /* R115 - AI DAC Control */
+	0x0000,     /* R116 */
+	0x0000,     /* R117 */
+	0x0000,     /* R118 */
+	0x0000,     /* R119 */
+	0x0000,     /* R120 */
+	0x0000,     /* R121 */
+	0x0000,     /* R122 */
+	0x0000,     /* R123 */
+	0x0000,     /* R124 */
+	0x0000,     /* R125 */
+	0x0000,     /* R126 */
+	0x0000,     /* R127 */
+	0x1FFF,     /* R128 - GPIO Debounce */
+	0x0000,     /* R129 - GPIO Pin pull up Control */
+	0x0000,     /* R130 - GPIO Pull down Control */
+	0x0000,     /* R131 - GPIO Interrupt Mode */
+	0x0000,     /* R132 */
+	0x0000,     /* R133 - GPIO Control */
+	0x0FFC,     /* R134 - GPIO Configuration (i/o) */
+	0x0FFC,     /* R135 - GPIO Pin Polarity / Type */
+	0x0000,     /* R136 */
+	0x0000,     /* R137 */
+	0x0000,     /* R138 */
+	0x0000,     /* R139 */
+	0x0013,     /* R140 - GPIO Function Select 1 */
+	0x0000,     /* R141 - GPIO Function Select 2 */
+	0x0000,     /* R142 - GPIO Function Select 3 */
+	0x0003,     /* R143 - GPIO Function Select 4 */
+	0x0000,     /* R144 - Digitiser Control (1) */
+	0x0002,     /* R145 - Digitiser Control (2) */
+	0x0000,     /* R146 */
+	0x0000,     /* R147 */
+	0x0000,     /* R148 */
+	0x0000,     /* R149 */
+	0x0000,     /* R150 */
+	0x0000,     /* R151 */
+	0x7000,     /* R152 - AUX1 Readback */
+	0x7000,     /* R153 - AUX2 Readback */
+	0x7000,     /* R154 - AUX3 Readback */
+	0x7000,     /* R155 - AUX4 Readback */
+	0x0000,     /* R156 - USB Voltage Readback */
+	0x0000,     /* R157 - LINE Voltage Readback */
+	0x0000,     /* R158 - BATT Voltage Readback */
+	0x0000,     /* R159 - Chip Temp Readback */
+	0x0000,     /* R160 */
+	0x0000,     /* R161 */
+	0x0000,     /* R162 */
+	0x0000,     /* R163 - Generic Comparator Control */
+	0x0000,     /* R164 - Generic comparator 1 */
+	0x0000,     /* R165 - Generic comparator 2 */
+	0x0000,     /* R166 - Generic comparator 3 */
+	0x0000,     /* R167 - Generic comparator 4 */
+	0xA00F,     /* R168 - Battery Charger Control 1 */
+	0x0B06,     /* R169 - Battery Charger Control 2 */
+	0x0000,     /* R170 - Battery Charger Control 3 */
+	0x0000,     /* R171 */
+	0x0000,     /* R172 - Current Sink Driver A */
+	0x0000,     /* R173 - CSA Flash control */
+	0x0000,     /* R174 - Current Sink Driver B */
+	0x0000,     /* R175 - CSB Flash control */
+	0x0000,     /* R176 - DCDC/LDO requested */
+	0x032D,     /* R177 - DCDC Active options */
+	0x0000,     /* R178 - DCDC Sleep options */
+	0x0025,     /* R179 - Power-check comparator */
+	0x000E,     /* R180 - DCDC1 Control */
+	0x0000,     /* R181 - DCDC1 Timeouts */
+	0x1006,     /* R182 - DCDC1 Low Power */
+	0x0018,     /* R183 - DCDC2 Control */
+	0x0000,     /* R184 - DCDC2 Timeouts */
+	0x0000,     /* R185 */
+	0x0000,     /* R186 - DCDC3 Control */
+	0x0000,     /* R187 - DCDC3 Timeouts */
+	0x0006,     /* R188 - DCDC3 Low Power */
+	0x0000,     /* R189 - DCDC4 Control */
+	0x0000,     /* R190 - DCDC4 Timeouts */
+	0x0006,     /* R191 - DCDC4 Low Power */
+	0x0008,     /* R192 - DCDC5 Control */
+	0x0000,     /* R193 - DCDC5 Timeouts */
+	0x0000,     /* R194 */
+	0x0000,     /* R195 - DCDC6 Control */
+	0x0000,     /* R196 - DCDC6 Timeouts */
+	0x0006,     /* R197 - DCDC6 Low Power */
+	0x0000,     /* R198 */
+	0x0003,     /* R199 - Limit Switch Control */
+	0x001C,     /* R200 - LDO1 Control */
+	0x0000,     /* R201 - LDO1 Timeouts */
+	0x001C,     /* R202 - LDO1 Low Power */
+	0x001B,     /* R203 - LDO2 Control */
+	0x0000,     /* R204 - LDO2 Timeouts */
+	0x001C,     /* R205 - LDO2 Low Power */
+	0x001B,     /* R206 - LDO3 Control */
+	0x0000,     /* R207 - LDO3 Timeouts */
+	0x001C,     /* R208 - LDO3 Low Power */
+	0x001B,     /* R209 - LDO4 Control */
+	0x0000,     /* R210 - LDO4 Timeouts */
+	0x001C,     /* R211 - LDO4 Low Power */
+	0x0000,     /* R212 */
+	0x0000,     /* R213 */
+	0x0000,     /* R214 */
+	0x0000,     /* R215 - VCC_FAULT Masks */
+	0x001F,     /* R216 - Main Bandgap Control */
+	0x0000,     /* R217 - OSC Control */
+	0x9000,     /* R218 - RTC Tick Control */
+	0x0000,     /* R219 - Security1 */
+	0x4000,     /* R220 */
+	0x0000,     /* R221 */
+	0x0000,     /* R222 */
+	0x0000,     /* R223 */
+	0x0000,     /* R224 - Signal overrides */
+	0x0000,     /* R225 - DCDC/LDO status */
+	0x0000,     /* R226 - Charger Overides/status */
+	0x0000,     /* R227 - misc overrides */
+	0x0000,     /* R228 - Supply overrides/status 1 */
+	0x0000,     /* R229 - Supply overrides/status 2 */
+	0xE000,     /* R230 - GPIO Pin Status */
+	0x0000,     /* R231 - comparotor overrides */
+	0x0000,     /* R232 */
+	0x0000,     /* R233 - State Machine status */
+	0x1200,     /* R234 */
+	0x0000,     /* R235 */
+	0x8000,     /* R236 */
+	0x0000,     /* R237 */
+	0x0000,     /* R238 */
+	0x0000,     /* R239 */
+	0x0003,     /* R240 */
+	0x0000,     /* R241 */
+	0x0000,     /* R242 */
+	0x0004,     /* R243 */
+	0x0300,     /* R244 */
+	0x0000,     /* R245 */
+	0x0200,     /* R246 */
+	0x0000,     /* R247 */
+	0x1000,     /* R248 - DCDC1 Test Controls */
+	0x5000,     /* R249 */
+	0x1000,     /* R250 - DCDC3 Test Controls */
+	0x1000,     /* R251 - DCDC4 Test Controls */
+	0x5100,     /* R252 */
+	0x1000,     /* R253 - DCDC6 Test Controls */
+};
+#endif
+
+#ifdef CONFIG_MFD_WM8352_CONFIG_MODE_1
+
+#undef WM8350_HAVE_CONFIG_MODE
+#define WM8350_HAVE_CONFIG_MODE
+
+const u16 wm8352_mode1_defaults[] = {
+	0x6143,     /* R0   - Reset/ID */
+	0x0000,     /* R1   - ID */
+	0x0002,     /* R2   - Revision */
+	0x1C02,     /* R3   - System Control 1 */
+	0x0204,     /* R4   - System Control 2 */
+	0x0000,     /* R5   - System Hibernate */
+	0x8A00,     /* R6   - Interface Control */
+	0x0000,     /* R7 */
+	0x8000,     /* R8   - Power mgmt (1) */
+	0x0000,     /* R9   - Power mgmt (2) */
+	0x0000,     /* R10  - Power mgmt (3) */
+	0x2000,     /* R11  - Power mgmt (4) */
+	0x0E00,     /* R12  - Power mgmt (5) */
+	0x0000,     /* R13  - Power mgmt (6) */
+	0x0000,     /* R14  - Power mgmt (7) */
+	0x0000,     /* R15 */
+	0x0000,     /* R16  - RTC Seconds/Minutes */
+	0x0100,     /* R17  - RTC Hours/Day */
+	0x0101,     /* R18  - RTC Date/Month */
+	0x1400,     /* R19  - RTC Year */
+	0x0000,     /* R20  - Alarm Seconds/Minutes */
+	0x0000,     /* R21  - Alarm Hours/Day */
+	0x0000,     /* R22  - Alarm Date/Month */
+	0x0320,     /* R23  - RTC Time Control */
+	0x0000,     /* R24  - System Interrupts */
+	0x0000,     /* R25  - Interrupt Status 1 */
+	0x0000,     /* R26  - Interrupt Status 2 */
+	0x0000,     /* R27 */
+	0x0000,     /* R28  - Under Voltage Interrupt status */
+	0x0000,     /* R29  - Over Current Interrupt status */
+	0x0000,     /* R30  - GPIO Interrupt Status */
+	0x0000,     /* R31  - Comparator Interrupt Status */
+	0x3FFF,     /* R32  - System Interrupts Mask */
+	0x0000,     /* R33  - Interrupt Status 1 Mask */
+	0x0000,     /* R34  - Interrupt Status 2 Mask */
+	0x0000,     /* R35 */
+	0x0000,     /* R36  - Under Voltage Interrupt status Mask */
+	0x0000,     /* R37  - Over Current Interrupt status Mask */
+	0x0000,     /* R38  - GPIO Interrupt Status Mask */
+	0x0000,     /* R39  - Comparator Interrupt Status Mask */
+	0x0040,     /* R40  - Clock Control 1 */
+	0x0000,     /* R41  - Clock Control 2 */
+	0x3A00,     /* R42  - FLL Control 1 */
+	0x7086,     /* R43  - FLL Control 2 */
+	0xC226,     /* R44  - FLL Control 3 */
+	0x0000,     /* R45  - FLL Control 4 */
+	0x0000,     /* R46 */
+	0x0000,     /* R47 */
+	0x0000,     /* R48  - DAC Control */
+	0x0000,     /* R49 */
+	0x00C0,     /* R50  - DAC Digital Volume L */
+	0x00C0,     /* R51  - DAC Digital Volume R */
+	0x0000,     /* R52 */
+	0x0040,     /* R53  - DAC LR Rate */
+	0x0000,     /* R54  - DAC Clock Control */
+	0x0000,     /* R55 */
+	0x0000,     /* R56 */
+	0x0000,     /* R57 */
+	0x4000,     /* R58  - DAC Mute */
+	0x0000,     /* R59  - DAC Mute Volume */
+	0x0000,     /* R60  - DAC Side */
+	0x0000,     /* R61 */
+	0x0000,     /* R62 */
+	0x0000,     /* R63 */
+	0x8000,     /* R64  - ADC Control */
+	0x0000,     /* R65 */
+	0x00C0,     /* R66  - ADC Digital Volume L */
+	0x00C0,     /* R67  - ADC Digital Volume R */
+	0x0000,     /* R68  - ADC Divider */
+	0x0000,     /* R69 */
+	0x0040,     /* R70  - ADC LR Rate */
+	0x0000,     /* R71 */
+	0x0303,     /* R72  - Input Control */
+	0x0000,     /* R73  - IN3 Input Control */
+	0x0000,     /* R74  - Mic Bias Control */
+	0x0000,     /* R75 */
+	0x0000,     /* R76  - Output Control */
+	0x0000,     /* R77  - Jack Detect */
+	0x0000,     /* R78  - Anti Pop Control */
+	0x0000,     /* R79 */
+	0x0040,     /* R80  - Left Input Volume */
+	0x0040,     /* R81  - Right Input Volume */
+	0x0000,     /* R82 */
+	0x0000,     /* R83 */
+	0x0000,     /* R84 */
+	0x0000,     /* R85 */
+	0x0000,     /* R86 */
+	0x0000,     /* R87 */
+	0x0800,     /* R88  - Left Mixer Control */
+	0x1000,     /* R89  - Right Mixer Control */
+	0x0000,     /* R90 */
+	0x0000,     /* R91 */
+	0x0000,     /* R92  - OUT3 Mixer Control */
+	0x0000,     /* R93  - OUT4 Mixer Control */
+	0x0000,     /* R94 */
+	0x0000,     /* R95 */
+	0x0000,     /* R96  - Output Left Mixer Volume */
+	0x0000,     /* R97  - Output Right Mixer Volume */
+	0x0000,     /* R98  - Input Mixer Volume L */
+	0x0000,     /* R99  - Input Mixer Volume R */
+	0x0000,     /* R100 - Input Mixer Volume */
+	0x0000,     /* R101 */
+	0x0000,     /* R102 */
+	0x0000,     /* R103 */
+	0x00E4,     /* R104 - OUT1L Volume */
+	0x00E4,     /* R105 - OUT1R Volume */
+	0x00E4,     /* R106 - OUT2L Volume */
+	0x02E4,     /* R107 - OUT2R Volume */
+	0x0000,     /* R108 */
+	0x0000,     /* R109 */
+	0x0000,     /* R110 */
+	0x0000,     /* R111 - BEEP Volume */
+	0x0A00,     /* R112 - AI Formating */
+	0x0000,     /* R113 - ADC DAC COMP */
+	0x0020,     /* R114 - AI ADC Control */
+	0x0020,     /* R115 - AI DAC Control */
+	0x0000,     /* R116 */
+	0x0000,     /* R117 */
+	0x0000,     /* R118 */
+	0x0000,     /* R119 */
+	0x0000,     /* R120 */
+	0x0000,     /* R121 */
+	0x0000,     /* R122 */
+	0x0000,     /* R123 */
+	0x0000,     /* R124 */
+	0x0000,     /* R125 */
+	0x0000,     /* R126 */
+	0x0000,     /* R127 */
+	0x1FFF,     /* R128 - GPIO Debounce */
+	0x0000,     /* R129 - GPIO Pin pull up Control */
+	0x0000,     /* R130 - GPIO Pull down Control */
+	0x0000,     /* R131 - GPIO Interrupt Mode */
+	0x0000,     /* R132 */
+	0x0000,     /* R133 - GPIO Control */
+	0x0BFB,     /* R134 - GPIO Configuration (i/o) */
+	0x0FFF,     /* R135 - GPIO Pin Polarity / Type */
+	0x0000,     /* R136 */
+	0x0000,     /* R137 */
+	0x0000,     /* R138 */
+	0x0000,     /* R139 */
+	0x0300,     /* R140 - GPIO Function Select 1 */
+	0x0000,     /* R141 - GPIO Function Select 2 */
+	0x2300,     /* R142 - GPIO Function Select 3 */
+	0x0003,     /* R143 - GPIO Function Select 4 */
+	0x0000,     /* R144 - Digitiser Control (1) */
+	0x0002,     /* R145 - Digitiser Control (2) */
+	0x0000,     /* R146 */
+	0x0000,     /* R147 */
+	0x0000,     /* R148 */
+	0x0000,     /* R149 */
+	0x0000,     /* R150 */
+	0x0000,     /* R151 */
+	0x7000,     /* R152 - AUX1 Readback */
+	0x7000,     /* R153 - AUX2 Readback */
+	0x7000,     /* R154 - AUX3 Readback */
+	0x7000,     /* R155 - AUX4 Readback */
+	0x0000,     /* R156 - USB Voltage Readback */
+	0x0000,     /* R157 - LINE Voltage Readback */
+	0x0000,     /* R158 - BATT Voltage Readback */
+	0x0000,     /* R159 - Chip Temp Readback */
+	0x0000,     /* R160 */
+	0x0000,     /* R161 */
+	0x0000,     /* R162 */
+	0x0000,     /* R163 - Generic Comparator Control */
+	0x0000,     /* R164 - Generic comparator 1 */
+	0x0000,     /* R165 - Generic comparator 2 */
+	0x0000,     /* R166 - Generic comparator 3 */
+	0x0000,     /* R167 - Generic comparator 4 */
+	0xA00F,     /* R168 - Battery Charger Control 1 */
+	0x0B06,     /* R169 - Battery Charger Control 2 */
+	0x0000,     /* R170 - Battery Charger Control 3 */
+	0x0000,     /* R171 */
+	0x0000,     /* R172 - Current Sink Driver A */
+	0x0000,     /* R173 - CSA Flash control */
+	0x0000,     /* R174 - Current Sink Driver B */
+	0x0000,     /* R175 - CSB Flash control */
+	0x0000,     /* R176 - DCDC/LDO requested */
+	0x032D,     /* R177 - DCDC Active options */
+	0x0000,     /* R178 - DCDC Sleep options */
+	0x0025,     /* R179 - Power-check comparator */
+	0x0062,     /* R180 - DCDC1 Control */
+	0x0400,     /* R181 - DCDC1 Timeouts */
+	0x1006,     /* R182 - DCDC1 Low Power */
+	0x0018,     /* R183 - DCDC2 Control */
+	0x0000,     /* R184 - DCDC2 Timeouts */
+	0x0000,     /* R185 */
+	0x0006,     /* R186 - DCDC3 Control */
+	0x0800,     /* R187 - DCDC3 Timeouts */
+	0x0006,     /* R188 - DCDC3 Low Power */
+	0x0006,     /* R189 - DCDC4 Control */
+	0x0C00,     /* R190 - DCDC4 Timeouts */
+	0x0006,     /* R191 - DCDC4 Low Power */
+	0x0008,     /* R192 - DCDC5 Control */
+	0x0000,     /* R193 - DCDC5 Timeouts */
+	0x0000,     /* R194 */
+	0x0026,     /* R195 - DCDC6 Control */
+	0x1000,     /* R196 - DCDC6 Timeouts */
+	0x0006,     /* R197 - DCDC6 Low Power */
+	0x0000,     /* R198 */
+	0x0003,     /* R199 - Limit Switch Control */
+	0x0002,     /* R200 - LDO1 Control */
+	0x0000,     /* R201 - LDO1 Timeouts */
+	0x001C,     /* R202 - LDO1 Low Power */
+	0x001A,     /* R203 - LDO2 Control */
+	0x0000,     /* R204 - LDO2 Timeouts */
+	0x001C,     /* R205 - LDO2 Low Power */
+	0x001F,     /* R206 - LDO3 Control */
+	0x0000,     /* R207 - LDO3 Timeouts */
+	0x001C,     /* R208 - LDO3 Low Power */
+	0x001F,     /* R209 - LDO4 Control */
+	0x0000,     /* R210 - LDO4 Timeouts */
+	0x001C,     /* R211 - LDO4 Low Power */
+	0x0000,     /* R212 */
+	0x0000,     /* R213 */
+	0x0000,     /* R214 */
+	0x0000,     /* R215 - VCC_FAULT Masks */
+	0x001F,     /* R216 - Main Bandgap Control */
+	0x0000,     /* R217 - OSC Control */
+	0x9000,     /* R218 - RTC Tick Control */
+	0x0000,     /* R219 - Security1 */
+	0x4000,     /* R220 */
+	0x0000,     /* R221 */
+	0x0000,     /* R222 */
+	0x0000,     /* R223 */
+	0x0000,     /* R224 - Signal overrides */
+	0x0000,     /* R225 - DCDC/LDO status */
+	0x0000,     /* R226 - Charger Overides/status */
+	0x0000,     /* R227 - misc overrides */
+	0x0000,     /* R228 - Supply overrides/status 1 */
+	0x0000,     /* R229 - Supply overrides/status 2 */
+	0xE000,     /* R230 - GPIO Pin Status */
+	0x0000,     /* R231 - comparotor overrides */
+	0x0000,     /* R232 */
+	0x0000,     /* R233 - State Machine status */
+	0x1200,     /* R234 */
+	0x0000,     /* R235 */
+	0x8000,     /* R236 */
+	0x0000,     /* R237 */
+	0x0000,     /* R238 */
+	0x0000,     /* R239 */
+	0x0003,     /* R240 */
+	0x0000,     /* R241 */
+	0x0000,     /* R242 */
+	0x0004,     /* R243 */
+	0x0300,     /* R244 */
+	0x0000,     /* R245 */
+	0x0200,     /* R246 */
+	0x0000,     /* R247 */
+	0x1000,     /* R248 - DCDC1 Test Controls */
+	0x5000,     /* R249 */
+	0x1000,     /* R250 - DCDC3 Test Controls */
+	0x1000,     /* R251 - DCDC4 Test Controls */
+	0x5100,     /* R252 */
+	0x1000,     /* R253 - DCDC6 Test Controls */
+};
+#endif
+
+#ifdef CONFIG_MFD_WM8352_CONFIG_MODE_2
+
+#undef WM8350_HAVE_CONFIG_MODE
+#define WM8350_HAVE_CONFIG_MODE
+
+const u16 wm8352_mode2_defaults[] = {
+	0x6143,     /* R0   - Reset/ID */
+	0x0000,     /* R1   - ID */
+	0x0002,     /* R2   - Revision */
+	0x1C02,     /* R3   - System Control 1 */
+	0x0204,     /* R4   - System Control 2 */
+	0x0000,     /* R5   - System Hibernate */
+	0x8A00,     /* R6   - Interface Control */
+	0x0000,     /* R7 */
+	0x8000,     /* R8   - Power mgmt (1) */
+	0x0000,     /* R9   - Power mgmt (2) */
+	0x0000,     /* R10  - Power mgmt (3) */
+	0x2000,     /* R11  - Power mgmt (4) */
+	0x0E00,     /* R12  - Power mgmt (5) */
+	0x0000,     /* R13  - Power mgmt (6) */
+	0x0000,     /* R14  - Power mgmt (7) */
+	0x0000,     /* R15 */
+	0x0000,     /* R16  - RTC Seconds/Minutes */
+	0x0100,     /* R17  - RTC Hours/Day */
+	0x0101,     /* R18  - RTC Date/Month */
+	0x1400,     /* R19  - RTC Year */
+	0x0000,     /* R20  - Alarm Seconds/Minutes */
+	0x0000,     /* R21  - Alarm Hours/Day */
+	0x0000,     /* R22  - Alarm Date/Month */
+	0x0320,     /* R23  - RTC Time Control */
+	0x0000,     /* R24  - System Interrupts */
+	0x0000,     /* R25  - Interrupt Status 1 */
+	0x0000,     /* R26  - Interrupt Status 2 */
+	0x0000,     /* R27 */
+	0x0000,     /* R28  - Under Voltage Interrupt status */
+	0x0000,     /* R29  - Over Current Interrupt status */
+	0x0000,     /* R30  - GPIO Interrupt Status */
+	0x0000,     /* R31  - Comparator Interrupt Status */
+	0x3FFF,     /* R32  - System Interrupts Mask */
+	0x0000,     /* R33  - Interrupt Status 1 Mask */
+	0x0000,     /* R34  - Interrupt Status 2 Mask */
+	0x0000,     /* R35 */
+	0x0000,     /* R36  - Under Voltage Interrupt status Mask */
+	0x0000,     /* R37  - Over Current Interrupt status Mask */
+	0x0000,     /* R38  - GPIO Interrupt Status Mask */
+	0x0000,     /* R39  - Comparator Interrupt Status Mask */
+	0x0040,     /* R40  - Clock Control 1 */
+	0x0000,     /* R41  - Clock Control 2 */
+	0x3A00,     /* R42  - FLL Control 1 */
+	0x7086,     /* R43  - FLL Control 2 */
+	0xC226,     /* R44  - FLL Control 3 */
+	0x0000,     /* R45  - FLL Control 4 */
+	0x0000,     /* R46 */
+	0x0000,     /* R47 */
+	0x0000,     /* R48  - DAC Control */
+	0x0000,     /* R49 */
+	0x00C0,     /* R50  - DAC Digital Volume L */
+	0x00C0,     /* R51  - DAC Digital Volume R */
+	0x0000,     /* R52 */
+	0x0040,     /* R53  - DAC LR Rate */
+	0x0000,     /* R54  - DAC Clock Control */
+	0x0000,     /* R55 */
+	0x0000,     /* R56 */
+	0x0000,     /* R57 */
+	0x4000,     /* R58  - DAC Mute */
+	0x0000,     /* R59  - DAC Mute Volume */
+	0x0000,     /* R60  - DAC Side */
+	0x0000,     /* R61 */
+	0x0000,     /* R62 */
+	0x0000,     /* R63 */
+	0x8000,     /* R64  - ADC Control */
+	0x0000,     /* R65 */
+	0x00C0,     /* R66  - ADC Digital Volume L */
+	0x00C0,     /* R67  - ADC Digital Volume R */
+	0x0000,     /* R68  - ADC Divider */
+	0x0000,     /* R69 */
+	0x0040,     /* R70  - ADC LR Rate */
+	0x0000,     /* R71 */
+	0x0303,     /* R72  - Input Control */
+	0x0000,     /* R73  - IN3 Input Control */
+	0x0000,     /* R74  - Mic Bias Control */
+	0x0000,     /* R75 */
+	0x0000,     /* R76  - Output Control */
+	0x0000,     /* R77  - Jack Detect */
+	0x0000,     /* R78  - Anti Pop Control */
+	0x0000,     /* R79 */
+	0x0040,     /* R80  - Left Input Volume */
+	0x0040,     /* R81  - Right Input Volume */
+	0x0000,     /* R82 */
+	0x0000,     /* R83 */
+	0x0000,     /* R84 */
+	0x0000,     /* R85 */
+	0x0000,     /* R86 */
+	0x0000,     /* R87 */
+	0x0800,     /* R88  - Left Mixer Control */
+	0x1000,     /* R89  - Right Mixer Control */
+	0x0000,     /* R90 */
+	0x0000,     /* R91 */
+	0x0000,     /* R92  - OUT3 Mixer Control */
+	0x0000,     /* R93  - OUT4 Mixer Control */
+	0x0000,     /* R94 */
+	0x0000,     /* R95 */
+	0x0000,     /* R96  - Output Left Mixer Volume */
+	0x0000,     /* R97  - Output Right Mixer Volume */
+	0x0000,     /* R98  - Input Mixer Volume L */
+	0x0000,     /* R99  - Input Mixer Volume R */
+	0x0000,     /* R100 - Input Mixer Volume */
+	0x0000,     /* R101 */
+	0x0000,     /* R102 */
+	0x0000,     /* R103 */
+	0x00E4,     /* R104 - OUT1L Volume */
+	0x00E4,     /* R105 - OUT1R Volume */
+	0x00E4,     /* R106 - OUT2L Volume */
+	0x02E4,     /* R107 - OUT2R Volume */
+	0x0000,     /* R108 */
+	0x0000,     /* R109 */
+	0x0000,     /* R110 */
+	0x0000,     /* R111 - BEEP Volume */
+	0x0A00,     /* R112 - AI Formating */
+	0x0000,     /* R113 - ADC DAC COMP */
+	0x0020,     /* R114 - AI ADC Control */
+	0x0020,     /* R115 - AI DAC Control */
+	0x0000,     /* R116 */
+	0x0000,     /* R117 */
+	0x0000,     /* R118 */
+	0x0000,     /* R119 */
+	0x0000,     /* R120 */
+	0x0000,     /* R121 */
+	0x0000,     /* R122 */
+	0x0000,     /* R123 */
+	0x0000,     /* R124 */
+	0x0000,     /* R125 */
+	0x0000,     /* R126 */
+	0x0000,     /* R127 */
+	0x1FFF,     /* R128 - GPIO Debounce */
+	0x0000,     /* R129 - GPIO Pin pull up Control */
+	0x0110,     /* R130 - GPIO Pull down Control */
+	0x0000,     /* R131 - GPIO Interrupt Mode */
+	0x0000,     /* R132 */
+	0x0000,     /* R133 - GPIO Control */
+	0x09DA,     /* R134 - GPIO Configuration (i/o) */
+	0x0DD6,     /* R135 - GPIO Pin Polarity / Type */
+	0x0000,     /* R136 */
+	0x0000,     /* R137 */
+	0x0000,     /* R138 */
+	0x0000,     /* R139 */
+	0x1310,     /* R140 - GPIO Function Select 1 */
+	0x0033,     /* R141 - GPIO Function Select 2 */
+	0x2000,     /* R142 - GPIO Function Select 3 */
+	0x0000,     /* R143 - GPIO Function Select 4 */
+	0x0000,     /* R144 - Digitiser Control (1) */
+	0x0002,     /* R145 - Digitiser Control (2) */
+	0x0000,     /* R146 */
+	0x0000,     /* R147 */
+	0x0000,     /* R148 */
+	0x0000,     /* R149 */
+	0x0000,     /* R150 */
+	0x0000,     /* R151 */
+	0x7000,     /* R152 - AUX1 Readback */
+	0x7000,     /* R153 - AUX2 Readback */
+	0x7000,     /* R154 - AUX3 Readback */
+	0x7000,     /* R155 - AUX4 Readback */
+	0x0000,     /* R156 - USB Voltage Readback */
+	0x0000,     /* R157 - LINE Voltage Readback */
+	0x0000,     /* R158 - BATT Voltage Readback */
+	0x0000,     /* R159 - Chip Temp Readback */
+	0x0000,     /* R160 */
+	0x0000,     /* R161 */
+	0x0000,     /* R162 */
+	0x0000,     /* R163 - Generic Comparator Control */
+	0x0000,     /* R164 - Generic comparator 1 */
+	0x0000,     /* R165 - Generic comparator 2 */
+	0x0000,     /* R166 - Generic comparator 3 */
+	0x0000,     /* R167 - Generic comparator 4 */
+	0xA00F,     /* R168 - Battery Charger Control 1 */
+	0x0B06,     /* R169 - Battery Charger Control 2 */
+	0x0000,     /* R170 - Battery Charger Control 3 */
+	0x0000,     /* R171 */
+	0x0000,     /* R172 - Current Sink Driver A */
+	0x0000,     /* R173 - CSA Flash control */
+	0x0000,     /* R174 - Current Sink Driver B */
+	0x0000,     /* R175 - CSB Flash control */
+	0x0000,     /* R176 - DCDC/LDO requested */
+	0x032D,     /* R177 - DCDC Active options */
+	0x0000,     /* R178 - DCDC Sleep options */
+	0x0025,     /* R179 - Power-check comparator */
+	0x000E,     /* R180 - DCDC1 Control */
+	0x0800,     /* R181 - DCDC1 Timeouts */
+	0x1006,     /* R182 - DCDC1 Low Power */
+	0x0018,     /* R183 - DCDC2 Control */
+	0x0000,     /* R184 - DCDC2 Timeouts */
+	0x0000,     /* R185 */
+	0x0056,     /* R186 - DCDC3 Control */
+	0x1800,     /* R187 - DCDC3 Timeouts */
+	0x0006,     /* R188 - DCDC3 Low Power */
+	0x000E,     /* R189 - DCDC4 Control */
+	0x1000,     /* R190 - DCDC4 Timeouts */
+	0x0006,     /* R191 - DCDC4 Low Power */
+	0x0008,     /* R192 - DCDC5 Control */
+	0x0000,     /* R193 - DCDC5 Timeouts */
+	0x0000,     /* R194 */
+	0x0026,     /* R195 - DCDC6 Control */
+	0x0C00,     /* R196 - DCDC6 Timeouts */
+	0x0006,     /* R197 - DCDC6 Low Power */
+	0x0000,     /* R198 */
+	0x0003,     /* R199 - Limit Switch Control */
+	0x001C,     /* R200 - LDO1 Control */
+	0x0000,     /* R201 - LDO1 Timeouts */
+	0x001C,     /* R202 - LDO1 Low Power */
+	0x0006,     /* R203 - LDO2 Control */
+	0x0400,     /* R204 - LDO2 Timeouts */
+	0x001C,     /* R205 - LDO2 Low Power */
+	0x001C,     /* R206 - LDO3 Control */
+	0x1400,     /* R207 - LDO3 Timeouts */
+	0x001C,     /* R208 - LDO3 Low Power */
+	0x001A,     /* R209 - LDO4 Control */
+	0x0000,     /* R210 - LDO4 Timeouts */
+	0x001C,     /* R211 - LDO4 Low Power */
+	0x0000,     /* R212 */
+	0x0000,     /* R213 */
+	0x0000,     /* R214 */
+	0x0000,     /* R215 - VCC_FAULT Masks */
+	0x001F,     /* R216 - Main Bandgap Control */
+	0x0000,     /* R217 - OSC Control */
+	0x9000,     /* R218 - RTC Tick Control */
+	0x0000,     /* R219 - Security1 */
+	0x4000,     /* R220 */
+	0x0000,     /* R221 */
+	0x0000,     /* R222 */
+	0x0000,     /* R223 */
+	0x0000,     /* R224 - Signal overrides */
+	0x0000,     /* R225 - DCDC/LDO status */
+	0x0000,     /* R226 - Charger Overides/status */
+	0x0000,     /* R227 - misc overrides */
+	0x0000,     /* R228 - Supply overrides/status 1 */
+	0x0000,     /* R229 - Supply overrides/status 2 */
+	0xE000,     /* R230 - GPIO Pin Status */
+	0x0000,     /* R231 - comparotor overrides */
+	0x0000,     /* R232 */
+	0x0000,     /* R233 - State Machine status */
+	0x1200,     /* R234 */
+	0x0000,     /* R235 */
+	0x8000,     /* R236 */
+	0x0000,     /* R237 */
+	0x0000,     /* R238 */
+	0x0000,     /* R239 */
+	0x0003,     /* R240 */
+	0x0000,     /* R241 */
+	0x0000,     /* R242 */
+	0x0004,     /* R243 */
+	0x0300,     /* R244 */
+	0x0000,     /* R245 */
+	0x0200,     /* R246 */
+	0x0000,     /* R247 */
+	0x1000,     /* R248 - DCDC1 Test Controls */
+	0x5000,     /* R249 */
+	0x1000,     /* R250 - DCDC3 Test Controls */
+	0x1000,     /* R251 - DCDC4 Test Controls */
+	0x5100,     /* R252 */
+	0x1000,     /* R253 - DCDC6 Test Controls */
+};
+#endif
+
+#ifdef CONFIG_MFD_WM8352_CONFIG_MODE_3
+
+#undef WM8350_HAVE_CONFIG_MODE
+#define WM8350_HAVE_CONFIG_MODE
+
+const u16 wm8352_mode3_defaults[] = {
+	0x6143,     /* R0   - Reset/ID */
+	0x0000,     /* R1   - ID */
+	0x0002,     /* R2   - Revision */
+	0x1C02,     /* R3   - System Control 1 */
+	0x0204,     /* R4   - System Control 2 */
+	0x0000,     /* R5   - System Hibernate */
+	0x8A00,     /* R6   - Interface Control */
+	0x0000,     /* R7 */
+	0x8000,     /* R8   - Power mgmt (1) */
+	0x0000,     /* R9   - Power mgmt (2) */
+	0x0000,     /* R10  - Power mgmt (3) */
+	0x2000,     /* R11  - Power mgmt (4) */
+	0x0E00,     /* R12  - Power mgmt (5) */
+	0x0000,     /* R13  - Power mgmt (6) */
+	0x0000,     /* R14  - Power mgmt (7) */
+	0x0000,     /* R15 */
+	0x0000,     /* R16  - RTC Seconds/Minutes */
+	0x0100,     /* R17  - RTC Hours/Day */
+	0x0101,     /* R18  - RTC Date/Month */
+	0x1400,     /* R19  - RTC Year */
+	0x0000,     /* R20  - Alarm Seconds/Minutes */
+	0x0000,     /* R21  - Alarm Hours/Day */
+	0x0000,     /* R22  - Alarm Date/Month */
+	0x0320,     /* R23  - RTC Time Control */
+	0x0000,     /* R24  - System Interrupts */
+	0x0000,     /* R25  - Interrupt Status 1 */
+	0x0000,     /* R26  - Interrupt Status 2 */
+	0x0000,     /* R27 */
+	0x0000,     /* R28  - Under Voltage Interrupt status */
+	0x0000,     /* R29  - Over Current Interrupt status */
+	0x0000,     /* R30  - GPIO Interrupt Status */
+	0x0000,     /* R31  - Comparator Interrupt Status */
+	0x3FFF,     /* R32  - System Interrupts Mask */
+	0x0000,     /* R33  - Interrupt Status 1 Mask */
+	0x0000,     /* R34  - Interrupt Status 2 Mask */
+	0x0000,     /* R35 */
+	0x0000,     /* R36  - Under Voltage Interrupt status Mask */
+	0x0000,     /* R37  - Over Current Interrupt status Mask */
+	0x0000,     /* R38  - GPIO Interrupt Status Mask */
+	0x0000,     /* R39  - Comparator Interrupt Status Mask */
+	0x0040,     /* R40  - Clock Control 1 */
+	0x0000,     /* R41  - Clock Control 2 */
+	0x3A00,     /* R42  - FLL Control 1 */
+	0x7086,     /* R43  - FLL Control 2 */
+	0xC226,     /* R44  - FLL Control 3 */
+	0x0000,     /* R45  - FLL Control 4 */
+	0x0000,     /* R46 */
+	0x0000,     /* R47 */
+	0x0000,     /* R48  - DAC Control */
+	0x0000,     /* R49 */
+	0x00C0,     /* R50  - DAC Digital Volume L */
+	0x00C0,     /* R51  - DAC Digital Volume R */
+	0x0000,     /* R52 */
+	0x0040,     /* R53  - DAC LR Rate */
+	0x0000,     /* R54  - DAC Clock Control */
+	0x0000,     /* R55 */
+	0x0000,     /* R56 */
+	0x0000,     /* R57 */
+	0x4000,     /* R58  - DAC Mute */
+	0x0000,     /* R59  - DAC Mute Volume */
+	0x0000,     /* R60  - DAC Side */
+	0x0000,     /* R61 */
+	0x0000,     /* R62 */
+	0x0000,     /* R63 */
+	0x8000,     /* R64  - ADC Control */
+	0x0000,     /* R65 */
+	0x00C0,     /* R66  - ADC Digital Volume L */
+	0x00C0,     /* R67  - ADC Digital Volume R */
+	0x0000,     /* R68  - ADC Divider */
+	0x0000,     /* R69 */
+	0x0040,     /* R70  - ADC LR Rate */
+	0x0000,     /* R71 */
+	0x0303,     /* R72  - Input Control */
+	0x0000,     /* R73  - IN3 Input Control */
+	0x0000,     /* R74  - Mic Bias Control */
+	0x0000,     /* R75 */
+	0x0000,     /* R76  - Output Control */
+	0x0000,     /* R77  - Jack Detect */
+	0x0000,     /* R78  - Anti Pop Control */
+	0x0000,     /* R79 */
+	0x0040,     /* R80  - Left Input Volume */
+	0x0040,     /* R81  - Right Input Volume */
+	0x0000,     /* R82 */
+	0x0000,     /* R83 */
+	0x0000,     /* R84 */
+	0x0000,     /* R85 */
+	0x0000,     /* R86 */
+	0x0000,     /* R87 */
+	0x0800,     /* R88  - Left Mixer Control */
+	0x1000,     /* R89  - Right Mixer Control */
+	0x0000,     /* R90 */
+	0x0000,     /* R91 */
+	0x0000,     /* R92  - OUT3 Mixer Control */
+	0x0000,     /* R93  - OUT4 Mixer Control */
+	0x0000,     /* R94 */
+	0x0000,     /* R95 */
+	0x0000,     /* R96  - Output Left Mixer Volume */
+	0x0000,     /* R97  - Output Right Mixer Volume */
+	0x0000,     /* R98  - Input Mixer Volume L */
+	0x0000,     /* R99  - Input Mixer Volume R */
+	0x0000,     /* R100 - Input Mixer Volume */
+	0x0000,     /* R101 */
+	0x0000,     /* R102 */
+	0x0000,     /* R103 */
+	0x00E4,     /* R104 - OUT1L Volume */
+	0x00E4,     /* R105 - OUT1R Volume */
+	0x00E4,     /* R106 - OUT2L Volume */
+	0x02E4,     /* R107 - OUT2R Volume */
+	0x0000,     /* R108 */
+	0x0000,     /* R109 */
+	0x0000,     /* R110 */
+	0x0000,     /* R111 - BEEP Volume */
+	0x0A00,     /* R112 - AI Formating */
+	0x0000,     /* R113 - ADC DAC COMP */
+	0x0020,     /* R114 - AI ADC Control */
+	0x0020,     /* R115 - AI DAC Control */
+	0x0000,     /* R116 */
+	0x0000,     /* R117 */
+	0x0000,     /* R118 */
+	0x0000,     /* R119 */
+	0x0000,     /* R120 */
+	0x0000,     /* R121 */
+	0x0000,     /* R122 */
+	0x0000,     /* R123 */
+	0x0000,     /* R124 */
+	0x0000,     /* R125 */
+	0x0000,     /* R126 */
+	0x0000,     /* R127 */
+	0x1FFF,     /* R128 - GPIO Debounce */
+	0x0010,     /* R129 - GPIO Pin pull up Control */
+	0x0000,     /* R130 - GPIO Pull down Control */
+	0x0000,     /* R131 - GPIO Interrupt Mode */
+	0x0000,     /* R132 */
+	0x0000,     /* R133 - GPIO Control */
+	0x0BFB,     /* R134 - GPIO Configuration (i/o) */
+	0x0FFD,     /* R135 - GPIO Pin Polarity / Type */
+	0x0000,     /* R136 */
+	0x0000,     /* R137 */
+	0x0000,     /* R138 */
+	0x0000,     /* R139 */
+	0x0310,     /* R140 - GPIO Function Select 1 */
+	0x0001,     /* R141 - GPIO Function Select 2 */
+	0x2300,     /* R142 - GPIO Function Select 3 */
+	0x0003,     /* R143 - GPIO Function Select 4 */
+	0x0000,     /* R144 - Digitiser Control (1) */
+	0x0002,     /* R145 - Digitiser Control (2) */
+	0x0000,     /* R146 */
+	0x0000,     /* R147 */
+	0x0000,     /* R148 */
+	0x0000,     /* R149 */
+	0x0000,     /* R150 */
+	0x0000,     /* R151 */
+	0x7000,     /* R152 - AUX1 Readback */
+	0x7000,     /* R153 - AUX2 Readback */
+	0x7000,     /* R154 - AUX3 Readback */
+	0x7000,     /* R155 - AUX4 Readback */
+	0x0000,     /* R156 - USB Voltage Readback */
+	0x0000,     /* R157 - LINE Voltage Readback */
+	0x0000,     /* R158 - BATT Voltage Readback */
+	0x0000,     /* R159 - Chip Temp Readback */
+	0x0000,     /* R160 */
+	0x0000,     /* R161 */
+	0x0000,     /* R162 */
+	0x0000,     /* R163 - Generic Comparator Control */
+	0x0000,     /* R164 - Generic comparator 1 */
+	0x0000,     /* R165 - Generic comparator 2 */
+	0x0000,     /* R166 - Generic comparator 3 */
+	0x0000,     /* R167 - Generic comparator 4 */
+	0xA00F,     /* R168 - Battery Charger Control 1 */
+	0x0B06,     /* R169 - Battery Charger Control 2 */
+	0x0000,     /* R170 - Battery Charger Control 3 */
+	0x0000,     /* R171 */
+	0x0000,     /* R172 - Current Sink Driver A */
+	0x0000,     /* R173 - CSA Flash control */
+	0x0000,     /* R174 - Current Sink Driver B */
+	0x0000,     /* R175 - CSB Flash control */
+	0x0000,     /* R176 - DCDC/LDO requested */
+	0x032D,     /* R177 - DCDC Active options */
+	0x0000,     /* R178 - DCDC Sleep options */
+	0x0025,     /* R179 - Power-check comparator */
+	0x0006,     /* R180 - DCDC1 Control */
+	0x0400,     /* R181 - DCDC1 Timeouts */
+	0x1006,     /* R182 - DCDC1 Low Power */
+	0x0018,     /* R183 - DCDC2 Control */
+	0x0000,     /* R184 - DCDC2 Timeouts */
+	0x0000,     /* R185 */
+	0x0050,     /* R186 - DCDC3 Control */
+	0x0C00,     /* R187 - DCDC3 Timeouts */
+	0x0006,     /* R188 - DCDC3 Low Power */
+	0x000E,     /* R189 - DCDC4 Control */
+	0x0400,     /* R190 - DCDC4 Timeouts */
+	0x0006,     /* R191 - DCDC4 Low Power */
+	0x0008,     /* R192 - DCDC5 Control */
+	0x0000,     /* R193 - DCDC5 Timeouts */
+	0x0000,     /* R194 */
+	0x0029,     /* R195 - DCDC6 Control */
+	0x0800,     /* R196 - DCDC6 Timeouts */
+	0x0006,     /* R197 - DCDC6 Low Power */
+	0x0000,     /* R198 */
+	0x0003,     /* R199 - Limit Switch Control */
+	0x001D,     /* R200 - LDO1 Control */
+	0x1000,     /* R201 - LDO1 Timeouts */
+	0x001C,     /* R202 - LDO1 Low Power */
+	0x0017,     /* R203 - LDO2 Control */
+	0x1000,     /* R204 - LDO2 Timeouts */
+	0x001C,     /* R205 - LDO2 Low Power */
+	0x0006,     /* R206 - LDO3 Control */
+	0x1000,     /* R207 - LDO3 Timeouts */
+	0x001C,     /* R208 - LDO3 Low Power */
+	0x0010,     /* R209 - LDO4 Control */
+	0x1000,     /* R210 - LDO4 Timeouts */
+	0x001C,     /* R211 - LDO4 Low Power */
+	0x0000,     /* R212 */
+	0x0000,     /* R213 */
+	0x0000,     /* R214 */
+	0x0000,     /* R215 - VCC_FAULT Masks */
+	0x001F,     /* R216 - Main Bandgap Control */
+	0x0000,     /* R217 - OSC Control */
+	0x9000,     /* R218 - RTC Tick Control */
+	0x0000,     /* R219 - Security1 */
+	0x4000,     /* R220 */
+	0x0000,     /* R221 */
+	0x0000,     /* R222 */
+	0x0000,     /* R223 */
+	0x0000,     /* R224 - Signal overrides */
+	0x0000,     /* R225 - DCDC/LDO status */
+	0x0000,     /* R226 - Charger Overides/status */
+	0x0000,     /* R227 - misc overrides */
+	0x0000,     /* R228 - Supply overrides/status 1 */
+	0x0000,     /* R229 - Supply overrides/status 2 */
+	0xE000,     /* R230 - GPIO Pin Status */
+	0x0000,     /* R231 - comparotor overrides */
+	0x0000,     /* R232 */
+	0x0000,     /* R233 - State Machine status */
+	0x1200,     /* R234 */
+	0x0000,     /* R235 */
+	0x8000,     /* R236 */
+	0x0000,     /* R237 */
+	0x0000,     /* R238 */
+	0x0000,     /* R239 */
+	0x0003,     /* R240 */
+	0x0000,     /* R241 */
+	0x0000,     /* R242 */
+	0x0004,     /* R243 */
+	0x0300,     /* R244 */
+	0x0000,     /* R245 */
+	0x0200,     /* R246 */
+	0x0000,     /* R247 */
+	0x1000,     /* R248 - DCDC1 Test Controls */
+	0x5000,     /* R249 */
+	0x1000,     /* R250 - DCDC3 Test Controls */
+	0x1000,     /* R251 - DCDC4 Test Controls */
+	0x5100,     /* R252 */
+	0x1000,     /* R253 - DCDC6 Test Controls */
+};
+#endif
+
 /* The register defaults for the config mode used must be compiled in but
  * due to the impact on kernel size it is possible to disable
  */
diff --git a/include/linux/mfd/wm8350/core.h b/include/linux/mfd/wm8350/core.h
index afeff6f1316c..737579086d01 100644
--- a/include/linux/mfd/wm8350/core.h
+++ b/include/linux/mfd/wm8350/core.h
@@ -589,6 +589,10 @@ extern const u16 wm8350_mode0_defaults[];
 extern const u16 wm8350_mode1_defaults[];
 extern const u16 wm8350_mode2_defaults[];
 extern const u16 wm8350_mode3_defaults[];
+extern const u16 wm8352_mode0_defaults[];
+extern const u16 wm8352_mode1_defaults[];
+extern const u16 wm8352_mode2_defaults[];
+extern const u16 wm8352_mode3_defaults[];
 
 struct wm8350;
 
-- 
cgit v1.2.3


From 645524a9c6e1e42dc4fe03217befb20e2fc4d43e Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Thu, 18 Dec 2008 23:12:16 +0100
Subject: mfd: Support configurable numbers of DCDCs and ISINKs on WM8350

Some WM8350 variants have fewer DCDCs and ISINKs. Identify these at
probe and refuse to use the absent DCDCs when running on these chips.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/wm8350-core.c            | 6 ++++++
 drivers/regulator/wm8350-regulator.c | 7 +++++++
 include/linux/mfd/wm8350/pmic.h      | 4 ++++
 3 files changed, 17 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/mfd/wm8350-core.c b/drivers/mfd/wm8350-core.c
index 03af3b12c020..56c363c240a9 100644
--- a/drivers/mfd/wm8350-core.c
+++ b/drivers/mfd/wm8350-core.c
@@ -1301,6 +1301,9 @@ int wm8350_device_init(struct wm8350 *wm8350, int irq,
 
 	switch (mask_rev) {
 	case 0:
+		wm8350->pmic.max_dcdc = WM8350_DCDC_6;
+		wm8350->pmic.max_isink = WM8350_ISINK_B;
+
 		switch (chip_rev) {
 		case WM8350_REV_E:
 			dev_info(wm8350->dev, "WM8350 Rev E\n");
@@ -1325,6 +1328,9 @@ int wm8350_device_init(struct wm8350 *wm8350, int irq,
 		break;
 
 	case 2:
+		wm8350->pmic.max_dcdc = WM8350_DCDC_6;
+		wm8350->pmic.max_isink = WM8350_ISINK_B;
+
 		switch (chip_rev) {
 		case 0:
 			dev_info(wm8350->dev, "WM8352 Rev A\n");
diff --git a/drivers/regulator/wm8350-regulator.c b/drivers/regulator/wm8350-regulator.c
index 1f44b17e23b1..c68c496b2c49 100644
--- a/drivers/regulator/wm8350-regulator.c
+++ b/drivers/regulator/wm8350-regulator.c
@@ -1380,6 +1380,13 @@ int wm8350_register_regulator(struct wm8350 *wm8350, int reg,
 	if (wm8350->pmic.pdev[reg])
 		return -EBUSY;
 
+	if (reg >= WM8350_DCDC_1 && reg <= WM8350_DCDC_6 &&
+	    reg > wm8350->pmic.max_dcdc)
+		return -ENODEV;
+	if (reg >= WM8350_ISINK_A && reg <= WM8350_ISINK_B &&
+	    reg > wm8350->pmic.max_isink)
+		return -ENODEV;
+
 	pdev = platform_device_alloc("wm8350-regulator", reg);
 	if (!pdev)
 		return -ENOMEM;
diff --git a/include/linux/mfd/wm8350/pmic.h b/include/linux/mfd/wm8350/pmic.h
index 69b69e07f62f..96acbfc8aa12 100644
--- a/include/linux/mfd/wm8350/pmic.h
+++ b/include/linux/mfd/wm8350/pmic.h
@@ -701,6 +701,10 @@ struct platform_device;
 struct regulator_init_data;
 
 struct wm8350_pmic {
+	/* Number of regulators of each type on this device */
+	int max_dcdc;
+	int max_isink;
+
 	/* ISINK to DCDC mapping */
 	int isink_A_dcdc;
 	int isink_B_dcdc;
-- 
cgit v1.2.3


From ca23f8c1b0aa15dc69565244fc5dffa67a72dd02 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Thu, 18 Dec 2008 23:12:28 +0100
Subject: mfd: Add WM8351 support

The WM8351 is a WM8350 variant. As well as register default changes the
WM8351 has fewer voltage and current regulators than the WM8350.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/Kconfig             |   16 +
 drivers/mfd/wm8350-core.c       |   47 ++
 drivers/mfd/wm8350-i2c.c        |    1 +
 drivers/mfd/wm8350-regmap.c     | 1044 +++++++++++++++++++++++++++++++++++++++
 include/linux/mfd/wm8350/core.h |    4 +
 5 files changed, 1112 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index 76a482dfc177..781a27955ff5 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -143,6 +143,22 @@ config MFD_WM8350_CONFIG_MODE_3
 	bool
 	depends on MFD_WM8350
 
+config MFD_WM8351_CONFIG_MODE_0
+	bool
+	depends on MFD_WM8350
+
+config MFD_WM8351_CONFIG_MODE_1
+	bool
+	depends on MFD_WM8350
+
+config MFD_WM8351_CONFIG_MODE_2
+	bool
+	depends on MFD_WM8350
+
+config MFD_WM8351_CONFIG_MODE_3
+	bool
+	depends on MFD_WM8350
+
 config MFD_WM8352_CONFIG_MODE_0
 	bool
 	depends on MFD_WM8350
diff --git a/drivers/mfd/wm8350-core.c b/drivers/mfd/wm8350-core.c
index 56c363c240a9..e03fe60b55bf 100644
--- a/drivers/mfd/wm8350-core.c
+++ b/drivers/mfd/wm8350-core.c
@@ -1169,6 +1169,36 @@ static int wm8350_create_cache(struct wm8350 *wm8350, int type, int mode)
 			return -EINVAL;
 		}
 
+	case 1:
+		switch (mode) {
+#ifdef CONFIG_MFD_WM8351_CONFIG_MODE_0
+		case 0:
+			reg_map = wm8351_mode0_defaults;
+			break;
+#endif
+#ifdef CONFIG_MFD_WM8351_CONFIG_MODE_1
+		case 1:
+			reg_map = wm8351_mode1_defaults;
+			break;
+#endif
+#ifdef CONFIG_MFD_WM8351_CONFIG_MODE_2
+		case 2:
+			reg_map = wm8351_mode2_defaults;
+			break;
+#endif
+#ifdef CONFIG_MFD_WM8351_CONFIG_MODE_3
+		case 3:
+			reg_map = wm8351_mode3_defaults;
+			break;
+#endif
+		default:
+			dev_err(wm8350->dev,
+				"WM8351 configuration mode %d not supported\n",
+				mode);
+			return -EINVAL;
+		}
+		break;
+
 	case 2:
 		switch (mode) {
 #ifdef CONFIG_MFD_WM8352_CONFIG_MODE_0
@@ -1327,6 +1357,23 @@ int wm8350_device_init(struct wm8350 *wm8350, int irq,
 		}
 		break;
 
+	case 1:
+		wm8350->pmic.max_dcdc = WM8350_DCDC_4;
+		wm8350->pmic.max_isink = WM8350_ISINK_A;
+
+		switch (chip_rev) {
+		case 0:
+			dev_info(wm8350->dev, "WM8351 Rev A\n");
+			wm8350->power.rev_g_coeff = 1;
+			break;
+
+		default:
+			dev_err(wm8350->dev, "Unknown WM8351 CHIP_REV\n");
+			ret = -ENODEV;
+			goto err;
+		}
+		break;
+
 	case 2:
 		wm8350->pmic.max_dcdc = WM8350_DCDC_6;
 		wm8350->pmic.max_isink = WM8350_ISINK_B;
diff --git a/drivers/mfd/wm8350-i2c.c b/drivers/mfd/wm8350-i2c.c
index 878051253084..8d8c93217572 100644
--- a/drivers/mfd/wm8350-i2c.c
+++ b/drivers/mfd/wm8350-i2c.c
@@ -97,6 +97,7 @@ static int wm8350_i2c_remove(struct i2c_client *i2c)
 
 static const struct i2c_device_id wm8350_i2c_id[] = {
        { "wm8350", 0 },
+       { "wm8351", 0 },
        { "wm8352", 0 },
        { }
 };
diff --git a/drivers/mfd/wm8350-regmap.c b/drivers/mfd/wm8350-regmap.c
index 3e2cc37961f3..68887b817d17 100644
--- a/drivers/mfd/wm8350-regmap.c
+++ b/drivers/mfd/wm8350-regmap.c
@@ -1074,6 +1074,1050 @@ const u16 wm8350_mode3_defaults[] = {
 };
 #endif
 
+#ifdef CONFIG_MFD_WM8351_CONFIG_MODE_0
+
+#undef WM8350_HAVE_CONFIG_MODE
+#define WM8350_HAVE_CONFIG_MODE
+
+const u16 wm8351_mode0_defaults[] = {
+	0x6143,     /* R0   - Reset/ID */
+	0x0000,     /* R1   - ID */
+	0x0001,     /* R2   - Revision */
+	0x1C02,     /* R3   - System Control 1 */
+	0x0004,     /* R4   - System Control 2 */
+	0x0000,     /* R5   - System Hibernate */
+	0x8A00,     /* R6   - Interface Control */
+	0x0000,     /* R7 */
+	0x8000,     /* R8   - Power mgmt (1) */
+	0x0000,     /* R9   - Power mgmt (2) */
+	0x0000,     /* R10  - Power mgmt (3) */
+	0x2000,     /* R11  - Power mgmt (4) */
+	0x0E00,     /* R12  - Power mgmt (5) */
+	0x0000,     /* R13  - Power mgmt (6) */
+	0x0000,     /* R14  - Power mgmt (7) */
+	0x0000,     /* R15 */
+	0x0000,     /* R16  - RTC Seconds/Minutes */
+	0x0100,     /* R17  - RTC Hours/Day */
+	0x0101,     /* R18  - RTC Date/Month */
+	0x1400,     /* R19  - RTC Year */
+	0x0000,     /* R20  - Alarm Seconds/Minutes */
+	0x0000,     /* R21  - Alarm Hours/Day */
+	0x0000,     /* R22  - Alarm Date/Month */
+	0x0320,     /* R23  - RTC Time Control */
+	0x0000,     /* R24  - System Interrupts */
+	0x0000,     /* R25  - Interrupt Status 1 */
+	0x0000,     /* R26  - Interrupt Status 2 */
+	0x0000,     /* R27 */
+	0x0000,     /* R28  - Under Voltage Interrupt status */
+	0x0000,     /* R29  - Over Current Interrupt status */
+	0x0000,     /* R30  - GPIO Interrupt Status */
+	0x0000,     /* R31  - Comparator Interrupt Status */
+	0x3FFF,     /* R32  - System Interrupts Mask */
+	0x0000,     /* R33  - Interrupt Status 1 Mask */
+	0x0000,     /* R34  - Interrupt Status 2 Mask */
+	0x0000,     /* R35 */
+	0x0000,     /* R36  - Under Voltage Interrupt status Mask */
+	0x0000,     /* R37  - Over Current Interrupt status Mask */
+	0x0000,     /* R38  - GPIO Interrupt Status Mask */
+	0x0000,     /* R39  - Comparator Interrupt Status Mask */
+	0x0040,     /* R40  - Clock Control 1 */
+	0x0000,     /* R41  - Clock Control 2 */
+	0x3A00,     /* R42  - FLL Control 1 */
+	0x7086,     /* R43  - FLL Control 2 */
+	0xC226,     /* R44  - FLL Control 3 */
+	0x0000,     /* R45  - FLL Control 4 */
+	0x0000,     /* R46 */
+	0x0000,     /* R47 */
+	0x0000,     /* R48  - DAC Control */
+	0x0000,     /* R49 */
+	0x00C0,     /* R50  - DAC Digital Volume L */
+	0x00C0,     /* R51  - DAC Digital Volume R */
+	0x0000,     /* R52 */
+	0x0040,     /* R53  - DAC LR Rate */
+	0x0000,     /* R54  - DAC Clock Control */
+	0x0000,     /* R55 */
+	0x0000,     /* R56 */
+	0x0000,     /* R57 */
+	0x4000,     /* R58  - DAC Mute */
+	0x0000,     /* R59  - DAC Mute Volume */
+	0x0000,     /* R60  - DAC Side */
+	0x0000,     /* R61 */
+	0x0000,     /* R62 */
+	0x0000,     /* R63 */
+	0x8000,     /* R64  - ADC Control */
+	0x0000,     /* R65 */
+	0x00C0,     /* R66  - ADC Digital Volume L */
+	0x00C0,     /* R67  - ADC Digital Volume R */
+	0x0000,     /* R68  - ADC Divider */
+	0x0000,     /* R69 */
+	0x0040,     /* R70  - ADC LR Rate */
+	0x0000,     /* R71 */
+	0x0303,     /* R72  - Input Control */
+	0x0000,     /* R73  - IN3 Input Control */
+	0x0000,     /* R74  - Mic Bias Control */
+	0x0000,     /* R75 */
+	0x0000,     /* R76  - Output Control */
+	0x0000,     /* R77  - Jack Detect */
+	0x0000,     /* R78  - Anti Pop Control */
+	0x0000,     /* R79 */
+	0x0040,     /* R80  - Left Input Volume */
+	0x0040,     /* R81  - Right Input Volume */
+	0x0000,     /* R82 */
+	0x0000,     /* R83 */
+	0x0000,     /* R84 */
+	0x0000,     /* R85 */
+	0x0000,     /* R86 */
+	0x0000,     /* R87 */
+	0x0800,     /* R88  - Left Mixer Control */
+	0x1000,     /* R89  - Right Mixer Control */
+	0x0000,     /* R90 */
+	0x0000,     /* R91 */
+	0x0000,     /* R92  - OUT3 Mixer Control */
+	0x0000,     /* R93  - OUT4 Mixer Control */
+	0x0000,     /* R94 */
+	0x0000,     /* R95 */
+	0x0000,     /* R96  - Output Left Mixer Volume */
+	0x0000,     /* R97  - Output Right Mixer Volume */
+	0x0000,     /* R98  - Input Mixer Volume L */
+	0x0000,     /* R99  - Input Mixer Volume R */
+	0x0000,     /* R100 - Input Mixer Volume */
+	0x0000,     /* R101 */
+	0x0000,     /* R102 */
+	0x0000,     /* R103 */
+	0x00E4,     /* R104 - OUT1L Volume */
+	0x00E4,     /* R105 - OUT1R Volume */
+	0x00E4,     /* R106 - OUT2L Volume */
+	0x02E4,     /* R107 - OUT2R Volume */
+	0x0000,     /* R108 */
+	0x0000,     /* R109 */
+	0x0000,     /* R110 */
+	0x0000,     /* R111 - BEEP Volume */
+	0x0A00,     /* R112 - AI Formating */
+	0x0000,     /* R113 - ADC DAC COMP */
+	0x0020,     /* R114 - AI ADC Control */
+	0x0020,     /* R115 - AI DAC Control */
+	0x0000,     /* R116 */
+	0x0000,     /* R117 */
+	0x0000,     /* R118 */
+	0x0000,     /* R119 */
+	0x0000,     /* R120 */
+	0x0000,     /* R121 */
+	0x0000,     /* R122 */
+	0x0000,     /* R123 */
+	0x0000,     /* R124 */
+	0x0000,     /* R125 */
+	0x0000,     /* R126 */
+	0x0000,     /* R127 */
+	0x1FFF,     /* R128 - GPIO Debounce */
+	0x0000,     /* R129 - GPIO Pin pull up Control */
+	0x0000,     /* R130 - GPIO Pull down Control */
+	0x0000,     /* R131 - GPIO Interrupt Mode */
+	0x0000,     /* R132 */
+	0x0000,     /* R133 - GPIO Control */
+	0x0FFC,     /* R134 - GPIO Configuration (i/o) */
+	0x0FFC,     /* R135 - GPIO Pin Polarity / Type */
+	0x0000,     /* R136 */
+	0x0000,     /* R137 */
+	0x0000,     /* R138 */
+	0x0000,     /* R139 */
+	0x0013,     /* R140 - GPIO Function Select 1 */
+	0x0000,     /* R141 - GPIO Function Select 2 */
+	0x0000,     /* R142 - GPIO Function Select 3 */
+	0x0003,     /* R143 - GPIO Function Select 4 */
+	0x0000,     /* R144 - Digitiser Control (1) */
+	0x0002,     /* R145 - Digitiser Control (2) */
+	0x0000,     /* R146 */
+	0x0000,     /* R147 */
+	0x0000,     /* R148 */
+	0x0000,     /* R149 */
+	0x0000,     /* R150 */
+	0x0000,     /* R151 */
+	0x7000,     /* R152 - AUX1 Readback */
+	0x7000,     /* R153 - AUX2 Readback */
+	0x7000,     /* R154 - AUX3 Readback */
+	0x7000,     /* R155 - AUX4 Readback */
+	0x0000,     /* R156 - USB Voltage Readback */
+	0x0000,     /* R157 - LINE Voltage Readback */
+	0x0000,     /* R158 - BATT Voltage Readback */
+	0x0000,     /* R159 - Chip Temp Readback */
+	0x0000,     /* R160 */
+	0x0000,     /* R161 */
+	0x0000,     /* R162 */
+	0x0000,     /* R163 - Generic Comparator Control */
+	0x0000,     /* R164 - Generic comparator 1 */
+	0x0000,     /* R165 - Generic comparator 2 */
+	0x0000,     /* R166 - Generic comparator 3 */
+	0x0000,     /* R167 - Generic comparator 4 */
+	0xA00F,     /* R168 - Battery Charger Control 1 */
+	0x0B06,     /* R169 - Battery Charger Control 2 */
+	0x0000,     /* R170 - Battery Charger Control 3 */
+	0x0000,     /* R171 */
+	0x0000,     /* R172 - Current Sink Driver A */
+	0x0000,     /* R173 - CSA Flash control */
+	0x0000,     /* R174 */
+	0x0000,     /* R175 */
+	0x0000,     /* R176 - DCDC/LDO requested */
+	0x032D,     /* R177 - DCDC Active options */
+	0x0000,     /* R178 - DCDC Sleep options */
+	0x0025,     /* R179 - Power-check comparator */
+	0x000E,     /* R180 - DCDC1 Control */
+	0x0000,     /* R181 - DCDC1 Timeouts */
+	0x1006,     /* R182 - DCDC1 Low Power */
+	0x0018,     /* R183 - DCDC2 Control */
+	0x0000,     /* R184 - DCDC2 Timeouts */
+	0x0000,     /* R185 */
+	0x0000,     /* R186 - DCDC3 Control */
+	0x0000,     /* R187 - DCDC3 Timeouts */
+	0x0006,     /* R188 - DCDC3 Low Power */
+	0x0000,     /* R189 - DCDC4 Control */
+	0x0000,     /* R190 - DCDC4 Timeouts */
+	0x0006,     /* R191 - DCDC4 Low Power */
+	0x0008,     /* R192 */
+	0x0000,     /* R193 */
+	0x0000,     /* R194 */
+	0x0000,     /* R195 */
+	0x0000,     /* R196 */
+	0x0006,     /* R197 */
+	0x0000,     /* R198 */
+	0x0003,     /* R199 - Limit Switch Control */
+	0x001C,     /* R200 - LDO1 Control */
+	0x0000,     /* R201 - LDO1 Timeouts */
+	0x001C,     /* R202 - LDO1 Low Power */
+	0x001B,     /* R203 - LDO2 Control */
+	0x0000,     /* R204 - LDO2 Timeouts */
+	0x001C,     /* R205 - LDO2 Low Power */
+	0x001B,     /* R206 - LDO3 Control */
+	0x0000,     /* R207 - LDO3 Timeouts */
+	0x001C,     /* R208 - LDO3 Low Power */
+	0x001B,     /* R209 - LDO4 Control */
+	0x0000,     /* R210 - LDO4 Timeouts */
+	0x001C,     /* R211 - LDO4 Low Power */
+	0x0000,     /* R212 */
+	0x0000,     /* R213 */
+	0x0000,     /* R214 */
+	0x0000,     /* R215 - VCC_FAULT Masks */
+	0x001F,     /* R216 - Main Bandgap Control */
+	0x0000,     /* R217 - OSC Control */
+	0x9000,     /* R218 - RTC Tick Control */
+	0x0000,     /* R219 - Security1 */
+	0x4000,     /* R220 */
+	0x0000,     /* R221 */
+	0x0000,     /* R222 */
+	0x0000,     /* R223 */
+	0x0000,     /* R224 - Signal overrides */
+	0x0000,     /* R225 - DCDC/LDO status */
+	0x0000,     /* R226 - Charger Overides/status */
+	0x0000,     /* R227 - misc overrides */
+	0x0000,     /* R228 - Supply overrides/status 1 */
+	0x0000,     /* R229 - Supply overrides/status 2 */
+	0xE000,     /* R230 - GPIO Pin Status */
+	0x0000,     /* R231 - comparotor overrides */
+	0x0000,     /* R232 */
+	0x0000,     /* R233 - State Machine status */
+	0x1200,     /* R234 - FLL Test 1 */
+	0x0000,     /* R235 */
+	0x8000,     /* R236 */
+	0x0000,     /* R237 */
+	0x0000,     /* R238 */
+	0x0000,     /* R239 */
+	0x0003,     /* R240 */
+	0x0000,     /* R241 */
+	0x0000,     /* R242 */
+	0x0004,     /* R243 */
+	0x0300,     /* R244 */
+	0x0000,     /* R245 */
+	0x0200,     /* R246 */
+	0x0000,     /* R247 */
+	0x1000,     /* R248 - DCDC1 Test Controls */
+	0x1000,     /* R249 */
+	0x1000,     /* R250 - DCDC3 Test Controls */
+	0x1000,     /* R251 - DCDC4 Test Controls */
+};
+#endif
+
+#ifdef CONFIG_MFD_WM8351_CONFIG_MODE_1
+
+#undef WM8350_HAVE_CONFIG_MODE
+#define WM8350_HAVE_CONFIG_MODE
+
+const u16 wm8351_mode1_defaults[] = {
+	0x6143,     /* R0   - Reset/ID */
+	0x0000,     /* R1   - ID */
+	0x0001,     /* R2   - Revision */
+	0x1C02,     /* R3   - System Control 1 */
+	0x0204,     /* R4   - System Control 2 */
+	0x0000,     /* R5   - System Hibernate */
+	0x8A00,     /* R6   - Interface Control */
+	0x0000,     /* R7 */
+	0x8000,     /* R8   - Power mgmt (1) */
+	0x0000,     /* R9   - Power mgmt (2) */
+	0x0000,     /* R10  - Power mgmt (3) */
+	0x2000,     /* R11  - Power mgmt (4) */
+	0x0E00,     /* R12  - Power mgmt (5) */
+	0x0000,     /* R13  - Power mgmt (6) */
+	0x0000,     /* R14  - Power mgmt (7) */
+	0x0000,     /* R15 */
+	0x0000,     /* R16  - RTC Seconds/Minutes */
+	0x0100,     /* R17  - RTC Hours/Day */
+	0x0101,     /* R18  - RTC Date/Month */
+	0x1400,     /* R19  - RTC Year */
+	0x0000,     /* R20  - Alarm Seconds/Minutes */
+	0x0000,     /* R21  - Alarm Hours/Day */
+	0x0000,     /* R22  - Alarm Date/Month */
+	0x0320,     /* R23  - RTC Time Control */
+	0x0000,     /* R24  - System Interrupts */
+	0x0000,     /* R25  - Interrupt Status 1 */
+	0x0000,     /* R26  - Interrupt Status 2 */
+	0x0000,     /* R27 */
+	0x0000,     /* R28  - Under Voltage Interrupt status */
+	0x0000,     /* R29  - Over Current Interrupt status */
+	0x0000,     /* R30  - GPIO Interrupt Status */
+	0x0000,     /* R31  - Comparator Interrupt Status */
+	0x3FFF,     /* R32  - System Interrupts Mask */
+	0x0000,     /* R33  - Interrupt Status 1 Mask */
+	0x0000,     /* R34  - Interrupt Status 2 Mask */
+	0x0000,     /* R35 */
+	0x0000,     /* R36  - Under Voltage Interrupt status Mask */
+	0x0000,     /* R37  - Over Current Interrupt status Mask */
+	0x0000,     /* R38  - GPIO Interrupt Status Mask */
+	0x0000,     /* R39  - Comparator Interrupt Status Mask */
+	0x0040,     /* R40  - Clock Control 1 */
+	0x0000,     /* R41  - Clock Control 2 */
+	0x3A00,     /* R42  - FLL Control 1 */
+	0x7086,     /* R43  - FLL Control 2 */
+	0xC226,     /* R44  - FLL Control 3 */
+	0x0000,     /* R45  - FLL Control 4 */
+	0x0000,     /* R46 */
+	0x0000,     /* R47 */
+	0x0000,     /* R48  - DAC Control */
+	0x0000,     /* R49 */
+	0x00C0,     /* R50  - DAC Digital Volume L */
+	0x00C0,     /* R51  - DAC Digital Volume R */
+	0x0000,     /* R52 */
+	0x0040,     /* R53  - DAC LR Rate */
+	0x0000,     /* R54  - DAC Clock Control */
+	0x0000,     /* R55 */
+	0x0000,     /* R56 */
+	0x0000,     /* R57 */
+	0x4000,     /* R58  - DAC Mute */
+	0x0000,     /* R59  - DAC Mute Volume */
+	0x0000,     /* R60  - DAC Side */
+	0x0000,     /* R61 */
+	0x0000,     /* R62 */
+	0x0000,     /* R63 */
+	0x8000,     /* R64  - ADC Control */
+	0x0000,     /* R65 */
+	0x00C0,     /* R66  - ADC Digital Volume L */
+	0x00C0,     /* R67  - ADC Digital Volume R */
+	0x0000,     /* R68  - ADC Divider */
+	0x0000,     /* R69 */
+	0x0040,     /* R70  - ADC LR Rate */
+	0x0000,     /* R71 */
+	0x0303,     /* R72  - Input Control */
+	0x0000,     /* R73  - IN3 Input Control */
+	0x0000,     /* R74  - Mic Bias Control */
+	0x0000,     /* R75 */
+	0x0000,     /* R76  - Output Control */
+	0x0000,     /* R77  - Jack Detect */
+	0x0000,     /* R78  - Anti Pop Control */
+	0x0000,     /* R79 */
+	0x0040,     /* R80  - Left Input Volume */
+	0x0040,     /* R81  - Right Input Volume */
+	0x0000,     /* R82 */
+	0x0000,     /* R83 */
+	0x0000,     /* R84 */
+	0x0000,     /* R85 */
+	0x0000,     /* R86 */
+	0x0000,     /* R87 */
+	0x0800,     /* R88  - Left Mixer Control */
+	0x1000,     /* R89  - Right Mixer Control */
+	0x0000,     /* R90 */
+	0x0000,     /* R91 */
+	0x0000,     /* R92  - OUT3 Mixer Control */
+	0x0000,     /* R93  - OUT4 Mixer Control */
+	0x0000,     /* R94 */
+	0x0000,     /* R95 */
+	0x0000,     /* R96  - Output Left Mixer Volume */
+	0x0000,     /* R97  - Output Right Mixer Volume */
+	0x0000,     /* R98  - Input Mixer Volume L */
+	0x0000,     /* R99  - Input Mixer Volume R */
+	0x0000,     /* R100 - Input Mixer Volume */
+	0x0000,     /* R101 */
+	0x0000,     /* R102 */
+	0x0000,     /* R103 */
+	0x00E4,     /* R104 - OUT1L Volume */
+	0x00E4,     /* R105 - OUT1R Volume */
+	0x00E4,     /* R106 - OUT2L Volume */
+	0x02E4,     /* R107 - OUT2R Volume */
+	0x0000,     /* R108 */
+	0x0000,     /* R109 */
+	0x0000,     /* R110 */
+	0x0000,     /* R111 - BEEP Volume */
+	0x0A00,     /* R112 - AI Formating */
+	0x0000,     /* R113 - ADC DAC COMP */
+	0x0020,     /* R114 - AI ADC Control */
+	0x0020,     /* R115 - AI DAC Control */
+	0x0000,     /* R116 */
+	0x0000,     /* R117 */
+	0x0000,     /* R118 */
+	0x0000,     /* R119 */
+	0x0000,     /* R120 */
+	0x0000,     /* R121 */
+	0x0000,     /* R122 */
+	0x0000,     /* R123 */
+	0x0000,     /* R124 */
+	0x0000,     /* R125 */
+	0x0000,     /* R126 */
+	0x0000,     /* R127 */
+	0x1FFF,     /* R128 - GPIO Debounce */
+	0x0000,     /* R129 - GPIO Pin pull up Control */
+	0x0000,     /* R130 - GPIO Pull down Control */
+	0x0000,     /* R131 - GPIO Interrupt Mode */
+	0x0000,     /* R132 */
+	0x0000,     /* R133 - GPIO Control */
+	0x0CFB,     /* R134 - GPIO Configuration (i/o) */
+	0x0C1F,     /* R135 - GPIO Pin Polarity / Type */
+	0x0000,     /* R136 */
+	0x0000,     /* R137 */
+	0x0000,     /* R138 */
+	0x0000,     /* R139 */
+	0x0300,     /* R140 - GPIO Function Select 1 */
+	0x1110,     /* R141 - GPIO Function Select 2 */
+	0x0013,     /* R142 - GPIO Function Select 3 */
+	0x0003,     /* R143 - GPIO Function Select 4 */
+	0x0000,     /* R144 - Digitiser Control (1) */
+	0x0002,     /* R145 - Digitiser Control (2) */
+	0x0000,     /* R146 */
+	0x0000,     /* R147 */
+	0x0000,     /* R148 */
+	0x0000,     /* R149 */
+	0x0000,     /* R150 */
+	0x0000,     /* R151 */
+	0x7000,     /* R152 - AUX1 Readback */
+	0x7000,     /* R153 - AUX2 Readback */
+	0x7000,     /* R154 - AUX3 Readback */
+	0x7000,     /* R155 - AUX4 Readback */
+	0x0000,     /* R156 - USB Voltage Readback */
+	0x0000,     /* R157 - LINE Voltage Readback */
+	0x0000,     /* R158 - BATT Voltage Readback */
+	0x0000,     /* R159 - Chip Temp Readback */
+	0x0000,     /* R160 */
+	0x0000,     /* R161 */
+	0x0000,     /* R162 */
+	0x0000,     /* R163 - Generic Comparator Control */
+	0x0000,     /* R164 - Generic comparator 1 */
+	0x0000,     /* R165 - Generic comparator 2 */
+	0x0000,     /* R166 - Generic comparator 3 */
+	0x0000,     /* R167 - Generic comparator 4 */
+	0xA00F,     /* R168 - Battery Charger Control 1 */
+	0x0B06,     /* R169 - Battery Charger Control 2 */
+	0x0000,     /* R170 - Battery Charger Control 3 */
+	0x0000,     /* R171 */
+	0x0000,     /* R172 - Current Sink Driver A */
+	0x0000,     /* R173 - CSA Flash control */
+	0x0000,     /* R174 */
+	0x0000,     /* R175 */
+	0x0000,     /* R176 - DCDC/LDO requested */
+	0x032D,     /* R177 - DCDC Active options */
+	0x0000,     /* R178 - DCDC Sleep options */
+	0x0025,     /* R179 - Power-check comparator */
+	0x000E,     /* R180 - DCDC1 Control */
+	0x0C00,     /* R181 - DCDC1 Timeouts */
+	0x1006,     /* R182 - DCDC1 Low Power */
+	0x0018,     /* R183 - DCDC2 Control */
+	0x0000,     /* R184 - DCDC2 Timeouts */
+	0x0000,     /* R185 */
+	0x0026,     /* R186 - DCDC3 Control */
+	0x0400,     /* R187 - DCDC3 Timeouts */
+	0x0006,     /* R188 - DCDC3 Low Power */
+	0x0062,     /* R189 - DCDC4 Control */
+	0x0800,     /* R190 - DCDC4 Timeouts */
+	0x0006,     /* R191 - DCDC4 Low Power */
+	0x0008,     /* R192 */
+	0x0000,     /* R193 */
+	0x0000,     /* R194 */
+	0x000A,     /* R195 */
+	0x1000,     /* R196 */
+	0x0006,     /* R197 */
+	0x0000,     /* R198 */
+	0x0003,     /* R199 - Limit Switch Control */
+	0x0006,     /* R200 - LDO1 Control */
+	0x0000,     /* R201 - LDO1 Timeouts */
+	0x001C,     /* R202 - LDO1 Low Power */
+	0x0010,     /* R203 - LDO2 Control */
+	0x0C00,     /* R204 - LDO2 Timeouts */
+	0x001C,     /* R205 - LDO2 Low Power */
+	0x001F,     /* R206 - LDO3 Control */
+	0x0800,     /* R207 - LDO3 Timeouts */
+	0x001C,     /* R208 - LDO3 Low Power */
+	0x000A,     /* R209 - LDO4 Control */
+	0x0800,     /* R210 - LDO4 Timeouts */
+	0x001C,     /* R211 - LDO4 Low Power */
+	0x0000,     /* R212 */
+	0x0000,     /* R213 */
+	0x0000,     /* R214 */
+	0x0000,     /* R215 - VCC_FAULT Masks */
+	0x001F,     /* R216 - Main Bandgap Control */
+	0x0000,     /* R217 - OSC Control */
+	0x9000,     /* R218 - RTC Tick Control */
+	0x0000,     /* R219 - Security1 */
+	0x4000,     /* R220 */
+	0x0000,     /* R221 */
+	0x0000,     /* R222 */
+	0x0000,     /* R223 */
+	0x0000,     /* R224 - Signal overrides */
+	0x0000,     /* R225 - DCDC/LDO status */
+	0x0000,     /* R226 - Charger Overides/status */
+	0x0000,     /* R227 - misc overrides */
+	0x0000,     /* R228 - Supply overrides/status 1 */
+	0x0000,     /* R229 - Supply overrides/status 2 */
+	0xE000,     /* R230 - GPIO Pin Status */
+	0x0000,     /* R231 - comparotor overrides */
+	0x0000,     /* R232 */
+	0x0000,     /* R233 - State Machine status */
+	0x1200,     /* R234 - FLL Test 1 */
+	0x0000,     /* R235 */
+	0x8000,     /* R236 */
+	0x0000,     /* R237 */
+	0x0000,     /* R238 */
+	0x0000,     /* R239 */
+	0x0003,     /* R240 */
+	0x0000,     /* R241 */
+	0x0000,     /* R242 */
+	0x0004,     /* R243 */
+	0x0300,     /* R244 */
+	0x0000,     /* R245 */
+	0x0200,     /* R246 */
+	0x1000,     /* R247 */
+	0x1000,     /* R248 - DCDC1 Test Controls */
+	0x1000,     /* R249 */
+	0x1000,     /* R250 - DCDC3 Test Controls */
+	0x1000,     /* R251 - DCDC4 Test Controls */
+};
+#endif
+
+#ifdef CONFIG_MFD_WM8351_CONFIG_MODE_2
+
+#undef WM8350_HAVE_CONFIG_MODE
+#define WM8350_HAVE_CONFIG_MODE
+
+const u16 wm8351_mode2_defaults[] = {
+	0x6143,     /* R0   - Reset/ID */
+	0x0000,     /* R1   - ID */
+	0x0001,     /* R2   - Revision */
+	0x1C02,     /* R3   - System Control 1 */
+	0x0214,     /* R4   - System Control 2 */
+	0x0000,     /* R5   - System Hibernate */
+	0x8A00,     /* R6   - Interface Control */
+	0x0000,     /* R7 */
+	0x8000,     /* R8   - Power mgmt (1) */
+	0x0000,     /* R9   - Power mgmt (2) */
+	0x0000,     /* R10  - Power mgmt (3) */
+	0x2000,     /* R11  - Power mgmt (4) */
+	0x0E00,     /* R12  - Power mgmt (5) */
+	0x0000,     /* R13  - Power mgmt (6) */
+	0x0000,     /* R14  - Power mgmt (7) */
+	0x0000,     /* R15 */
+	0x0000,     /* R16  - RTC Seconds/Minutes */
+	0x0100,     /* R17  - RTC Hours/Day */
+	0x0101,     /* R18  - RTC Date/Month */
+	0x1400,     /* R19  - RTC Year */
+	0x0000,     /* R20  - Alarm Seconds/Minutes */
+	0x0000,     /* R21  - Alarm Hours/Day */
+	0x0000,     /* R22  - Alarm Date/Month */
+	0x0320,     /* R23  - RTC Time Control */
+	0x0000,     /* R24  - System Interrupts */
+	0x0000,     /* R25  - Interrupt Status 1 */
+	0x0000,     /* R26  - Interrupt Status 2 */
+	0x0000,     /* R27 */
+	0x0000,     /* R28  - Under Voltage Interrupt status */
+	0x0000,     /* R29  - Over Current Interrupt status */
+	0x0000,     /* R30  - GPIO Interrupt Status */
+	0x0000,     /* R31  - Comparator Interrupt Status */
+	0x3FFF,     /* R32  - System Interrupts Mask */
+	0x0000,     /* R33  - Interrupt Status 1 Mask */
+	0x0000,     /* R34  - Interrupt Status 2 Mask */
+	0x0000,     /* R35 */
+	0x0000,     /* R36  - Under Voltage Interrupt status Mask */
+	0x0000,     /* R37  - Over Current Interrupt status Mask */
+	0x0000,     /* R38  - GPIO Interrupt Status Mask */
+	0x0000,     /* R39  - Comparator Interrupt Status Mask */
+	0x0040,     /* R40  - Clock Control 1 */
+	0x0000,     /* R41  - Clock Control 2 */
+	0x3A00,     /* R42  - FLL Control 1 */
+	0x7086,     /* R43  - FLL Control 2 */
+	0xC226,     /* R44  - FLL Control 3 */
+	0x0000,     /* R45  - FLL Control 4 */
+	0x0000,     /* R46 */
+	0x0000,     /* R47 */
+	0x0000,     /* R48  - DAC Control */
+	0x0000,     /* R49 */
+	0x00C0,     /* R50  - DAC Digital Volume L */
+	0x00C0,     /* R51  - DAC Digital Volume R */
+	0x0000,     /* R52 */
+	0x0040,     /* R53  - DAC LR Rate */
+	0x0000,     /* R54  - DAC Clock Control */
+	0x0000,     /* R55 */
+	0x0000,     /* R56 */
+	0x0000,     /* R57 */
+	0x4000,     /* R58  - DAC Mute */
+	0x0000,     /* R59  - DAC Mute Volume */
+	0x0000,     /* R60  - DAC Side */
+	0x0000,     /* R61 */
+	0x0000,     /* R62 */
+	0x0000,     /* R63 */
+	0x8000,     /* R64  - ADC Control */
+	0x0000,     /* R65 */
+	0x00C0,     /* R66  - ADC Digital Volume L */
+	0x00C0,     /* R67  - ADC Digital Volume R */
+	0x0000,     /* R68  - ADC Divider */
+	0x0000,     /* R69 */
+	0x0040,     /* R70  - ADC LR Rate */
+	0x0000,     /* R71 */
+	0x0303,     /* R72  - Input Control */
+	0x0000,     /* R73  - IN3 Input Control */
+	0x0000,     /* R74  - Mic Bias Control */
+	0x0000,     /* R75 */
+	0x0000,     /* R76  - Output Control */
+	0x0000,     /* R77  - Jack Detect */
+	0x0000,     /* R78  - Anti Pop Control */
+	0x0000,     /* R79 */
+	0x0040,     /* R80  - Left Input Volume */
+	0x0040,     /* R81  - Right Input Volume */
+	0x0000,     /* R82 */
+	0x0000,     /* R83 */
+	0x0000,     /* R84 */
+	0x0000,     /* R85 */
+	0x0000,     /* R86 */
+	0x0000,     /* R87 */
+	0x0800,     /* R88  - Left Mixer Control */
+	0x1000,     /* R89  - Right Mixer Control */
+	0x0000,     /* R90 */
+	0x0000,     /* R91 */
+	0x0000,     /* R92  - OUT3 Mixer Control */
+	0x0000,     /* R93  - OUT4 Mixer Control */
+	0x0000,     /* R94 */
+	0x0000,     /* R95 */
+	0x0000,     /* R96  - Output Left Mixer Volume */
+	0x0000,     /* R97  - Output Right Mixer Volume */
+	0x0000,     /* R98  - Input Mixer Volume L */
+	0x0000,     /* R99  - Input Mixer Volume R */
+	0x0000,     /* R100 - Input Mixer Volume */
+	0x0000,     /* R101 */
+	0x0000,     /* R102 */
+	0x0000,     /* R103 */
+	0x00E4,     /* R104 - OUT1L Volume */
+	0x00E4,     /* R105 - OUT1R Volume */
+	0x00E4,     /* R106 - OUT2L Volume */
+	0x02E4,     /* R107 - OUT2R Volume */
+	0x0000,     /* R108 */
+	0x0000,     /* R109 */
+	0x0000,     /* R110 */
+	0x0000,     /* R111 - BEEP Volume */
+	0x0A00,     /* R112 - AI Formating */
+	0x0000,     /* R113 - ADC DAC COMP */
+	0x0020,     /* R114 - AI ADC Control */
+	0x0020,     /* R115 - AI DAC Control */
+	0x0000,     /* R116 */
+	0x0000,     /* R117 */
+	0x0000,     /* R118 */
+	0x0000,     /* R119 */
+	0x0000,     /* R120 */
+	0x0000,     /* R121 */
+	0x0000,     /* R122 */
+	0x0000,     /* R123 */
+	0x0000,     /* R124 */
+	0x0000,     /* R125 */
+	0x0000,     /* R126 */
+	0x0000,     /* R127 */
+	0x1FFF,     /* R128 - GPIO Debounce */
+	0x0000,     /* R129 - GPIO Pin pull up Control */
+	0x0110,     /* R130 - GPIO Pull down Control */
+	0x0000,     /* R131 - GPIO Interrupt Mode */
+	0x0000,     /* R132 */
+	0x0000,     /* R133 - GPIO Control */
+	0x09FA,     /* R134 - GPIO Configuration (i/o) */
+	0x0DF6,     /* R135 - GPIO Pin Polarity / Type */
+	0x0000,     /* R136 */
+	0x0000,     /* R137 */
+	0x0000,     /* R138 */
+	0x0000,     /* R139 */
+	0x1310,     /* R140 - GPIO Function Select 1 */
+	0x0003,     /* R141 - GPIO Function Select 2 */
+	0x2000,     /* R142 - GPIO Function Select 3 */
+	0x0000,     /* R143 - GPIO Function Select 4 */
+	0x0000,     /* R144 - Digitiser Control (1) */
+	0x0002,     /* R145 - Digitiser Control (2) */
+	0x0000,     /* R146 */
+	0x0000,     /* R147 */
+	0x0000,     /* R148 */
+	0x0000,     /* R149 */
+	0x0000,     /* R150 */
+	0x0000,     /* R151 */
+	0x7000,     /* R152 - AUX1 Readback */
+	0x7000,     /* R153 - AUX2 Readback */
+	0x7000,     /* R154 - AUX3 Readback */
+	0x7000,     /* R155 - AUX4 Readback */
+	0x0000,     /* R156 - USB Voltage Readback */
+	0x0000,     /* R157 - LINE Voltage Readback */
+	0x0000,     /* R158 - BATT Voltage Readback */
+	0x0000,     /* R159 - Chip Temp Readback */
+	0x0000,     /* R160 */
+	0x0000,     /* R161 */
+	0x0000,     /* R162 */
+	0x0000,     /* R163 - Generic Comparator Control */
+	0x0000,     /* R164 - Generic comparator 1 */
+	0x0000,     /* R165 - Generic comparator 2 */
+	0x0000,     /* R166 - Generic comparator 3 */
+	0x0000,     /* R167 - Generic comparator 4 */
+	0xA00F,     /* R168 - Battery Charger Control 1 */
+	0x0B06,     /* R169 - Battery Charger Control 2 */
+	0x0000,     /* R170 - Battery Charger Control 3 */
+	0x0000,     /* R171 */
+	0x0000,     /* R172 - Current Sink Driver A */
+	0x0000,     /* R173 - CSA Flash control */
+	0x0000,     /* R174 */
+	0x0000,     /* R175 */
+	0x0000,     /* R176 - DCDC/LDO requested */
+	0x032D,     /* R177 - DCDC Active options */
+	0x0000,     /* R178 - DCDC Sleep options */
+	0x0025,     /* R179 - Power-check comparator */
+	0x001A,     /* R180 - DCDC1 Control */
+	0x0800,     /* R181 - DCDC1 Timeouts */
+	0x1006,     /* R182 - DCDC1 Low Power */
+	0x0018,     /* R183 - DCDC2 Control */
+	0x0000,     /* R184 - DCDC2 Timeouts */
+	0x0000,     /* R185 */
+	0x0056,     /* R186 - DCDC3 Control */
+	0x0400,     /* R187 - DCDC3 Timeouts */
+	0x0006,     /* R188 - DCDC3 Low Power */
+	0x0026,     /* R189 - DCDC4 Control */
+	0x0C00,     /* R190 - DCDC4 Timeouts */
+	0x0006,     /* R191 - DCDC4 Low Power */
+	0x0008,     /* R192 */
+	0x0000,     /* R193 */
+	0x0000,     /* R194 */
+	0x0026,     /* R195 */
+	0x0C00,     /* R196 */
+	0x0006,     /* R197 */
+	0x0000,     /* R198 */
+	0x0003,     /* R199 - Limit Switch Control */
+	0x001C,     /* R200 - LDO1 Control */
+	0x0400,     /* R201 - LDO1 Timeouts */
+	0x001C,     /* R202 - LDO1 Low Power */
+	0x0010,     /* R203 - LDO2 Control */
+	0x0C00,     /* R204 - LDO2 Timeouts */
+	0x001C,     /* R205 - LDO2 Low Power */
+	0x0015,     /* R206 - LDO3 Control */
+	0x0000,     /* R207 - LDO3 Timeouts */
+	0x001C,     /* R208 - LDO3 Low Power */
+	0x001A,     /* R209 - LDO4 Control */
+	0x0000,     /* R210 - LDO4 Timeouts */
+	0x001C,     /* R211 - LDO4 Low Power */
+	0x0000,     /* R212 */
+	0x0000,     /* R213 */
+	0x0000,     /* R214 */
+	0x0000,     /* R215 - VCC_FAULT Masks */
+	0x001F,     /* R216 - Main Bandgap Control */
+	0x0000,     /* R217 - OSC Control */
+	0x9000,     /* R218 - RTC Tick Control */
+	0x0000,     /* R219 - Security1 */
+	0x4000,     /* R220 */
+	0x0000,     /* R221 */
+	0x0000,     /* R222 */
+	0x0000,     /* R223 */
+	0x0000,     /* R224 - Signal overrides */
+	0x0000,     /* R225 - DCDC/LDO status */
+	0x0000,     /* R226 - Charger Overides/status */
+	0x0000,     /* R227 - misc overrides */
+	0x0000,     /* R228 - Supply overrides/status 1 */
+	0x0000,     /* R229 - Supply overrides/status 2 */
+	0xE000,     /* R230 - GPIO Pin Status */
+	0x0000,     /* R231 - comparotor overrides */
+	0x0000,     /* R232 */
+	0x0000,     /* R233 - State Machine status */
+	0x1200,     /* R234 - FLL Test 1 */
+	0x0000,     /* R235 */
+	0x8000,     /* R236 */
+	0x0000,     /* R237 */
+	0x0000,     /* R238 */
+	0x0000,     /* R239 */
+	0x0003,     /* R240 */
+	0x0000,     /* R241 */
+	0x0000,     /* R242 */
+	0x0004,     /* R243 */
+	0x0300,     /* R244 */
+	0x0000,     /* R245 */
+	0x0200,     /* R246 */
+	0x0000,     /* R247 */
+	0x1000,     /* R248 - DCDC1 Test Controls */
+	0x1000,     /* R249 */
+	0x1000,     /* R250 - DCDC3 Test Controls */
+	0x1000,     /* R251 - DCDC4 Test Controls */
+};
+#endif
+
+#ifdef CONFIG_MFD_WM8351_CONFIG_MODE_3
+
+#undef WM8350_HAVE_CONFIG_MODE
+#define WM8350_HAVE_CONFIG_MODE
+
+const u16 wm8351_mode3_defaults[] = {
+	0x6143,     /* R0   - Reset/ID */
+	0x0000,     /* R1   - ID */
+	0x0001,     /* R2   - Revision */
+	0x1C02,     /* R3   - System Control 1 */
+	0x0204,     /* R4   - System Control 2 */
+	0x0000,     /* R5   - System Hibernate */
+	0x8A00,     /* R6   - Interface Control */
+	0x0000,     /* R7 */
+	0x8000,     /* R8   - Power mgmt (1) */
+	0x0000,     /* R9   - Power mgmt (2) */
+	0x0000,     /* R10  - Power mgmt (3) */
+	0x2000,     /* R11  - Power mgmt (4) */
+	0x0E00,     /* R12  - Power mgmt (5) */
+	0x0000,     /* R13  - Power mgmt (6) */
+	0x0000,     /* R14  - Power mgmt (7) */
+	0x0000,     /* R15 */
+	0x0000,     /* R16  - RTC Seconds/Minutes */
+	0x0100,     /* R17  - RTC Hours/Day */
+	0x0101,     /* R18  - RTC Date/Month */
+	0x1400,     /* R19  - RTC Year */
+	0x0000,     /* R20  - Alarm Seconds/Minutes */
+	0x0000,     /* R21  - Alarm Hours/Day */
+	0x0000,     /* R22  - Alarm Date/Month */
+	0x0320,     /* R23  - RTC Time Control */
+	0x0000,     /* R24  - System Interrupts */
+	0x0000,     /* R25  - Interrupt Status 1 */
+	0x0000,     /* R26  - Interrupt Status 2 */
+	0x0000,     /* R27 */
+	0x0000,     /* R28  - Under Voltage Interrupt status */
+	0x0000,     /* R29  - Over Current Interrupt status */
+	0x0000,     /* R30  - GPIO Interrupt Status */
+	0x0000,     /* R31  - Comparator Interrupt Status */
+	0x3FFF,     /* R32  - System Interrupts Mask */
+	0x0000,     /* R33  - Interrupt Status 1 Mask */
+	0x0000,     /* R34  - Interrupt Status 2 Mask */
+	0x0000,     /* R35 */
+	0x0000,     /* R36  - Under Voltage Interrupt status Mask */
+	0x0000,     /* R37  - Over Current Interrupt status Mask */
+	0x0000,     /* R38  - GPIO Interrupt Status Mask */
+	0x0000,     /* R39  - Comparator Interrupt Status Mask */
+	0x0040,     /* R40  - Clock Control 1 */
+	0x0000,     /* R41  - Clock Control 2 */
+	0x3A00,     /* R42  - FLL Control 1 */
+	0x7086,     /* R43  - FLL Control 2 */
+	0xC226,     /* R44  - FLL Control 3 */
+	0x0000,     /* R45  - FLL Control 4 */
+	0x0000,     /* R46 */
+	0x0000,     /* R47 */
+	0x0000,     /* R48  - DAC Control */
+	0x0000,     /* R49 */
+	0x00C0,     /* R50  - DAC Digital Volume L */
+	0x00C0,     /* R51  - DAC Digital Volume R */
+	0x0000,     /* R52 */
+	0x0040,     /* R53  - DAC LR Rate */
+	0x0000,     /* R54  - DAC Clock Control */
+	0x0000,     /* R55 */
+	0x0000,     /* R56 */
+	0x0000,     /* R57 */
+	0x4000,     /* R58  - DAC Mute */
+	0x0000,     /* R59  - DAC Mute Volume */
+	0x0000,     /* R60  - DAC Side */
+	0x0000,     /* R61 */
+	0x0000,     /* R62 */
+	0x0000,     /* R63 */
+	0x8000,     /* R64  - ADC Control */
+	0x0000,     /* R65 */
+	0x00C0,     /* R66  - ADC Digital Volume L */
+	0x00C0,     /* R67  - ADC Digital Volume R */
+	0x0000,     /* R68  - ADC Divider */
+	0x0000,     /* R69 */
+	0x0040,     /* R70  - ADC LR Rate */
+	0x0000,     /* R71 */
+	0x0303,     /* R72  - Input Control */
+	0x0000,     /* R73  - IN3 Input Control */
+	0x0000,     /* R74  - Mic Bias Control */
+	0x0000,     /* R75 */
+	0x0000,     /* R76  - Output Control */
+	0x0000,     /* R77  - Jack Detect */
+	0x0000,     /* R78  - Anti Pop Control */
+	0x0000,     /* R79 */
+	0x0040,     /* R80  - Left Input Volume */
+	0x0040,     /* R81  - Right Input Volume */
+	0x0000,     /* R82 */
+	0x0000,     /* R83 */
+	0x0000,     /* R84 */
+	0x0000,     /* R85 */
+	0x0000,     /* R86 */
+	0x0000,     /* R87 */
+	0x0800,     /* R88  - Left Mixer Control */
+	0x1000,     /* R89  - Right Mixer Control */
+	0x0000,     /* R90 */
+	0x0000,     /* R91 */
+	0x0000,     /* R92  - OUT3 Mixer Control */
+	0x0000,     /* R93  - OUT4 Mixer Control */
+	0x0000,     /* R94 */
+	0x0000,     /* R95 */
+	0x0000,     /* R96  - Output Left Mixer Volume */
+	0x0000,     /* R97  - Output Right Mixer Volume */
+	0x0000,     /* R98  - Input Mixer Volume L */
+	0x0000,     /* R99  - Input Mixer Volume R */
+	0x0000,     /* R100 - Input Mixer Volume */
+	0x0000,     /* R101 */
+	0x0000,     /* R102 */
+	0x0000,     /* R103 */
+	0x00E4,     /* R104 - OUT1L Volume */
+	0x00E4,     /* R105 - OUT1R Volume */
+	0x00E4,     /* R106 - OUT2L Volume */
+	0x02E4,     /* R107 - OUT2R Volume */
+	0x0000,     /* R108 */
+	0x0000,     /* R109 */
+	0x0000,     /* R110 */
+	0x0000,     /* R111 - BEEP Volume */
+	0x0A00,     /* R112 - AI Formating */
+	0x0000,     /* R113 - ADC DAC COMP */
+	0x0020,     /* R114 - AI ADC Control */
+	0x0020,     /* R115 - AI DAC Control */
+	0x0000,     /* R116 */
+	0x0000,     /* R117 */
+	0x0000,     /* R118 */
+	0x0000,     /* R119 */
+	0x0000,     /* R120 */
+	0x0000,     /* R121 */
+	0x0000,     /* R122 */
+	0x0000,     /* R123 */
+	0x0000,     /* R124 */
+	0x0000,     /* R125 */
+	0x0000,     /* R126 */
+	0x0000,     /* R127 */
+	0x1FFF,     /* R128 - GPIO Debounce */
+	0x0010,     /* R129 - GPIO Pin pull up Control */
+	0x0000,     /* R130 - GPIO Pull down Control */
+	0x0000,     /* R131 - GPIO Interrupt Mode */
+	0x0000,     /* R132 */
+	0x0000,     /* R133 - GPIO Control */
+	0x0BFB,     /* R134 - GPIO Configuration (i/o) */
+	0x0FFD,     /* R135 - GPIO Pin Polarity / Type */
+	0x0000,     /* R136 */
+	0x0000,     /* R137 */
+	0x0000,     /* R138 */
+	0x0000,     /* R139 */
+	0x0310,     /* R140 - GPIO Function Select 1 */
+	0x0001,     /* R141 - GPIO Function Select 2 */
+	0x2300,     /* R142 - GPIO Function Select 3 */
+	0x0003,     /* R143 - GPIO Function Select 4 */
+	0x0000,     /* R144 - Digitiser Control (1) */
+	0x0002,     /* R145 - Digitiser Control (2) */
+	0x0000,     /* R146 */
+	0x0000,     /* R147 */
+	0x0000,     /* R148 */
+	0x0000,     /* R149 */
+	0x0000,     /* R150 */
+	0x0000,     /* R151 */
+	0x7000,     /* R152 - AUX1 Readback */
+	0x7000,     /* R153 - AUX2 Readback */
+	0x7000,     /* R154 - AUX3 Readback */
+	0x7000,     /* R155 - AUX4 Readback */
+	0x0000,     /* R156 - USB Voltage Readback */
+	0x0000,     /* R157 - LINE Voltage Readback */
+	0x0000,     /* R158 - BATT Voltage Readback */
+	0x0000,     /* R159 - Chip Temp Readback */
+	0x0000,     /* R160 */
+	0x0000,     /* R161 */
+	0x0000,     /* R162 */
+	0x0000,     /* R163 - Generic Comparator Control */
+	0x0000,     /* R164 - Generic comparator 1 */
+	0x0000,     /* R165 - Generic comparator 2 */
+	0x0000,     /* R166 - Generic comparator 3 */
+	0x0000,     /* R167 - Generic comparator 4 */
+	0xA00F,     /* R168 - Battery Charger Control 1 */
+	0x0B06,     /* R169 - Battery Charger Control 2 */
+	0x0000,     /* R170 - Battery Charger Control 3 */
+	0x0000,     /* R171 */
+	0x0000,     /* R172 - Current Sink Driver A */
+	0x0000,     /* R173 - CSA Flash control */
+	0x0000,     /* R174 */
+	0x0000,     /* R175 */
+	0x0000,     /* R176 - DCDC/LDO requested */
+	0x032D,     /* R177 - DCDC Active options */
+	0x0000,     /* R178 - DCDC Sleep options */
+	0x0025,     /* R179 - Power-check comparator */
+	0x000E,     /* R180 - DCDC1 Control */
+	0x0400,     /* R181 - DCDC1 Timeouts */
+	0x1006,     /* R182 - DCDC1 Low Power */
+	0x0018,     /* R183 - DCDC2 Control */
+	0x0000,     /* R184 - DCDC2 Timeouts */
+	0x0000,     /* R185 */
+	0x0026,     /* R186 - DCDC3 Control */
+	0x0800,     /* R187 - DCDC3 Timeouts */
+	0x0006,     /* R188 - DCDC3 Low Power */
+	0x0062,     /* R189 - DCDC4 Control */
+	0x1400,     /* R190 - DCDC4 Timeouts */
+	0x0006,     /* R191 - DCDC4 Low Power */
+	0x0008,     /* R192 */
+	0x0000,     /* R193 */
+	0x0000,     /* R194 */
+	0x0026,     /* R195 */
+	0x0400,     /* R196 */
+	0x0006,     /* R197 */
+	0x0000,     /* R198 */
+	0x0003,     /* R199 - Limit Switch Control */
+	0x0006,     /* R200 - LDO1 Control */
+	0x0C00,     /* R201 - LDO1 Timeouts */
+	0x001C,     /* R202 - LDO1 Low Power */
+	0x0016,     /* R203 - LDO2 Control */
+	0x0000,     /* R204 - LDO2 Timeouts */
+	0x001C,     /* R205 - LDO2 Low Power */
+	0x0019,     /* R206 - LDO3 Control */
+	0x0000,     /* R207 - LDO3 Timeouts */
+	0x001C,     /* R208 - LDO3 Low Power */
+	0x001A,     /* R209 - LDO4 Control */
+	0x1000,     /* R210 - LDO4 Timeouts */
+	0x001C,     /* R211 - LDO4 Low Power */
+	0x0000,     /* R212 */
+	0x0000,     /* R213 */
+	0x0000,     /* R214 */
+	0x0000,     /* R215 - VCC_FAULT Masks */
+	0x001F,     /* R216 - Main Bandgap Control */
+	0x0000,     /* R217 - OSC Control */
+	0x9000,     /* R218 - RTC Tick Control */
+	0x0000,     /* R219 - Security1 */
+	0x4000,     /* R220 */
+	0x0000,     /* R221 */
+	0x0000,     /* R222 */
+	0x0000,     /* R223 */
+	0x0000,     /* R224 - Signal overrides */
+	0x0000,     /* R225 - DCDC/LDO status */
+	0x0000,     /* R226 - Charger Overides/status */
+	0x0000,     /* R227 - misc overrides */
+	0x0000,     /* R228 - Supply overrides/status 1 */
+	0x0000,     /* R229 - Supply overrides/status 2 */
+	0xE000,     /* R230 - GPIO Pin Status */
+	0x0000,     /* R231 - comparotor overrides */
+	0x0000,     /* R232 */
+	0x0000,     /* R233 - State Machine status */
+	0x1200,     /* R234 - FLL Test 1 */
+	0x0000,     /* R235 */
+	0x8000,     /* R236 */
+	0x0000,     /* R237 */
+	0x0000,     /* R238 */
+	0x0000,     /* R239 */
+	0x0003,     /* R240 */
+	0x0000,     /* R241 */
+	0x0000,     /* R242 */
+	0x0004,     /* R243 */
+	0x0300,     /* R244 */
+	0x0000,     /* R245 */
+	0x0200,     /* R246 */
+	0x0000,     /* R247 */
+	0x1000,     /* R248 - DCDC1 Test Controls */
+	0x1000,     /* R249 */
+	0x1000,     /* R250 - DCDC3 Test Controls */
+	0x1000,     /* R251 - DCDC4 Test Controls */
+};
+#endif
+
 #ifdef CONFIG_MFD_WM8352_CONFIG_MODE_0
 
 #undef WM8350_HAVE_CONFIG_MODE
diff --git a/include/linux/mfd/wm8350/core.h b/include/linux/mfd/wm8350/core.h
index 737579086d01..980669d50dca 100644
--- a/include/linux/mfd/wm8350/core.h
+++ b/include/linux/mfd/wm8350/core.h
@@ -589,6 +589,10 @@ extern const u16 wm8350_mode0_defaults[];
 extern const u16 wm8350_mode1_defaults[];
 extern const u16 wm8350_mode2_defaults[];
 extern const u16 wm8350_mode3_defaults[];
+extern const u16 wm8351_mode0_defaults[];
+extern const u16 wm8351_mode1_defaults[];
+extern const u16 wm8351_mode2_defaults[];
+extern const u16 wm8351_mode3_defaults[];
 extern const u16 wm8352_mode0_defaults[];
 extern const u16 wm8352_mode1_defaults[];
 extern const u16 wm8352_mode2_defaults[];
-- 
cgit v1.2.3


From 0931a4c6dbfab03f2bfd22a9170130f7b155d53a Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Mon, 22 Dec 2008 12:05:27 +0100
Subject: mfd: dm355evm msp430 driver

Basic MFD framework for the MSP430 microcontroller firmware used
on the dm355evm board:

 - Provides an interface for other drivers: register read/write
   utilities, and register declarations.

 - Directly exports:
     * Many signals through the GPIO framework
         + LEDs
         + SW6 through gpio sysfs
	 + NTSC/nPAL jumper through gpio sysfs
	 + ... more could be added later, e.g. MMC signals
     * Child devices:
	+ LEDs, via leds-gpio child (and default triggers)
	+ RTC, via rtc-dm355evm child device
	+ Buttons and IR control, via dm355evm_keys

 - Supports power-off system call.  Use the reset button to power
   the board back up; the power supply LED will be on, but the
   MSP430 waits to re-activate the regulators.

 - On probe() this:
     * Announces firmware revision
     * Turns off the banked LEDs
     * Exports the resources noted above
     * Hooks the power-off support
     * Muxes tvp5146 -or- imager for video input

Unless the new tvp514x driver (tracked for mainline) is configured,
this assumes that some custom imager driver handles video-in.

This completely ignores the registers reporting the output voltages
on the various power supplies.  Someone could add a hwmon interface
if that seems useful.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/Kconfig              |   8 +
 drivers/mfd/Makefile             |   2 +
 drivers/mfd/dm355evm_msp.c       | 420 +++++++++++++++++++++++++++++++++++++++
 include/linux/i2c/dm355evm_msp.h |  79 ++++++++
 4 files changed, 509 insertions(+)
 create mode 100644 drivers/mfd/dm355evm_msp.c
 create mode 100644 include/linux/i2c/dm355evm_msp.h

(limited to 'include/linux')

diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index 781a27955ff5..02e9146ca44a 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -34,6 +34,14 @@ config MFD_ASIC3
 	  This driver supports the ASIC3 multifunction chip found on many
 	  PDAs (mainly iPAQ and HTC based ones)
 
+config MFD_DM355EVM_MSP
+	bool "DaVinci DM355 EVM microcontroller"
+	depends on I2C && MACH_DAVINCI_DM355_EVM
+	help
+	  This driver supports the MSP430 microcontroller used on these
+	  boards.  MSP430 firmware manages resets and power sequencing,
+	  inputs from buttons and the IR remote, LEDs, an RTC, and more.
+
 config HTC_EGPIO
 	bool "HTC EGPIO support"
 	depends on GENERIC_HARDIRQS && GPIOLIB && ARM
diff --git a/drivers/mfd/Makefile b/drivers/mfd/Makefile
index 9a5ad8af9116..8f6cd5c48930 100644
--- a/drivers/mfd/Makefile
+++ b/drivers/mfd/Makefile
@@ -8,6 +8,8 @@ obj-$(CONFIG_MFD_ASIC3)		+= asic3.o
 obj-$(CONFIG_HTC_EGPIO)		+= htc-egpio.o
 obj-$(CONFIG_HTC_PASIC3)	+= htc-pasic3.o
 
+obj-$(CONFIG_MFD_DM355EVM_MSP)	+= dm355evm_msp.o
+
 obj-$(CONFIG_MFD_T7L66XB)	+= t7l66xb.o
 obj-$(CONFIG_MFD_TC6387XB)	+= tc6387xb.o
 obj-$(CONFIG_MFD_TC6393XB)	+= tc6393xb.o
diff --git a/drivers/mfd/dm355evm_msp.c b/drivers/mfd/dm355evm_msp.c
new file mode 100644
index 000000000000..4214b3f72426
--- /dev/null
+++ b/drivers/mfd/dm355evm_msp.c
@@ -0,0 +1,420 @@
+/*
+ * dm355evm_msp.c - driver for MSP430 firmware on DM355EVM board
+ *
+ * Copyright (C) 2008 David Brownell
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/init.h>
+#include <linux/mutex.h>
+#include <linux/platform_device.h>
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/gpio.h>
+#include <linux/leds.h>
+#include <linux/i2c.h>
+#include <linux/i2c/dm355evm_msp.h>
+
+
+/*
+ * The DM355 is a DaVinci chip with video support but no C64+ DSP.  Its
+ * EVM board has an MSP430 programmed with firmware for various board
+ * support functions.  This driver exposes some of them directly, and
+ * supports other drivers (e.g. RTC, input) for more complex access.
+ *
+ * Because this firmware is entirely board-specific, this file embeds
+ * knowledge that would be passed as platform_data in a generic driver.
+ *
+ * This driver was tested with firmware revision A4.
+ */
+
+#if defined(CONFIG_KEYBOARD_DM355EVM) \
+		|| defined(CONFIG_KEYBOARD_DM355EVM_MODULE)
+#define msp_has_keyboard()	true
+#else
+#define msp_has_keyboard()	false
+#endif
+
+#if defined(CONFIG_LEDS_GPIO) || defined(CONFIG_LEDS_GPIO_MODULE)
+#define msp_has_leds()		true
+#else
+#define msp_has_leds()		false
+#endif
+
+#if defined(CONFIG_RTC_DRV_DM355EVM) || defined(CONFIG_RTC_DRV_DM355EVM_MODULE)
+#define msp_has_rtc()		true
+#else
+#define msp_has_rtc()		false
+#endif
+
+#if defined(CONFIG_VIDEO_TVP514X) || defined(CONFIG_VIDEO_TVP514X_MODULE)
+#define msp_has_tvp()		true
+#else
+#define msp_has_tvp()		false
+#endif
+
+
+/*----------------------------------------------------------------------*/
+
+/* REVISIT for paranoia's sake, retry reads/writes on error */
+
+static struct i2c_client *msp430;
+
+/**
+ * dm355evm_msp_write - Writes a register in dm355evm_msp
+ * @value: the value to be written
+ * @reg: register address
+ *
+ * Returns result of operation - 0 is success, else negative errno
+ */
+int dm355evm_msp_write(u8 value, u8 reg)
+{
+	return i2c_smbus_write_byte_data(msp430, reg, value);
+}
+EXPORT_SYMBOL(dm355evm_msp_write);
+
+/**
+ * dm355evm_msp_read - Reads a register from dm355evm_msp
+ * @reg: register address
+ *
+ * Returns result of operation - value, or negative errno
+ */
+int dm355evm_msp_read(u8 reg)
+{
+	return i2c_smbus_read_byte_data(msp430, reg);
+}
+EXPORT_SYMBOL(dm355evm_msp_read);
+
+/*----------------------------------------------------------------------*/
+
+/*
+ * Many of the msp430 pins are just used as fixed-direction GPIOs.
+ * We could export a few more of them this way, if we wanted.
+ */
+#define MSP_GPIO(bit,reg)	((DM355EVM_MSP_ ## reg) << 3 | (bit))
+
+static const u8 msp_gpios[] = {
+	/* eight leds */
+	MSP_GPIO(0, LED), MSP_GPIO(1, LED),
+	MSP_GPIO(2, LED), MSP_GPIO(3, LED),
+	MSP_GPIO(4, LED), MSP_GPIO(5, LED),
+	MSP_GPIO(6, LED), MSP_GPIO(7, LED),
+	/* SW6 and the NTSC/nPAL jumper */
+	MSP_GPIO(0, SWITCH1), MSP_GPIO(1, SWITCH1),
+	MSP_GPIO(2, SWITCH1), MSP_GPIO(3, SWITCH1),
+	MSP_GPIO(4, SWITCH1),
+};
+
+#define MSP_GPIO_REG(offset)	(msp_gpios[(offset)] >> 3)
+#define MSP_GPIO_MASK(offset)	BIT(msp_gpios[(offset)] & 0x07)
+
+static int msp_gpio_in(struct gpio_chip *chip, unsigned offset)
+{
+	switch (MSP_GPIO_REG(offset)) {
+	case DM355EVM_MSP_SWITCH1:
+	case DM355EVM_MSP_SWITCH2:
+	case DM355EVM_MSP_SDMMC:
+		return 0;
+	default:
+		return -EINVAL;
+	}
+}
+
+static u8 msp_led_cache;
+
+static int msp_gpio_get(struct gpio_chip *chip, unsigned offset)
+{
+	int reg, status;
+
+	reg = MSP_GPIO_REG(offset);
+	status = dm355evm_msp_read(reg);
+	if (status < 0)
+		return status;
+	if (reg == DM355EVM_MSP_LED)
+		msp_led_cache = status;
+	return status & MSP_GPIO_MASK(offset);
+}
+
+static int msp_gpio_out(struct gpio_chip *chip, unsigned offset, int value)
+{
+	int mask, bits;
+
+	/* NOTE:  there are some other signals that could be
+	 * packaged as output GPIOs, but they aren't as useful
+	 * as the LEDs ... so for now we don't.
+	 */
+	if (MSP_GPIO_REG(offset) != DM355EVM_MSP_LED)
+		return -EINVAL;
+
+	mask = MSP_GPIO_MASK(offset);
+	bits = msp_led_cache;
+
+	bits &= ~mask;
+	if (value)
+		bits |= mask;
+	msp_led_cache = bits;
+
+	return dm355evm_msp_write(bits, DM355EVM_MSP_LED);
+}
+
+static void msp_gpio_set(struct gpio_chip *chip, unsigned offset, int value)
+{
+	msp_gpio_out(chip, offset, value);
+}
+
+static struct gpio_chip dm355evm_msp_gpio = {
+	.label			= "dm355evm_msp",
+	.owner			= THIS_MODULE,
+	.direction_input	= msp_gpio_in,
+	.get			= msp_gpio_get,
+	.direction_output	= msp_gpio_out,
+	.set			= msp_gpio_set,
+	.base			= -EINVAL,		/* dynamic assignment */
+	.ngpio			= ARRAY_SIZE(msp_gpios),
+	.can_sleep		= true,
+};
+
+/*----------------------------------------------------------------------*/
+
+static struct device *add_child(struct i2c_client *client, const char *name,
+		void *pdata, unsigned pdata_len,
+		bool can_wakeup, int irq)
+{
+	struct platform_device	*pdev;
+	int			status;
+
+	pdev = platform_device_alloc(name, -1);
+	if (!pdev) {
+		dev_dbg(&client->dev, "can't alloc dev\n");
+		status = -ENOMEM;
+		goto err;
+	}
+
+	device_init_wakeup(&pdev->dev, can_wakeup);
+	pdev->dev.parent = &client->dev;
+
+	if (pdata) {
+		status = platform_device_add_data(pdev, pdata, pdata_len);
+		if (status < 0) {
+			dev_dbg(&pdev->dev, "can't add platform_data\n");
+			goto err;
+		}
+	}
+
+	if (irq) {
+		struct resource r = {
+			.start = irq,
+			.flags = IORESOURCE_IRQ,
+		};
+
+		status = platform_device_add_resources(pdev, &r, 1);
+		if (status < 0) {
+			dev_dbg(&pdev->dev, "can't add irq\n");
+			goto err;
+		}
+	}
+
+	status = platform_device_add(pdev);
+
+err:
+	if (status < 0) {
+		platform_device_put(pdev);
+		dev_err(&client->dev, "can't add %s dev\n", name);
+		return ERR_PTR(status);
+	}
+	return &pdev->dev;
+}
+
+static int add_children(struct i2c_client *client)
+{
+	static const struct {
+		int offset;
+		char *label;
+	} config_inputs[] = {
+		/* 8 == right after the LEDs */
+		{ 8 + 0, "sw6_1", },
+		{ 8 + 1, "sw6_2", },
+		{ 8 + 2, "sw6_3", },
+		{ 8 + 3, "sw6_4", },
+		{ 8 + 4, "NTSC/nPAL", },
+	};
+
+	struct device	*child;
+	int		status;
+	int		i;
+
+	/* GPIO-ish stuff */
+	dm355evm_msp_gpio.dev = &client->dev;
+	status = gpiochip_add(&dm355evm_msp_gpio);
+	if (status < 0)
+		return status;
+
+	/* LED output */
+	if (msp_has_leds()) {
+#define GPIO_LED(l)	.name = l, .active_low = true
+		static struct gpio_led evm_leds[] = {
+			{ GPIO_LED("dm355evm::ds14"),
+				.default_trigger = "heartbeat", },
+			{ GPIO_LED("dm355evm::ds15"),
+				.default_trigger = "mmc0", },
+			{ GPIO_LED("dm355evm::ds16"),
+				/* could also be a CE-ATA drive */
+				.default_trigger = "mmc1", },
+			{ GPIO_LED("dm355evm::ds17"),
+				.default_trigger = "nand-disk", },
+			{ GPIO_LED("dm355evm::ds18"), },
+			{ GPIO_LED("dm355evm::ds19"), },
+			{ GPIO_LED("dm355evm::ds20"), },
+			{ GPIO_LED("dm355evm::ds21"), },
+		};
+#undef GPIO_LED
+
+		struct gpio_led_platform_data evm_led_data = {
+			.num_leds	= ARRAY_SIZE(evm_leds),
+			.leds		= evm_leds,
+		};
+
+		for (i = 0; i < ARRAY_SIZE(evm_leds); i++)
+			evm_leds[i].gpio = i + dm355evm_msp_gpio.base;
+
+		/* NOTE:  these are the only fully programmable LEDs
+		 * on the board, since GPIO-61/ds22 (and many signals
+		 * going to DC7) must be used for AEMIF address lines
+		 * unless the top 1 GB of NAND is unused...
+		 */
+		child = add_child(client, "leds-gpio",
+				&evm_led_data, sizeof(evm_led_data),
+				false, 0);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+	}
+
+	/* configuration inputs */
+	for (i = 0; i < ARRAY_SIZE(config_inputs); i++) {
+		int gpio = dm355evm_msp_gpio.base + config_inputs[i].offset;
+
+		gpio_request(gpio, config_inputs[i].label);
+		gpio_direction_input(gpio);
+
+		/* make it easy for userspace to see these */
+		gpio_export(gpio, false);
+	}
+
+	/* RTC is a 32 bit counter, no alarm */
+	if (msp_has_rtc()) {
+		child = add_child(client, "rtc-dm355evm",
+				NULL, 0, false, 0);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+	}
+
+	/* input from buttons and IR remote (uses the IRQ) */
+	if (msp_has_keyboard()) {
+		child = add_child(client, "dm355evm_keys",
+				NULL, 0, true, client->irq);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+	}
+
+	return 0;
+}
+
+/*----------------------------------------------------------------------*/
+
+static void dm355evm_command(unsigned command)
+{
+	int status;
+
+	status = dm355evm_msp_write(command, DM355EVM_MSP_COMMAND);
+	if (status < 0)
+		dev_err(&msp430->dev, "command %d failure %d\n",
+				command, status);
+}
+
+static void dm355evm_power_off(void)
+{
+	dm355evm_command(MSP_COMMAND_POWEROFF);
+}
+
+static int dm355evm_msp_remove(struct i2c_client *client)
+{
+	pm_power_off = NULL;
+	msp430 = NULL;
+	return 0;
+}
+
+static int
+dm355evm_msp_probe(struct i2c_client *client, const struct i2c_device_id *id)
+{
+	int		status;
+	const char	*video = msp_has_tvp() ? "TVP5146" : "imager";
+
+	if (msp430)
+		return -EBUSY;
+	msp430 = client;
+
+	/* display revision status; doubles as sanity check */
+	status = dm355evm_msp_read(DM355EVM_MSP_FIRMREV);
+	if (status < 0)
+		goto fail;
+	dev_info(&client->dev, "firmware v.%02X, %s as video-in\n",
+			status, video);
+
+	/* mux video input:  either tvp5146 or some external imager */
+	status = dm355evm_msp_write(msp_has_tvp() ? 0 : MSP_VIDEO_IMAGER,
+			DM355EVM_MSP_VIDEO_IN);
+	if (status < 0)
+		dev_warn(&client->dev, "error %d muxing %s as video-in\n",
+			status, video);
+
+	/* init LED cache, and turn off the LEDs */
+	msp_led_cache = 0xff;
+	dm355evm_msp_write(msp_led_cache, DM355EVM_MSP_LED);
+
+	/* export capabilities we support */
+	status = add_children(client);
+	if (status < 0)
+		goto fail;
+
+	/* PM hookup */
+	pm_power_off = dm355evm_power_off;
+
+	return 0;
+
+fail:
+	/* FIXME remove children ... */
+	dm355evm_msp_remove(client);
+	return status;
+}
+
+static const struct i2c_device_id dm355evm_msp_ids[] = {
+	{ "dm355evm_msp", 0 },
+	{ /* end of list */ },
+};
+MODULE_DEVICE_TABLE(i2c, dm355evm_msp_ids);
+
+static struct i2c_driver dm355evm_msp_driver = {
+	.driver.name	= "dm355evm_msp",
+	.id_table	= dm355evm_msp_ids,
+	.probe		= dm355evm_msp_probe,
+	.remove		= dm355evm_msp_remove,
+};
+
+static int __init dm355evm_msp_init(void)
+{
+	return i2c_add_driver(&dm355evm_msp_driver);
+}
+subsys_initcall(dm355evm_msp_init);
+
+static void __exit dm355evm_msp_exit(void)
+{
+	i2c_del_driver(&dm355evm_msp_driver);
+}
+module_exit(dm355evm_msp_exit);
+
+MODULE_DESCRIPTION("Interface to MSP430 firmware on DM355EVM");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/i2c/dm355evm_msp.h b/include/linux/i2c/dm355evm_msp.h
new file mode 100644
index 000000000000..372470350fab
--- /dev/null
+++ b/include/linux/i2c/dm355evm_msp.h
@@ -0,0 +1,79 @@
+/*
+ * dm355evm_msp.h - support MSP430 microcontroller on DM355EVM board
+ */
+#ifndef __LINUX_I2C_DM355EVM_MSP
+#define __LINUX_I2C_DM355EVM_MSP
+
+/*
+ * Written against Spectrum's writeup for the A4 firmware revision,
+ * and tweaked to match source and rev D2 schematics by removing CPLD
+ * and NOR flash hooks (which were last appropriate in rev B boards).
+ *
+ * Note that the firmware supports a flavor of write posting ... to be
+ * sure a write completes, issue another read or write.
+ */
+
+/* utilities to access "registers" emulated by msp430 firmware */
+extern int dm355evm_msp_write(u8 value, u8 reg);
+extern int dm355evm_msp_read(u8 reg);
+
+
+/* command/control registers */
+#define DM355EVM_MSP_COMMAND		0x00
+#	define MSP_COMMAND_NULL		0
+#	define MSP_COMMAND_RESET_COLD	1
+#	define MSP_COMMAND_RESET_WARM	2
+#	define MSP_COMMAND_RESET_WARM_I	3
+#	define MSP_COMMAND_POWEROFF	4
+#	define MSP_COMMAND_IR_REINIT	5
+#define DM355EVM_MSP_STATUS		0x01
+#	define MSP_STATUS_BAD_OFFSET	BIT(0)
+#	define MSP_STATUS_BAD_COMMAND	BIT(1)
+#	define MSP_STATUS_POWER_ERROR	BIT(2)
+#	define MSP_STATUS_RXBUF_OVERRUN	BIT(3)
+#define DM355EVM_MSP_RESET		0x02	/* 0 bits == in reset */
+#	define MSP_RESET_DC5		BIT(0)
+#	define MSP_RESET_TVP5154	BIT(2)
+#	define MSP_RESET_IMAGER		BIT(3)
+#	define MSP_RESET_ETHERNET	BIT(4)
+#	define MSP_RESET_SYS		BIT(5)
+#	define MSP_RESET_AIC33		BIT(7)
+
+/* GPIO registers ... bit patterns mostly match the source MSP ports */
+#define DM355EVM_MSP_LED		0x03	/* active low (MSP P4) */
+#define DM355EVM_MSP_SWITCH1		0x04	/* (MSP P5, masked) */
+#	define MSP_SWITCH1_SW6_1	BIT(0)
+#	define MSP_SWITCH1_SW6_2	BIT(1)
+#	define MSP_SWITCH1_SW6_3	BIT(2)
+#	define MSP_SWITCH1_SW6_4	BIT(3)
+#	define MSP_SWITCH1_J1		BIT(4)	/* NTSC/PAL */
+#	define MSP_SWITCH1_MSP_INT	BIT(5)	/* active low */
+#define DM355EVM_MSP_SWITCH2		0x05	/* (MSP P6, masked) */
+#	define MSP_SWITCH2_SW10		BIT(3)
+#	define MSP_SWITCH2_SW11		BIT(4)
+#	define MSP_SWITCH2_SW12		BIT(5)
+#	define MSP_SWITCH2_SW13		BIT(6)
+#	define MSP_SWITCH2_SW14		BIT(7)
+#define DM355EVM_MSP_SDMMC		0x06	/* (MSP P2, masked) */
+#	define MSP_SDMMC_0_WP		BIT(1)
+#	define MSP_SDMMC_0_CD		BIT(2)	/* active low */
+#	define MSP_SDMMC_1_WP		BIT(3)
+#	define MSP_SDMMC_1_CD		BIT(4)	/* active low */
+#define DM355EVM_MSP_FIRMREV		0x07	/* not a GPIO (out of order) */
+#define DM355EVM_MSP_VIDEO_IN		0x08	/* (MSP P3, masked) */
+#	define MSP_VIDEO_IMAGER		BIT(7)	/* low == tvp5146 */
+
+/* power supply registers are currently omitted */
+
+/* RTC registers */
+#define DM355EVM_MSP_RTC_0		0x12	/* LSB */
+#define DM355EVM_MSP_RTC_1		0x13
+#define DM355EVM_MSP_RTC_2		0x14
+#define DM355EVM_MSP_RTC_3		0x15	/* MSB */
+
+/* input event queue registers; code == ((HIGH << 8) | LOW) */
+#define DM355EVM_MSP_INPUT_COUNT	0x16	/* decrement by reading LOW */
+#define DM355EVM_MSP_INPUT_HIGH		0x17
+#define DM355EVM_MSP_INPUT_LOW		0x18
+
+#endif /* __LINUX_I2C_DM355EVM_MSP */
-- 
cgit v1.2.3


From f3298dc4f2277874d40cb4fc3a6e277317d6603b Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 10 Dec 2008 03:16:51 -0500
Subject: sanitize audit_socketcall

* don't bother with allocations
* now that it can't fail, make it return void

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/audit.h |  4 ++--
 kernel/auditsc.c      | 66 +++++++++++++++++++++++++++++----------------------
 net/socket.c          |  4 +---
 3 files changed, 41 insertions(+), 33 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index 26c4f6f65a46..466a953d4bf6 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -446,7 +446,7 @@ extern void audit_log_task_context(struct audit_buffer *ab);
 extern int __audit_ipc_obj(struct kern_ipc_perm *ipcp);
 extern int __audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode);
 extern int audit_bprm(struct linux_binprm *bprm);
-extern int audit_socketcall(int nargs, unsigned long *args);
+extern void audit_socketcall(int nargs, unsigned long *args);
 extern int audit_sockaddr(int len, void *addr);
 extern int __audit_fd_pair(int fd1, int fd2);
 extern int audit_set_macxattr(const char *name);
@@ -549,7 +549,7 @@ extern int audit_signals;
 #define audit_ipc_obj(i) ({ 0; })
 #define audit_ipc_set_perm(q,u,g,m) ({ 0; })
 #define audit_bprm(p) ({ 0; })
-#define audit_socketcall(n,a) ({ 0; })
+#define audit_socketcall(n,a) ((void)0)
 #define audit_fd_pair(n,a) ({ 0; })
 #define audit_sockaddr(len, addr) ({ 0; })
 #define audit_set_macxattr(n) do { ; } while (0)
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index c2e43ebb1b68..5cda66466e14 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -168,12 +168,6 @@ struct audit_aux_data_execve {
 	struct mm_struct *mm;
 };
 
-struct audit_aux_data_socketcall {
-	struct audit_aux_data	d;
-	int			nargs;
-	unsigned long		args[0];
-};
-
 struct audit_aux_data_fd_pair {
 	struct	audit_aux_data d;
 	int	fd[2];
@@ -247,6 +241,14 @@ struct audit_context {
 	struct audit_tree_refs *trees, *first_trees;
 	int tree_count;
 
+	int type;
+	union {
+		struct {
+			int nargs;
+			long args[6];
+		} socketcall;
+	};
+
 #if AUDIT_DEBUG
 	int		    put_count;
 	int		    ino_count;
@@ -1226,6 +1228,27 @@ static void audit_log_fcaps(struct audit_buffer *ab, struct audit_names *name)
 		audit_log_format(ab, " cap_fe=%d cap_fver=%x", name->fcap.fE, name->fcap_ver);
 }
 
+static void show_special(struct audit_context *context)
+{
+	struct audit_buffer *ab;
+	int i;
+
+	ab = audit_log_start(context, GFP_KERNEL, context->type);
+	if (!ab)
+		return;
+
+	switch (context->type) {
+	case AUDIT_SOCKETCALL: {
+		int nargs = context->socketcall.nargs;
+		audit_log_format(ab, "nargs=%d", nargs);
+		for (i = 0; i < nargs; i++)
+			audit_log_format(ab, " a%d=%lx", i,
+				context->socketcall.args[i]);
+		break; }
+	}
+	audit_log_end(ab);
+}
+
 static void audit_log_exit(struct audit_context *context, struct task_struct *tsk)
 {
 	const struct cred *cred;
@@ -1372,13 +1395,6 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 			audit_log_execve_info(context, &ab, axi);
 			break; }
 
-		case AUDIT_SOCKETCALL: {
-			struct audit_aux_data_socketcall *axs = (void *)aux;
-			audit_log_format(ab, "nargs=%d", axs->nargs);
-			for (i=0; i<axs->nargs; i++)
-				audit_log_format(ab, " a%d=%lx", i, axs->args[i]);
-			break; }
-
 		case AUDIT_FD_PAIR: {
 			struct audit_aux_data_fd_pair *axs = (void *)aux;
 			audit_log_format(ab, "fd0=%d fd1=%d", axs->fd[0], axs->fd[1]);
@@ -1410,6 +1426,9 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 		audit_log_end(ab);
 	}
 
+	if (context->type)
+		show_special(context);
+
 	if (context->sockaddr_len) {
 		ab = audit_log_start(context, GFP_KERNEL, AUDIT_SOCKADDR);
 		if (ab) {
@@ -1689,6 +1708,7 @@ void audit_syscall_exit(int valid, long return_code)
 		context->target_pid = 0;
 		context->target_sid = 0;
 		context->sockaddr_len = 0;
+		context->type = 0;
 		kfree(context->filterkey);
 		context->filterkey = NULL;
 		tsk->audit_context = context;
@@ -2406,27 +2426,17 @@ int audit_bprm(struct linux_binprm *bprm)
  * @nargs: number of args
  * @args: args array
  *
- * Returns 0 for success or NULL context or < 0 on error.
  */
-int audit_socketcall(int nargs, unsigned long *args)
+void audit_socketcall(int nargs, unsigned long *args)
 {
-	struct audit_aux_data_socketcall *ax;
 	struct audit_context *context = current->audit_context;
 
 	if (likely(!context || context->dummy))
-		return 0;
-
-	ax = kmalloc(sizeof(*ax) + nargs * sizeof(unsigned long), GFP_KERNEL);
-	if (!ax)
-		return -ENOMEM;
-
-	ax->nargs = nargs;
-	memcpy(ax->args, args, nargs * sizeof(unsigned long));
+		return;
 
-	ax->d.type = AUDIT_SOCKETCALL;
-	ax->d.next = context->aux;
-	context->aux = (void *)ax;
-	return 0;
+	context->type = AUDIT_SOCKETCALL;
+	context->socketcall.nargs = nargs;
+	memcpy(context->socketcall.args, args, nargs * sizeof(unsigned long));
 }
 
 /**
diff --git a/net/socket.c b/net/socket.c
index 2c730fc718ab..b41a92093e40 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -2065,9 +2065,7 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args)
 	if (copy_from_user(a, args, nargs[call]))
 		return -EFAULT;
 
-	err = audit_socketcall(nargs[call] / sizeof(unsigned long), a);
-	if (err)
-		return err;
+	audit_socketcall(nargs[call] / sizeof(unsigned long), a);
 
 	a0 = a[0];
 	a1 = a[1];
-- 
cgit v1.2.3


From a33e6751003c5ade603737d828b1519d980ce392 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 10 Dec 2008 03:40:06 -0500
Subject: sanitize audit_ipc_obj()

* get rid of allocations
* make it return void
* simplify callers

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/audit.h |  9 +++---
 ipc/shm.c             |  4 +--
 ipc/util.c            |  9 ++----
 kernel/auditsc.c      | 88 ++++++++++++++++++++++-----------------------------
 4 files changed, 45 insertions(+), 65 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index 466a953d4bf6..f8578b9088e1 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -443,7 +443,7 @@ extern int  audit_set_loginuid(struct task_struct *task, uid_t loginuid);
 #define audit_get_loginuid(t) ((t)->loginuid)
 #define audit_get_sessionid(t) ((t)->sessionid)
 extern void audit_log_task_context(struct audit_buffer *ab);
-extern int __audit_ipc_obj(struct kern_ipc_perm *ipcp);
+extern void __audit_ipc_obj(struct kern_ipc_perm *ipcp);
 extern int __audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode);
 extern int audit_bprm(struct linux_binprm *bprm);
 extern void audit_socketcall(int nargs, unsigned long *args);
@@ -460,11 +460,10 @@ extern int __audit_log_bprm_fcaps(struct linux_binprm *bprm,
 				  const struct cred *old);
 extern int __audit_log_capset(pid_t pid, const struct cred *new, const struct cred *old);
 
-static inline int audit_ipc_obj(struct kern_ipc_perm *ipcp)
+static inline void audit_ipc_obj(struct kern_ipc_perm *ipcp)
 {
 	if (unlikely(!audit_dummy_context()))
-		return __audit_ipc_obj(ipcp);
-	return 0;
+		__audit_ipc_obj(ipcp);
 }
 static inline int audit_fd_pair(int fd1, int fd2)
 {
@@ -546,7 +545,7 @@ extern int audit_signals;
 #define audit_get_loginuid(t) (-1)
 #define audit_get_sessionid(t) (-1)
 #define audit_log_task_context(b) do { ; } while (0)
-#define audit_ipc_obj(i) ({ 0; })
+#define audit_ipc_obj(i) ((void)0)
 #define audit_ipc_set_perm(q,u,g,m) ({ 0; })
 #define audit_bprm(p) ({ 0; })
 #define audit_socketcall(n,a) ((void)0)
diff --git a/ipc/shm.c b/ipc/shm.c
index 38a055758a9b..57dd50046cef 100644
--- a/ipc/shm.c
+++ b/ipc/shm.c
@@ -747,9 +747,7 @@ asmlinkage long sys_shmctl(int shmid, int cmd, struct shmid_ds __user *buf)
 			goto out;
 		}
 
-		err = audit_ipc_obj(&(shp->shm_perm));
-		if (err)
-			goto out_unlock;
+		audit_ipc_obj(&(shp->shm_perm));
 
 		if (!capable(CAP_IPC_LOCK)) {
 			uid_t euid = current_euid();
diff --git a/ipc/util.c b/ipc/util.c
index 5a1808c774a2..579552abd50a 100644
--- a/ipc/util.c
+++ b/ipc/util.c
@@ -624,10 +624,9 @@ void ipc_rcu_putref(void *ptr)
 int ipcperms (struct kern_ipc_perm *ipcp, short flag)
 {	/* flag will most probably be 0 or S_...UGO from <linux/stat.h> */
 	uid_t euid = current_euid();
-	int requested_mode, granted_mode, err;
+	int requested_mode, granted_mode;
 
-	if (unlikely((err = audit_ipc_obj(ipcp))))
-		return err;
+	audit_ipc_obj(ipcp);
 	requested_mode = (flag >> 6) | (flag >> 3) | flag;
 	granted_mode = ipcp->mode;
 	if (euid == ipcp->cuid ||
@@ -803,9 +802,7 @@ struct kern_ipc_perm *ipcctl_pre_down(struct ipc_ids *ids, int id, int cmd,
 		goto out_up;
 	}
 
-	err = audit_ipc_obj(ipcp);
-	if (err)
-		goto out_unlock;
+	audit_ipc_obj(ipcp);
 
 	if (cmd == IPC_SET) {
 		err = audit_ipc_set_perm(extra_perm, perm->uid,
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 5cda66466e14..73504313264f 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -247,6 +247,12 @@ struct audit_context {
 			int nargs;
 			long args[6];
 		} socketcall;
+		struct {
+			uid_t			uid;
+			gid_t			gid;
+			mode_t			mode;
+			u32			osid;
+		} ipc;
 	};
 
 #if AUDIT_DEBUG
@@ -605,19 +611,12 @@ static int audit_filter_rules(struct task_struct *tsk,
 					}
 				}
 				/* Find ipc objects that match */
-				if (ctx) {
-					struct audit_aux_data *aux;
-					for (aux = ctx->aux; aux;
-					     aux = aux->next) {
-						if (aux->type == AUDIT_IPC) {
-							struct audit_aux_data_ipcctl *axi = (void *)aux;
-							if (security_audit_rule_match(axi->osid, f->type, f->op, f->lsm_rule, ctx)) {
-								++result;
-								break;
-							}
-						}
-					}
-				}
+				if (!ctx || ctx->type != AUDIT_IPC)
+					break;
+				if (security_audit_rule_match(ctx->ipc.osid,
+							      f->type, f->op,
+							      f->lsm_rule, ctx))
+					++result;
 			}
 			break;
 		case AUDIT_ARG0:
@@ -1228,7 +1227,7 @@ static void audit_log_fcaps(struct audit_buffer *ab, struct audit_names *name)
 		audit_log_format(ab, " cap_fe=%d cap_fver=%x", name->fcap.fE, name->fcap_ver);
 }
 
-static void show_special(struct audit_context *context)
+static void show_special(struct audit_context *context, int *call_panic)
 {
 	struct audit_buffer *ab;
 	int i;
@@ -1245,6 +1244,23 @@ static void show_special(struct audit_context *context)
 			audit_log_format(ab, " a%d=%lx", i,
 				context->socketcall.args[i]);
 		break; }
+	case AUDIT_IPC: {
+		u32 osid = context->ipc.osid;
+
+		audit_log_format(ab, "ouid=%u ogid=%u mode=%#o",
+			 context->ipc.uid, context->ipc.gid, context->ipc.mode);
+		if (osid) {
+			char *ctx = NULL;
+			u32 len;
+			if (security_secid_to_secctx(osid, &ctx, &len)) {
+				audit_log_format(ab, " osid=%u", osid);
+				*call_panic = 1;
+			} else {
+				audit_log_format(ab, " obj=%s", ctx);
+				security_release_secctx(ctx, len);
+			}
+		}
+		break; }
 	}
 	audit_log_end(ab);
 }
@@ -1363,26 +1379,6 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 				axi->mqstat.mq_msgsize, axi->mqstat.mq_curmsgs);
 			break; }
 
-		case AUDIT_IPC: {
-			struct audit_aux_data_ipcctl *axi = (void *)aux;
-			audit_log_format(ab, 
-				 "ouid=%u ogid=%u mode=%#o",
-				 axi->uid, axi->gid, axi->mode);
-			if (axi->osid != 0) {
-				char *ctx = NULL;
-				u32 len;
-				if (security_secid_to_secctx(
-						axi->osid, &ctx, &len)) {
-					audit_log_format(ab, " osid=%u",
-							axi->osid);
-					call_panic = 1;
-				} else {
-					audit_log_format(ab, " obj=%s", ctx);
-					security_release_secctx(ctx, len);
-				}
-			}
-			break; }
-
 		case AUDIT_IPC_SET_PERM: {
 			struct audit_aux_data_ipcctl *axi = (void *)aux;
 			audit_log_format(ab,
@@ -1427,7 +1423,7 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 	}
 
 	if (context->type)
-		show_special(context);
+		show_special(context, &call_panic);
 
 	if (context->sockaddr_len) {
 		ab = audit_log_start(context, GFP_KERNEL, AUDIT_SOCKADDR);
@@ -2349,25 +2345,15 @@ int __audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat)
  * audit_ipc_obj - record audit data for ipc object
  * @ipcp: ipc permissions
  *
- * Returns 0 for success or NULL context or < 0 on error.
  */
-int __audit_ipc_obj(struct kern_ipc_perm *ipcp)
+void __audit_ipc_obj(struct kern_ipc_perm *ipcp)
 {
-	struct audit_aux_data_ipcctl *ax;
 	struct audit_context *context = current->audit_context;
-
-	ax = kmalloc(sizeof(*ax), GFP_ATOMIC);
-	if (!ax)
-		return -ENOMEM;
-
-	ax->uid = ipcp->uid;
-	ax->gid = ipcp->gid;
-	ax->mode = ipcp->mode;
-	security_ipc_getsecid(ipcp, &ax->osid);
-	ax->d.type = AUDIT_IPC;
-	ax->d.next = context->aux;
-	context->aux = (void *)ax;
-	return 0;
+	context->ipc.uid = ipcp->uid;
+	context->ipc.gid = ipcp->gid;
+	context->ipc.mode = ipcp->mode;
+	security_ipc_getsecid(ipcp, &context->ipc.osid);
+	context->type = AUDIT_IPC;
 }
 
 /**
-- 
cgit v1.2.3


From e816f370cbadd2afea9f1a42f232d0636137d563 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 10 Dec 2008 03:47:15 -0500
Subject: sanitize audit_ipc_set_perm()

* get rid of allocations
* make it return void
* simplify callers

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/audit.h |  9 ++++----
 ipc/util.c            |  9 ++------
 kernel/auditsc.c      | 59 +++++++++++++++++++++++----------------------------
 3 files changed, 32 insertions(+), 45 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index f8578b9088e1..b7abfe0d6737 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -444,7 +444,7 @@ extern int  audit_set_loginuid(struct task_struct *task, uid_t loginuid);
 #define audit_get_sessionid(t) ((t)->sessionid)
 extern void audit_log_task_context(struct audit_buffer *ab);
 extern void __audit_ipc_obj(struct kern_ipc_perm *ipcp);
-extern int __audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode);
+extern void __audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode);
 extern int audit_bprm(struct linux_binprm *bprm);
 extern void audit_socketcall(int nargs, unsigned long *args);
 extern int audit_sockaddr(int len, void *addr);
@@ -471,11 +471,10 @@ static inline int audit_fd_pair(int fd1, int fd2)
 		return __audit_fd_pair(fd1, fd2);
 	return 0;
 }
-static inline int audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode)
+static inline void audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode)
 {
 	if (unlikely(!audit_dummy_context()))
-		return __audit_ipc_set_perm(qbytes, uid, gid, mode);
-	return 0;
+		__audit_ipc_set_perm(qbytes, uid, gid, mode);
 }
 static inline int audit_mq_open(int oflag, mode_t mode, struct mq_attr __user *u_attr)
 {
@@ -546,7 +545,7 @@ extern int audit_signals;
 #define audit_get_sessionid(t) (-1)
 #define audit_log_task_context(b) do { ; } while (0)
 #define audit_ipc_obj(i) ((void)0)
-#define audit_ipc_set_perm(q,u,g,m) ({ 0; })
+#define audit_ipc_set_perm(q,u,g,m) ((void)0)
 #define audit_bprm(p) ({ 0; })
 #define audit_socketcall(n,a) ((void)0)
 #define audit_fd_pair(n,a) ({ 0; })
diff --git a/ipc/util.c b/ipc/util.c
index 579552abd50a..7585a72e259b 100644
--- a/ipc/util.c
+++ b/ipc/util.c
@@ -803,13 +803,9 @@ struct kern_ipc_perm *ipcctl_pre_down(struct ipc_ids *ids, int id, int cmd,
 	}
 
 	audit_ipc_obj(ipcp);
-
-	if (cmd == IPC_SET) {
-		err = audit_ipc_set_perm(extra_perm, perm->uid,
+	if (cmd == IPC_SET)
+		audit_ipc_set_perm(extra_perm, perm->uid,
 					 perm->gid, perm->mode);
-		if (err)
-			goto out_unlock;
-	}
 
 	euid = current_euid();
 	if (euid == ipcp->cuid ||
@@ -817,7 +813,6 @@ struct kern_ipc_perm *ipcctl_pre_down(struct ipc_ids *ids, int id, int cmd,
 		return ipcp;
 
 	err = -EPERM;
-out_unlock:
 	ipc_unlock(ipcp);
 out_up:
 	up_write(&ids->rw_mutex);
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 73504313264f..fbed62e05bce 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -151,16 +151,6 @@ struct audit_aux_data_mq_getsetattr {
 	struct mq_attr 		mqstat;
 };
 
-struct audit_aux_data_ipcctl {
-	struct audit_aux_data	d;
-	struct ipc_perm		p;
-	unsigned long		qbytes;
-	uid_t			uid;
-	gid_t			gid;
-	mode_t			mode;
-	u32			osid;
-};
-
 struct audit_aux_data_execve {
 	struct audit_aux_data	d;
 	int argc;
@@ -252,6 +242,11 @@ struct audit_context {
 			gid_t			gid;
 			mode_t			mode;
 			u32			osid;
+			int			has_perm;
+			uid_t			perm_uid;
+			gid_t			perm_gid;
+			mode_t			perm_mode;
+			unsigned long		qbytes;
 		} ipc;
 	};
 
@@ -1260,6 +1255,19 @@ static void show_special(struct audit_context *context, int *call_panic)
 				security_release_secctx(ctx, len);
 			}
 		}
+		if (context->ipc.has_perm) {
+			audit_log_end(ab);
+			ab = audit_log_start(context, GFP_KERNEL,
+					     AUDIT_IPC_SET_PERM);
+			audit_log_format(ab,
+				"qbytes=%lx ouid=%u ogid=%u mode=%#o",
+				context->ipc.qbytes,
+				context->ipc.perm_uid,
+				context->ipc.perm_gid,
+				context->ipc.perm_mode);
+			if (!ab)
+				return;
+		}
 		break; }
 	}
 	audit_log_end(ab);
@@ -1379,13 +1387,6 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 				axi->mqstat.mq_msgsize, axi->mqstat.mq_curmsgs);
 			break; }
 
-		case AUDIT_IPC_SET_PERM: {
-			struct audit_aux_data_ipcctl *axi = (void *)aux;
-			audit_log_format(ab,
-				"qbytes=%lx ouid=%u ogid=%u mode=%#o",
-				axi->qbytes, axi->uid, axi->gid, axi->mode);
-			break; }
-
 		case AUDIT_EXECVE: {
 			struct audit_aux_data_execve *axi = (void *)aux;
 			audit_log_execve_info(context, &ab, axi);
@@ -2352,6 +2353,7 @@ void __audit_ipc_obj(struct kern_ipc_perm *ipcp)
 	context->ipc.uid = ipcp->uid;
 	context->ipc.gid = ipcp->gid;
 	context->ipc.mode = ipcp->mode;
+	context->ipc.has_perm = 0;
 	security_ipc_getsecid(ipcp, &context->ipc.osid);
 	context->type = AUDIT_IPC;
 }
@@ -2363,26 +2365,17 @@ void __audit_ipc_obj(struct kern_ipc_perm *ipcp)
  * @gid: msgq group id
  * @mode: msgq mode (permissions)
  *
- * Returns 0 for success or NULL context or < 0 on error.
+ * Called only after audit_ipc_obj().
  */
-int __audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode)
+void __audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode)
 {
-	struct audit_aux_data_ipcctl *ax;
 	struct audit_context *context = current->audit_context;
 
-	ax = kmalloc(sizeof(*ax), GFP_ATOMIC);
-	if (!ax)
-		return -ENOMEM;
-
-	ax->qbytes = qbytes;
-	ax->uid = uid;
-	ax->gid = gid;
-	ax->mode = mode;
-
-	ax->d.type = AUDIT_IPC_SET_PERM;
-	ax->d.next = context->aux;
-	context->aux = (void *)ax;
-	return 0;
+	context->ipc.qbytes = qbytes;
+	context->ipc.perm_uid = uid;
+	context->ipc.perm_gid = gid;
+	context->ipc.perm_mode = mode;
+	context->ipc.has_perm = 1;
 }
 
 int audit_bprm(struct linux_binprm *bprm)
-- 
cgit v1.2.3


From 7392906ea915b9a2c14dea32b3604b4e178f82f7 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 10 Dec 2008 06:58:59 -0500
Subject: sanitize audit_mq_getsetattr()

* get rid of allocations
* make it return void
* don't duplicate parts of audit_dummy_context()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/audit.h |  9 ++++-----
 ipc/mqueue.c          |  6 +-----
 kernel/auditsc.c      | 54 ++++++++++++++++-----------------------------------
 3 files changed, 22 insertions(+), 47 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index b7abfe0d6737..b7707e577b80 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -454,7 +454,7 @@ extern int __audit_mq_open(int oflag, mode_t mode, struct mq_attr __user *u_attr
 extern int __audit_mq_timedsend(mqd_t mqdes, size_t msg_len, unsigned int msg_prio, const struct timespec __user *u_abs_timeout);
 extern int __audit_mq_timedreceive(mqd_t mqdes, size_t msg_len, unsigned int __user *u_msg_prio, const struct timespec __user *u_abs_timeout);
 extern int __audit_mq_notify(mqd_t mqdes, const struct sigevent __user *u_notification);
-extern int __audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat);
+extern void __audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat);
 extern int __audit_log_bprm_fcaps(struct linux_binprm *bprm,
 				  const struct cred *new,
 				  const struct cred *old);
@@ -500,11 +500,10 @@ static inline int audit_mq_notify(mqd_t mqdes, const struct sigevent __user *u_n
 		return __audit_mq_notify(mqdes, u_notification);
 	return 0;
 }
-static inline int audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat)
+static inline void audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat)
 {
 	if (unlikely(!audit_dummy_context()))
-		return __audit_mq_getsetattr(mqdes, mqstat);
-	return 0;
+		__audit_mq_getsetattr(mqdes, mqstat);
 }
 
 static inline int audit_log_bprm_fcaps(struct linux_binprm *bprm,
@@ -555,7 +554,7 @@ extern int audit_signals;
 #define audit_mq_timedsend(d,l,p,t) ({ 0; })
 #define audit_mq_timedreceive(d,l,p,t) ({ 0; })
 #define audit_mq_notify(d,n) ({ 0; })
-#define audit_mq_getsetattr(d,s) ({ 0; })
+#define audit_mq_getsetattr(d,s) ((void)0)
 #define audit_log_bprm_fcaps(b, ncr, ocr) ({ 0; })
 #define audit_log_capset(pid, ncr, ocr) ({ 0; })
 #define audit_ptrace(t) ((void)0)
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index d9393f8e4c3e..7563611c6615 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -1150,11 +1150,7 @@ asmlinkage long sys_mq_getsetattr(mqd_t mqdes,
 	omqstat = info->attr;
 	omqstat.mq_flags = filp->f_flags & O_NONBLOCK;
 	if (u_mqstat) {
-		ret = audit_mq_getsetattr(mqdes, &mqstat);
-		if (ret != 0) {
-			spin_unlock(&info->lock);
-			goto out_fput;
-		}
+		audit_mq_getsetattr(mqdes, &mqstat);
 		if (mqstat.mq_flags & O_NONBLOCK)
 			filp->f_flags |= O_NONBLOCK;
 		else
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index fbed62e05bce..c50178c7e245 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -145,12 +145,6 @@ struct audit_aux_data_mq_notify {
 	struct sigevent 	notification;
 };
 
-struct audit_aux_data_mq_getsetattr {
-	struct audit_aux_data	d;
-	mqd_t			mqdes;
-	struct mq_attr 		mqstat;
-};
-
 struct audit_aux_data_execve {
 	struct audit_aux_data	d;
 	int argc;
@@ -248,6 +242,10 @@ struct audit_context {
 			mode_t			perm_mode;
 			unsigned long		qbytes;
 		} ipc;
+		struct {
+			mqd_t			mqdes;
+			struct mq_attr 		mqstat;
+		} mq_getsetattr;
 	};
 
 #if AUDIT_DEBUG
@@ -1269,6 +1267,15 @@ static void show_special(struct audit_context *context, int *call_panic)
 				return;
 		}
 		break; }
+	case AUDIT_MQ_GETSETATTR: {
+		struct mq_attr *attr = &context->mq_getsetattr.mqstat;
+		audit_log_format(ab,
+			"mqdes=%d mq_flags=0x%lx mq_maxmsg=%ld mq_msgsize=%ld "
+			"mq_curmsgs=%ld ",
+			context->mq_getsetattr.mqdes,
+			attr->mq_flags, attr->mq_maxmsg,
+			attr->mq_msgsize, attr->mq_curmsgs);
+		break; }
 	}
 	audit_log_end(ab);
 }
@@ -1377,16 +1384,6 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 				axi->notification.sigev_signo);
 			break; }
 
-		case AUDIT_MQ_GETSETATTR: {
-			struct audit_aux_data_mq_getsetattr *axi = (void *)aux;
-			audit_log_format(ab,
-				"mqdes=%d mq_flags=0x%lx mq_maxmsg=%ld mq_msgsize=%ld "
-				"mq_curmsgs=%ld ",
-				axi->mqdes,
-				axi->mqstat.mq_flags, axi->mqstat.mq_maxmsg,
-				axi->mqstat.mq_msgsize, axi->mqstat.mq_curmsgs);
-			break; }
-
 		case AUDIT_EXECVE: {
 			struct audit_aux_data_execve *axi = (void *)aux;
 			audit_log_execve_info(context, &ab, axi);
@@ -2316,30 +2313,13 @@ int __audit_mq_notify(mqd_t mqdes, const struct sigevent __user *u_notification)
  * @mqdes: MQ descriptor
  * @mqstat: MQ flags
  *
- * Returns 0 for success or NULL context or < 0 on error.
  */
-int __audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat)
+void __audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat)
 {
-	struct audit_aux_data_mq_getsetattr *ax;
 	struct audit_context *context = current->audit_context;
-
-	if (!audit_enabled)
-		return 0;
-
-	if (likely(!context))
-		return 0;
-
-	ax = kmalloc(sizeof(*ax), GFP_ATOMIC);
-	if (!ax)
-		return -ENOMEM;
-
-	ax->mqdes = mqdes;
-	ax->mqstat = *mqstat;
-
-	ax->d.type = AUDIT_MQ_GETSETATTR;
-	ax->d.next = context->aux;
-	context->aux = (void *)ax;
-	return 0;
+	context->mq_getsetattr.mqdes = mqdes;
+	context->mq_getsetattr.mqstat = *mqstat;
+	context->type = AUDIT_MQ_GETSETATTR;
 }
 
 /**
-- 
cgit v1.2.3


From 20114f71b27cafeb7c7e41d2b0f0b68c3fbb022b Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 10 Dec 2008 07:16:12 -0500
Subject: sanitize audit_mq_notify()

* don't copy_from_user() twice
* don't bother with allocations
* don't duplicate parts of audit_dummy_context()
* make it return void

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/audit.h |  9 ++++-----
 ipc/mqueue.c          | 14 ++++++-------
 kernel/auditsc.c      | 56 +++++++++++++++------------------------------------
 3 files changed, 27 insertions(+), 52 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index b7707e577b80..8101d2c4a995 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -453,7 +453,7 @@ extern int audit_set_macxattr(const char *name);
 extern int __audit_mq_open(int oflag, mode_t mode, struct mq_attr __user *u_attr);
 extern int __audit_mq_timedsend(mqd_t mqdes, size_t msg_len, unsigned int msg_prio, const struct timespec __user *u_abs_timeout);
 extern int __audit_mq_timedreceive(mqd_t mqdes, size_t msg_len, unsigned int __user *u_msg_prio, const struct timespec __user *u_abs_timeout);
-extern int __audit_mq_notify(mqd_t mqdes, const struct sigevent __user *u_notification);
+extern void __audit_mq_notify(mqd_t mqdes, const struct sigevent *notification);
 extern void __audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat);
 extern int __audit_log_bprm_fcaps(struct linux_binprm *bprm,
 				  const struct cred *new,
@@ -494,11 +494,10 @@ static inline int audit_mq_timedreceive(mqd_t mqdes, size_t msg_len, unsigned in
 		return __audit_mq_timedreceive(mqdes, msg_len, u_msg_prio, u_abs_timeout);
 	return 0;
 }
-static inline int audit_mq_notify(mqd_t mqdes, const struct sigevent __user *u_notification)
+static inline void audit_mq_notify(mqd_t mqdes, const struct sigevent *notification)
 {
 	if (unlikely(!audit_dummy_context()))
-		return __audit_mq_notify(mqdes, u_notification);
-	return 0;
+		__audit_mq_notify(mqdes, notification);
 }
 static inline void audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat)
 {
@@ -553,7 +552,7 @@ extern int audit_signals;
 #define audit_mq_open(o,m,a) ({ 0; })
 #define audit_mq_timedsend(d,l,p,t) ({ 0; })
 #define audit_mq_timedreceive(d,l,p,t) ({ 0; })
-#define audit_mq_notify(d,n) ({ 0; })
+#define audit_mq_notify(d,n) ((void)0)
 #define audit_mq_getsetattr(d,s) ((void)0)
 #define audit_log_bprm_fcaps(b, ncr, ocr) ({ 0; })
 #define audit_log_capset(pid, ncr, ocr) ({ 0; })
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 7563611c6615..e7b2f68f8d77 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -1003,17 +1003,17 @@ asmlinkage long sys_mq_notify(mqd_t mqdes,
 	struct mqueue_inode_info *info;
 	struct sk_buff *nc;
 
-	ret = audit_mq_notify(mqdes, u_notification);
-	if (ret != 0)
-		return ret;
-
-	nc = NULL;
-	sock = NULL;
-	if (u_notification != NULL) {
+	if (u_notification) {
 		if (copy_from_user(&notification, u_notification,
 					sizeof(struct sigevent)))
 			return -EFAULT;
+	}
+
+	audit_mq_notify(mqdes, u_notification ? &notification : NULL);
 
+	nc = NULL;
+	sock = NULL;
+	if (u_notification != NULL) {
 		if (unlikely(notification.sigev_notify != SIGEV_NONE &&
 			     notification.sigev_notify != SIGEV_SIGNAL &&
 			     notification.sigev_notify != SIGEV_THREAD))
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index c50178c7e245..3ece960de894 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -139,12 +139,6 @@ struct audit_aux_data_mq_sendrecv {
 	struct timespec		abs_timeout;
 };
 
-struct audit_aux_data_mq_notify {
-	struct audit_aux_data	d;
-	mqd_t			mqdes;
-	struct sigevent 	notification;
-};
-
 struct audit_aux_data_execve {
 	struct audit_aux_data	d;
 	int argc;
@@ -246,6 +240,10 @@ struct audit_context {
 			mqd_t			mqdes;
 			struct mq_attr 		mqstat;
 		} mq_getsetattr;
+		struct {
+			mqd_t			mqdes;
+			int			sigev_signo;
+		} mq_notify;
 	};
 
 #if AUDIT_DEBUG
@@ -1267,6 +1265,11 @@ static void show_special(struct audit_context *context, int *call_panic)
 				return;
 		}
 		break; }
+	case AUDIT_MQ_NOTIFY: {
+		audit_log_format(ab, "mqdes=%d sigev_signo=%d",
+				context->mq_notify.mqdes,
+				context->mq_notify.sigev_signo);
+		break; }
 	case AUDIT_MQ_GETSETATTR: {
 		struct mq_attr *attr = &context->mq_getsetattr.mqstat;
 		audit_log_format(ab,
@@ -1376,14 +1379,6 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 				axi->abs_timeout.tv_sec, axi->abs_timeout.tv_nsec);
 			break; }
 
-		case AUDIT_MQ_NOTIFY: {
-			struct audit_aux_data_mq_notify *axi = (void *)aux;
-			audit_log_format(ab,
-				"mqdes=%d sigev_signo=%d",
-				axi->mqdes,
-				axi->notification.sigev_signo);
-			break; }
-
 		case AUDIT_EXECVE: {
 			struct audit_aux_data_execve *axi = (void *)aux;
 			audit_log_execve_info(context, &ab, axi);
@@ -2274,38 +2269,19 @@ int __audit_mq_timedreceive(mqd_t mqdes, size_t msg_len,
  * @mqdes: MQ descriptor
  * @u_notification: Notification event
  *
- * Returns 0 for success or NULL context or < 0 on error.
  */
 
-int __audit_mq_notify(mqd_t mqdes, const struct sigevent __user *u_notification)
+void __audit_mq_notify(mqd_t mqdes, const struct sigevent *notification)
 {
-	struct audit_aux_data_mq_notify *ax;
 	struct audit_context *context = current->audit_context;
 
-	if (!audit_enabled)
-		return 0;
-
-	if (likely(!context))
-		return 0;
-
-	ax = kmalloc(sizeof(*ax), GFP_ATOMIC);
-	if (!ax)
-		return -ENOMEM;
-
-	if (u_notification != NULL) {
-		if (copy_from_user(&ax->notification, u_notification, sizeof(ax->notification))) {
-			kfree(ax);
-			return -EFAULT;
-		}
-	} else
-		memset(&ax->notification, 0, sizeof(ax->notification));
-
-	ax->mqdes = mqdes;
+	if (notification)
+		context->mq_notify.sigev_signo = notification->sigev_signo;
+	else
+		context->mq_notify.sigev_signo = 0;
 
-	ax->d.type = AUDIT_MQ_NOTIFY;
-	ax->d.next = context->aux;
-	context->aux = (void *)ax;
-	return 0;
+	context->mq_notify.mqdes = mqdes;
+	context->type = AUDIT_MQ_NOTIFY;
 }
 
 /**
-- 
cgit v1.2.3


From c32c8af43b9adde8d6f938d8e6328c13b8de79ac Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sun, 14 Dec 2008 03:46:48 -0500
Subject: sanitize AUDIT_MQ_SENDRECV

* logging the original value of *msg_prio in mq_timedreceive(2)
  is insane - the argument is write-only (i.e. syscall always
  ignores the original value and only overwrites it).
* merge __audit_mq_timed{send,receive}
* don't do copy_from_user() twice
* don't mess with allocations in auditsc part
* ... and don't bother checking !audit_enabled and !context in there -
  we'd already checked for audit_dummy_context().

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/audit.h |  17 ++-----
 ipc/mqueue.c          |  54 +++++++++++----------
 kernel/auditsc.c      | 127 ++++++++++++--------------------------------------
 3 files changed, 63 insertions(+), 135 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index 8101d2c4a995..67f0cdd991ba 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -451,8 +451,7 @@ extern int audit_sockaddr(int len, void *addr);
 extern int __audit_fd_pair(int fd1, int fd2);
 extern int audit_set_macxattr(const char *name);
 extern int __audit_mq_open(int oflag, mode_t mode, struct mq_attr __user *u_attr);
-extern int __audit_mq_timedsend(mqd_t mqdes, size_t msg_len, unsigned int msg_prio, const struct timespec __user *u_abs_timeout);
-extern int __audit_mq_timedreceive(mqd_t mqdes, size_t msg_len, unsigned int __user *u_msg_prio, const struct timespec __user *u_abs_timeout);
+extern void __audit_mq_sendrecv(mqd_t mqdes, size_t msg_len, unsigned int msg_prio, const struct timespec *abs_timeout);
 extern void __audit_mq_notify(mqd_t mqdes, const struct sigevent *notification);
 extern void __audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat);
 extern int __audit_log_bprm_fcaps(struct linux_binprm *bprm,
@@ -482,17 +481,10 @@ static inline int audit_mq_open(int oflag, mode_t mode, struct mq_attr __user *u
 		return __audit_mq_open(oflag, mode, u_attr);
 	return 0;
 }
-static inline int audit_mq_timedsend(mqd_t mqdes, size_t msg_len, unsigned int msg_prio, const struct timespec __user *u_abs_timeout)
+static inline void audit_mq_sendrecv(mqd_t mqdes, size_t msg_len, unsigned int msg_prio, const struct timespec *abs_timeout)
 {
 	if (unlikely(!audit_dummy_context()))
-		return __audit_mq_timedsend(mqdes, msg_len, msg_prio, u_abs_timeout);
-	return 0;
-}
-static inline int audit_mq_timedreceive(mqd_t mqdes, size_t msg_len, unsigned int __user *u_msg_prio, const struct timespec __user *u_abs_timeout)
-{
-	if (unlikely(!audit_dummy_context()))
-		return __audit_mq_timedreceive(mqdes, msg_len, u_msg_prio, u_abs_timeout);
-	return 0;
+		__audit_mq_sendrecv(mqdes, msg_len, msg_prio, abs_timeout);
 }
 static inline void audit_mq_notify(mqd_t mqdes, const struct sigevent *notification)
 {
@@ -550,8 +542,7 @@ extern int audit_signals;
 #define audit_sockaddr(len, addr) ({ 0; })
 #define audit_set_macxattr(n) do { ; } while (0)
 #define audit_mq_open(o,m,a) ({ 0; })
-#define audit_mq_timedsend(d,l,p,t) ({ 0; })
-#define audit_mq_timedreceive(d,l,p,t) ({ 0; })
+#define audit_mq_sendrecv(d,l,p,t) ((void)0)
 #define audit_mq_notify(d,n) ((void)0)
 #define audit_mq_getsetattr(d,s) ((void)0)
 #define audit_log_bprm_fcaps(b, ncr, ocr) ({ 0; })
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index e7b2f68f8d77..192da806c283 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -524,31 +524,27 @@ static void __do_notify(struct mqueue_inode_info *info)
 	wake_up(&info->wait_q);
 }
 
-static long prepare_timeout(const struct timespec __user *u_arg)
+static long prepare_timeout(struct timespec *p)
 {
-	struct timespec ts, nowts;
+	struct timespec nowts;
 	long timeout;
 
-	if (u_arg) {
-		if (unlikely(copy_from_user(&ts, u_arg,
-					sizeof(struct timespec))))
-			return -EFAULT;
-
-		if (unlikely(ts.tv_nsec < 0 || ts.tv_sec < 0
-			|| ts.tv_nsec >= NSEC_PER_SEC))
+	if (p) {
+		if (unlikely(p->tv_nsec < 0 || p->tv_sec < 0
+			|| p->tv_nsec >= NSEC_PER_SEC))
 			return -EINVAL;
 		nowts = CURRENT_TIME;
 		/* first subtract as jiffies can't be too big */
-		ts.tv_sec -= nowts.tv_sec;
-		if (ts.tv_nsec < nowts.tv_nsec) {
-			ts.tv_nsec += NSEC_PER_SEC;
-			ts.tv_sec--;
+		p->tv_sec -= nowts.tv_sec;
+		if (p->tv_nsec < nowts.tv_nsec) {
+			p->tv_nsec += NSEC_PER_SEC;
+			p->tv_sec--;
 		}
-		ts.tv_nsec -= nowts.tv_nsec;
-		if (ts.tv_sec < 0)
+		p->tv_nsec -= nowts.tv_nsec;
+		if (p->tv_sec < 0)
 			return 0;
 
-		timeout = timespec_to_jiffies(&ts) + 1;
+		timeout = timespec_to_jiffies(p) + 1;
 	} else
 		return MAX_SCHEDULE_TIMEOUT;
 
@@ -829,17 +825,22 @@ asmlinkage long sys_mq_timedsend(mqd_t mqdes, const char __user *u_msg_ptr,
 	struct ext_wait_queue *receiver;
 	struct msg_msg *msg_ptr;
 	struct mqueue_inode_info *info;
+	struct timespec ts, *p = NULL;
 	long timeout;
 	int ret;
 
-	ret = audit_mq_timedsend(mqdes, msg_len, msg_prio, u_abs_timeout);
-	if (ret != 0)
-		return ret;
+	if (u_abs_timeout) {
+		if (copy_from_user(&ts, u_abs_timeout, 
+					sizeof(struct timespec)))
+			return -EFAULT;
+		p = &ts;
+	}
 
 	if (unlikely(msg_prio >= (unsigned long) MQ_PRIO_MAX))
 		return -EINVAL;
 
-	timeout = prepare_timeout(u_abs_timeout);
+	audit_mq_sendrecv(mqdes, msg_len, msg_prio, p);
+	timeout = prepare_timeout(p);
 
 	ret = -EBADF;
 	filp = fget(mqdes);
@@ -918,12 +919,17 @@ asmlinkage ssize_t sys_mq_timedreceive(mqd_t mqdes, char __user *u_msg_ptr,
 	struct inode *inode;
 	struct mqueue_inode_info *info;
 	struct ext_wait_queue wait;
+	struct timespec ts, *p = NULL;
 
-	ret = audit_mq_timedreceive(mqdes, msg_len, u_msg_prio, u_abs_timeout);
-	if (ret != 0)
-		return ret;
+	if (u_abs_timeout) {
+		if (copy_from_user(&ts, u_abs_timeout, 
+					sizeof(struct timespec)))
+			return -EFAULT;
+		p = &ts;
+	}
 
-	timeout = prepare_timeout(u_abs_timeout);
+	audit_mq_sendrecv(mqdes, msg_len, 0, p);
+	timeout = prepare_timeout(p);
 
 	ret = -EBADF;
 	filp = fget(mqdes);
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 3ece960de894..140c47453470 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -131,14 +131,6 @@ struct audit_aux_data_mq_open {
 	struct mq_attr		attr;
 };
 
-struct audit_aux_data_mq_sendrecv {
-	struct audit_aux_data	d;
-	mqd_t			mqdes;
-	size_t			msg_len;
-	unsigned int		msg_prio;
-	struct timespec		abs_timeout;
-};
-
 struct audit_aux_data_execve {
 	struct audit_aux_data	d;
 	int argc;
@@ -244,6 +236,12 @@ struct audit_context {
 			mqd_t			mqdes;
 			int			sigev_signo;
 		} mq_notify;
+		struct {
+			mqd_t			mqdes;
+			size_t			msg_len;
+			unsigned int		msg_prio;
+			struct timespec		abs_timeout;
+		} mq_sendrecv;
 	};
 
 #if AUDIT_DEBUG
@@ -1265,6 +1263,16 @@ static void show_special(struct audit_context *context, int *call_panic)
 				return;
 		}
 		break; }
+	case AUDIT_MQ_SENDRECV: {
+		audit_log_format(ab,
+			"mqdes=%d msg_len=%zd msg_prio=%u "
+			"abs_timeout_sec=%ld abs_timeout_nsec=%ld",
+			context->mq_sendrecv.mqdes,
+			context->mq_sendrecv.msg_len,
+			context->mq_sendrecv.msg_prio,
+			context->mq_sendrecv.abs_timeout.tv_sec,
+			context->mq_sendrecv.abs_timeout.tv_nsec);
+		break; }
 	case AUDIT_MQ_NOTIFY: {
 		audit_log_format(ab, "mqdes=%d sigev_signo=%d",
 				context->mq_notify.mqdes,
@@ -1370,15 +1378,6 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 				axi->attr.mq_curmsgs);
 			break; }
 
-		case AUDIT_MQ_SENDRECV: {
-			struct audit_aux_data_mq_sendrecv *axi = (void *)aux;
-			audit_log_format(ab,
-				"mqdes=%d msg_len=%zd msg_prio=%u "
-				"abs_timeout_sec=%ld abs_timeout_nsec=%ld",
-				axi->mqdes, axi->msg_len, axi->msg_prio,
-				axi->abs_timeout.tv_sec, axi->abs_timeout.tv_nsec);
-			break; }
-
 		case AUDIT_EXECVE: {
 			struct audit_aux_data_execve *axi = (void *)aux;
 			audit_log_execve_info(context, &ab, axi);
@@ -2171,97 +2170,29 @@ int __audit_mq_open(int oflag, mode_t mode, struct mq_attr __user *u_attr)
 }
 
 /**
- * __audit_mq_timedsend - record audit data for a POSIX MQ timed send
+ * __audit_mq_sendrecv - record audit data for a POSIX MQ timed send/receive
  * @mqdes: MQ descriptor
  * @msg_len: Message length
  * @msg_prio: Message priority
- * @u_abs_timeout: Message timeout in absolute time
- *
- * Returns 0 for success or NULL context or < 0 on error.
- */
-int __audit_mq_timedsend(mqd_t mqdes, size_t msg_len, unsigned int msg_prio,
-			const struct timespec __user *u_abs_timeout)
-{
-	struct audit_aux_data_mq_sendrecv *ax;
-	struct audit_context *context = current->audit_context;
-
-	if (!audit_enabled)
-		return 0;
-
-	if (likely(!context))
-		return 0;
-
-	ax = kmalloc(sizeof(*ax), GFP_ATOMIC);
-	if (!ax)
-		return -ENOMEM;
-
-	if (u_abs_timeout != NULL) {
-		if (copy_from_user(&ax->abs_timeout, u_abs_timeout, sizeof(ax->abs_timeout))) {
-			kfree(ax);
-			return -EFAULT;
-		}
-	} else
-		memset(&ax->abs_timeout, 0, sizeof(ax->abs_timeout));
-
-	ax->mqdes = mqdes;
-	ax->msg_len = msg_len;
-	ax->msg_prio = msg_prio;
-
-	ax->d.type = AUDIT_MQ_SENDRECV;
-	ax->d.next = context->aux;
-	context->aux = (void *)ax;
-	return 0;
-}
-
-/**
- * __audit_mq_timedreceive - record audit data for a POSIX MQ timed receive
- * @mqdes: MQ descriptor
- * @msg_len: Message length
- * @u_msg_prio: Message priority
- * @u_abs_timeout: Message timeout in absolute time
+ * @abs_timeout: Message timeout in absolute time
  *
- * Returns 0 for success or NULL context or < 0 on error.
  */
-int __audit_mq_timedreceive(mqd_t mqdes, size_t msg_len,
-				unsigned int __user *u_msg_prio,
-				const struct timespec __user *u_abs_timeout)
+void __audit_mq_sendrecv(mqd_t mqdes, size_t msg_len, unsigned int msg_prio,
+			const struct timespec *abs_timeout)
 {
-	struct audit_aux_data_mq_sendrecv *ax;
 	struct audit_context *context = current->audit_context;
+	struct timespec *p = &context->mq_sendrecv.abs_timeout;
 
-	if (!audit_enabled)
-		return 0;
-
-	if (likely(!context))
-		return 0;
-
-	ax = kmalloc(sizeof(*ax), GFP_ATOMIC);
-	if (!ax)
-		return -ENOMEM;
-
-	if (u_msg_prio != NULL) {
-		if (get_user(ax->msg_prio, u_msg_prio)) {
-			kfree(ax);
-			return -EFAULT;
-		}
-	} else
-		ax->msg_prio = 0;
-
-	if (u_abs_timeout != NULL) {
-		if (copy_from_user(&ax->abs_timeout, u_abs_timeout, sizeof(ax->abs_timeout))) {
-			kfree(ax);
-			return -EFAULT;
-		}
-	} else
-		memset(&ax->abs_timeout, 0, sizeof(ax->abs_timeout));
+	if (abs_timeout)
+		memcpy(p, abs_timeout, sizeof(struct timespec));
+	else
+		memset(p, 0, sizeof(struct timespec));
 
-	ax->mqdes = mqdes;
-	ax->msg_len = msg_len;
+	context->mq_sendrecv.mqdes = mqdes;
+	context->mq_sendrecv.msg_len = msg_len;
+	context->mq_sendrecv.msg_prio = msg_prio;
 
-	ax->d.type = AUDIT_MQ_SENDRECV;
-	ax->d.next = context->aux;
-	context->aux = (void *)ax;
-	return 0;
+	context->type = AUDIT_MQ_SENDRECV;
 }
 
 /**
-- 
cgit v1.2.3


From 564f6993ffef656aebaf46cf2f1f6cb4f5c97207 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sun, 14 Dec 2008 04:02:26 -0500
Subject: sanitize audit_mq_open()

* don't bother with allocations
* don't do double copy_from_user()
* don't duplicate parts of check for audit_dummy_context()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/audit.h |  9 ++++---
 ipc/mqueue.c          | 23 +++++++++---------
 kernel/auditsc.c      | 65 ++++++++++++++++++---------------------------------
 3 files changed, 38 insertions(+), 59 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index 67f0cdd991ba..54978bdd2bd4 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -450,7 +450,7 @@ extern void audit_socketcall(int nargs, unsigned long *args);
 extern int audit_sockaddr(int len, void *addr);
 extern int __audit_fd_pair(int fd1, int fd2);
 extern int audit_set_macxattr(const char *name);
-extern int __audit_mq_open(int oflag, mode_t mode, struct mq_attr __user *u_attr);
+extern void __audit_mq_open(int oflag, mode_t mode, struct mq_attr *attr);
 extern void __audit_mq_sendrecv(mqd_t mqdes, size_t msg_len, unsigned int msg_prio, const struct timespec *abs_timeout);
 extern void __audit_mq_notify(mqd_t mqdes, const struct sigevent *notification);
 extern void __audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat);
@@ -475,11 +475,10 @@ static inline void audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid
 	if (unlikely(!audit_dummy_context()))
 		__audit_ipc_set_perm(qbytes, uid, gid, mode);
 }
-static inline int audit_mq_open(int oflag, mode_t mode, struct mq_attr __user *u_attr)
+static inline void audit_mq_open(int oflag, mode_t mode, struct mq_attr *attr)
 {
 	if (unlikely(!audit_dummy_context()))
-		return __audit_mq_open(oflag, mode, u_attr);
-	return 0;
+		__audit_mq_open(oflag, mode, attr);
 }
 static inline void audit_mq_sendrecv(mqd_t mqdes, size_t msg_len, unsigned int msg_prio, const struct timespec *abs_timeout)
 {
@@ -541,7 +540,7 @@ extern int audit_signals;
 #define audit_fd_pair(n,a) ({ 0; })
 #define audit_sockaddr(len, addr) ({ 0; })
 #define audit_set_macxattr(n) do { ; } while (0)
-#define audit_mq_open(o,m,a) ({ 0; })
+#define audit_mq_open(o,m,a) ((void)0)
 #define audit_mq_sendrecv(d,l,p,t) ((void)0)
 #define audit_mq_notify(d,n) ((void)0)
 #define audit_mq_getsetattr(d,s) ((void)0)
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 192da806c283..d448b69672b5 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -588,22 +588,18 @@ static int mq_attr_ok(struct mq_attr *attr)
  * Invoked when creating a new queue via sys_mq_open
  */
 static struct file *do_create(struct dentry *dir, struct dentry *dentry,
-			int oflag, mode_t mode, struct mq_attr __user *u_attr)
+			int oflag, mode_t mode, struct mq_attr *attr)
 {
 	const struct cred *cred = current_cred();
-	struct mq_attr attr;
 	struct file *result;
 	int ret;
 
-	if (u_attr) {
-		ret = -EFAULT;
-		if (copy_from_user(&attr, u_attr, sizeof(attr)))
-			goto out;
+	if (attr) {
 		ret = -EINVAL;
-		if (!mq_attr_ok(&attr))
+		if (!mq_attr_ok(attr))
 			goto out;
 		/* store for use during create */
-		dentry->d_fsdata = &attr;
+		dentry->d_fsdata = attr;
 	}
 
 	mode &= ~current->fs->umask;
@@ -660,11 +656,13 @@ asmlinkage long sys_mq_open(const char __user *u_name, int oflag, mode_t mode,
 	struct dentry *dentry;
 	struct file *filp;
 	char *name;
+	struct mq_attr attr;
 	int fd, error;
 
-	error = audit_mq_open(oflag, mode, u_attr);
-	if (error != 0)
-		return error;
+	if (u_attr && copy_from_user(&attr, u_attr, sizeof(struct mq_attr)))
+		return -EFAULT;
+
+	audit_mq_open(oflag, mode, u_attr ? &attr : NULL);
 
 	if (IS_ERR(name = getname(u_name)))
 		return PTR_ERR(name);
@@ -690,7 +688,8 @@ asmlinkage long sys_mq_open(const char __user *u_name, int oflag, mode_t mode,
 			filp = do_open(dentry, oflag);
 		} else {
 			filp = do_create(mqueue_mnt->mnt_root, dentry,
-						oflag, mode, u_attr);
+						oflag, mode,
+						u_attr ? &attr : NULL);
 		}
 	} else {
 		error = -ENOENT;
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 140c47453470..83e946f1cdde 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -124,13 +124,6 @@ struct audit_aux_data {
 /* Number of target pids per aux struct. */
 #define AUDIT_AUX_PIDS	16
 
-struct audit_aux_data_mq_open {
-	struct audit_aux_data	d;
-	int			oflag;
-	mode_t			mode;
-	struct mq_attr		attr;
-};
-
 struct audit_aux_data_execve {
 	struct audit_aux_data	d;
 	int argc;
@@ -242,6 +235,11 @@ struct audit_context {
 			unsigned int		msg_prio;
 			struct timespec		abs_timeout;
 		} mq_sendrecv;
+		struct {
+			int			oflag;
+			mode_t			mode;
+			struct mq_attr		attr;
+		} mq_open;
 	};
 
 #if AUDIT_DEBUG
@@ -1263,6 +1261,16 @@ static void show_special(struct audit_context *context, int *call_panic)
 				return;
 		}
 		break; }
+	case AUDIT_MQ_OPEN: {
+		audit_log_format(ab,
+			"oflag=0x%x mode=%#o mq_flags=0x%lx mq_maxmsg=%ld "
+			"mq_msgsize=%ld mq_curmsgs=%ld",
+			context->mq_open.oflag, context->mq_open.mode,
+			context->mq_open.attr.mq_flags,
+			context->mq_open.attr.mq_maxmsg,
+			context->mq_open.attr.mq_msgsize,
+			context->mq_open.attr.mq_curmsgs);
+		break; }
 	case AUDIT_MQ_SENDRECV: {
 		audit_log_format(ab,
 			"mqdes=%d msg_len=%zd msg_prio=%u "
@@ -1368,15 +1376,6 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 			continue; /* audit_panic has been called */
 
 		switch (aux->type) {
-		case AUDIT_MQ_OPEN: {
-			struct audit_aux_data_mq_open *axi = (void *)aux;
-			audit_log_format(ab,
-				"oflag=0x%x mode=%#o mq_flags=0x%lx mq_maxmsg=%ld "
-				"mq_msgsize=%ld mq_curmsgs=%ld",
-				axi->oflag, axi->mode, axi->attr.mq_flags,
-				axi->attr.mq_maxmsg, axi->attr.mq_msgsize,
-				axi->attr.mq_curmsgs);
-			break; }
 
 		case AUDIT_EXECVE: {
 			struct audit_aux_data_execve *axi = (void *)aux;
@@ -2135,38 +2134,20 @@ int audit_set_loginuid(struct task_struct *task, uid_t loginuid)
  * @mode: mode bits
  * @u_attr: queue attributes
  *
- * Returns 0 for success or NULL context or < 0 on error.
  */
-int __audit_mq_open(int oflag, mode_t mode, struct mq_attr __user *u_attr)
+void __audit_mq_open(int oflag, mode_t mode, struct mq_attr *attr)
 {
-	struct audit_aux_data_mq_open *ax;
 	struct audit_context *context = current->audit_context;
 
-	if (!audit_enabled)
-		return 0;
-
-	if (likely(!context))
-		return 0;
-
-	ax = kmalloc(sizeof(*ax), GFP_ATOMIC);
-	if (!ax)
-		return -ENOMEM;
-
-	if (u_attr != NULL) {
-		if (copy_from_user(&ax->attr, u_attr, sizeof(ax->attr))) {
-			kfree(ax);
-			return -EFAULT;
-		}
-	} else
-		memset(&ax->attr, 0, sizeof(ax->attr));
+	if (attr)
+		memcpy(&context->mq_open.attr, attr, sizeof(struct mq_attr));
+	else
+		memset(&context->mq_open.attr, 0, sizeof(struct mq_attr));
 
-	ax->oflag = oflag;
-	ax->mode = mode;
+	context->mq_open.oflag = oflag;
+	context->mq_open.mode = mode;
 
-	ax->d.type = AUDIT_MQ_OPEN;
-	ax->d.next = context->aux;
-	context->aux = (void *)ax;
-	return 0;
+	context->type = AUDIT_MQ_OPEN;
 }
 
 /**
-- 
cgit v1.2.3


From 157cf649a735a2f7e8dba0ed08e6e38b6c30d886 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sun, 14 Dec 2008 04:57:47 -0500
Subject: sanitize audit_fd_pair()

* no allocations
* return void

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/pipe.c             |  7 +------
 include/linux/audit.h |  9 ++++-----
 kernel/auditsc.c      | 44 ++++++++++++++------------------------------
 net/socket.c          |  9 +--------
 4 files changed, 20 insertions(+), 49 deletions(-)

(limited to 'include/linux')

diff --git a/fs/pipe.c b/fs/pipe.c
index aaf797bd57b9..891697112f66 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -1016,10 +1016,7 @@ int do_pipe_flags(int *fd, int flags)
 		goto err_fdr;
 	fdw = error;
 
-	error = audit_fd_pair(fdr, fdw);
-	if (error < 0)
-		goto err_fdw;
-
+	audit_fd_pair(fdr, fdw);
 	fd_install(fdr, fr);
 	fd_install(fdw, fw);
 	fd[0] = fdr;
@@ -1027,8 +1024,6 @@ int do_pipe_flags(int *fd, int flags)
 
 	return 0;
 
- err_fdw:
-	put_unused_fd(fdw);
  err_fdr:
 	put_unused_fd(fdr);
  err_read_pipe:
diff --git a/include/linux/audit.h b/include/linux/audit.h
index 54978bdd2bd4..bd59cd1e3219 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -448,7 +448,7 @@ extern void __audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mod
 extern int audit_bprm(struct linux_binprm *bprm);
 extern void audit_socketcall(int nargs, unsigned long *args);
 extern int audit_sockaddr(int len, void *addr);
-extern int __audit_fd_pair(int fd1, int fd2);
+extern void __audit_fd_pair(int fd1, int fd2);
 extern int audit_set_macxattr(const char *name);
 extern void __audit_mq_open(int oflag, mode_t mode, struct mq_attr *attr);
 extern void __audit_mq_sendrecv(mqd_t mqdes, size_t msg_len, unsigned int msg_prio, const struct timespec *abs_timeout);
@@ -464,11 +464,10 @@ static inline void audit_ipc_obj(struct kern_ipc_perm *ipcp)
 	if (unlikely(!audit_dummy_context()))
 		__audit_ipc_obj(ipcp);
 }
-static inline int audit_fd_pair(int fd1, int fd2)
+static inline void audit_fd_pair(int fd1, int fd2)
 {
 	if (unlikely(!audit_dummy_context()))
-		return __audit_fd_pair(fd1, fd2);
-	return 0;
+		__audit_fd_pair(fd1, fd2);
 }
 static inline void audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode)
 {
@@ -537,7 +536,7 @@ extern int audit_signals;
 #define audit_ipc_set_perm(q,u,g,m) ((void)0)
 #define audit_bprm(p) ({ 0; })
 #define audit_socketcall(n,a) ((void)0)
-#define audit_fd_pair(n,a) ({ 0; })
+#define audit_fd_pair(n,a) ((void)0)
 #define audit_sockaddr(len, addr) ({ 0; })
 #define audit_set_macxattr(n) do { ; } while (0)
 #define audit_mq_open(o,m,a) ((void)0)
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 83e946f1cdde..327e65d50674 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -131,11 +131,6 @@ struct audit_aux_data_execve {
 	struct mm_struct *mm;
 };
 
-struct audit_aux_data_fd_pair {
-	struct	audit_aux_data d;
-	int	fd[2];
-};
-
 struct audit_aux_data_pids {
 	struct audit_aux_data	d;
 	pid_t			target_pid[AUDIT_AUX_PIDS];
@@ -241,6 +236,7 @@ struct audit_context {
 			struct mq_attr		attr;
 		} mq_open;
 	};
+	int fds[2];
 
 #if AUDIT_DEBUG
 	int		    put_count;
@@ -1382,11 +1378,6 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 			audit_log_execve_info(context, &ab, axi);
 			break; }
 
-		case AUDIT_FD_PAIR: {
-			struct audit_aux_data_fd_pair *axs = (void *)aux;
-			audit_log_format(ab, "fd0=%d fd1=%d", axs->fd[0], axs->fd[1]);
-			break; }
-
 		case AUDIT_BPRM_FCAPS: {
 			struct audit_aux_data_bprm_fcaps *axs = (void *)aux;
 			audit_log_format(ab, "fver=%x", axs->fcap_ver);
@@ -1416,6 +1407,15 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 	if (context->type)
 		show_special(context, &call_panic);
 
+	if (context->fds[0] >= 0) {
+		ab = audit_log_start(context, GFP_KERNEL, AUDIT_FD_PAIR);
+		if (ab) {
+			audit_log_format(ab, "fd0=%d fd1=%d",
+					context->fds[0], context->fds[1]);
+			audit_log_end(ab);
+		}
+	}
+
 	if (context->sockaddr_len) {
 		ab = audit_log_start(context, GFP_KERNEL, AUDIT_SOCKADDR);
 		if (ab) {
@@ -1696,6 +1696,7 @@ void audit_syscall_exit(int valid, long return_code)
 		context->target_sid = 0;
 		context->sockaddr_len = 0;
 		context->type = 0;
+		context->fds[0] = -1;
 		kfree(context->filterkey);
 		context->filterkey = NULL;
 		tsk->audit_context = context;
@@ -2291,29 +2292,12 @@ void audit_socketcall(int nargs, unsigned long *args)
  * @fd1: the first file descriptor
  * @fd2: the second file descriptor
  *
- * Returns 0 for success or NULL context or < 0 on error.
  */
-int __audit_fd_pair(int fd1, int fd2)
+void __audit_fd_pair(int fd1, int fd2)
 {
 	struct audit_context *context = current->audit_context;
-	struct audit_aux_data_fd_pair *ax;
-
-	if (likely(!context)) {
-		return 0;
-	}
-
-	ax = kmalloc(sizeof(*ax), GFP_KERNEL);
-	if (!ax) {
-		return -ENOMEM;
-	}
-
-	ax->fd[0] = fd1;
-	ax->fd[1] = fd2;
-
-	ax->d.type = AUDIT_FD_PAIR;
-	ax->d.next = context->aux;
-	context->aux = (void *)ax;
-	return 0;
+	context->fds[0] = fd1;
+	context->fds[1] = fd2;
 }
 
 /**
diff --git a/net/socket.c b/net/socket.c
index b41a92093e40..06603d73c411 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -1313,13 +1313,7 @@ asmlinkage long sys_socketpair(int family, int type, int protocol,
 		goto out_fd1;
 	}
 
-	err = audit_fd_pair(fd1, fd2);
-	if (err < 0) {
-		fput(newfile1);
-		fput(newfile2);
-		goto out_fd;
-	}
-
+	audit_fd_pair(fd1, fd2);
 	fd_install(fd1, newfile1);
 	fd_install(fd2, newfile2);
 	/* fd1 and fd2 may be already another descriptors.
@@ -1349,7 +1343,6 @@ out_fd2:
 out_fd1:
 	put_filp(newfile2);
 	sock_release(sock2);
-out_fd:
 	put_unused_fd(fd1);
 	put_unused_fd(fd2);
 	goto out;
-- 
cgit v1.2.3


From 57f71a0af4244d9ba3c0bce74b1d2e66e8d520bd Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sun, 4 Jan 2009 14:52:57 -0500
Subject: sanitize audit_log_capset()

* no allocations
* return void
* don't duplicate checked for dummy context

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/audit.h |  9 ++++-----
 kernel/auditsc.c      | 44 ++++++++++++++++----------------------------
 kernel/capability.c   |  4 +---
 3 files changed, 21 insertions(+), 36 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index bd59cd1e3219..7ddcb6a29eb1 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -457,7 +457,7 @@ extern void __audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat);
 extern int __audit_log_bprm_fcaps(struct linux_binprm *bprm,
 				  const struct cred *new,
 				  const struct cred *old);
-extern int __audit_log_capset(pid_t pid, const struct cred *new, const struct cred *old);
+extern void __audit_log_capset(pid_t pid, const struct cred *new, const struct cred *old);
 
 static inline void audit_ipc_obj(struct kern_ipc_perm *ipcp)
 {
@@ -504,12 +504,11 @@ static inline int audit_log_bprm_fcaps(struct linux_binprm *bprm,
 	return 0;
 }
 
-static inline int audit_log_capset(pid_t pid, const struct cred *new,
+static inline void audit_log_capset(pid_t pid, const struct cred *new,
 				   const struct cred *old)
 {
 	if (unlikely(!audit_dummy_context()))
-		return __audit_log_capset(pid, new, old);
-	return 0;
+		__audit_log_capset(pid, new, old);
 }
 
 extern int audit_n_rules;
@@ -544,7 +543,7 @@ extern int audit_signals;
 #define audit_mq_notify(d,n) ((void)0)
 #define audit_mq_getsetattr(d,s) ((void)0)
 #define audit_log_bprm_fcaps(b, ncr, ocr) ({ 0; })
-#define audit_log_capset(pid, ncr, ocr) ({ 0; })
+#define audit_log_capset(pid, ncr, ocr) ((void)0)
 #define audit_ptrace(t) ((void)0)
 #define audit_n_rules 0
 #define audit_signals 0
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 327e65d50674..c76a58215f54 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -235,6 +235,10 @@ struct audit_context {
 			mode_t			mode;
 			struct mq_attr		attr;
 		} mq_open;
+		struct {
+			pid_t			pid;
+			struct audit_cap_data	cap;
+		} capset;
 	};
 	int fds[2];
 
@@ -1291,6 +1295,12 @@ static void show_special(struct audit_context *context, int *call_panic)
 			attr->mq_flags, attr->mq_maxmsg,
 			attr->mq_msgsize, attr->mq_curmsgs);
 		break; }
+	case AUDIT_CAPSET: {
+		audit_log_format(ab, "pid=%d", context->capset.pid);
+		audit_log_cap(ab, "cap_pi", &context->capset.cap.inheritable);
+		audit_log_cap(ab, "cap_pp", &context->capset.cap.permitted);
+		audit_log_cap(ab, "cap_pe", &context->capset.cap.effective);
+		break; }
 	}
 	audit_log_end(ab);
 }
@@ -1392,14 +1402,6 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 			audit_log_cap(ab, "new_pe", &axs->new_pcap.effective);
 			break; }
 
-		case AUDIT_CAPSET: {
-			struct audit_aux_data_capset *axs = (void *)aux;
-			audit_log_format(ab, "pid=%d", axs->pid);
-			audit_log_cap(ab, "cap_pi", &axs->cap.inheritable);
-			audit_log_cap(ab, "cap_pp", &axs->cap.permitted);
-			audit_log_cap(ab, "cap_pe", &axs->cap.effective);
-			break; }
-
 		}
 		audit_log_end(ab);
 	}
@@ -2456,29 +2458,15 @@ int __audit_log_bprm_fcaps(struct linux_binprm *bprm,
  * Record the aguments userspace sent to sys_capset for later printing by the
  * audit system if applicable
  */
-int __audit_log_capset(pid_t pid,
+void __audit_log_capset(pid_t pid,
 		       const struct cred *new, const struct cred *old)
 {
-	struct audit_aux_data_capset *ax;
 	struct audit_context *context = current->audit_context;
-
-	if (likely(!audit_enabled || !context || context->dummy))
-		return 0;
-
-	ax = kmalloc(sizeof(*ax), GFP_KERNEL);
-	if (!ax)
-		return -ENOMEM;
-
-	ax->d.type = AUDIT_CAPSET;
-	ax->d.next = context->aux;
-	context->aux = (void *)ax;
-
-	ax->pid = pid;
-	ax->cap.effective   = new->cap_effective;
-	ax->cap.inheritable = new->cap_effective;
-	ax->cap.permitted   = new->cap_permitted;
-
-	return 0;
+	context->capset.pid = pid;
+	context->capset.cap.effective   = new->cap_effective;
+	context->capset.cap.inheritable = new->cap_effective;
+	context->capset.cap.permitted   = new->cap_permitted;
+	context->type = AUDIT_CAPSET;
 }
 
 /**
diff --git a/kernel/capability.c b/kernel/capability.c
index 36b4b4daebec..c598d9d5be4f 100644
--- a/kernel/capability.c
+++ b/kernel/capability.c
@@ -280,9 +280,7 @@ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data)
 	if (ret < 0)
 		goto error;
 
-	ret = audit_log_capset(pid, new, current_cred());
-	if (ret < 0)
-		return ret;
+	audit_log_capset(pid, new, current_cred());
 
 	return commit_creds(new);
 
-- 
cgit v1.2.3


From 0590b9335a1c72a3f0defcc6231287f7817e07c8 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sun, 14 Dec 2008 23:45:27 -0500
Subject: fixing audit rule ordering mess, part 1

Problem: ordering between the rules on exit chain is currently lost;
all watch and inode rules are listed after everything else _and_
exit,never on one kind doesn't stop exit,always on another from
being matched.

Solution: assign priorities to rules, keep track of the current
highest-priority matching rule and its result (always/never).

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/audit.h |  1 +
 kernel/audit.h        |  5 +---
 kernel/auditfilter.c  | 17 +++++++++--
 kernel/auditsc.c      | 79 ++++++++++++++++++++++++++++-----------------------
 4 files changed, 59 insertions(+), 43 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index 7ddcb6a29eb1..5b47eeb00d53 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -373,6 +373,7 @@ struct audit_krule {
 	struct audit_watch	*watch;	/* associated watch */
 	struct audit_tree	*tree;	/* associated watched tree */
 	struct list_head	rlist;	/* entry in audit_{watch,tree}.rules list */
+	u64			prio;
 };
 
 struct audit_field {
diff --git a/kernel/audit.h b/kernel/audit.h
index 9d6717412fec..16f18cac661b 100644
--- a/kernel/audit.h
+++ b/kernel/audit.h
@@ -159,11 +159,8 @@ static inline int audit_signal_info(int sig, struct task_struct *t)
 		return __audit_signal_info(sig, t);
 	return 0;
 }
-extern enum audit_state audit_filter_inodes(struct task_struct *,
-					    struct audit_context *);
-extern void audit_set_auditable(struct audit_context *);
+extern void audit_filter_inodes(struct task_struct *, struct audit_context *);
 #else
 #define audit_signal_info(s,t) AUDIT_DISABLED
 #define audit_filter_inodes(t,c) AUDIT_DISABLED
-#define audit_set_auditable(c)
 #endif
diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index 0febaa0f784c..995a2e86808d 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -919,6 +919,7 @@ static struct audit_entry *audit_dupe_rule(struct audit_krule *old,
 	new->action = old->action;
 	for (i = 0; i < AUDIT_BITMASK_SIZE; i++)
 		new->mask[i] = old->mask[i];
+	new->prio = old->prio;
 	new->buflen = old->buflen;
 	new->inode_f = old->inode_f;
 	new->watch = NULL;
@@ -987,9 +988,8 @@ static void audit_update_watch(struct audit_parent *parent,
 
 		/* If the update involves invalidating rules, do the inode-based
 		 * filtering now, so we don't omit records. */
-		if (invalidating && current->audit_context &&
-		    audit_filter_inodes(current, current->audit_context) == AUDIT_RECORD_CONTEXT)
-			audit_set_auditable(current->audit_context);
+		if (invalidating && current->audit_context)
+			audit_filter_inodes(current, current->audit_context);
 
 		nwatch = audit_dupe_watch(owatch);
 		if (IS_ERR(nwatch)) {
@@ -1258,6 +1258,9 @@ static int audit_add_watch(struct audit_krule *krule, struct nameidata *ndp,
 	return ret;
 }
 
+static u64 prio_low = ~0ULL/2;
+static u64 prio_high = ~0ULL/2 - 1;
+
 /* Add rule to given filterlist if not a duplicate. */
 static inline int audit_add_rule(struct audit_entry *entry,
 				 struct list_head *list)
@@ -1319,6 +1322,14 @@ static inline int audit_add_rule(struct audit_entry *entry,
 		}
 	}
 
+	entry->rule.prio = ~0ULL;
+	if (entry->rule.listnr == AUDIT_FILTER_EXIT) {
+		if (entry->rule.flags & AUDIT_FILTER_PREPEND)
+			entry->rule.prio = ++prio_high;
+		else
+			entry->rule.prio = --prio_low;
+	}
+
 	if (entry->rule.flags & AUDIT_FILTER_PREPEND) {
 		list_add_rcu(&entry->list, list);
 		entry->rule.flags &= ~AUDIT_FILTER_PREPEND;
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index c76a58215f54..19d2c2747c8d 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -165,14 +165,14 @@ struct audit_tree_refs {
 struct audit_context {
 	int		    dummy;	/* must be the first element */
 	int		    in_syscall;	/* 1 if task is in a syscall */
-	enum audit_state    state;
+	enum audit_state    state, current_state;
 	unsigned int	    serial;     /* serial number for record */
 	struct timespec	    ctime;      /* time of syscall entry */
 	int		    major;      /* syscall number */
 	unsigned long	    argv[4];    /* syscall arguments */
 	int		    return_valid; /* return code is valid */
 	long		    return_code;/* syscall return code */
-	int		    auditable;  /* 1 if record should be written */
+	u64		    prio;
 	int		    name_count;
 	struct audit_names  names[AUDIT_NAMES];
 	char *		    filterkey;	/* key for rule that triggered record */
@@ -630,8 +630,16 @@ static int audit_filter_rules(struct task_struct *tsk,
 			return 0;
 		}
 	}
-	if (rule->filterkey && ctx)
-		ctx->filterkey = kstrdup(rule->filterkey, GFP_ATOMIC);
+
+	if (ctx) {
+		if (rule->prio <= ctx->prio)
+			return 0;
+		if (rule->filterkey) {
+			kfree(ctx->filterkey);
+			ctx->filterkey = kstrdup(rule->filterkey, GFP_ATOMIC);
+		}
+		ctx->prio = rule->prio;
+	}
 	switch (rule->action) {
 	case AUDIT_NEVER:    *state = AUDIT_DISABLED;	    break;
 	case AUDIT_ALWAYS:   *state = AUDIT_RECORD_CONTEXT; break;
@@ -685,6 +693,7 @@ static enum audit_state audit_filter_syscall(struct task_struct *tsk,
 			    audit_filter_rules(tsk, &e->rule, ctx, NULL,
 					       &state)) {
 				rcu_read_unlock();
+				ctx->current_state = state;
 				return state;
 			}
 		}
@@ -698,15 +707,14 @@ static enum audit_state audit_filter_syscall(struct task_struct *tsk,
  * buckets applicable to the inode numbers in audit_names[].
  * Regarding audit_state, same rules apply as for audit_filter_syscall().
  */
-enum audit_state audit_filter_inodes(struct task_struct *tsk,
-				     struct audit_context *ctx)
+void audit_filter_inodes(struct task_struct *tsk, struct audit_context *ctx)
 {
 	int i;
 	struct audit_entry *e;
 	enum audit_state state;
 
 	if (audit_pid && tsk->tgid == audit_pid)
-		return AUDIT_DISABLED;
+		return;
 
 	rcu_read_lock();
 	for (i = 0; i < ctx->name_count; i++) {
@@ -723,17 +731,20 @@ enum audit_state audit_filter_inodes(struct task_struct *tsk,
 			if ((e->rule.mask[word] & bit) == bit &&
 			    audit_filter_rules(tsk, &e->rule, ctx, n, &state)) {
 				rcu_read_unlock();
-				return state;
+				ctx->current_state = state;
+				return;
 			}
 		}
 	}
 	rcu_read_unlock();
-	return AUDIT_BUILD_CONTEXT;
 }
 
-void audit_set_auditable(struct audit_context *ctx)
+static void audit_set_auditable(struct audit_context *ctx)
 {
-	ctx->auditable = 1;
+	if (!ctx->prio) {
+		ctx->prio = 1;
+		ctx->current_state = AUDIT_RECORD_CONTEXT;
+	}
 }
 
 static inline struct audit_context *audit_get_context(struct task_struct *tsk,
@@ -764,23 +775,11 @@ static inline struct audit_context *audit_get_context(struct task_struct *tsk,
 	else
 		context->return_code  = return_code;
 
-	if (context->in_syscall && !context->dummy && !context->auditable) {
-		enum audit_state state;
-
-		state = audit_filter_syscall(tsk, context, &audit_filter_list[AUDIT_FILTER_EXIT]);
-		if (state == AUDIT_RECORD_CONTEXT) {
-			context->auditable = 1;
-			goto get_context;
-		}
-
-		state = audit_filter_inodes(tsk, context);
-		if (state == AUDIT_RECORD_CONTEXT)
-			context->auditable = 1;
-
+	if (context->in_syscall && !context->dummy) {
+		audit_filter_syscall(tsk, context, &audit_filter_list[AUDIT_FILTER_EXIT]);
+		audit_filter_inodes(tsk, context);
 	}
 
-get_context:
-
 	tsk->audit_context = NULL;
 	return context;
 }
@@ -790,8 +789,7 @@ static inline void audit_free_names(struct audit_context *context)
 	int i;
 
 #if AUDIT_DEBUG == 2
-	if (context->auditable
-	    ||context->put_count + context->ino_count != context->name_count) {
+	if (context->put_count + context->ino_count != context->name_count) {
 		printk(KERN_ERR "%s:%d(:%d): major=%d in_syscall=%d"
 		       " name_count=%d put_count=%d"
 		       " ino_count=%d [NOT freeing]\n",
@@ -842,6 +840,7 @@ static inline void audit_zero_context(struct audit_context *context,
 {
 	memset(context, 0, sizeof(*context));
 	context->state      = state;
+	context->prio = state == AUDIT_RECORD_CONTEXT ? ~0ULL : 0;
 }
 
 static inline struct audit_context *audit_alloc_context(enum audit_state state)
@@ -1543,7 +1542,7 @@ void audit_free(struct task_struct *tsk)
 	 * We use GFP_ATOMIC here because we might be doing this
 	 * in the context of the idle thread */
 	/* that can happen only if we are called from do_exit() */
-	if (context->in_syscall && context->auditable)
+	if (context->in_syscall && context->current_state == AUDIT_RECORD_CONTEXT)
 		audit_log_exit(context, tsk);
 
 	audit_free_context(context);
@@ -1627,15 +1626,17 @@ void audit_syscall_entry(int arch, int major,
 
 	state = context->state;
 	context->dummy = !audit_n_rules;
-	if (!context->dummy && (state == AUDIT_SETUP_CONTEXT || state == AUDIT_BUILD_CONTEXT))
+	if (!context->dummy && state == AUDIT_BUILD_CONTEXT) {
+		context->prio = 0;
 		state = audit_filter_syscall(tsk, context, &audit_filter_list[AUDIT_FILTER_ENTRY]);
+	}
 	if (likely(state == AUDIT_DISABLED))
 		return;
 
 	context->serial     = 0;
 	context->ctime      = CURRENT_TIME;
 	context->in_syscall = 1;
-	context->auditable  = !!(state == AUDIT_RECORD_CONTEXT);
+	context->current_state  = state;
 	context->ppid       = 0;
 }
 
@@ -1643,17 +1644,20 @@ void audit_finish_fork(struct task_struct *child)
 {
 	struct audit_context *ctx = current->audit_context;
 	struct audit_context *p = child->audit_context;
-	if (!p || !ctx || !ctx->auditable)
+	if (!p || !ctx)
+		return;
+	if (!ctx->in_syscall || ctx->current_state != AUDIT_RECORD_CONTEXT)
 		return;
 	p->arch = ctx->arch;
 	p->major = ctx->major;
 	memcpy(p->argv, ctx->argv, sizeof(ctx->argv));
 	p->ctime = ctx->ctime;
 	p->dummy = ctx->dummy;
-	p->auditable = ctx->auditable;
 	p->in_syscall = ctx->in_syscall;
 	p->filterkey = kstrdup(ctx->filterkey, GFP_KERNEL);
 	p->ppid = current->pid;
+	p->prio = ctx->prio;
+	p->current_state = ctx->current_state;
 }
 
 /**
@@ -1677,11 +1681,11 @@ void audit_syscall_exit(int valid, long return_code)
 	if (likely(!context))
 		return;
 
-	if (context->in_syscall && context->auditable)
+	if (context->in_syscall && context->current_state == AUDIT_RECORD_CONTEXT)
 		audit_log_exit(context, tsk);
 
 	context->in_syscall = 0;
-	context->auditable  = 0;
+	context->prio = context->state == AUDIT_RECORD_CONTEXT ? ~0ULL : 0;
 
 	if (context->previous) {
 		struct audit_context *new_context = context->previous;
@@ -2091,7 +2095,10 @@ int auditsc_get_stamp(struct audit_context *ctx,
 	t->tv_sec  = ctx->ctime.tv_sec;
 	t->tv_nsec = ctx->ctime.tv_nsec;
 	*serial    = ctx->serial;
-	ctx->auditable = 1;
+	if (!ctx->prio) {
+		ctx->prio = 1;
+		ctx->current_state = AUDIT_RECORD_CONTEXT;
+	}
 	return 1;
 }
 
-- 
cgit v1.2.3


From e45aa212ea81d39b38ba158df344dc3a500153e5 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 15 Dec 2008 01:17:50 -0500
Subject: audit rules ordering, part 2

Fix the actual rule listing; add per-type lists _not_ used for matching,
with all exit,... sitting on one such list.  Simplifies "do something
for all rules" logics, while we are at it...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/audit.h |  1 +
 kernel/audit_tree.c   |  1 +
 kernel/auditfilter.c  | 95 +++++++++++++++++++++------------------------------
 3 files changed, 41 insertions(+), 56 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index 5b47eeb00d53..cc71fdb56ae2 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -373,6 +373,7 @@ struct audit_krule {
 	struct audit_watch	*watch;	/* associated watch */
 	struct audit_tree	*tree;	/* associated watched tree */
 	struct list_head	rlist;	/* entry in audit_{watch,tree}.rules list */
+	struct list_head	list;	/* for AUDIT_LIST* purposes only */
 	u64			prio;
 };
 
diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c
index 8b509441f49a..48bddad2a3dc 100644
--- a/kernel/audit_tree.c
+++ b/kernel/audit_tree.c
@@ -450,6 +450,7 @@ static void kill_rules(struct audit_tree *tree)
 			audit_log_end(ab);
 			rule->tree = NULL;
 			list_del_rcu(&entry->list);
+			list_del(&entry->rule.list);
 			call_rcu(&entry->rcu, audit_free_rule_rcu);
 		}
 	}
diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index 995a2e86808d..5d4edc6f7a32 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -86,6 +86,14 @@ struct list_head audit_filter_list[AUDIT_NR_FILTERS] = {
 #error Fix audit_filter_list initialiser
 #endif
 };
+static struct list_head audit_rules_list[AUDIT_NR_FILTERS] = {
+	LIST_HEAD_INIT(audit_rules_list[0]),
+	LIST_HEAD_INIT(audit_rules_list[1]),
+	LIST_HEAD_INIT(audit_rules_list[2]),
+	LIST_HEAD_INIT(audit_rules_list[3]),
+	LIST_HEAD_INIT(audit_rules_list[4]),
+	LIST_HEAD_INIT(audit_rules_list[5]),
+};
 
 DEFINE_MUTEX(audit_filter_mutex);
 
@@ -1007,12 +1015,15 @@ static void audit_update_watch(struct audit_parent *parent,
 			list_del_rcu(&oentry->list);
 
 			nentry = audit_dupe_rule(&oentry->rule, nwatch);
-			if (IS_ERR(nentry))
+			if (IS_ERR(nentry)) {
+				list_del(&oentry->rule.list);
 				audit_panic("error updating watch, removing");
-			else {
+			} else {
 				int h = audit_hash_ino((u32)ino);
 				list_add(&nentry->rule.rlist, &nwatch->rules);
 				list_add_rcu(&nentry->list, &audit_inode_hash[h]);
+				list_replace(&oentry->rule.list,
+					     &nentry->rule.list);
 			}
 
 			call_rcu(&oentry->rcu, audit_free_rule_rcu);
@@ -1077,6 +1088,7 @@ static void audit_remove_parent_watches(struct audit_parent *parent)
 				audit_log_end(ab);
 			}
 			list_del(&r->rlist);
+			list_del(&r->list);
 			list_del_rcu(&e->list);
 			call_rcu(&e->rcu, audit_free_rule_rcu);
 		}
@@ -1331,9 +1343,13 @@ static inline int audit_add_rule(struct audit_entry *entry,
 	}
 
 	if (entry->rule.flags & AUDIT_FILTER_PREPEND) {
+		list_add(&entry->rule.list,
+			 &audit_rules_list[entry->rule.listnr]);
 		list_add_rcu(&entry->list, list);
 		entry->rule.flags &= ~AUDIT_FILTER_PREPEND;
 	} else {
+		list_add_tail(&entry->rule.list,
+			      &audit_rules_list[entry->rule.listnr]);
 		list_add_tail_rcu(&entry->list, list);
 	}
 #ifdef CONFIG_AUDITSYSCALL
@@ -1415,6 +1431,7 @@ static inline int audit_del_rule(struct audit_entry *entry,
 		audit_remove_tree_rule(&e->rule);
 
 	list_del_rcu(&e->list);
+	list_del(&e->rule.list);
 	call_rcu(&e->rcu, audit_free_rule_rcu);
 
 #ifdef CONFIG_AUDITSYSCALL
@@ -1443,30 +1460,16 @@ out:
 static void audit_list(int pid, int seq, struct sk_buff_head *q)
 {
 	struct sk_buff *skb;
-	struct audit_entry *entry;
+	struct audit_krule *r;
 	int i;
 
 	/* This is a blocking read, so use audit_filter_mutex instead of rcu
 	 * iterator to sync with list writers. */
 	for (i=0; i<AUDIT_NR_FILTERS; i++) {
-		list_for_each_entry(entry, &audit_filter_list[i], list) {
-			struct audit_rule *rule;
-
-			rule = audit_krule_to_rule(&entry->rule);
-			if (unlikely(!rule))
-				break;
-			skb = audit_make_reply(pid, seq, AUDIT_LIST, 0, 1,
-					 rule, sizeof(*rule));
-			if (skb)
-				skb_queue_tail(q, skb);
-			kfree(rule);
-		}
-	}
-	for (i = 0; i < AUDIT_INODE_BUCKETS; i++) {
-		list_for_each_entry(entry, &audit_inode_hash[i], list) {
+		list_for_each_entry(r, &audit_rules_list[i], list) {
 			struct audit_rule *rule;
 
-			rule = audit_krule_to_rule(&entry->rule);
+			rule = audit_krule_to_rule(r);
 			if (unlikely(!rule))
 				break;
 			skb = audit_make_reply(pid, seq, AUDIT_LIST, 0, 1,
@@ -1485,30 +1488,16 @@ static void audit_list(int pid, int seq, struct sk_buff_head *q)
 static void audit_list_rules(int pid, int seq, struct sk_buff_head *q)
 {
 	struct sk_buff *skb;
-	struct audit_entry *e;
+	struct audit_krule *r;
 	int i;
 
 	/* This is a blocking read, so use audit_filter_mutex instead of rcu
 	 * iterator to sync with list writers. */
 	for (i=0; i<AUDIT_NR_FILTERS; i++) {
-		list_for_each_entry(e, &audit_filter_list[i], list) {
-			struct audit_rule_data *data;
-
-			data = audit_krule_to_data(&e->rule);
-			if (unlikely(!data))
-				break;
-			skb = audit_make_reply(pid, seq, AUDIT_LIST_RULES, 0, 1,
-					 data, sizeof(*data) + data->buflen);
-			if (skb)
-				skb_queue_tail(q, skb);
-			kfree(data);
-		}
-	}
-	for (i=0; i< AUDIT_INODE_BUCKETS; i++) {
-		list_for_each_entry(e, &audit_inode_hash[i], list) {
+		list_for_each_entry(r, &audit_rules_list[i], list) {
 			struct audit_rule_data *data;
 
-			data = audit_krule_to_data(&e->rule);
+			data = audit_krule_to_data(r);
 			if (unlikely(!data))
 				break;
 			skb = audit_make_reply(pid, seq, AUDIT_LIST_RULES, 0, 1,
@@ -1789,35 +1778,37 @@ unlock_and_return:
 	return result;
 }
 
-static int update_lsm_rule(struct audit_entry *entry)
+static int update_lsm_rule(struct audit_krule *r)
 {
+	struct audit_entry *entry = container_of(r, struct audit_entry, rule);
 	struct audit_entry *nentry;
 	struct audit_watch *watch;
 	struct audit_tree *tree;
 	int err = 0;
 
-	if (!security_audit_rule_known(&entry->rule))
+	if (!security_audit_rule_known(r))
 		return 0;
 
-	watch = entry->rule.watch;
-	tree = entry->rule.tree;
-	nentry = audit_dupe_rule(&entry->rule, watch);
+	watch = r->watch;
+	tree = r->tree;
+	nentry = audit_dupe_rule(r, watch);
 	if (IS_ERR(nentry)) {
 		/* save the first error encountered for the
 		 * return value */
 		err = PTR_ERR(nentry);
 		audit_panic("error updating LSM filters");
 		if (watch)
-			list_del(&entry->rule.rlist);
+			list_del(&r->rlist);
 		list_del_rcu(&entry->list);
+		list_del(&r->list);
 	} else {
 		if (watch) {
 			list_add(&nentry->rule.rlist, &watch->rules);
-			list_del(&entry->rule.rlist);
+			list_del(&r->rlist);
 		} else if (tree)
-			list_replace_init(&entry->rule.rlist,
-				     &nentry->rule.rlist);
+			list_replace_init(&r->rlist, &nentry->rule.rlist);
 		list_replace_rcu(&entry->list, &nentry->list);
+		list_replace(&r->list, &nentry->rule.list);
 	}
 	call_rcu(&entry->rcu, audit_free_rule_rcu);
 
@@ -1831,27 +1822,19 @@ static int update_lsm_rule(struct audit_entry *entry)
  * updated rule. */
 int audit_update_lsm_rules(void)
 {
-	struct audit_entry *e, *n;
+	struct audit_krule *r, *n;
 	int i, err = 0;
 
 	/* audit_filter_mutex synchronizes the writers */
 	mutex_lock(&audit_filter_mutex);
 
 	for (i = 0; i < AUDIT_NR_FILTERS; i++) {
-		list_for_each_entry_safe(e, n, &audit_filter_list[i], list) {
-			int res = update_lsm_rule(e);
-			if (!err)
-				err = res;
-		}
-	}
-	for (i=0; i< AUDIT_INODE_BUCKETS; i++) {
-		list_for_each_entry_safe(e, n, &audit_inode_hash[i], list) {
-			int res = update_lsm_rule(e);
+		list_for_each_entry_safe(r, n, &audit_rules_list[i], list) {
+			int res = update_lsm_rule(r);
 			if (!err)
 				err = res;
 		}
 	}
-
 	mutex_unlock(&audit_filter_mutex);
 
 	return err;
-- 
cgit v1.2.3


From 5af75d8d58d0f9f7b7c0515b35786b22892d5f12 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 16 Dec 2008 05:59:26 -0500
Subject: audit: validate comparison operations, store them in sane form

Don't store the field->op in the messy (and very inconvenient for e.g.
audit_comparator()) form; translate to dense set of values and do full
validation of userland-submitted value while we are at it.

->audit_init_rule() and ->audit_match_rule() get new values now; in-tree
instances updated.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/audit.h          |  12 ++++
 kernel/audit_tree.c            |   2 +-
 kernel/auditfilter.c           | 132 ++++++++++++++++++++---------------------
 security/selinux/ss/services.c |  26 ++++----
 security/smack/smack_lsm.c     |   6 +-
 5 files changed, 94 insertions(+), 84 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/audit.h b/include/linux/audit.h
index cc71fdb56ae2..67e5dbfc2961 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -247,6 +247,18 @@
 #define AUDIT_GREATER_THAN_OR_EQUAL	(AUDIT_GREATER_THAN|AUDIT_EQUAL)
 #define AUDIT_OPERATORS			(AUDIT_EQUAL|AUDIT_NOT_EQUAL|AUDIT_BIT_MASK)
 
+enum {
+	Audit_equal,
+	Audit_not_equal,
+	Audit_bitmask,
+	Audit_bittest,
+	Audit_lt,
+	Audit_gt,
+	Audit_le,
+	Audit_ge,
+	Audit_bad
+};
+
 /* Status symbols */
 				/* Mask values */
 #define AUDIT_STATUS_ENABLED		0x0001
diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c
index 48bddad2a3dc..8ad9545b8db9 100644
--- a/kernel/audit_tree.c
+++ b/kernel/audit_tree.c
@@ -618,7 +618,7 @@ int audit_make_tree(struct audit_krule *rule, char *pathname, u32 op)
 
 	if (pathname[0] != '/' ||
 	    rule->listnr != AUDIT_FILTER_EXIT ||
-	    op & ~AUDIT_EQUAL ||
+	    op != Audit_equal ||
 	    rule->inode_f || rule->watch || rule->tree)
 		return -EINVAL;
 	rule->tree = alloc_tree(pathname);
diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index e6e3829cadd1..fbf24d121d97 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -252,7 +252,8 @@ static inline int audit_to_inode(struct audit_krule *krule,
 				 struct audit_field *f)
 {
 	if (krule->listnr != AUDIT_FILTER_EXIT ||
-	    krule->watch || krule->inode_f || krule->tree)
+	    krule->watch || krule->inode_f || krule->tree ||
+	    (f->op != Audit_equal && f->op != Audit_not_equal))
 		return -EINVAL;
 
 	krule->inode_f = f;
@@ -270,7 +271,7 @@ static int audit_to_watch(struct audit_krule *krule, char *path, int len,
 
 	if (path[0] != '/' || path[len-1] == '/' ||
 	    krule->listnr != AUDIT_FILTER_EXIT ||
-	    op & ~AUDIT_EQUAL ||
+	    op != Audit_equal ||
 	    krule->inode_f || krule->watch || krule->tree)
 		return -EINVAL;
 
@@ -420,12 +421,32 @@ exit_err:
 	return ERR_PTR(err);
 }
 
+static u32 audit_ops[] =
+{
+	[Audit_equal] = AUDIT_EQUAL,
+	[Audit_not_equal] = AUDIT_NOT_EQUAL,
+	[Audit_bitmask] = AUDIT_BIT_MASK,
+	[Audit_bittest] = AUDIT_BIT_TEST,
+	[Audit_lt] = AUDIT_LESS_THAN,
+	[Audit_gt] = AUDIT_GREATER_THAN,
+	[Audit_le] = AUDIT_LESS_THAN_OR_EQUAL,
+	[Audit_ge] = AUDIT_GREATER_THAN_OR_EQUAL,
+};
+
+static u32 audit_to_op(u32 op)
+{
+	u32 n;
+	for (n = Audit_equal; n < Audit_bad && audit_ops[n] != op; n++)
+		;
+	return n;
+}
+
+
 /* Translate struct audit_rule to kernel's rule respresentation.
  * Exists for backward compatibility with userspace. */
 static struct audit_entry *audit_rule_to_entry(struct audit_rule *rule)
 {
 	struct audit_entry *entry;
-	struct audit_field *ino_f;
 	int err = 0;
 	int i;
 
@@ -435,12 +456,28 @@ static struct audit_entry *audit_rule_to_entry(struct audit_rule *rule)
 
 	for (i = 0; i < rule->field_count; i++) {
 		struct audit_field *f = &entry->rule.fields[i];
+		u32 n;
+
+		n = rule->fields[i] & (AUDIT_NEGATE|AUDIT_OPERATORS);
+
+		/* Support for legacy operators where
+		 * AUDIT_NEGATE bit signifies != and otherwise assumes == */
+		if (n & AUDIT_NEGATE)
+			f->op = Audit_not_equal;
+		else if (!n)
+			f->op = Audit_equal;
+		else
+			f->op = audit_to_op(n);
+
+		entry->rule.vers_ops = (n & AUDIT_OPERATORS) ? 2 : 1;
 
-		f->op = rule->fields[i] & (AUDIT_NEGATE|AUDIT_OPERATORS);
 		f->type = rule->fields[i] & ~(AUDIT_NEGATE|AUDIT_OPERATORS);
 		f->val = rule->values[i];
 
 		err = -EINVAL;
+		if (f->op == Audit_bad)
+			goto exit_free;
+
 		switch(f->type) {
 		default:
 			goto exit_free;
@@ -462,11 +499,8 @@ static struct audit_entry *audit_rule_to_entry(struct audit_rule *rule)
 		case AUDIT_EXIT:
 		case AUDIT_SUCCESS:
 			/* bit ops are only useful on syscall args */
-			if (f->op == AUDIT_BIT_MASK ||
-						f->op == AUDIT_BIT_TEST) {
-				err = -EINVAL;
+			if (f->op == Audit_bitmask || f->op == Audit_bittest)
 				goto exit_free;
-			}
 			break;
 		case AUDIT_ARG0:
 		case AUDIT_ARG1:
@@ -475,11 +509,8 @@ static struct audit_entry *audit_rule_to_entry(struct audit_rule *rule)
 			break;
 		/* arch is only allowed to be = or != */
 		case AUDIT_ARCH:
-			if ((f->op != AUDIT_NOT_EQUAL) && (f->op != AUDIT_EQUAL)
-					&& (f->op != AUDIT_NEGATE) && (f->op)) {
-				err = -EINVAL;
+			if (f->op != Audit_not_equal && f->op != Audit_equal)
 				goto exit_free;
-			}
 			entry->rule.arch_f = f;
 			break;
 		case AUDIT_PERM:
@@ -496,33 +527,10 @@ static struct audit_entry *audit_rule_to_entry(struct audit_rule *rule)
 				goto exit_free;
 			break;
 		}
-
-		entry->rule.vers_ops = (f->op & AUDIT_OPERATORS) ? 2 : 1;
-
-		/* Support for legacy operators where
-		 * AUDIT_NEGATE bit signifies != and otherwise assumes == */
-		if (f->op & AUDIT_NEGATE)
-			f->op = AUDIT_NOT_EQUAL;
-		else if (!f->op)
-			f->op = AUDIT_EQUAL;
-		else if (f->op == AUDIT_OPERATORS) {
-			err = -EINVAL;
-			goto exit_free;
-		}
 	}
 
-	ino_f = entry->rule.inode_f;
-	if (ino_f) {
-		switch(ino_f->op) {
-		case AUDIT_NOT_EQUAL:
-			entry->rule.inode_f = NULL;
-		case AUDIT_EQUAL:
-			break;
-		default:
-			err = -EINVAL;
-			goto exit_free;
-		}
-	}
+	if (entry->rule.inode_f && entry->rule.inode_f->op == Audit_not_equal)
+		entry->rule.inode_f = NULL;
 
 exit_nofree:
 	return entry;
@@ -538,7 +546,6 @@ static struct audit_entry *audit_data_to_entry(struct audit_rule_data *data,
 {
 	int err = 0;
 	struct audit_entry *entry;
-	struct audit_field *ino_f;
 	void *bufp;
 	size_t remain = datasz - sizeof(struct audit_rule_data);
 	int i;
@@ -554,11 +561,11 @@ static struct audit_entry *audit_data_to_entry(struct audit_rule_data *data,
 		struct audit_field *f = &entry->rule.fields[i];
 
 		err = -EINVAL;
-		if (!(data->fieldflags[i] & AUDIT_OPERATORS) ||
-		    data->fieldflags[i] & ~AUDIT_OPERATORS)
+
+		f->op = audit_to_op(data->fieldflags[i]);
+		if (f->op == Audit_bad)
 			goto exit_free;
 
-		f->op = data->fieldflags[i] & AUDIT_OPERATORS;
 		f->type = data->fields[i];
 		f->val = data->values[i];
 		f->lsm_str = NULL;
@@ -670,18 +677,8 @@ static struct audit_entry *audit_data_to_entry(struct audit_rule_data *data,
 		}
 	}
 
-	ino_f = entry->rule.inode_f;
-	if (ino_f) {
-		switch(ino_f->op) {
-		case AUDIT_NOT_EQUAL:
-			entry->rule.inode_f = NULL;
-		case AUDIT_EQUAL:
-			break;
-		default:
-			err = -EINVAL;
-			goto exit_free;
-		}
-	}
+	if (entry->rule.inode_f && entry->rule.inode_f->op == Audit_not_equal)
+		entry->rule.inode_f = NULL;
 
 exit_nofree:
 	return entry;
@@ -721,10 +718,10 @@ static struct audit_rule *audit_krule_to_rule(struct audit_krule *krule)
 		rule->fields[i] = krule->fields[i].type;
 
 		if (krule->vers_ops == 1) {
-			if (krule->fields[i].op & AUDIT_NOT_EQUAL)
+			if (krule->fields[i].op == Audit_not_equal)
 				rule->fields[i] |= AUDIT_NEGATE;
 		} else {
-			rule->fields[i] |= krule->fields[i].op;
+			rule->fields[i] |= audit_ops[krule->fields[i].op];
 		}
 	}
 	for (i = 0; i < AUDIT_BITMASK_SIZE; i++) rule->mask[i] = krule->mask[i];
@@ -752,7 +749,7 @@ static struct audit_rule_data *audit_krule_to_data(struct audit_krule *krule)
 		struct audit_field *f = &krule->fields[i];
 
 		data->fields[i] = f->type;
-		data->fieldflags[i] = f->op;
+		data->fieldflags[i] = audit_ops[f->op];
 		switch(f->type) {
 		case AUDIT_SUBJ_USER:
 		case AUDIT_SUBJ_ROLE:
@@ -1626,28 +1623,29 @@ int audit_receive_filter(int type, int pid, int uid, int seq, void *data,
 	return err;
 }
 
-int audit_comparator(const u32 left, const u32 op, const u32 right)
+int audit_comparator(u32 left, u32 op, u32 right)
 {
 	switch (op) {
-	case AUDIT_EQUAL:
+	case Audit_equal:
 		return (left == right);
-	case AUDIT_NOT_EQUAL:
+	case Audit_not_equal:
 		return (left != right);
-	case AUDIT_LESS_THAN:
+	case Audit_lt:
 		return (left < right);
-	case AUDIT_LESS_THAN_OR_EQUAL:
+	case Audit_le:
 		return (left <= right);
-	case AUDIT_GREATER_THAN:
+	case Audit_gt:
 		return (left > right);
-	case AUDIT_GREATER_THAN_OR_EQUAL:
+	case Audit_ge:
 		return (left >= right);
-	case AUDIT_BIT_MASK:
+	case Audit_bitmask:
 		return (left & right);
-	case AUDIT_BIT_TEST:
+	case Audit_bittest:
 		return ((left & right) == right);
+	default:
+		BUG();
+		return 0;
 	}
-	BUG();
-	return 0;
 }
 
 /* Compare given dentry name with last component in given path,
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index 343c8ab14af0..c65e4fe4a0f1 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -2602,7 +2602,7 @@ int selinux_audit_rule_init(u32 field, u32 op, char *rulestr, void **vrule)
 	case AUDIT_OBJ_ROLE:
 	case AUDIT_OBJ_TYPE:
 		/* only 'equals' and 'not equals' fit user, role, and type */
-		if (op != AUDIT_EQUAL && op != AUDIT_NOT_EQUAL)
+		if (op != Audit_equal && op != Audit_not_equal)
 			return -EINVAL;
 		break;
 	case AUDIT_SUBJ_SEN:
@@ -2736,10 +2736,10 @@ int selinux_audit_rule_match(u32 sid, u32 field, u32 op, void *vrule,
 	case AUDIT_SUBJ_USER:
 	case AUDIT_OBJ_USER:
 		switch (op) {
-		case AUDIT_EQUAL:
+		case Audit_equal:
 			match = (ctxt->user == rule->au_ctxt.user);
 			break;
-		case AUDIT_NOT_EQUAL:
+		case Audit_not_equal:
 			match = (ctxt->user != rule->au_ctxt.user);
 			break;
 		}
@@ -2747,10 +2747,10 @@ int selinux_audit_rule_match(u32 sid, u32 field, u32 op, void *vrule,
 	case AUDIT_SUBJ_ROLE:
 	case AUDIT_OBJ_ROLE:
 		switch (op) {
-		case AUDIT_EQUAL:
+		case Audit_equal:
 			match = (ctxt->role == rule->au_ctxt.role);
 			break;
-		case AUDIT_NOT_EQUAL:
+		case Audit_not_equal:
 			match = (ctxt->role != rule->au_ctxt.role);
 			break;
 		}
@@ -2758,10 +2758,10 @@ int selinux_audit_rule_match(u32 sid, u32 field, u32 op, void *vrule,
 	case AUDIT_SUBJ_TYPE:
 	case AUDIT_OBJ_TYPE:
 		switch (op) {
-		case AUDIT_EQUAL:
+		case Audit_equal:
 			match = (ctxt->type == rule->au_ctxt.type);
 			break;
-		case AUDIT_NOT_EQUAL:
+		case Audit_not_equal:
 			match = (ctxt->type != rule->au_ctxt.type);
 			break;
 		}
@@ -2774,31 +2774,31 @@ int selinux_audit_rule_match(u32 sid, u32 field, u32 op, void *vrule,
 			  field == AUDIT_OBJ_LEV_LOW) ?
 			 &ctxt->range.level[0] : &ctxt->range.level[1]);
 		switch (op) {
-		case AUDIT_EQUAL:
+		case Audit_equal:
 			match = mls_level_eq(&rule->au_ctxt.range.level[0],
 					     level);
 			break;
-		case AUDIT_NOT_EQUAL:
+		case Audit_not_equal:
 			match = !mls_level_eq(&rule->au_ctxt.range.level[0],
 					      level);
 			break;
-		case AUDIT_LESS_THAN:
+		case Audit_lt:
 			match = (mls_level_dom(&rule->au_ctxt.range.level[0],
 					       level) &&
 				 !mls_level_eq(&rule->au_ctxt.range.level[0],
 					       level));
 			break;
-		case AUDIT_LESS_THAN_OR_EQUAL:
+		case Audit_le:
 			match = mls_level_dom(&rule->au_ctxt.range.level[0],
 					      level);
 			break;
-		case AUDIT_GREATER_THAN:
+		case Audit_gt:
 			match = (mls_level_dom(level,
 					      &rule->au_ctxt.range.level[0]) &&
 				 !mls_level_eq(level,
 					       &rule->au_ctxt.range.level[0]));
 			break;
-		case AUDIT_GREATER_THAN_OR_EQUAL:
+		case Audit_ge:
 			match = mls_level_dom(level,
 					      &rule->au_ctxt.range.level[0]);
 			break;
diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c
index 1b5551dfc1f7..848212fd4845 100644
--- a/security/smack/smack_lsm.c
+++ b/security/smack/smack_lsm.c
@@ -2492,7 +2492,7 @@ static int smack_audit_rule_init(u32 field, u32 op, char *rulestr, void **vrule)
 	if (field != AUDIT_SUBJ_USER && field != AUDIT_OBJ_USER)
 		return -EINVAL;
 
-	if (op != AUDIT_EQUAL && op != AUDIT_NOT_EQUAL)
+	if (op != Audit_equal && op != Audit_not_equal)
 		return -EINVAL;
 
 	*rule = smk_import(rulestr, 0);
@@ -2556,9 +2556,9 @@ static int smack_audit_rule_match(u32 secid, u32 field, u32 op, void *vrule,
 	 * both pointers will point to the same smack_known
 	 * label.
 	 */
-	if (op == AUDIT_EQUAL)
+	if (op == Audit_equal)
 		return (rule == smack);
-	if (op == AUDIT_NOT_EQUAL)
+	if (op == Audit_not_equal)
 		return (rule != smack);
 
 	return 0;
-- 
cgit v1.2.3


From 0a30c5cefa53cbac429dcb2de906c0637b646253 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Sun, 4 Jan 2009 12:00:47 -0800
Subject: spi.h uses/needs device.h

Include header files as used/needed:

  In file included from drivers/leds/leds-dac124s085.c:16:
  include/linux/spi/spi.h:66: error: field 'dev' has incomplete type
  include/linux/spi/spi.h: In function 'to_spi_device':
  include/linux/spi/spi.h:100: warning: type defaults to 'int' in declaration of '__mptr'
  ...

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/spi/spi.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h
index 4be01bb44377..82229317753d 100644
--- a/include/linux/spi/spi.h
+++ b/include/linux/spi/spi.h
@@ -19,6 +19,8 @@
 #ifndef __LINUX_SPI_H
 #define __LINUX_SPI_H
 
+#include <linux/device.h>
+
 /*
  * INTERFACES between SPI master-side drivers and SPI infrastructure.
  * (There's no SPI slave support for Linux yet...)
-- 
cgit v1.2.3


From c644f0e4b56f9a2fc066cd0d75a18074d130e4a3 Mon Sep 17 00:00:00 2001
From: Pekka Enberg <penberg@cs.helsinki.fi>
Date: Sun, 4 Jan 2009 12:00:48 -0800
Subject: fs: introduce bgl_lock_ptr()

As suggested by Andreas Dilger, introduce a bgl_lock_ptr() helper in
<linux/blockgroup_lock.h> and add separate sb_bgl_lock() helpers to
filesystem specific header files to break the hidden dependency to
struct ext[234]_sb_info.

Also, while at it, convert the macros to static inlines to try make up
for all the times I broke Andrew Morton's tree.

Acked-by: Andreas Dilger <adilger@sun.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext4/ext4_sb.h               | 6 ++++++
 include/linux/blockgroup_lock.h | 7 +++++--
 include/linux/ext2_fs_sb.h      | 6 ++++++
 include/linux/ext3_fs_sb.h      | 6 ++++++
 4 files changed, 23 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/fs/ext4/ext4_sb.h b/fs/ext4/ext4_sb.h
index 445fde603df8..b21f16713db0 100644
--- a/fs/ext4/ext4_sb.h
+++ b/fs/ext4/ext4_sb.h
@@ -146,4 +146,10 @@ struct ext4_sb_info {
 	struct flex_groups *s_flex_groups;
 };
 
+static inline spinlock_t *
+sb_bgl_lock(struct ext4_sb_info *sbi, unsigned int block_group)
+{
+	return bgl_lock_ptr(&sbi->s_blockgroup_lock, block_group);
+}
+
 #endif	/* _EXT4_SB */
diff --git a/include/linux/blockgroup_lock.h b/include/linux/blockgroup_lock.h
index 8607312983bd..e44b88ba552b 100644
--- a/include/linux/blockgroup_lock.h
+++ b/include/linux/blockgroup_lock.h
@@ -53,7 +53,10 @@ static inline void bgl_lock_init(struct blockgroup_lock *bgl)
  * The accessor is a macro so we can embed a blockgroup_lock into different
  * superblock types
  */
-#define sb_bgl_lock(sb, block_group) \
-	(&(sb)->s_blockgroup_lock.locks[(block_group) & (NR_BG_LOCKS-1)].lock)
+static inline spinlock_t *
+bgl_lock_ptr(struct blockgroup_lock *bgl, unsigned int block_group)
+{
+	return &bgl->locks[(block_group) & (NR_BG_LOCKS-1)].lock;
+}
 
 #endif
diff --git a/include/linux/ext2_fs_sb.h b/include/linux/ext2_fs_sb.h
index f273415ab6f1..dc541f3653d1 100644
--- a/include/linux/ext2_fs_sb.h
+++ b/include/linux/ext2_fs_sb.h
@@ -108,4 +108,10 @@ struct ext2_sb_info {
 	struct ext2_reserve_window_node s_rsv_window_head;
 };
 
+static inline spinlock_t *
+sb_bgl_lock(struct ext2_sb_info *sbi, unsigned int block_group)
+{
+	return bgl_lock_ptr(&sbi->s_blockgroup_lock, block_group);
+}
+
 #endif	/* _LINUX_EXT2_FS_SB */
diff --git a/include/linux/ext3_fs_sb.h b/include/linux/ext3_fs_sb.h
index b65f0288b842..e024e38248ff 100644
--- a/include/linux/ext3_fs_sb.h
+++ b/include/linux/ext3_fs_sb.h
@@ -83,4 +83,10 @@ struct ext3_sb_info {
 #endif
 };
 
+static inline spinlock_t *
+sb_bgl_lock(struct ext3_sb_info *sbi, unsigned int block_group)
+{
+	return bgl_lock_ptr(&sbi->s_blockgroup_lock, block_group);
+}
+
 #endif	/* _LINUX_EXT3_FS_SB */
-- 
cgit v1.2.3


From 54566b2c1594c2326a645a3551f9d989f7ba3c5e Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Sun, 4 Jan 2009 12:00:53 -0800
Subject: fs: symlink write_begin allocation context fix

With the write_begin/write_end aops, page_symlink was broken because it
could no longer pass a GFP_NOFS type mask into the point where the
allocations happened.  They are done in write_begin, which would always
assume that the filesystem can be entered from reclaim.  This bug could
cause filesystem deadlocks.

The funny thing with having a gfp_t mask there is that it doesn't really
allow the caller to arbitrarily tinker with the context in which it can be
called.  It couldn't ever be GFP_ATOMIC, for example, because it needs to
take the page lock.  The only thing any callers care about is __GFP_FS
anyway, so turn that into a single flag.

Add a new flag for write_begin, AOP_FLAG_NOFS.  Filesystems can now act on
this flag in their write_begin function.  Change __grab_cache_page to
accept a nofs argument as well, to honour that flag (while we're there,
change the name to grab_cache_page_write_begin which is more instructive
and does away with random leading underscores).

This is really a more flexible way to go in the end anyway -- if a
filesystem happens to want any extra allocations aside from the pagecache
ones in ints write_begin function, it may now use GFP_KERNEL (rather than
GFP_NOFS) for common case allocations (eg.  ocfs2_alloc_write_ctxt, for a
random example).

[kosaki.motohiro@jp.fujitsu.com: fix ubifs]
[kosaki.motohiro@jp.fujitsu.com: fix fuse]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: <stable@kernel.org>		[2.6.28.x]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
[ Cleaned up the calling convention: just pass in the AOP flags
  untouched to the grab_cache_page_write_begin() function.  That
  just simplifies everybody, and may even allow future expansion of the
  logic.   - Linus ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/affs/file.c          |  2 +-
 fs/afs/write.c          |  2 +-
 fs/buffer.c             |  4 ++--
 fs/cifs/file.c          |  2 +-
 fs/ecryptfs/mmap.c      |  2 +-
 fs/ext3/inode.c         |  2 +-
 fs/ext3/namei.c         |  3 +--
 fs/ext4/inode.c         |  4 ++--
 fs/ext4/namei.c         |  3 +--
 fs/fuse/file.c          |  4 ++--
 fs/gfs2/ops_address.c   |  2 +-
 fs/hostfs/hostfs_kern.c |  2 +-
 fs/jffs2/file.c         |  2 +-
 fs/libfs.c              |  2 +-
 fs/namei.c              | 13 +++++++++----
 fs/nfs/file.c           |  2 +-
 fs/reiserfs/inode.c     |  2 +-
 fs/smbfs/file.c         |  2 +-
 fs/ubifs/file.c         |  9 +++++----
 include/linux/fs.h      |  5 ++++-
 include/linux/pagemap.h |  3 ++-
 mm/filemap.c            | 13 +++++++++----
 22 files changed, 49 insertions(+), 36 deletions(-)

(limited to 'include/linux')

diff --git a/fs/affs/file.c b/fs/affs/file.c
index 1377b1240b6e..9246cb4aa018 100644
--- a/fs/affs/file.c
+++ b/fs/affs/file.c
@@ -628,7 +628,7 @@ static int affs_write_begin_ofs(struct file *file, struct address_space *mapping
 	}
 
 	index = pos >> PAGE_CACHE_SHIFT;
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page)
 		return -ENOMEM;
 	*pagep = page;
diff --git a/fs/afs/write.c b/fs/afs/write.c
index d6b85dab35fc..3fb36d433621 100644
--- a/fs/afs/write.c
+++ b/fs/afs/write.c
@@ -144,7 +144,7 @@ int afs_write_begin(struct file *file, struct address_space *mapping,
 	candidate->state = AFS_WBACK_PENDING;
 	init_waitqueue_head(&candidate->waitq);
 
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page) {
 		kfree(candidate);
 		return -ENOMEM;
diff --git a/fs/buffer.c b/fs/buffer.c
index 776ae091d3b0..a13f09b696f7 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -1996,7 +1996,7 @@ int block_write_begin(struct file *file, struct address_space *mapping,
 	page = *pagep;
 	if (page == NULL) {
 		ownpage = 1;
-		page = __grab_cache_page(mapping, index);
+		page = grab_cache_page_write_begin(mapping, index, flags);
 		if (!page) {
 			status = -ENOMEM;
 			goto out;
@@ -2502,7 +2502,7 @@ int nobh_write_begin(struct file *file, struct address_space *mapping,
 	from = pos & (PAGE_CACHE_SIZE - 1);
 	to = from + len;
 
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page)
 		return -ENOMEM;
 	*pagep = page;
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index b1e1fc6a6e6a..12bb656fbe75 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -2074,7 +2074,7 @@ static int cifs_write_begin(struct file *file, struct address_space *mapping,
 
 	cFYI(1, ("write_begin from %lld len %d", (long long)pos, len));
 
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page) {
 		rc = -ENOMEM;
 		goto out;
diff --git a/fs/ecryptfs/mmap.c b/fs/ecryptfs/mmap.c
index 04d7b3fa1ac6..46cec2b69796 100644
--- a/fs/ecryptfs/mmap.c
+++ b/fs/ecryptfs/mmap.c
@@ -288,7 +288,7 @@ static int ecryptfs_write_begin(struct file *file,
 	loff_t prev_page_end_size;
 	int rc = 0;
 
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page)
 		return -ENOMEM;
 	*pagep = page;
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index c4bdccf976b5..5fa453b49a64 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -1161,7 +1161,7 @@ static int ext3_write_begin(struct file *file, struct address_space *mapping,
 	to = from + len;
 
 retry:
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page)
 		return -ENOMEM;
 	*pagep = page;
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 297ea8dfac7c..1dd2abe6313e 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -2175,8 +2175,7 @@ retry:
 		 * We have a transaction open.  All is sweetness.  It also sets
 		 * i_size in generic_commit_write().
 		 */
-		err = __page_symlink(inode, symname, l,
-				mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS);
+		err = __page_symlink(inode, symname, l, 1);
 		if (err) {
 			drop_nlink(inode);
 			unlock_new_inode(inode);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 7c3325e0b005..6702a49992a6 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -1346,7 +1346,7 @@ retry:
 		goto out;
 	}
 
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page) {
 		ext4_journal_stop(handle);
 		ret = -ENOMEM;
@@ -2550,7 +2550,7 @@ retry:
 		goto out;
 	}
 
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page) {
 		ext4_journal_stop(handle);
 		ret = -ENOMEM;
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index da98a9012fa5..9fd2a5e1be4d 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -2212,8 +2212,7 @@ retry:
 		 * We have a transaction open.  All is sweetness.  It also sets
 		 * i_size in generic_commit_write().
 		 */
-		err = __page_symlink(inode, symname, l,
-				mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS);
+		err = __page_symlink(inode, symname, l, 1);
 		if (err) {
 			clear_nlink(inode);
 			unlock_new_inode(inode);
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 34930a964b82..4c9ee7011265 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -646,7 +646,7 @@ static int fuse_write_begin(struct file *file, struct address_space *mapping,
 {
 	pgoff_t index = pos >> PAGE_CACHE_SHIFT;
 
-	*pagep = __grab_cache_page(mapping, index);
+	*pagep = grab_cache_page_write_begin(mapping, index, flags);
 	if (!*pagep)
 		return -ENOMEM;
 	return 0;
@@ -779,7 +779,7 @@ static ssize_t fuse_fill_write_pages(struct fuse_req *req,
 			break;
 
 		err = -ENOMEM;
-		page = __grab_cache_page(mapping, index);
+		page = grab_cache_page_write_begin(mapping, index, 0);
 		if (!page)
 			break;
 
diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c
index 27563816e1c5..15f710f2d4da 100644
--- a/fs/gfs2/ops_address.c
+++ b/fs/gfs2/ops_address.c
@@ -675,7 +675,7 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping,
 		goto out_trans_fail;
 
 	error = -ENOMEM;
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	*pagep = page;
 	if (unlikely(!page))
 		goto out_endtrans;
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index 3a31451ac170..5c538e0ec14b 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -501,7 +501,7 @@ int hostfs_write_begin(struct file *file, struct address_space *mapping,
 {
 	pgoff_t index = pos >> PAGE_CACHE_SHIFT;
 
-	*pagep = __grab_cache_page(mapping, index);
+	*pagep = grab_cache_page_write_begin(mapping, index, flags);
 	if (!*pagep)
 		return -ENOMEM;
 	return 0;
diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c
index 5a98aa87c853..5edc2bf20581 100644
--- a/fs/jffs2/file.c
+++ b/fs/jffs2/file.c
@@ -132,7 +132,7 @@ static int jffs2_write_begin(struct file *filp, struct address_space *mapping,
 	uint32_t pageofs = index << PAGE_CACHE_SHIFT;
 	int ret = 0;
 
-	pg = __grab_cache_page(mapping, index);
+	pg = grab_cache_page_write_begin(mapping, index, flags);
 	if (!pg)
 		return -ENOMEM;
 	*pagep = pg;
diff --git a/fs/libfs.c b/fs/libfs.c
index e960a8321902..bdaec17fa388 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -360,7 +360,7 @@ int simple_write_begin(struct file *file, struct address_space *mapping,
 	index = pos >> PAGE_CACHE_SHIFT;
 	from = pos & (PAGE_CACHE_SIZE - 1);
 
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page)
 		return -ENOMEM;
 
diff --git a/fs/namei.c b/fs/namei.c
index dd5c9f0bf829..df2d3df4f049 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2817,18 +2817,23 @@ void page_put_link(struct dentry *dentry, struct nameidata *nd, void *cookie)
 	}
 }
 
-int __page_symlink(struct inode *inode, const char *symname, int len,
-		gfp_t gfp_mask)
+/*
+ * The nofs argument instructs pagecache_write_begin to pass AOP_FLAG_NOFS
+ */
+int __page_symlink(struct inode *inode, const char *symname, int len, int nofs)
 {
 	struct address_space *mapping = inode->i_mapping;
 	struct page *page;
 	void *fsdata;
 	int err;
 	char *kaddr;
+	unsigned int flags = AOP_FLAG_UNINTERRUPTIBLE;
+	if (nofs)
+		flags |= AOP_FLAG_NOFS;
 
 retry:
 	err = pagecache_write_begin(NULL, mapping, 0, len-1,
-				AOP_FLAG_UNINTERRUPTIBLE, &page, &fsdata);
+				flags, &page, &fsdata);
 	if (err)
 		goto fail;
 
@@ -2852,7 +2857,7 @@ fail:
 int page_symlink(struct inode *inode, const char *symname, int len)
 {
 	return __page_symlink(inode, symname, len,
-			mapping_gfp_mask(inode->i_mapping));
+			!(mapping_gfp_mask(inode->i_mapping) & __GFP_FS));
 }
 
 const struct inode_operations page_symlink_inode_operations = {
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index d319b49f8f06..90f292b520d2 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -354,7 +354,7 @@ static int nfs_write_begin(struct file *file, struct address_space *mapping,
 		file->f_path.dentry->d_name.name,
 		mapping->host->i_ino, len, (long long) pos);
 
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page)
 		return -ENOMEM;
 	*pagep = page;
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 145c2d3e5e01..ed04f47007f8 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -2561,7 +2561,7 @@ static int reiserfs_write_begin(struct file *file,
 	}
 
 	index = pos >> PAGE_CACHE_SHIFT;
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page)
 		return -ENOMEM;
 	*pagep = page;
diff --git a/fs/smbfs/file.c b/fs/smbfs/file.c
index e4f8d51a5553..92d5e8ffb639 100644
--- a/fs/smbfs/file.c
+++ b/fs/smbfs/file.c
@@ -297,7 +297,7 @@ static int smb_write_begin(struct file *file, struct address_space *mapping,
 			struct page **pagep, void **fsdata)
 {
 	pgoff_t index = pos >> PAGE_CACHE_SHIFT;
-	*pagep = __grab_cache_page(mapping, index);
+	*pagep = grab_cache_page_write_begin(mapping, index, flags);
 	if (!*pagep)
 		return -ENOMEM;
 	return 0;
diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index fe82d2464d46..bf37374567fa 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -219,7 +219,8 @@ static void release_existing_page_budget(struct ubifs_info *c)
 }
 
 static int write_begin_slow(struct address_space *mapping,
-			    loff_t pos, unsigned len, struct page **pagep)
+			    loff_t pos, unsigned len, struct page **pagep,
+			    unsigned flags)
 {
 	struct inode *inode = mapping->host;
 	struct ubifs_info *c = inode->i_sb->s_fs_info;
@@ -247,7 +248,7 @@ static int write_begin_slow(struct address_space *mapping,
 	if (unlikely(err))
 		return err;
 
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (unlikely(!page)) {
 		ubifs_release_budget(c, &req);
 		return -ENOMEM;
@@ -438,7 +439,7 @@ static int ubifs_write_begin(struct file *file, struct address_space *mapping,
 		return -EROFS;
 
 	/* Try out the fast-path part first */
-	page = __grab_cache_page(mapping, index);
+	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (unlikely(!page))
 		return -ENOMEM;
 
@@ -483,7 +484,7 @@ static int ubifs_write_begin(struct file *file, struct address_space *mapping,
 		unlock_page(page);
 		page_cache_release(page);
 
-		return write_begin_slow(mapping, pos, len, pagep);
+		return write_begin_slow(mapping, pos, len, pagep, flags);
 	}
 
 	/*
diff --git a/include/linux/fs.h b/include/linux/fs.h
index e2170ee21e18..f2a3010140e3 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -423,6 +423,9 @@ enum positive_aop_returns {
 
 #define AOP_FLAG_UNINTERRUPTIBLE	0x0001 /* will not do a short write */
 #define AOP_FLAG_CONT_EXPAND		0x0002 /* called from cont_expand */
+#define AOP_FLAG_NOFS			0x0004 /* used by filesystem to direct
+						* helper code (eg buffer layer)
+						* to clear GFP_FS from alloc */
 
 /*
  * oh the beauties of C type declarations.
@@ -2035,7 +2038,7 @@ extern int page_readlink(struct dentry *, char __user *, int);
 extern void *page_follow_link_light(struct dentry *, struct nameidata *);
 extern void page_put_link(struct dentry *, struct nameidata *, void *);
 extern int __page_symlink(struct inode *inode, const char *symname, int len,
-		gfp_t gfp_mask);
+		int nofs);
 extern int page_symlink(struct inode *inode, const char *symname, int len);
 extern const struct inode_operations page_symlink_inode_operations;
 extern int generic_readlink(struct dentry *, char __user *, int);
diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index 709742be02f0..01ca0856caff 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -241,7 +241,8 @@ unsigned find_get_pages_contig(struct address_space *mapping, pgoff_t start,
 unsigned find_get_pages_tag(struct address_space *mapping, pgoff_t *index,
 			int tag, unsigned int nr_pages, struct page **pages);
 
-struct page *__grab_cache_page(struct address_space *mapping, pgoff_t index);
+struct page *grab_cache_page_write_begin(struct address_space *mapping,
+			pgoff_t index, unsigned flags);
 
 /*
  * Returns locked page at given index in given cache, creating it if needed.
diff --git a/mm/filemap.c b/mm/filemap.c
index f3e5f8944d17..f8c69273c37f 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -2140,19 +2140,24 @@ EXPORT_SYMBOL(generic_file_direct_write);
  * Find or create a page at the given pagecache position. Return the locked
  * page. This function is specifically for buffered writes.
  */
-struct page *__grab_cache_page(struct address_space *mapping, pgoff_t index)
+struct page *grab_cache_page_write_begin(struct address_space *mapping,
+					pgoff_t index, unsigned flags)
 {
 	int status;
 	struct page *page;
+	gfp_t gfp_notmask = 0;
+	if (flags & AOP_FLAG_NOFS)
+		gfp_notmask = __GFP_FS;
 repeat:
 	page = find_lock_page(mapping, index);
 	if (likely(page))
 		return page;
 
-	page = page_cache_alloc(mapping);
+	page = __page_cache_alloc(mapping_gfp_mask(mapping) & ~gfp_notmask);
 	if (!page)
 		return NULL;
-	status = add_to_page_cache_lru(page, mapping, index, GFP_KERNEL);
+	status = add_to_page_cache_lru(page, mapping, index,
+						GFP_KERNEL & ~gfp_notmask);
 	if (unlikely(status)) {
 		page_cache_release(page);
 		if (status == -EEXIST)
@@ -2161,7 +2166,7 @@ repeat:
 	}
 	return page;
 }
-EXPORT_SYMBOL(__grab_cache_page);
+EXPORT_SYMBOL(grab_cache_page_write_begin);
 
 static ssize_t generic_perform_write(struct file *file,
 				struct iov_iter *i, loff_t pos)
-- 
cgit v1.2.3


From 099e657625e801adf82054c8050dde5aceb68452 Mon Sep 17 00:00:00 2001
From: Alessandro Zummo <a.zummo@towertech.it>
Date: Sun, 4 Jan 2009 12:00:54 -0800
Subject: rtc: add alarm/update irq interfaces

Add standard interfaces for alarm/update irqs enabling.  Drivers are no
more required to implement equivalent ioctl code as rtc-dev will provide
it.

UIE emulation should now be handled correctly and will work even for those
RTC drivers who cannot be configured to do both UIE and AIE.

Signed-off-by: Alessandro Zummo <a.zummo@towertech.it>
Cc: David Brownell <david-b@pacbell.net>
Cc: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/Kconfig     |  6 +++++-
 drivers/rtc/interface.c | 54 +++++++++++++++++++++++++++++++++++++++++++++++++
 drivers/rtc/rtc-dev.c   | 51 +++++++++++++++++++++++++++++++---------------
 include/linux/rtc.h     |  8 +++++++-
 4 files changed, 101 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig
index 123092d8a984..165a81843357 100644
--- a/drivers/rtc/Kconfig
+++ b/drivers/rtc/Kconfig
@@ -102,9 +102,13 @@ config RTC_INTF_DEV_UIE_EMUL
 	depends on RTC_INTF_DEV
 	help
 	  Provides an emulation for RTC_UIE if the underlying rtc chip
-	  driver does not expose RTC_UIE ioctls.  Those requests generate
+	  driver does not expose RTC_UIE ioctls. Those requests generate
 	  once-per-second update interrupts, used for synchronization.
 
+	  The emulation code will read the time from the hardware
+	  clock several times per second, please enable this option
+	  only if you know that you really need it.
+
 config RTC_DRV_TEST
 	tristate "Test driver/device"
 	help
diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c
index a04c1b6b1575..fd2c652504ff 100644
--- a/drivers/rtc/interface.c
+++ b/drivers/rtc/interface.c
@@ -307,6 +307,60 @@ int rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm)
 }
 EXPORT_SYMBOL_GPL(rtc_set_alarm);
 
+int rtc_alarm_irq_enable(struct rtc_device *rtc, unsigned int enabled)
+{
+	int err = mutex_lock_interruptible(&rtc->ops_lock);
+	if (err)
+		return err;
+
+	if (!rtc->ops)
+		err = -ENODEV;
+	else if (!rtc->ops->alarm_irq_enable)
+		err = -EINVAL;
+	else
+		err = rtc->ops->alarm_irq_enable(rtc->dev.parent, enabled);
+
+	mutex_unlock(&rtc->ops_lock);
+	return err;
+}
+EXPORT_SYMBOL_GPL(rtc_alarm_irq_enable);
+
+int rtc_update_irq_enable(struct rtc_device *rtc, unsigned int enabled)
+{
+	int err = mutex_lock_interruptible(&rtc->ops_lock);
+	if (err)
+		return err;
+
+#ifdef CONFIG_RTC_INTF_DEV_UIE_EMUL
+	if (enabled == 0 && rtc->uie_irq_active) {
+		mutex_unlock(&rtc->ops_lock);
+		return rtc_dev_update_irq_enable_emul(rtc, enabled);
+	}
+#endif
+
+	if (!rtc->ops)
+		err = -ENODEV;
+	else if (!rtc->ops->update_irq_enable)
+		err = -EINVAL;
+	else
+		err = rtc->ops->update_irq_enable(rtc->dev.parent, enabled);
+
+	mutex_unlock(&rtc->ops_lock);
+
+#ifdef CONFIG_RTC_INTF_DEV_UIE_EMUL
+	/*
+	 * Enable emulation if the driver did not provide
+	 * the update_irq_enable function pointer or if returned
+	 * -EINVAL to signal that it has been configured without
+	 * interrupts or that are not available at the moment.
+	 */
+	if (err == -EINVAL)
+		err = rtc_dev_update_irq_enable_emul(rtc, enabled);
+#endif
+	return err;
+}
+EXPORT_SYMBOL_GPL(rtc_update_irq_enable);
+
 /**
  * rtc_update_irq - report RTC periodic, alarm, and/or update irqs
  * @rtc: the rtc device
diff --git a/drivers/rtc/rtc-dev.c b/drivers/rtc/rtc-dev.c
index ecdea44ae4e5..45152f4952d6 100644
--- a/drivers/rtc/rtc-dev.c
+++ b/drivers/rtc/rtc-dev.c
@@ -92,10 +92,10 @@ static void rtc_uie_timer(unsigned long data)
 	spin_unlock_irqrestore(&rtc->irq_lock, flags);
 }
 
-static void clear_uie(struct rtc_device *rtc)
+static int clear_uie(struct rtc_device *rtc)
 {
 	spin_lock_irq(&rtc->irq_lock);
-	if (rtc->irq_active) {
+	if (rtc->uie_irq_active) {
 		rtc->stop_uie_polling = 1;
 		if (rtc->uie_timer_active) {
 			spin_unlock_irq(&rtc->irq_lock);
@@ -108,9 +108,10 @@ static void clear_uie(struct rtc_device *rtc)
 			flush_scheduled_work();
 			spin_lock_irq(&rtc->irq_lock);
 		}
-		rtc->irq_active = 0;
+		rtc->uie_irq_active = 0;
 	}
 	spin_unlock_irq(&rtc->irq_lock);
+	return 0;
 }
 
 static int set_uie(struct rtc_device *rtc)
@@ -122,8 +123,8 @@ static int set_uie(struct rtc_device *rtc)
 	if (err)
 		return err;
 	spin_lock_irq(&rtc->irq_lock);
-	if (!rtc->irq_active) {
-		rtc->irq_active = 1;
+	if (!rtc->uie_irq_active) {
+		rtc->uie_irq_active = 1;
 		rtc->stop_uie_polling = 0;
 		rtc->oldsecs = tm.tm_sec;
 		rtc->uie_task_active = 1;
@@ -134,6 +135,16 @@ static int set_uie(struct rtc_device *rtc)
 	spin_unlock_irq(&rtc->irq_lock);
 	return 0;
 }
+
+int rtc_dev_update_irq_enable_emul(struct rtc_device *rtc, unsigned int enabled)
+{
+	if (enabled)
+		return set_uie(rtc);
+	else
+		return clear_uie(rtc);
+}
+EXPORT_SYMBOL(rtc_dev_update_irq_enable_emul);
+
 #endif /* CONFIG_RTC_INTF_DEV_UIE_EMUL */
 
 static ssize_t
@@ -357,6 +368,22 @@ static long rtc_dev_ioctl(struct file *file,
 		err = rtc_irq_set_state(rtc, NULL, 0);
 		break;
 
+	case RTC_AIE_ON:
+		mutex_unlock(&rtc->ops_lock);
+		return rtc_alarm_irq_enable(rtc, 1);
+
+	case RTC_AIE_OFF:
+		mutex_unlock(&rtc->ops_lock);
+		return rtc_alarm_irq_enable(rtc, 0);
+
+	case RTC_UIE_ON:
+		mutex_unlock(&rtc->ops_lock);
+		return rtc_update_irq_enable(rtc, 1);
+
+	case RTC_UIE_OFF:
+		mutex_unlock(&rtc->ops_lock);
+		return rtc_update_irq_enable(rtc, 0);
+
 	case RTC_IRQP_SET:
 		err = rtc_irq_set_freq(rtc, NULL, arg);
 		break;
@@ -401,17 +428,6 @@ static long rtc_dev_ioctl(struct file *file,
 			err = -EFAULT;
 		return err;
 
-#ifdef CONFIG_RTC_INTF_DEV_UIE_EMUL
-	case RTC_UIE_OFF:
-		mutex_unlock(&rtc->ops_lock);
-		clear_uie(rtc);
-		return 0;
-
-	case RTC_UIE_ON:
-		mutex_unlock(&rtc->ops_lock);
-		err = set_uie(rtc);
-		return err;
-#endif
 	default:
 		err = -ENOTTY;
 		break;
@@ -440,7 +456,10 @@ static int rtc_dev_release(struct inode *inode, struct file *file)
 	 * Leave the alarm alone; it may be set to trigger a system wakeup
 	 * later, or be used by kernel code, and is a one-shot event anyway.
 	 */
+
+	/* Keep ioctl until all drivers are converted */
 	rtc_dev_ioctl(file, RTC_UIE_OFF, 0);
+	rtc_update_irq_enable(rtc, 0);
 	rtc_irq_set_state(rtc, NULL, 0);
 
 	if (rtc->ops->release)
diff --git a/include/linux/rtc.h b/include/linux/rtc.h
index 91f597ad6acc..4046b75563c1 100644
--- a/include/linux/rtc.h
+++ b/include/linux/rtc.h
@@ -145,6 +145,8 @@ struct rtc_class_ops {
 	int (*irq_set_state)(struct device *, int enabled);
 	int (*irq_set_freq)(struct device *, int freq);
 	int (*read_callback)(struct device *, int data);
+	int (*alarm_irq_enable)(struct device *, unsigned int enabled);
+	int (*update_irq_enable)(struct device *, unsigned int enabled);
 };
 
 #define RTC_DEVICE_NAME_SIZE 20
@@ -181,7 +183,7 @@ struct rtc_device
 	struct timer_list uie_timer;
 	/* Those fields are protected by rtc->irq_lock */
 	unsigned int oldsecs;
-	unsigned int irq_active:1;
+	unsigned int uie_irq_active:1;
 	unsigned int stop_uie_polling:1;
 	unsigned int uie_task_active:1;
 	unsigned int uie_timer_active:1;
@@ -216,6 +218,10 @@ extern int rtc_irq_set_state(struct rtc_device *rtc,
 				struct rtc_task *task, int enabled);
 extern int rtc_irq_set_freq(struct rtc_device *rtc,
 				struct rtc_task *task, int freq);
+extern int rtc_update_irq_enable(struct rtc_device *rtc, unsigned int enabled);
+extern int rtc_alarm_irq_enable(struct rtc_device *rtc, unsigned int enabled);
+extern int rtc_dev_update_irq_enable_emul(struct rtc_device *rtc,
+						unsigned int enabled);
 
 typedef struct rtc_task {
 	void (*func)(void *private_data);
-- 
cgit v1.2.3


From 088af9a6e05d51e7c3dc85d45d8b7a52c3ee08d7 Mon Sep 17 00:00:00 2001
From: Helge Deller <deller@gmx.de>
Date: Wed, 31 Dec 2008 12:31:18 +0100
Subject: module: fix module loading failure of large kernel modules for parisc

When creating the final layout of a kernel module in memory, allow the
module loader to reserve some additional memory in front of a given section.
This is currently only needed for the parisc port which needs to put the
stub entries there to fulfill the 17/22bit PCREL relocations with large
kernel modules like xfs.

Signed-off-by: Helge Deller <deller@gmx.de>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (renamed fn)
---
 include/linux/moduleloader.h |  3 +++
 kernel/module.c              | 16 +++++++++++++---
 2 files changed, 16 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/moduleloader.h b/include/linux/moduleloader.h
index eb1033957486..c1f40c2f7ffb 100644
--- a/include/linux/moduleloader.h
+++ b/include/linux/moduleloader.h
@@ -13,6 +13,9 @@ int module_frob_arch_sections(Elf_Ehdr *hdr,
 			      char *secstrings,
 			      struct module *mod);
 
+/* Additional bytes needed by arch in front of individual sections */
+unsigned int arch_mod_section_prepend(struct module *mod, unsigned int section);
+
 /* Allocator used for allocating struct module, core sections and init
    sections.  Returns NULL on failure. */
 void *module_alloc(unsigned long size);
diff --git a/kernel/module.c b/kernel/module.c
index d3d254571bda..4299aefc20b8 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -1578,11 +1578,21 @@ static int simplify_symbols(Elf_Shdr *sechdrs,
 	return ret;
 }
 
+/* Additional bytes needed by arch in front of individual sections */
+unsigned int __weak arch_mod_section_prepend(struct module *mod,
+					     unsigned int section)
+{
+	/* default implementation just returns zero */
+	return 0;
+}
+
 /* Update size with this section: return offset. */
-static long get_offset(unsigned int *size, Elf_Shdr *sechdr)
+static long get_offset(struct module *mod, unsigned int *size,
+		       Elf_Shdr *sechdr, unsigned int section)
 {
 	long ret;
 
+	*size += arch_mod_section_prepend(mod, section);
 	ret = ALIGN(*size, sechdr->sh_addralign ?: 1);
 	*size = ret + sechdr->sh_size;
 	return ret;
@@ -1622,7 +1632,7 @@ static void layout_sections(struct module *mod,
 			    || strncmp(secstrings + s->sh_name,
 				       ".init", 5) == 0)
 				continue;
-			s->sh_entsize = get_offset(&mod->core_size, s);
+			s->sh_entsize = get_offset(mod, &mod->core_size, s, i);
 			DEBUGP("\t%s\n", secstrings + s->sh_name);
 		}
 		if (m == 0)
@@ -1640,7 +1650,7 @@ static void layout_sections(struct module *mod,
 			    || strncmp(secstrings + s->sh_name,
 				       ".init", 5) != 0)
 				continue;
-			s->sh_entsize = (get_offset(&mod->init_size, s)
+			s->sh_entsize = (get_offset(mod, &mod->init_size, s, i)
 					 | INIT_OFFSET_MASK);
 			DEBUGP("\t%s\n", secstrings + s->sh_name);
 		}
-- 
cgit v1.2.3


From 9ea09af3bd3090e8349ca2899ca2011bd94cda85 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 22 Dec 2008 12:36:30 +0100
Subject: stop_machine: introduce stop_machine_create/destroy.

Introduce stop_machine_create/destroy. With this interface subsystems
that need a non-failing stop_machine environment can create the
stop_machine machine threads before actually calling stop_machine.
When the threads aren't needed anymore they can be killed with
stop_machine_destroy again.

When stop_machine gets called and the threads aren't present they
will be created and destroyed automatically. This restores the old
behaviour of stop_machine.

This patch also converts cpu hotplug to the new interface since it
is special: cpu_down calls __stop_machine instead of stop_machine.
However the kstop threads will only be created when stop_machine
gets called.

Changing the code so that the threads would be created automatically
on __stop_machine is currently not possible: when __stop_machine gets
called we hold cpu_add_remove_lock, which is the same lock that
create_rt_workqueue would take. So the workqueue needs to be created
before the cpu hotplug code locks cpu_add_remove_lock.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 include/linux/stop_machine.h | 22 ++++++++++++++++++
 kernel/cpu.c                 |  6 ++++-
 kernel/stop_machine.c        | 55 ++++++++++++++++++++++++++++++++++++--------
 3 files changed, 72 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/stop_machine.h b/include/linux/stop_machine.h
index 74d59a641362..baba3a23a814 100644
--- a/include/linux/stop_machine.h
+++ b/include/linux/stop_machine.h
@@ -35,6 +35,24 @@ int stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus);
  * won't come or go while it's being called.  Used by hotplug cpu.
  */
 int __stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus);
+
+/**
+ * stop_machine_create: create all stop_machine threads
+ *
+ * Description: This causes all stop_machine threads to be created before
+ * stop_machine actually gets called. This can be used by subsystems that
+ * need a non failing stop_machine infrastructure.
+ */
+int stop_machine_create(void);
+
+/**
+ * stop_machine_destroy: destroy all stop_machine threads
+ *
+ * Description: This causes all stop_machine threads which were created with
+ * stop_machine_create to be destroyed again.
+ */
+void stop_machine_destroy(void);
+
 #else
 
 static inline int stop_machine(int (*fn)(void *), void *data,
@@ -46,5 +64,9 @@ static inline int stop_machine(int (*fn)(void *), void *data,
 	local_irq_enable();
 	return ret;
 }
+
+static inline int stop_machine_create(void) { return 0; }
+static inline void stop_machine_destroy(void) { }
+
 #endif /* CONFIG_SMP */
 #endif /* _LINUX_STOP_MACHINE */
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 47fff3b63cbf..30e74dd6d01b 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -269,8 +269,11 @@ out_release:
 
 int __ref cpu_down(unsigned int cpu)
 {
-	int err = 0;
+	int err;
 
+	err = stop_machine_create();
+	if (err)
+		return err;
 	cpu_maps_update_begin();
 
 	if (cpu_hotplug_disabled) {
@@ -297,6 +300,7 @@ int __ref cpu_down(unsigned int cpu)
 
 out:
 	cpu_maps_update_done();
+	stop_machine_destroy();
 	return err;
 }
 EXPORT_SYMBOL(cpu_down);
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index 286c41722e8c..0cd415ee62a2 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -38,7 +38,10 @@ struct stop_machine_data {
 static unsigned int num_threads;
 static atomic_t thread_ack;
 static DEFINE_MUTEX(lock);
-
+/* setup_lock protects refcount, stop_machine_wq and stop_machine_work. */
+static DEFINE_MUTEX(setup_lock);
+/* Users of stop_machine. */
+static int refcount;
 static struct workqueue_struct *stop_machine_wq;
 static struct stop_machine_data active, idle;
 static const cpumask_t *active_cpus;
@@ -109,6 +112,43 @@ static int chill(void *unused)
 	return 0;
 }
 
+int stop_machine_create(void)
+{
+	mutex_lock(&setup_lock);
+	if (refcount)
+		goto done;
+	stop_machine_wq = create_rt_workqueue("kstop");
+	if (!stop_machine_wq)
+		goto err_out;
+	stop_machine_work = alloc_percpu(struct work_struct);
+	if (!stop_machine_work)
+		goto err_out;
+done:
+	refcount++;
+	mutex_unlock(&setup_lock);
+	return 0;
+
+err_out:
+	if (stop_machine_wq)
+		destroy_workqueue(stop_machine_wq);
+	mutex_unlock(&setup_lock);
+	return -ENOMEM;
+}
+EXPORT_SYMBOL_GPL(stop_machine_create);
+
+void stop_machine_destroy(void)
+{
+	mutex_lock(&setup_lock);
+	refcount--;
+	if (refcount)
+		goto done;
+	destroy_workqueue(stop_machine_wq);
+	free_percpu(stop_machine_work);
+done:
+	mutex_unlock(&setup_lock);
+}
+EXPORT_SYMBOL_GPL(stop_machine_destroy);
+
 int __stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus)
 {
 	struct work_struct *sm_work;
@@ -146,19 +186,14 @@ int stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus)
 {
 	int ret;
 
+	ret = stop_machine_create();
+	if (ret)
+		return ret;
 	/* No CPUs can come up or down during this. */
 	get_online_cpus();
 	ret = __stop_machine(fn, data, cpus);
 	put_online_cpus();
-
+	stop_machine_destroy();
 	return ret;
 }
 EXPORT_SYMBOL_GPL(stop_machine);
-
-static int __init stop_machine_init(void)
-{
-	stop_machine_wq = create_rt_workqueue("kstop");
-	stop_machine_work = alloc_percpu(struct work_struct);
-	return 0;
-}
-core_initcall(stop_machine_init);
-- 
cgit v1.2.3


From 5d38a079ce3971f932bbdc0dc5b887806fabd5dc Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 4 Jan 2009 16:13:40 -0800
Subject: gro: Add page frag support

This patch allows GRO to merge page frags (skb_shinfo(skb)->frags)
in one skb, rather than using the less efficient frag_list.

It also adds a new interface, napi_gro_frags to allow drivers
to inject page frags directly into the stack without allocating
an skb.  This is intended to be the GRO equivalent for LRO's
lro_receive_frags interface.

The existing GSO interface can already handle page frags with
or without an appended frag_list so nothing needs to be changed
there.

The merging itself is rather simple.  We store any new frag entries
after the last existing entry, without checking whether the first
new entry can be merged with the last existing entry.  Making this
check would actually be easy but since no existing driver can
produce contiguous frags anyway it would just be mental masturbation.

If the total number of entries would exceed the capacity of a
single skb, we simply resort to using frag_list as we do now.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 16 ++++++++-
 net/core/dev.c            | 91 ++++++++++++++++++++++++++++++++++++++++++++---
 net/core/skbuff.c         | 14 +++++++-
 3 files changed, 114 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 41e1224651cf..c28bbba3c23d 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -313,10 +313,11 @@ struct napi_struct {
 #ifdef CONFIG_NETPOLL
 	spinlock_t		poll_lock;
 	int			poll_owner;
-	struct net_device	*dev;
 #endif
+	struct net_device	*dev;
 	struct list_head	dev_list;
 	struct sk_buff		*gro_list;
+	struct sk_buff		*skb;
 };
 
 enum
@@ -990,6 +991,9 @@ struct napi_gro_cb {
 
 	/* Number of segments aggregated. */
 	int count;
+
+	/* Free the skb? */
+	int free;
 };
 
 #define NAPI_GRO_CB(skb) ((struct napi_gro_cb *)(skb)->cb)
@@ -1011,6 +1015,14 @@ struct packet_type {
 	struct list_head	list;
 };
 
+struct napi_gro_fraginfo {
+	skb_frag_t frags[MAX_SKB_FRAGS];
+	unsigned int nr_frags;
+	unsigned int ip_summed;
+	unsigned int len;
+	__wsum csum;
+};
+
 #include <linux/interrupt.h>
 #include <linux/notifier.h>
 
@@ -1363,6 +1375,8 @@ extern int		netif_receive_skb(struct sk_buff *skb);
 extern void		napi_gro_flush(struct napi_struct *napi);
 extern int		napi_gro_receive(struct napi_struct *napi,
 					 struct sk_buff *skb);
+extern int		napi_gro_frags(struct napi_struct *napi,
+				       struct napi_gro_fraginfo *info);
 extern void		netif_nit_deliver(struct sk_buff *skb);
 extern int		dev_valid_name(const char *name);
 extern int		dev_ioctl(struct net *net, unsigned int cmd, void __user *);
diff --git a/net/core/dev.c b/net/core/dev.c
index 1e1a68066457..382df6c09eec 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -132,6 +132,9 @@
 /* Instead of increasing this, you should create a hash table. */
 #define MAX_GRO_SKBS 8
 
+/* This should be increased if a protocol with a bigger head is added. */
+#define GRO_MAX_HEAD (MAX_HEADER + 128)
+
 /*
  *	The list of packet types we will receive (as opposed to discard)
  *	and the routines to invoke.
@@ -2345,7 +2348,7 @@ static int napi_gro_complete(struct sk_buff *skb)
 	struct list_head *head = &ptype_base[ntohs(type) & PTYPE_HASH_MASK];
 	int err = -ENOENT;
 
-	if (!skb_shinfo(skb)->frag_list)
+	if (NAPI_GRO_CB(skb)->count == 1)
 		goto out;
 
 	rcu_read_lock();
@@ -2384,7 +2387,7 @@ void napi_gro_flush(struct napi_struct *napi)
 }
 EXPORT_SYMBOL(napi_gro_flush);
 
-int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
+static int __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 {
 	struct sk_buff **pp = NULL;
 	struct packet_type *ptype;
@@ -2393,6 +2396,7 @@ int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 	int count = 0;
 	int same_flow;
 	int mac_len;
+	int free;
 
 	if (!(skb->dev->features & NETIF_F_GRO))
 		goto normal;
@@ -2409,6 +2413,7 @@ int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 		skb->mac_len = mac_len;
 		NAPI_GRO_CB(skb)->same_flow = 0;
 		NAPI_GRO_CB(skb)->flush = 0;
+		NAPI_GRO_CB(skb)->free = 0;
 
 		for (p = napi->gro_list; p; p = p->next) {
 			count++;
@@ -2428,6 +2433,7 @@ int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 		goto normal;
 
 	same_flow = NAPI_GRO_CB(skb)->same_flow;
+	free = NAPI_GRO_CB(skb)->free;
 
 	if (pp) {
 		struct sk_buff *nskb = *pp;
@@ -2452,13 +2458,86 @@ int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 	napi->gro_list = skb;
 
 ok:
-	return NET_RX_SUCCESS;
+	return free;
 
 normal:
-	return netif_receive_skb(skb);
+	return -1;
+}
+
+int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
+{
+	switch (__napi_gro_receive(napi, skb)) {
+	case -1:
+		return netif_receive_skb(skb);
+
+	case 1:
+		kfree_skb(skb);
+		break;
+	}
+
+	return NET_RX_SUCCESS;
 }
 EXPORT_SYMBOL(napi_gro_receive);
 
+int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info)
+{
+	struct net_device *dev = napi->dev;
+	struct sk_buff *skb = napi->skb;
+	int err = NET_RX_DROP;
+
+	napi->skb = NULL;
+
+	if (!skb) {
+		skb = netdev_alloc_skb(dev, GRO_MAX_HEAD + NET_IP_ALIGN);
+		if (!skb)
+			goto out;
+
+		skb_reserve(skb, NET_IP_ALIGN);
+	}
+
+	BUG_ON(info->nr_frags > MAX_SKB_FRAGS);
+	skb_shinfo(skb)->nr_frags = info->nr_frags;
+	memcpy(skb_shinfo(skb)->frags, info->frags, sizeof(info->frags));
+
+	skb->data_len = info->len;
+	skb->len += info->len;
+	skb->truesize += info->len;
+
+	if (!pskb_may_pull(skb, ETH_HLEN))
+		goto reuse;
+
+	err = NET_RX_SUCCESS;
+
+	skb->protocol = eth_type_trans(skb, dev);
+
+	skb->ip_summed = info->ip_summed;
+	skb->csum = info->csum;
+
+	switch (__napi_gro_receive(napi, skb)) {
+	case -1:
+		return netif_receive_skb(skb);
+
+	case 0:
+		goto out;
+	}
+
+reuse:
+	skb_shinfo(skb)->nr_frags = 0;
+
+	skb->len -= skb->data_len;
+	skb->truesize -= skb->data_len;
+	skb->data_len = 0;
+
+	__skb_pull(skb, skb_headlen(skb));
+	skb_reserve(skb, NET_IP_ALIGN - skb_headroom(skb));
+
+	napi->skb = skb;
+
+out:
+	return err;
+}
+EXPORT_SYMBOL(napi_gro_frags);
+
 static int process_backlog(struct napi_struct *napi, int quota)
 {
 	int work = 0;
@@ -2537,11 +2616,12 @@ void netif_napi_add(struct net_device *dev, struct napi_struct *napi,
 {
 	INIT_LIST_HEAD(&napi->poll_list);
 	napi->gro_list = NULL;
+	napi->skb = NULL;
 	napi->poll = poll;
 	napi->weight = weight;
 	list_add(&napi->dev_list, &dev->napi_list);
-#ifdef CONFIG_NETPOLL
 	napi->dev = dev;
+#ifdef CONFIG_NETPOLL
 	spin_lock_init(&napi->poll_lock);
 	napi->poll_owner = -1;
 #endif
@@ -2554,6 +2634,7 @@ void netif_napi_del(struct napi_struct *napi)
 	struct sk_buff *skb, *next;
 
 	list_del_init(&napi->dev_list);
+	kfree(napi->skb);
 
 	for (skb = napi->gro_list; skb; skb = next) {
 		next = skb->next;
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 3aafb10325b8..5110b359c758 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -2594,6 +2594,17 @@ int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb)
 
 	if (skb_shinfo(p)->frag_list)
 		goto merge;
+	else if (!skb_headlen(p) && !skb_headlen(skb) &&
+		 skb_shinfo(p)->nr_frags + skb_shinfo(skb)->nr_frags <
+		 MAX_SKB_FRAGS) {
+		memcpy(skb_shinfo(p)->frags + skb_shinfo(p)->nr_frags,
+		       skb_shinfo(skb)->frags,
+		       skb_shinfo(skb)->nr_frags * sizeof(skb_frag_t));
+
+		skb_shinfo(p)->nr_frags += skb_shinfo(skb)->nr_frags;
+		NAPI_GRO_CB(skb)->free = 1;
+		goto done;
+	}
 
 	headroom = skb_headroom(p);
 	nskb = netdev_alloc_skb(p->dev, headroom);
@@ -2628,11 +2639,12 @@ int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb)
 	p = nskb;
 
 merge:
-	NAPI_GRO_CB(p)->count++;
 	p->prev->next = skb;
 	p->prev = skb;
 	skb_header_release(skb);
 
+done:
+	NAPI_GRO_CB(p)->count++;
 	p->data_len += skb->len;
 	p->truesize += skb->len;
 	p->len += skb->len;
-- 
cgit v1.2.3


From e9079cce201784632aed4b1a3121ee38c1ced0b6 Mon Sep 17 00:00:00 2001
From: Steven Whitehouse <swhiteho@redhat.com>
Date: Tue, 14 Oct 2008 14:43:29 +0100
Subject: GFS2: Support for FIEMAP ioctl

This patch implements the FIEMAP ioctl for GFS2. We can use the generic
code (aside from a lock order issue, solved as per Ted Tso's suggestion)
for which I've introduced a new variant of the generic function. We also
have one exception to deal with, namely stuffed files, so we do that
"by hand", setting all the required flags.

This has been tested with a modified (I could only find an old version) of
Eric's test program, and appears to work correctly.

This patch does not currently support FIEMAP of xattrs, but the plan is to add
that feature at some future point.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Cc: Theodore Tso <tytso@mit.edu>
Cc: Eric Sandeen <sandeen@redhat.com>
---
 fs/gfs2/ops_inode.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++
 fs/ioctl.c          | 44 ++++++++++++++++++++++++++++++++++----------
 include/linux/fs.h  |  3 +++
 3 files changed, 83 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c
index d232991b9046..1e24b65e1d23 100644
--- a/fs/gfs2/ops_inode.c
+++ b/fs/gfs2/ops_inode.c
@@ -19,6 +19,7 @@
 #include <linux/gfs2_ondisk.h>
 #include <linux/crc32.h>
 #include <linux/lm_interface.h>
+#include <linux/fiemap.h>
 #include <asm/uaccess.h>
 
 #include "gfs2.h"
@@ -1212,6 +1213,48 @@ static int gfs2_removexattr(struct dentry *dentry, const char *name)
 	return gfs2_ea_remove(GFS2_I(dentry->d_inode), &er);
 }
 
+static int gfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+		       u64 start, u64 len)
+{
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_holder gh;
+	int ret;
+
+	ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC);
+	if (ret)
+		return ret;
+
+	mutex_lock(&inode->i_mutex);
+
+	ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, 0, &gh);
+	if (ret)
+		goto out;
+
+	if (gfs2_is_stuffed(ip)) {
+		u64 phys = ip->i_no_addr << inode->i_blkbits;
+		u64 size = i_size_read(inode);
+		u32 flags = FIEMAP_EXTENT_LAST|FIEMAP_EXTENT_NOT_ALIGNED|
+			    FIEMAP_EXTENT_DATA_INLINE;
+		phys += sizeof(struct gfs2_dinode);
+		phys += start;
+		if (start + len > size)
+			len = size - start;
+		if (start < size)
+			ret = fiemap_fill_next_extent(fieinfo, start, phys,
+						      len, flags);
+		if (ret == 1)
+			ret = 0;
+	} else {
+		ret = __generic_block_fiemap(inode, fieinfo, start, len,
+					     gfs2_block_map);
+	}
+
+	gfs2_glock_dq_uninit(&gh);
+out:
+	mutex_unlock(&inode->i_mutex);
+	return ret;
+}
+
 const struct inode_operations gfs2_file_iops = {
 	.permission = gfs2_permission,
 	.setattr = gfs2_setattr,
@@ -1220,6 +1263,7 @@ const struct inode_operations gfs2_file_iops = {
 	.getxattr = gfs2_getxattr,
 	.listxattr = gfs2_listxattr,
 	.removexattr = gfs2_removexattr,
+	.fiemap = gfs2_fiemap,
 };
 
 const struct inode_operations gfs2_dir_iops = {
@@ -1239,6 +1283,7 @@ const struct inode_operations gfs2_dir_iops = {
 	.getxattr = gfs2_getxattr,
 	.listxattr = gfs2_listxattr,
 	.removexattr = gfs2_removexattr,
+	.fiemap = gfs2_fiemap,
 };
 
 const struct inode_operations gfs2_symlink_iops = {
@@ -1251,5 +1296,6 @@ const struct inode_operations gfs2_symlink_iops = {
 	.getxattr = gfs2_getxattr,
 	.listxattr = gfs2_listxattr,
 	.removexattr = gfs2_removexattr,
+	.fiemap = gfs2_fiemap,
 };
 
diff --git a/fs/ioctl.c b/fs/ioctl.c
index 43e8b2c0664b..cc3f1aa1cf7b 100644
--- a/fs/ioctl.c
+++ b/fs/ioctl.c
@@ -231,7 +231,8 @@ static int ioctl_fiemap(struct file *filp, unsigned long arg)
 #define blk_to_logical(inode, blk) (blk << (inode)->i_blkbits)
 #define logical_to_blk(inode, offset) (offset >> (inode)->i_blkbits);
 
-/*
+/**
+ * __generic_block_fiemap - FIEMAP for block based inodes (no locking)
  * @inode - the inode to map
  * @arg - the pointer to userspace where we copy everything to
  * @get_block - the fs's get_block function
@@ -242,11 +243,15 @@ static int ioctl_fiemap(struct file *filp, unsigned long arg)
  *
  * If it is possible to have data blocks beyond a hole past @inode->i_size, then
  * please do not use this function, it will stop at the first unmapped block
- * beyond i_size
+ * beyond i_size.
+ *
+ * If you use this function directly, you need to do your own locking. Use
+ * generic_block_fiemap if you want the locking done for you.
  */
-int generic_block_fiemap(struct inode *inode,
-			 struct fiemap_extent_info *fieinfo, u64 start,
-			 u64 len, get_block_t *get_block)
+
+int __generic_block_fiemap(struct inode *inode,
+			   struct fiemap_extent_info *fieinfo, u64 start,
+			   u64 len, get_block_t *get_block)
 {
 	struct buffer_head tmp;
 	unsigned int start_blk;
@@ -260,9 +265,6 @@ int generic_block_fiemap(struct inode *inode,
 
 	start_blk = logical_to_blk(inode, start);
 
-	/* guard against change */
-	mutex_lock(&inode->i_mutex);
-
 	length = (long long)min_t(u64, len, i_size_read(inode));
 	map_len = length;
 
@@ -334,14 +336,36 @@ int generic_block_fiemap(struct inode *inode,
 		cond_resched();
 	} while (1);
 
-	mutex_unlock(&inode->i_mutex);
-
 	/* if ret is 1 then we just hit the end of the extent array */
 	if (ret == 1)
 		ret = 0;
 
 	return ret;
 }
+EXPORT_SYMBOL(__generic_block_fiemap);
+
+/**
+ * generic_block_fiemap - FIEMAP for block based inodes
+ * @inode: The inode to map
+ * @fieinfo: The mapping information
+ * @start: The initial block to map
+ * @len: The length of the extect to attempt to map
+ * @get_block: The block mapping function for the fs
+ *
+ * Calls __generic_block_fiemap to map the inode, after taking
+ * the inode's mutex lock.
+ */
+
+int generic_block_fiemap(struct inode *inode,
+			 struct fiemap_extent_info *fieinfo, u64 start,
+			 u64 len, get_block_t *get_block)
+{
+	int ret;
+	mutex_lock(&inode->i_mutex);
+	ret = __generic_block_fiemap(inode, fieinfo, start, len, get_block);
+	mutex_unlock(&inode->i_mutex);
+	return ret;
+}
 EXPORT_SYMBOL(generic_block_fiemap);
 
 #endif  /*  CONFIG_BLOCK  */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index f2a3010140e3..e34bc6925fdf 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2059,6 +2059,9 @@ extern int vfs_fstat(unsigned int, struct kstat *);
 
 extern int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd,
 		    unsigned long arg);
+extern int __generic_block_fiemap(struct inode *inode,
+				  struct fiemap_extent_info *fieinfo, u64 start,
+				  u64 len, get_block_t *get_block);
 extern int generic_block_fiemap(struct inode *inode,
 				struct fiemap_extent_info *fieinfo, u64 start,
 				u64 len, get_block_t *get_block);
-- 
cgit v1.2.3


From ea7d3fef4222cd98556a0b386598268d4dbf6670 Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Sun, 4 Jan 2009 13:03:02 -0800
Subject: rcu: eliminate synchronize_rcu_xxx macro

Impact: cleanup

Expand macro into two files.

The synchronize_rcu_xxx macro is quite ugly and it's only used by two
callers, so expand it instead.  This makes this code easier to change.

Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/rcupdate.h | 12 ------------
 kernel/rcupdate.c        | 11 +++++++++--
 kernel/rcupreempt.c      | 11 ++++++++++-
 3 files changed, 19 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 1168fbcea8d4..921340a7b71c 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -204,18 +204,6 @@ struct rcu_synchronize {
 
 extern void wakeme_after_rcu(struct rcu_head  *head);
 
-#define synchronize_rcu_xxx(name, func) \
-void name(void) \
-{ \
-	struct rcu_synchronize rcu; \
-	\
-	init_completion(&rcu.completion); \
-	/* Will wake me after RCU finished. */ \
-	func(&rcu.head, wakeme_after_rcu); \
-	/* Wait for it. */ \
-	wait_for_completion(&rcu.completion); \
-}
-
 /**
  * synchronize_sched - block until all CPUs have exited any non-preemptive
  * kernel code sequences.
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index ad63af8b2521..d92a76a881aa 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -77,8 +77,15 @@ void wakeme_after_rcu(struct rcu_head  *head)
  * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
  * and may be nested.
  */
-void synchronize_rcu(void);	/* Makes kernel-doc tools happy */
-synchronize_rcu_xxx(synchronize_rcu, call_rcu)
+void synchronize_rcu(void)
+{
+	struct rcu_synchronize rcu;
+	init_completion(&rcu.completion);
+	/* Will wake me after RCU finished. */
+	call_rcu(&rcu.head, wakeme_after_rcu);
+	/* Wait for it. */
+	wait_for_completion(&rcu.completion);
+}
 EXPORT_SYMBOL_GPL(synchronize_rcu);
 
 static void rcu_barrier_callback(struct rcu_head *notused)
diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c
index f9dc8f3720f6..33cfc50781f9 100644
--- a/kernel/rcupreempt.c
+++ b/kernel/rcupreempt.c
@@ -1177,7 +1177,16 @@ EXPORT_SYMBOL_GPL(call_rcu_sched);
  * in -rt this does -not- necessarily result in all currently executing
  * interrupt -handlers- having completed.
  */
-synchronize_rcu_xxx(__synchronize_sched, call_rcu_sched)
+void __synchronize_sched(void)
+{
+	struct rcu_synchronize rcu;
+
+	init_completion(&rcu.completion);
+	/* Will wake me after RCU finished. */
+	call_rcu_sched(&rcu.head, wakeme_after_rcu);
+	/* Wait for it. */
+	wait_for_completion(&rcu.completion);
+}
 EXPORT_SYMBOL_GPL(__synchronize_sched);
 
 /*
-- 
cgit v1.2.3


From a6037b61c2f5fc99c57c15b26d7cfa58bbb34008 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 5 Jan 2009 11:28:22 +0100
Subject: hrtimer: fix recursion deadlock by re-introducing the softirq

Impact: fix rare runtime deadlock

There are a few sites that do:

  spin_lock_irq(&foo)
  hrtimer_start(&bar)
    __run_hrtimer(&bar)
      func()
        spin_lock(&foo)

which obviously deadlocks. In order to avoid this, never call __run_hrtimer()
from hrtimer_start*() context, but instead defer this to softirq context.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/interrupt.h |  3 ++-
 kernel/hrtimer.c          | 60 +++++++++++++++++++++--------------------------
 2 files changed, 29 insertions(+), 34 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index 0702c4d7bdf0..2062833f5f7a 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -253,7 +253,8 @@ enum
 	BLOCK_SOFTIRQ,
 	TASKLET_SOFTIRQ,
 	SCHED_SOFTIRQ,
-	RCU_SOFTIRQ, 	/* Preferable RCU should always be the last softirq */
+	HRTIMER_SOFTIRQ,
+	RCU_SOFTIRQ,	/* Preferable RCU should always be the last softirq */
 
 	NR_SOFTIRQS
 };
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 8010a67cead0..b68e98f4e4c1 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -634,7 +634,6 @@ static inline void hrtimer_init_timer_hres(struct hrtimer *timer)
 {
 }
 
-static void __run_hrtimer(struct hrtimer *timer);
 
 /*
  * When High resolution timers are active, try to reprogram. Note, that in case
@@ -646,13 +645,9 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer,
 					    struct hrtimer_clock_base *base)
 {
 	if (base->cpu_base->hres_active && hrtimer_reprogram(timer, base)) {
-		/*
-		 * XXX: recursion check?
-		 * hrtimer_forward() should round up with timer granularity
-		 * so that we never get into inf recursion here,
-		 * it doesn't do that though
-		 */
-		__run_hrtimer(timer);
+		spin_unlock(&base->cpu_base->lock);
+		raise_softirq_irqoff(HRTIMER_SOFTIRQ);
+		spin_lock(&base->cpu_base->lock);
 		return 1;
 	}
 	return 0;
@@ -705,11 +700,6 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer,
 }
 static inline void hrtimer_init_hres(struct hrtimer_cpu_base *base) { }
 static inline void hrtimer_init_timer_hres(struct hrtimer *timer) { }
-static inline int hrtimer_reprogram(struct hrtimer *timer,
-				    struct hrtimer_clock_base *base)
-{
-	return 0;
-}
 
 #endif /* CONFIG_HIGH_RES_TIMERS */
 
@@ -780,9 +770,11 @@ EXPORT_SYMBOL_GPL(hrtimer_forward);
  *
  * The timer is inserted in expiry order. Insertion into the
  * red black tree is O(log(n)). Must hold the base lock.
+ *
+ * Returns 1 when the new timer is the leftmost timer in the tree.
  */
-static void enqueue_hrtimer(struct hrtimer *timer,
-			    struct hrtimer_clock_base *base, int reprogram)
+static int enqueue_hrtimer(struct hrtimer *timer,
+			   struct hrtimer_clock_base *base)
 {
 	struct rb_node **link = &base->active.rb_node;
 	struct rb_node *parent = NULL;
@@ -814,20 +806,8 @@ static void enqueue_hrtimer(struct hrtimer *timer,
 	 * Insert the timer to the rbtree and check whether it
 	 * replaces the first pending timer
 	 */
-	if (leftmost) {
-		/*
-		 * Reprogram the clock event device. When the timer is already
-		 * expired hrtimer_enqueue_reprogram has either called the
-		 * callback or added it to the pending list and raised the
-		 * softirq.
-		 *
-		 * This is a NOP for !HIGHRES
-		 */
-		if (reprogram && hrtimer_enqueue_reprogram(timer, base))
-			return;
-
+	if (leftmost)
 		base->first = &timer->node;
-	}
 
 	rb_link_node(&timer->node, parent, link);
 	rb_insert_color(&timer->node, &base->active);
@@ -836,6 +816,8 @@ static void enqueue_hrtimer(struct hrtimer *timer,
 	 * state of a possibly running callback.
 	 */
 	timer->state |= HRTIMER_STATE_ENQUEUED;
+
+	return leftmost;
 }
 
 /*
@@ -912,7 +894,7 @@ hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, unsigned long delta_n
 {
 	struct hrtimer_clock_base *base, *new_base;
 	unsigned long flags;
-	int ret;
+	int ret, leftmost;
 
 	base = lock_hrtimer_base(timer, &flags);
 
@@ -940,12 +922,16 @@ hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, unsigned long delta_n
 
 	timer_stats_hrtimer_set_start_info(timer);
 
+	leftmost = enqueue_hrtimer(timer, new_base);
+
 	/*
 	 * Only allow reprogramming if the new base is on this CPU.
 	 * (it might still be on another CPU if the timer was pending)
+	 *
+	 * XXX send_remote_softirq() ?
 	 */
-	enqueue_hrtimer(timer, new_base,
-			new_base->cpu_base == &__get_cpu_var(hrtimer_bases));
+	if (leftmost && new_base->cpu_base == &__get_cpu_var(hrtimer_bases))
+		hrtimer_enqueue_reprogram(timer, new_base);
 
 	unlock_hrtimer_base(timer, &flags);
 
@@ -1163,7 +1149,7 @@ static void __run_hrtimer(struct hrtimer *timer)
 	 */
 	if (restart != HRTIMER_NORESTART) {
 		BUG_ON(timer->state != HRTIMER_STATE_CALLBACK);
-		enqueue_hrtimer(timer, base, 0);
+		enqueue_hrtimer(timer, base);
 	}
 	timer->state &= ~HRTIMER_STATE_CALLBACK;
 }
@@ -1277,6 +1263,11 @@ void hrtimer_peek_ahead_timers(void)
 	local_irq_restore(flags);
 }
 
+static void run_hrtimer_softirq(struct softirq_action *h)
+{
+	hrtimer_peek_ahead_timers();
+}
+
 #endif	/* CONFIG_HIGH_RES_TIMERS */
 
 /*
@@ -1532,7 +1523,7 @@ static void migrate_hrtimer_list(struct hrtimer_clock_base *old_base,
 		 * is done, which will run all expired timers and re-programm
 		 * the timer device.
 		 */
-		enqueue_hrtimer(timer, new_base, 0);
+		enqueue_hrtimer(timer, new_base);
 
 		/* Clear the migration state bit */
 		timer->state &= ~HRTIMER_STATE_MIGRATE;
@@ -1610,6 +1601,9 @@ void __init hrtimers_init(void)
 	hrtimer_cpu_notify(&hrtimers_nb, (unsigned long)CPU_UP_PREPARE,
 			  (void *)(long)smp_processor_id());
 	register_cpu_notifier(&hrtimers_nb);
+#ifdef CONFIG_HIGH_RES_TIMERS
+	open_softirq(HRTIMER_SOFTIRQ, run_hrtimer_softirq);
+#endif
 }
 
 /**
-- 
cgit v1.2.3


From c70f22d203fc02c805b6ed4a3483b740dc36786b Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Mon, 5 Jan 2009 19:07:50 +0800
Subject: sched: clean up arch_reinit_sched_domains()

- Make arch_reinit_sched_domains() static. It was exported to be used in
  s390, but now rebuild_sched_domains() is used instead.

- Make it return void.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h | 1 -
 kernel/sched.c        | 9 +++------
 2 files changed, 3 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 38a3f4b15394..91207df702e8 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -912,7 +912,6 @@ static inline struct cpumask *sched_domain_span(struct sched_domain *sd)
 
 extern void partition_sched_domains(int ndoms_new, struct cpumask *doms_new,
 				    struct sched_domain_attr *dattr_new);
-extern int arch_reinit_sched_domains(void);
 
 /* Test a flag in parent sched domain */
 static inline int test_sd_parent(struct sched_domain *sd, int flag)
diff --git a/kernel/sched.c b/kernel/sched.c
index 9a8e296959c1..c5019a5dcaa4 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -7987,7 +7987,7 @@ match2:
 }
 
 #if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
-int arch_reinit_sched_domains(void)
+static void arch_reinit_sched_domains(void)
 {
 	get_online_cpus();
 
@@ -7996,13 +7996,10 @@ int arch_reinit_sched_domains(void)
 
 	rebuild_sched_domains();
 	put_online_cpus();
-
-	return 0;
 }
 
 static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt)
 {
-	int ret;
 	unsigned int level = 0;
 
 	if (sscanf(buf, "%u", &level) != 1)
@@ -8023,9 +8020,9 @@ static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt)
 	else
 		sched_mc_power_savings = level;
 
-	ret = arch_reinit_sched_domains();
+	arch_reinit_sched_domains();
 
-	return ret ? ret : count;
+	return count;
 }
 
 #ifdef CONFIG_SCHED_MC
-- 
cgit v1.2.3


From 922ab535bbe73975ce62f71ab9bf8ec9bce71c29 Mon Sep 17 00:00:00 2001
From: Alexey Korolev <akorolev@infradead.org>
Date: Tue, 16 Dec 2008 18:13:58 +0000
Subject: [MTD] LPDDR QINFO records definitions

There are declaraton of structures and macros definitions
necessary for operations with QINFO in this patch.

Signed-off-by: Alexey Korolev <akorolev@infradead.org>
Acked-by: Jared Hulbert <jaredeh@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 include/linux/mtd/qinfo.h | 91 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 91 insertions(+)
 create mode 100644 include/linux/mtd/qinfo.h

(limited to 'include/linux')

diff --git a/include/linux/mtd/qinfo.h b/include/linux/mtd/qinfo.h
new file mode 100644
index 000000000000..7b3d487d8b3f
--- /dev/null
+++ b/include/linux/mtd/qinfo.h
@@ -0,0 +1,91 @@
+#ifndef __LINUX_MTD_QINFO_H
+#define __LINUX_MTD_QINFO_H
+
+#include <linux/mtd/map.h>
+#include <linux/wait.h>
+#include <linux/spinlock.h>
+#include <linux/delay.h>
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/flashchip.h>
+#include <linux/mtd/partitions.h>
+
+/* lpddr_private describes lpddr flash chip in memory map
+ * @ManufactId - Chip Manufacture ID
+ * @DevId - Chip Device ID
+ * @qinfo - pointer to qinfo records describing the chip
+ * @numchips - number of chips including virual RWW partitions
+ * @chipshift - Chip/partiton size 2^chipshift
+ * @chips - per-chip data structure
+ */
+struct lpddr_private {
+	uint16_t ManufactId;
+	uint16_t DevId;
+	struct qinfo_chip *qinfo;
+	int numchips;
+	unsigned long chipshift;
+	struct flchip chips[0];
+};
+
+/* qinfo_query_info structure contains request information for
+ * each qinfo record
+ * @major - major number of qinfo record
+ * @major - minor number of qinfo record
+ * @id_str - descriptive string to access the record
+ * @desc - detailed description for the qinfo record
+ */
+struct qinfo_query_info {
+	uint8_t	major;
+	uint8_t	minor;
+	char *id_str;
+	char *desc;
+};
+
+/*
+ * qinfo_chip structure contains necessary qinfo records data
+ * @DevSizeShift - Device size 2^n bytes
+ * @BufSizeShift - Program buffer size 2^n bytes
+ * @TotalBlocksNum - Total number of blocks
+ * @UniformBlockSizeShift - Uniform block size 2^UniformBlockSizeShift bytes
+ * @HWPartsNum - Number of hardware partitions
+ * @SuspEraseSupp - Suspend erase supported
+ * @SingleWordProgTime - Single word program 2^SingleWordProgTime u-sec
+ * @ProgBufferTime - Program buffer write 2^ProgBufferTime u-sec
+ * @BlockEraseTime - Block erase 2^BlockEraseTime m-sec
+ */
+struct qinfo_chip {
+	/* General device info */
+	uint16_t DevSizeShift;
+	uint16_t BufSizeShift;
+	/* Erase block information */
+	uint16_t TotalBlocksNum;
+	uint16_t UniformBlockSizeShift;
+	/* Partition information */
+	uint16_t HWPartsNum;
+	/* Optional features */
+	uint16_t SuspEraseSupp;
+	/* Operation typical time */
+	uint16_t SingleWordProgTime;
+	uint16_t ProgBufferTime;
+	uint16_t BlockEraseTime;
+};
+
+/* defines for fixup usage */
+#define LPDDR_MFR_ANY		0xffff
+#define LPDDR_ID_ANY		0xffff
+#define NUMONYX_MFGR_ID		0x0089
+#define R18_DEVICE_ID_1G	0x893c
+
+static inline map_word lpddr_build_cmd(u_long cmd, struct map_info *map)
+{
+	map_word val = { {0} };
+	val.x[0] = cmd;
+	return val;
+}
+
+#define CMD(x) lpddr_build_cmd(x, map)
+#define CMDVAL(cmd) cmd.x[0]
+
+struct mtd_info *lpddr_cmdset(struct map_info *);
+
+#endif
+
-- 
cgit v1.2.3


From eb3db27507f74b99241abfa11824d8b6d92b84ef Mon Sep 17 00:00:00 2001
From: Alexey Korolev <akorolev@infradead.org>
Date: Tue, 16 Dec 2008 18:15:33 +0000
Subject: [MTD] LPDDR PFOW definition

LPDDR chips use PFOW window for sending commands, reading status and
capabilites requesting.
This pfow.h - contains definitions for PFOW window fileds, possible commands,
error flags and some common macro function to avoid code duplications.

Signed-off-by: Alexey Korolev <akorolev@infradead.org>
Acked-by: Jared Hulbert <jaredeh@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 include/linux/mtd/pfow.h | 159 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 159 insertions(+)
 create mode 100644 include/linux/mtd/pfow.h

(limited to 'include/linux')

diff --git a/include/linux/mtd/pfow.h b/include/linux/mtd/pfow.h
new file mode 100644
index 000000000000..b730d4f84655
--- /dev/null
+++ b/include/linux/mtd/pfow.h
@@ -0,0 +1,159 @@
+/* Primary function overlay window definitions
+ * and service functions used by LPDDR chips
+ */
+#ifndef __LINUX_MTD_PFOW_H
+#define __LINUX_MTD_PFOW_H
+
+#include <linux/mtd/qinfo.h>
+
+/* PFOW registers addressing */
+/* Address of symbol "P" */
+#define PFOW_QUERY_STRING_P			0x0000
+/* Address of symbol "F" */
+#define PFOW_QUERY_STRING_F			0x0002
+/* Address of symbol "O" */
+#define PFOW_QUERY_STRING_O			0x0004
+/* Address of symbol "W" */
+#define PFOW_QUERY_STRING_W			0x0006
+/* Identification info for LPDDR chip */
+#define PFOW_MANUFACTURER_ID			0x0020
+#define PFOW_DEVICE_ID				0x0022
+/* Address in PFOW where prog buffer can can be found */
+#define PFOW_PROGRAM_BUFFER_OFFSET		0x0040
+/* Size of program buffer in words */
+#define PFOW_PROGRAM_BUFFER_SIZE		0x0042
+/* Address command code register */
+#define PFOW_COMMAND_CODE			0x0080
+/* command data register */
+#define PFOW_COMMAND_DATA			0x0084
+/* command address register lower address bits */
+#define PFOW_COMMAND_ADDRESS_L			0x0088
+/* command address register upper address bits */
+#define PFOW_COMMAND_ADDRESS_H			0x008a
+/* number of bytes to be proggrammed lower address bits */
+#define PFOW_DATA_COUNT_L			0x0090
+/* number of bytes to be proggrammed higher address bits */
+#define PFOW_DATA_COUNT_H			0x0092
+/* command execution register, the only possible value is 0x01 */
+#define PFOW_COMMAND_EXECUTE			0x00c0
+/* 0x01 should be written at this address to clear buffer */
+#define PFOW_CLEAR_PROGRAM_BUFFER		0x00c4
+/* device program/erase suspend register */
+#define PFOW_PROGRAM_ERASE_SUSPEND		0x00c8
+/* device status register */
+#define PFOW_DSR				0x00cc
+
+/* LPDDR memory device command codes */
+/* They are possible values of PFOW command code register */
+#define LPDDR_WORD_PROGRAM		0x0041
+#define LPDDR_BUFF_PROGRAM		0x00E9
+#define LPDDR_BLOCK_ERASE		0x0020
+#define LPDDR_LOCK_BLOCK		0x0061
+#define LPDDR_UNLOCK_BLOCK		0x0062
+#define LPDDR_READ_BLOCK_LOCK_STATUS	0x0065
+#define LPDDR_INFO_QUERY		0x0098
+#define LPDDR_READ_OTP			0x0097
+#define LPDDR_PROG_OTP			0x00C0
+#define LPDDR_RESUME			0x00D0
+
+/* Defines possible value of PFOW command execution register */
+#define LPDDR_START_EXECUTION			0x0001
+
+/* Defines possible value of PFOW program/erase suspend register */
+#define LPDDR_SUSPEND				0x0001
+
+/* Possible values of PFOW device status register */
+/* access R - read; RC read & clearable */
+#define DSR_DPS			(1<<1) /* RC; device protect status
+					* 0 - not protected 1 - locked */
+#define DSR_PSS			(1<<2) /* R; program suspend status;
+					* 0-prog in progress/completed,
+					* 1- prog suspended */
+#define DSR_VPPS		(1<<3) /* RC; 0-Vpp OK, * 1-Vpp low */
+#define DSR_PROGRAM_STATUS	(1<<4) /* RC; 0-successful, 1-error */
+#define DSR_ERASE_STATUS	(1<<5) /* RC; erase or blank check status;
+					* 0-success erase/blank check,
+					* 1 blank check error */
+#define DSR_ESS			(1<<6) /* R; erase suspend status;
+					* 0-erase in progress/complete,
+					* 1 erase suspended */
+#define DSR_READY_STATUS	(1<<7) /* R; Device status
+					* 0-busy,
+					* 1-ready */
+#define DSR_RPS			(0x3<<8) /* RC;  region program status
+					* 00 - Success,
+					* 01-re-program attempt in region with
+					* object mode data,
+					* 10-object mode program w attempt in
+					* region with control mode data
+					* 11-attempt to program invalid half
+					* with 0x41 command */
+#define DSR_AOS			(1<<12) /* RC; 1- AO related failure */
+#define DSR_AVAILABLE		(1<<15) /* R; Device availbility
+					* 1 - Device available
+					* 0 - not available */
+
+/* The superset of all possible error bits in DSR */
+#define DSR_ERR			0x133A
+
+static inline void send_pfow_command(struct map_info *map,
+				unsigned long cmd_code, unsigned long adr,
+				unsigned long len, map_word *datum)
+{
+	int bits_per_chip = map_bankwidth(map) * 8;
+	int chipnum;
+	struct lpddr_private *lpddr = map->fldrv_priv;
+	chipnum = adr >> lpddr->chipshift;
+
+	map_write(map, CMD(cmd_code), map->pfow_base + PFOW_COMMAND_CODE);
+	map_write(map, CMD(adr & ((1<<bits_per_chip) - 1)),
+				map->pfow_base + PFOW_COMMAND_ADDRESS_L);
+	map_write(map, CMD(adr>>bits_per_chip),
+				map->pfow_base + PFOW_COMMAND_ADDRESS_H);
+	if (len) {
+		map_write(map, CMD(len & ((1<<bits_per_chip) - 1)),
+					map->pfow_base + PFOW_DATA_COUNT_L);
+		map_write(map, CMD(len>>bits_per_chip),
+					map->pfow_base + PFOW_DATA_COUNT_H);
+	}
+	if (datum)
+		map_write(map, *datum, map->pfow_base + PFOW_COMMAND_DATA);
+
+	/* Command execution start */
+	map_write(map, CMD(LPDDR_START_EXECUTION),
+			map->pfow_base + PFOW_COMMAND_EXECUTE);
+}
+
+static inline void print_drs_error(unsigned dsr)
+{
+	int prog_status = (dsr & DSR_RPS) >> 8;
+
+	if (!(dsr & DSR_AVAILABLE))
+		printk(KERN_NOTICE"DSR.15: (0) Device not Available\n");
+	if (prog_status & 0x03)
+		printk(KERN_NOTICE"DSR.9,8: (11) Attempt to program invalid "
+						"half with 41h command\n");
+	else if (prog_status & 0x02)
+		printk(KERN_NOTICE"DSR.9,8: (10) Object Mode Program attempt "
+					"in region with Control Mode data\n");
+	else if (prog_status &  0x01)
+		printk(KERN_NOTICE"DSR.9,8: (01) Program attempt in region "
+						"with Object Mode data\n");
+	if (!(dsr & DSR_READY_STATUS))
+		printk(KERN_NOTICE"DSR.7: (0) Device is Busy\n");
+	if (dsr & DSR_ESS)
+		printk(KERN_NOTICE"DSR.6: (1) Erase Suspended\n");
+	if (dsr & DSR_ERASE_STATUS)
+		printk(KERN_NOTICE"DSR.5: (1) Erase/Blank check error\n");
+	if (dsr & DSR_PROGRAM_STATUS)
+		printk(KERN_NOTICE"DSR.4: (1) Program Error\n");
+	if (dsr & DSR_VPPS)
+		printk(KERN_NOTICE"DSR.3: (1) Vpp low detect, operation "
+					"aborted\n");
+	if (dsr & DSR_PSS)
+		printk(KERN_NOTICE"DSR.2: (1) Program suspended\n");
+	if (dsr & DSR_DPS)
+		printk(KERN_NOTICE"DSR.1: (1) Aborted Erase/Program attempt "
+					"on locked block\n");
+}
+#endif /* __LINUX_MTD_PFOW_H */
-- 
cgit v1.2.3


From d13e51e747fee301b404dffcf4a7e1bdc558969b Mon Sep 17 00:00:00 2001
From: Alexey Korolev <akorolev@infradead.org>
Date: Tue, 16 Dec 2008 18:21:10 +0000
Subject: [MTD] LPDDR added new pfow_base parameter

We need to supply additional parameter to mapping driver and tell
LPDDR drivers where PFOW window is in chip mapping.
It leads to necessity of map_info structure extendoing.

Signed-off-by: Alexey Korolev <akorolev@infradead.org>
Acked-by: Jared Hulbert <jaredeh@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 include/linux/mtd/map.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/mtd/map.h b/include/linux/mtd/map.h
index aa30244492c6..b981b8772217 100644
--- a/include/linux/mtd/map.h
+++ b/include/linux/mtd/map.h
@@ -223,6 +223,7 @@ struct map_info {
 	   must leave it enabled. */
 	void (*set_vpp)(struct map_info *, int);
 
+	unsigned long pfow_base;
 	unsigned long map_priv_1;
 	unsigned long map_priv_2;
 	void *fldrv_priv;
-- 
cgit v1.2.3


From d81408304b06a71c28417445202af9cd6673168d Mon Sep 17 00:00:00 2001
From: Alexey Korolev <akorolev@infradead.org>
Date: Tue, 16 Dec 2008 18:22:39 +0000
Subject: [MTD] LPDDR extended physmap driver to support LPDDR flash

Physmap is a generic map driver for different platforms and flash types.
We added support of LPDDR to physmap.
All changes here are related to introduction of new pfow_base parameter.
This parameter is valid in case of LPDDR chips only.

Signed-off-by: Alexey Korolev <akorolev@infradead.org>
Acked-by: Jared Hulbert <jaredeh@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/maps/Kconfig    | 4 ++--
 drivers/mtd/maps/physmap.c  | 8 +++++++-
 include/linux/mtd/physmap.h | 1 +
 3 files changed, 10 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mtd/maps/Kconfig b/drivers/mtd/maps/Kconfig
index 3788a548336c..0225cbbf22de 100644
--- a/drivers/mtd/maps/Kconfig
+++ b/drivers/mtd/maps/Kconfig
@@ -10,8 +10,8 @@ config MTD_COMPLEX_MAPPINGS
 	  paged mappings of flash chips.
 
 config MTD_PHYSMAP
-	tristate "CFI Flash device in physical memory map"
-	depends on MTD_CFI || MTD_JEDECPROBE || MTD_ROM
+	tristate "Flash device in physical memory map"
+	depends on MTD_CFI || MTD_JEDECPROBE || MTD_ROM || MTD_LPDDR
 	help
 	  This provides a 'mapping' driver which allows the NOR Flash and
 	  ROM driver code to communicate with chips which are mapped
diff --git a/drivers/mtd/maps/physmap.c b/drivers/mtd/maps/physmap.c
index d3a2acc7e9be..87743661d48e 100644
--- a/drivers/mtd/maps/physmap.c
+++ b/drivers/mtd/maps/physmap.c
@@ -68,7 +68,12 @@ static int physmap_flash_remove(struct platform_device *dev)
 	return 0;
 }
 
-static const char *rom_probe_types[] = { "cfi_probe", "jedec_probe", "map_rom", NULL };
+static const char *rom_probe_types[] = {
+					"cfi_probe",
+					"jedec_probe",
+					"qinfo_probe",
+					"map_rom",
+					NULL };
 #ifdef CONFIG_MTD_PARTITIONS
 static const char *part_probe_types[] = { "cmdlinepart", "RedBoot", NULL };
 #endif
@@ -117,6 +122,7 @@ static int physmap_flash_probe(struct platform_device *dev)
 		info->map[i].size = dev->resource[i].end - dev->resource[i].start + 1;
 		info->map[i].bankwidth = physmap_data->width;
 		info->map[i].set_vpp = physmap_data->set_vpp;
+		info->map[i].pfow_base = physmap_data->pfow_base;
 
 		info->map[i].virt = devm_ioremap(&dev->dev, info->map[i].phys,
 						 info->map[i].size);
diff --git a/include/linux/mtd/physmap.h b/include/linux/mtd/physmap.h
index c8e63a5ee72e..76f7cabf07d3 100644
--- a/include/linux/mtd/physmap.h
+++ b/include/linux/mtd/physmap.h
@@ -24,6 +24,7 @@ struct physmap_flash_data {
 	unsigned int		width;
 	void			(*set_vpp)(struct map_info *, int);
 	unsigned int		nr_parts;
+	unsigned int		pfow_base;
 	struct mtd_partition	*parts;
 };
 
-- 
cgit v1.2.3


From be92d7af38fb8a91f8575ab2272e00f2e51667ff Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 5 Jan 2009 14:34:42 +0100
Subject: genirq: provide irq_to_desc() to non-genirq architectures too

Impact: build fix on non-genirq architectures

Sam Ravnborg reported this build failure on sparc32 allmodconfig,
the GPIO drivers assume the presence of irq_to_desc():

 drivers/gpio/gpiolib.c: In function `gpiolib_dbg_show':
 drivers/gpio/gpiolib.c:1146: error: implicit declaration of function 'irq_to_desc'

Add it in the !genirq case too.

Reported-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Tested-by: Sam Ravnborg <sam@ravnborg.org>
---
 include/linux/irqnr.h | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/irqnr.h b/include/linux/irqnr.h
index 5504a5c97836..86af92e9e84c 100644
--- a/include/linux/irqnr.h
+++ b/include/linux/irqnr.h
@@ -8,7 +8,12 @@
 
 #ifndef CONFIG_GENERIC_HARDIRQS
 #include <asm/irq.h>
-# define nr_irqs		NR_IRQS
+
+/*
+ * Wrappers for non-genirq architectures:
+ */
+#define nr_irqs			NR_IRQS
+#define irq_to_desc(irq)	(&irq_desc[irq])
 
 # define for_each_irq_desc(irq, desc)		\
 	for (irq = 0; irq < nr_irqs; irq++)
-- 
cgit v1.2.3


From c42aa775cc8a8ca558db0cc75979fb8e16667447 Mon Sep 17 00:00:00 2001
From: Nicolas Ferre <nicolas.ferre@atmel.com>
Date: Thu, 20 Nov 2008 15:59:12 +0100
Subject: atmel-mci: move atmel-mci.h file to include/linux

Needed to use the atmel-mci driver in an architecture
independant maner.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
---
 arch/avr32/boards/atngw100/setup.c      |  2 +-
 arch/avr32/boards/atstk1000/atstk1002.c |  2 +-
 arch/avr32/boards/atstk1000/atstk1003.c |  2 +-
 arch/avr32/boards/atstk1000/atstk1004.c |  2 +-
 arch/avr32/boards/mimc200/setup.c       |  2 +-
 arch/avr32/include/asm/atmel-mci.h      | 39 ---------------------------------
 arch/avr32/mach-at32ap/at32ap700x.c     |  2 +-
 drivers/mmc/host/atmel-mci.c            |  2 +-
 include/linux/atmel-mci.h               | 39 +++++++++++++++++++++++++++++++++
 9 files changed, 46 insertions(+), 46 deletions(-)
 delete mode 100644 arch/avr32/include/asm/atmel-mci.h
 create mode 100644 include/linux/atmel-mci.h

(limited to 'include/linux')

diff --git a/arch/avr32/boards/atngw100/setup.c b/arch/avr32/boards/atngw100/setup.c
index 32fb9ba0fbdf..05d3722fff18 100644
--- a/arch/avr32/boards/atngw100/setup.c
+++ b/arch/avr32/boards/atngw100/setup.c
@@ -19,8 +19,8 @@
 #include <linux/types.h>
 #include <linux/leds.h>
 #include <linux/spi/spi.h>
+#include <linux/atmel-mci.h>
 
-#include <asm/atmel-mci.h>
 #include <asm/io.h>
 #include <asm/setup.h>
 
diff --git a/arch/avr32/boards/atstk1000/atstk1002.c b/arch/avr32/boards/atstk1000/atstk1002.c
index 5c5cdf3b464f..6d94f74bc5c7 100644
--- a/arch/avr32/boards/atstk1000/atstk1002.c
+++ b/arch/avr32/boards/atstk1000/atstk1002.c
@@ -16,12 +16,12 @@
 #include <linux/types.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/at73c213.h>
+#include <linux/atmel-mci.h>
 
 #include <video/atmel_lcdc.h>
 
 #include <asm/io.h>
 #include <asm/setup.h>
-#include <asm/atmel-mci.h>
 
 #include <mach/at32ap700x.h>
 #include <mach/board.h>
diff --git a/arch/avr32/boards/atstk1000/atstk1003.c b/arch/avr32/boards/atstk1000/atstk1003.c
index 134b566630b0..db2038cd2e5b 100644
--- a/arch/avr32/boards/atstk1000/atstk1003.c
+++ b/arch/avr32/boards/atstk1000/atstk1003.c
@@ -17,9 +17,9 @@
 
 #include <linux/spi/at73c213.h>
 #include <linux/spi/spi.h>
+#include <linux/atmel-mci.h>
 
 #include <asm/setup.h>
-#include <asm/atmel-mci.h>
 
 #include <mach/at32ap700x.h>
 #include <mach/board.h>
diff --git a/arch/avr32/boards/atstk1000/atstk1004.c b/arch/avr32/boards/atstk1000/atstk1004.c
index cb32eb844aa7..29b35aca96cd 100644
--- a/arch/avr32/boards/atstk1000/atstk1004.c
+++ b/arch/avr32/boards/atstk1000/atstk1004.c
@@ -17,11 +17,11 @@
 
 #include <linux/spi/at73c213.h>
 #include <linux/spi/spi.h>
+#include <linux/atmel-mci.h>
 
 #include <video/atmel_lcdc.h>
 
 #include <asm/setup.h>
-#include <asm/atmel-mci.h>
 
 #include <mach/at32ap700x.h>
 #include <mach/board.h>
diff --git a/arch/avr32/boards/mimc200/setup.c b/arch/avr32/boards/mimc200/setup.c
index 397cbb8f44c8..811d73027f77 100644
--- a/arch/avr32/boards/mimc200/setup.c
+++ b/arch/avr32/boards/mimc200/setup.c
@@ -24,7 +24,7 @@ extern struct atmel_lcdfb_info mimc200_lcdc_data;
 #include <video/atmel_lcdc.h>
 #include <linux/fb.h>
 
-#include <asm/atmel-mci.h>
+#include <linux/atmel-mci.h>
 #include <linux/io.h>
 #include <asm/setup.h>
 
diff --git a/arch/avr32/include/asm/atmel-mci.h b/arch/avr32/include/asm/atmel-mci.h
deleted file mode 100644
index 59f3fadd0b68..000000000000
--- a/arch/avr32/include/asm/atmel-mci.h
+++ /dev/null
@@ -1,39 +0,0 @@
-#ifndef __ASM_AVR32_ATMEL_MCI_H
-#define __ASM_AVR32_ATMEL_MCI_H
-
-#define ATMEL_MCI_MAX_NR_SLOTS	2
-
-struct dma_slave;
-
-/**
- * struct mci_slot_pdata - board-specific per-slot configuration
- * @bus_width: Number of data lines wired up the slot
- * @detect_pin: GPIO pin wired to the card detect switch
- * @wp_pin: GPIO pin wired to the write protect sensor
- *
- * If a given slot is not present on the board, @bus_width should be
- * set to 0. The other fields are ignored in this case.
- *
- * Any pins that aren't available should be set to a negative value.
- *
- * Note that support for multiple slots is experimental -- some cards
- * might get upset if we don't get the clock management exactly right.
- * But in most cases, it should work just fine.
- */
-struct mci_slot_pdata {
-	unsigned int		bus_width;
-	int			detect_pin;
-	int			wp_pin;
-};
-
-/**
- * struct mci_platform_data - board-specific MMC/SDcard configuration
- * @dma_slave: DMA slave interface to use in data transfers, or NULL.
- * @slot: Per-slot configuration data.
- */
-struct mci_platform_data {
-	struct dma_slave	*dma_slave;
-	struct mci_slot_pdata	slot[ATMEL_MCI_MAX_NR_SLOTS];
-};
-
-#endif /* __ASM_AVR32_ATMEL_MCI_H */
diff --git a/arch/avr32/mach-at32ap/at32ap700x.c b/arch/avr32/mach-at32ap/at32ap700x.c
index 066252eebf61..3cf49553e4f9 100644
--- a/arch/avr32/mach-at32ap/at32ap700x.c
+++ b/arch/avr32/mach-at32ap/at32ap700x.c
@@ -15,8 +15,8 @@
 #include <linux/gpio.h>
 #include <linux/spi/spi.h>
 #include <linux/usb/atmel_usba_udc.h>
+#include <linux/atmel-mci.h>
 
-#include <asm/atmel-mci.h>
 #include <asm/io.h>
 #include <asm/irq.h>
 
diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
index 7a3f2436b011..1e97916914ad 100644
--- a/drivers/mmc/host/atmel-mci.c
+++ b/drivers/mmc/host/atmel-mci.c
@@ -25,8 +25,8 @@
 #include <linux/stat.h>
 
 #include <linux/mmc/host.h>
+#include <linux/atmel-mci.h>
 
-#include <asm/atmel-mci.h>
 #include <asm/io.h>
 #include <asm/unaligned.h>
 
diff --git a/include/linux/atmel-mci.h b/include/linux/atmel-mci.h
new file mode 100644
index 000000000000..2a2213eefd85
--- /dev/null
+++ b/include/linux/atmel-mci.h
@@ -0,0 +1,39 @@
+#ifndef __LINUX_ATMEL_MCI_H
+#define __LINUX_ATMEL_MCI_H
+
+#define ATMEL_MCI_MAX_NR_SLOTS	2
+
+struct dma_slave;
+
+/**
+ * struct mci_slot_pdata - board-specific per-slot configuration
+ * @bus_width: Number of data lines wired up the slot
+ * @detect_pin: GPIO pin wired to the card detect switch
+ * @wp_pin: GPIO pin wired to the write protect sensor
+ *
+ * If a given slot is not present on the board, @bus_width should be
+ * set to 0. The other fields are ignored in this case.
+ *
+ * Any pins that aren't available should be set to a negative value.
+ *
+ * Note that support for multiple slots is experimental -- some cards
+ * might get upset if we don't get the clock management exactly right.
+ * But in most cases, it should work just fine.
+ */
+struct mci_slot_pdata {
+	unsigned int		bus_width;
+	int			detect_pin;
+	int			wp_pin;
+};
+
+/**
+ * struct mci_platform_data - board-specific MMC/SDcard configuration
+ * @dma_slave: DMA slave interface to use in data transfers, or NULL.
+ * @slot: Per-slot configuration data.
+ */
+struct mci_platform_data {
+	struct dma_slave	*dma_slave;
+	struct mci_slot_pdata	slot[ATMEL_MCI_MAX_NR_SLOTS];
+};
+
+#endif /* __LINUX_ATMEL_MCI_H */
-- 
cgit v1.2.3


From 74f783af95c982aef6d3a1415275650dcf511666 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Tue, 19 Aug 2008 14:51:22 +0200
Subject: quota: Add callbacks for allocating and destroying dquot structures

Some filesystems would like to keep private information together with each
dquot. Add callbacks alloc_dquot and destroy_dquot allowing filesystem to
allocate larger dquots from their private slab in a similar fashion we
currently allocate inodes.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/dquot.c            | 27 ++++++++++++++++++++++-----
 include/linux/quota.h |  2 ++
 2 files changed, 24 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/fs/dquot.c b/fs/dquot.c
index c237ccc8581c..1b5fc4b7fbeb 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -415,6 +415,16 @@ out_dqlock:
 	return ret;
 }
 
+static void dquot_destroy(struct dquot *dquot)
+{
+	kmem_cache_free(dquot_cachep, dquot);
+}
+
+static inline void do_destroy_dquot(struct dquot *dquot)
+{
+	dquot->dq_sb->dq_op->destroy_dquot(dquot);
+}
+
 /* Invalidate all dquots on the list. Note that this function is called after
  * quota is disabled and pointers from inodes removed so there cannot be new
  * quota users. There can still be some users of quotas due to inodes being
@@ -463,7 +473,7 @@ restart:
 		remove_dquot_hash(dquot);
 		remove_free_dquot(dquot);
 		remove_inuse(dquot);
-		kmem_cache_free(dquot_cachep, dquot);
+		do_destroy_dquot(dquot);
 	}
 	spin_unlock(&dq_list_lock);
 }
@@ -527,7 +537,7 @@ static void prune_dqcache(int count)
 		remove_dquot_hash(dquot);
 		remove_free_dquot(dquot);
 		remove_inuse(dquot);
-		kmem_cache_free(dquot_cachep, dquot);
+		do_destroy_dquot(dquot);
 		count--;
 		head = free_dquots.prev;
 	}
@@ -625,11 +635,16 @@ we_slept:
 	spin_unlock(&dq_list_lock);
 }
 
+static struct dquot *dquot_alloc(struct super_block *sb, int type)
+{
+	return kmem_cache_zalloc(dquot_cachep, GFP_NOFS);
+}
+
 static struct dquot *get_empty_dquot(struct super_block *sb, int type)
 {
 	struct dquot *dquot;
 
-	dquot = kmem_cache_zalloc(dquot_cachep, GFP_NOFS);
+	dquot = sb->dq_op->alloc_dquot(sb, type);
 	if(!dquot)
 		return NODQUOT;
 
@@ -682,7 +697,7 @@ we_slept:
 		dqstats.lookups++;
 		spin_unlock(&dq_list_lock);
 		if (empty)
-			kmem_cache_free(dquot_cachep, empty);
+			do_destroy_dquot(empty);
 	}
 	/* Wait for dq_lock - after this we know that either dquot_release() is already
 	 * finished or it will be canceled due to dq_count > 1 test */
@@ -1533,7 +1548,9 @@ struct dquot_operations dquot_operations = {
 	.acquire_dquot	= dquot_acquire,
 	.release_dquot	= dquot_release,
 	.mark_dirty	= dquot_mark_dquot_dirty,
-	.write_info	= dquot_commit_info
+	.write_info	= dquot_commit_info,
+	.alloc_dquot	= dquot_alloc,
+	.destroy_dquot	= dquot_destroy,
 };
 
 static inline void set_enable_flags(struct quota_info *dqopt, int type)
diff --git a/include/linux/quota.h b/include/linux/quota.h
index 40401b554484..3ce708c2cb3c 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -292,6 +292,8 @@ struct dquot_operations {
 	int (*free_inode) (const struct inode *, unsigned long);
 	int (*transfer) (struct inode *, struct iattr *);
 	int (*write_dquot) (struct dquot *);		/* Ordinary dquot write */
+	struct dquot *(*alloc_dquot)(struct super_block *, int);	/* Allocate memory for new dquot */
+	void (*destroy_dquot)(struct dquot *);		/* Free memory for dquot */
 	int (*acquire_dquot) (struct dquot *);		/* Quota is going to be created on disk */
 	int (*release_dquot) (struct dquot *);		/* Quota is going to be deleted from disk */
 	int (*mark_dirty) (struct dquot *);		/* Dquot is marked dirty */
-- 
cgit v1.2.3


From 12095460f7f315f8ef67a55b2194195d325d48d7 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Wed, 20 Aug 2008 14:45:12 +0200
Subject: quota: Increase size of variables for limits and inode usage

So far quota was fine with quota block limits and inode limits/numbers in
a 32-bit type. Now with rapid increase in storage sizes there are coming
requests to be able to handle quota limits above 4TB / more that 2^32 inodes.
So bump up sizes of types in mem_dqblk structure to 64-bits to be able to
handle this. Also update inode allocation / checking functions to use qsize_t
and make global structure keep quota limits in bytes so that things are
consistent.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/dquot.c               | 50 ++++++++++++++++++++++++++++--------------------
 fs/quota_v1.c            | 25 ++++++++++++++++++------
 fs/quota_v2.c            | 21 ++++++++++++++++----
 include/linux/quota.h    | 28 ++++++++++++---------------
 include/linux/quotaops.h |  4 ++--
 5 files changed, 79 insertions(+), 49 deletions(-)

(limited to 'include/linux')

diff --git a/fs/dquot.c b/fs/dquot.c
index 1b5fc4b7fbeb..c02223b6aeb2 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -835,7 +835,7 @@ static void drop_dquot_ref(struct super_block *sb, int type)
 	}
 }
 
-static inline void dquot_incr_inodes(struct dquot *dquot, unsigned long number)
+static inline void dquot_incr_inodes(struct dquot *dquot, qsize_t number)
 {
 	dquot->dq_dqb.dqb_curinodes += number;
 }
@@ -845,7 +845,7 @@ static inline void dquot_incr_space(struct dquot *dquot, qsize_t number)
 	dquot->dq_dqb.dqb_curspace += number;
 }
 
-static inline void dquot_decr_inodes(struct dquot *dquot, unsigned long number)
+static inline void dquot_decr_inodes(struct dquot *dquot, qsize_t number)
 {
 	if (dquot->dq_dqb.dqb_curinodes > number)
 		dquot->dq_dqb.dqb_curinodes -= number;
@@ -862,7 +862,7 @@ static inline void dquot_decr_space(struct dquot *dquot, qsize_t number)
 		dquot->dq_dqb.dqb_curspace -= number;
 	else
 		dquot->dq_dqb.dqb_curspace = 0;
-	if (toqb(dquot->dq_dqb.dqb_curspace) <= dquot->dq_dqb.dqb_bsoftlimit)
+	if (dquot->dq_dqb.dqb_curspace <= dquot->dq_dqb.dqb_bsoftlimit)
 		dquot->dq_dqb.dqb_btime = (time_t) 0;
 	clear_bit(DQ_BLKS_B, &dquot->dq_flags);
 }
@@ -1038,7 +1038,7 @@ static inline char ignore_hardlimit(struct dquot *dquot)
 }
 
 /* needs dq_data_lock */
-static int check_idq(struct dquot *dquot, ulong inodes, char *warntype)
+static int check_idq(struct dquot *dquot, qsize_t inodes, char *warntype)
 {
 	*warntype = QUOTA_NL_NOWARN;
 	if (inodes <= 0 || test_bit(DQ_FAKE_B, &dquot->dq_flags))
@@ -1077,7 +1077,7 @@ static int check_bdq(struct dquot *dquot, qsize_t space, int prealloc, char *war
 		return QUOTA_OK;
 
 	if (dquot->dq_dqb.dqb_bhardlimit &&
-	   toqb(dquot->dq_dqb.dqb_curspace + space) > dquot->dq_dqb.dqb_bhardlimit &&
+	    dquot->dq_dqb.dqb_curspace + space > dquot->dq_dqb.dqb_bhardlimit &&
             !ignore_hardlimit(dquot)) {
 		if (!prealloc)
 			*warntype = QUOTA_NL_BHARDWARN;
@@ -1085,7 +1085,7 @@ static int check_bdq(struct dquot *dquot, qsize_t space, int prealloc, char *war
 	}
 
 	if (dquot->dq_dqb.dqb_bsoftlimit &&
-	   toqb(dquot->dq_dqb.dqb_curspace + space) > dquot->dq_dqb.dqb_bsoftlimit &&
+	    dquot->dq_dqb.dqb_curspace + space > dquot->dq_dqb.dqb_bsoftlimit &&
 	    dquot->dq_dqb.dqb_btime && get_seconds() >= dquot->dq_dqb.dqb_btime &&
             !ignore_hardlimit(dquot)) {
 		if (!prealloc)
@@ -1094,7 +1094,7 @@ static int check_bdq(struct dquot *dquot, qsize_t space, int prealloc, char *war
 	}
 
 	if (dquot->dq_dqb.dqb_bsoftlimit &&
-	   toqb(dquot->dq_dqb.dqb_curspace + space) > dquot->dq_dqb.dqb_bsoftlimit &&
+	    dquot->dq_dqb.dqb_curspace + space > dquot->dq_dqb.dqb_bsoftlimit &&
 	    dquot->dq_dqb.dqb_btime == 0) {
 		if (!prealloc) {
 			*warntype = QUOTA_NL_BSOFTWARN;
@@ -1111,7 +1111,7 @@ static int check_bdq(struct dquot *dquot, qsize_t space, int prealloc, char *war
 	return QUOTA_OK;
 }
 
-static int info_idq_free(struct dquot *dquot, ulong inodes)
+static int info_idq_free(struct dquot *dquot, qsize_t inodes)
 {
 	if (test_bit(DQ_FAKE_B, &dquot->dq_flags) ||
 	    dquot->dq_dqb.dqb_curinodes <= dquot->dq_dqb.dqb_isoftlimit)
@@ -1128,15 +1128,13 @@ static int info_idq_free(struct dquot *dquot, ulong inodes)
 static int info_bdq_free(struct dquot *dquot, qsize_t space)
 {
 	if (test_bit(DQ_FAKE_B, &dquot->dq_flags) ||
-	    toqb(dquot->dq_dqb.dqb_curspace) <= dquot->dq_dqb.dqb_bsoftlimit)
+	    dquot->dq_dqb.dqb_curspace <= dquot->dq_dqb.dqb_bsoftlimit)
 		return QUOTA_NL_NOWARN;
 
-	if (toqb(dquot->dq_dqb.dqb_curspace - space) <=
-	    dquot->dq_dqb.dqb_bsoftlimit)
+	if (dquot->dq_dqb.dqb_curspace - space <= dquot->dq_dqb.dqb_bsoftlimit)
 		return QUOTA_NL_BSOFTBELOW;
-	if (toqb(dquot->dq_dqb.dqb_curspace) >= dquot->dq_dqb.dqb_bhardlimit &&
-	    toqb(dquot->dq_dqb.dqb_curspace - space) <
-						dquot->dq_dqb.dqb_bhardlimit)
+	if (dquot->dq_dqb.dqb_curspace >= dquot->dq_dqb.dqb_bhardlimit &&
+	    dquot->dq_dqb.dqb_curspace - space < dquot->dq_dqb.dqb_bhardlimit)
 		return QUOTA_NL_BHARDBELOW;
 	return QUOTA_NL_NOWARN;
 }
@@ -1279,7 +1277,7 @@ warn_put_all:
 /*
  * This operation can block, but only after everything is updated
  */
-int dquot_alloc_inode(const struct inode *inode, unsigned long number)
+int dquot_alloc_inode(const struct inode *inode, qsize_t number)
 {
 	int cnt, ret = NO_QUOTA;
 	char warntype[MAXQUOTAS];
@@ -1364,7 +1362,7 @@ out_sub:
 /*
  * This operation can block, but only after everything is updated
  */
-int dquot_free_inode(const struct inode *inode, unsigned long number)
+int dquot_free_inode(const struct inode *inode, qsize_t number)
 {
 	unsigned int cnt;
 	char warntype[MAXQUOTAS];
@@ -1883,14 +1881,24 @@ int vfs_dq_quota_on_remount(struct super_block *sb)
 	return ret;
 }
 
+static inline qsize_t qbtos(qsize_t blocks)
+{
+	return blocks << QIF_DQBLKSIZE_BITS;
+}
+
+static inline qsize_t stoqb(qsize_t space)
+{
+	return (space + QIF_DQBLKSIZE - 1) >> QIF_DQBLKSIZE_BITS;
+}
+
 /* Generic routine for getting common part of quota structure */
 static void do_get_dqblk(struct dquot *dquot, struct if_dqblk *di)
 {
 	struct mem_dqblk *dm = &dquot->dq_dqb;
 
 	spin_lock(&dq_data_lock);
-	di->dqb_bhardlimit = dm->dqb_bhardlimit;
-	di->dqb_bsoftlimit = dm->dqb_bsoftlimit;
+	di->dqb_bhardlimit = stoqb(dm->dqb_bhardlimit);
+	di->dqb_bsoftlimit = stoqb(dm->dqb_bsoftlimit);
 	di->dqb_curspace = dm->dqb_curspace;
 	di->dqb_ihardlimit = dm->dqb_ihardlimit;
 	di->dqb_isoftlimit = dm->dqb_isoftlimit;
@@ -1937,8 +1945,8 @@ static int do_set_dqblk(struct dquot *dquot, struct if_dqblk *di)
 		check_blim = 1;
 	}
 	if (di->dqb_valid & QIF_BLIMITS) {
-		dm->dqb_bsoftlimit = di->dqb_bsoftlimit;
-		dm->dqb_bhardlimit = di->dqb_bhardlimit;
+		dm->dqb_bsoftlimit = qbtos(di->dqb_bsoftlimit);
+		dm->dqb_bhardlimit = qbtos(di->dqb_bhardlimit);
 		check_blim = 1;
 	}
 	if (di->dqb_valid & QIF_INODES) {
@@ -1956,7 +1964,7 @@ static int do_set_dqblk(struct dquot *dquot, struct if_dqblk *di)
 		dm->dqb_itime = di->dqb_itime;
 
 	if (check_blim) {
-		if (!dm->dqb_bsoftlimit || toqb(dm->dqb_curspace) < dm->dqb_bsoftlimit) {
+		if (!dm->dqb_bsoftlimit || dm->dqb_curspace < dm->dqb_bsoftlimit) {
 			dm->dqb_btime = 0;
 			clear_bit(DQ_BLKS_B, &dquot->dq_flags);
 		}
diff --git a/fs/quota_v1.c b/fs/quota_v1.c
index 5ae15b13eeb0..3e078eee5644 100644
--- a/fs/quota_v1.c
+++ b/fs/quota_v1.c
@@ -14,14 +14,27 @@ MODULE_AUTHOR("Jan Kara");
 MODULE_DESCRIPTION("Old quota format support");
 MODULE_LICENSE("GPL");
 
+#define QUOTABLOCK_BITS 10
+#define QUOTABLOCK_SIZE (1 << QUOTABLOCK_BITS)
+
+static inline qsize_t v1_stoqb(qsize_t space)
+{
+	return (space + QUOTABLOCK_SIZE - 1) >> QUOTABLOCK_BITS;
+}
+
+static inline qsize_t v1_qbtos(qsize_t blocks)
+{
+	return blocks << QUOTABLOCK_BITS;
+}
+
 static void v1_disk2mem_dqblk(struct mem_dqblk *m, struct v1_disk_dqblk *d)
 {
 	m->dqb_ihardlimit = d->dqb_ihardlimit;
 	m->dqb_isoftlimit = d->dqb_isoftlimit;
 	m->dqb_curinodes = d->dqb_curinodes;
-	m->dqb_bhardlimit = d->dqb_bhardlimit;
-	m->dqb_bsoftlimit = d->dqb_bsoftlimit;
-	m->dqb_curspace = ((qsize_t)d->dqb_curblocks) << QUOTABLOCK_BITS;
+	m->dqb_bhardlimit = v1_qbtos(d->dqb_bhardlimit);
+	m->dqb_bsoftlimit = v1_qbtos(d->dqb_bsoftlimit);
+	m->dqb_curspace = v1_qbtos(d->dqb_curblocks);
 	m->dqb_itime = d->dqb_itime;
 	m->dqb_btime = d->dqb_btime;
 }
@@ -31,9 +44,9 @@ static void v1_mem2disk_dqblk(struct v1_disk_dqblk *d, struct mem_dqblk *m)
 	d->dqb_ihardlimit = m->dqb_ihardlimit;
 	d->dqb_isoftlimit = m->dqb_isoftlimit;
 	d->dqb_curinodes = m->dqb_curinodes;
-	d->dqb_bhardlimit = m->dqb_bhardlimit;
-	d->dqb_bsoftlimit = m->dqb_bsoftlimit;
-	d->dqb_curblocks = toqb(m->dqb_curspace);
+	d->dqb_bhardlimit = v1_stoqb(m->dqb_bhardlimit);
+	d->dqb_bsoftlimit = v1_stoqb(m->dqb_bsoftlimit);
+	d->dqb_curblocks = v1_stoqb(m->dqb_curspace);
 	d->dqb_itime = m->dqb_itime;
 	d->dqb_btime = m->dqb_btime;
 }
diff --git a/fs/quota_v2.c b/fs/quota_v2.c
index b53827dc02d9..51c4717f7c6a 100644
--- a/fs/quota_v2.c
+++ b/fs/quota_v2.c
@@ -26,6 +26,19 @@ typedef char *dqbuf_t;
 #define GETIDINDEX(id, depth) (((id) >> ((V2_DQTREEDEPTH-(depth)-1)*8)) & 0xff)
 #define GETENTRIES(buf) ((struct v2_disk_dqblk *)(((char *)buf)+sizeof(struct v2_disk_dqdbheader)))
 
+#define QUOTABLOCK_BITS 10
+#define QUOTABLOCK_SIZE (1 << QUOTABLOCK_BITS)
+
+static inline qsize_t v2_stoqb(qsize_t space)
+{
+	return (space + QUOTABLOCK_SIZE - 1) >> QUOTABLOCK_BITS;
+}
+
+static inline qsize_t v2_qbtos(qsize_t blocks)
+{
+	return blocks << QUOTABLOCK_BITS;
+}
+
 /* Check whether given file is really vfsv0 quotafile */
 static int v2_check_quota_file(struct super_block *sb, int type)
 {
@@ -104,8 +117,8 @@ static void disk2memdqb(struct mem_dqblk *m, struct v2_disk_dqblk *d)
 	m->dqb_isoftlimit = le32_to_cpu(d->dqb_isoftlimit);
 	m->dqb_curinodes = le32_to_cpu(d->dqb_curinodes);
 	m->dqb_itime = le64_to_cpu(d->dqb_itime);
-	m->dqb_bhardlimit = le32_to_cpu(d->dqb_bhardlimit);
-	m->dqb_bsoftlimit = le32_to_cpu(d->dqb_bsoftlimit);
+	m->dqb_bhardlimit = v2_qbtos(le32_to_cpu(d->dqb_bhardlimit));
+	m->dqb_bsoftlimit = v2_qbtos(le32_to_cpu(d->dqb_bsoftlimit));
 	m->dqb_curspace = le64_to_cpu(d->dqb_curspace);
 	m->dqb_btime = le64_to_cpu(d->dqb_btime);
 }
@@ -116,8 +129,8 @@ static void mem2diskdqb(struct v2_disk_dqblk *d, struct mem_dqblk *m, qid_t id)
 	d->dqb_isoftlimit = cpu_to_le32(m->dqb_isoftlimit);
 	d->dqb_curinodes = cpu_to_le32(m->dqb_curinodes);
 	d->dqb_itime = cpu_to_le64(m->dqb_itime);
-	d->dqb_bhardlimit = cpu_to_le32(m->dqb_bhardlimit);
-	d->dqb_bsoftlimit = cpu_to_le32(m->dqb_bsoftlimit);
+	d->dqb_bhardlimit = cpu_to_le32(v2_qbtos(m->dqb_bhardlimit));
+	d->dqb_bsoftlimit = cpu_to_le32(v2_qbtos(m->dqb_bsoftlimit));
 	d->dqb_curspace = cpu_to_le64(m->dqb_curspace);
 	d->dqb_btime = cpu_to_le64(m->dqb_btime);
 	d->dqb_id = cpu_to_le32(id);
diff --git a/include/linux/quota.h b/include/linux/quota.h
index 3ce708c2cb3c..9ea468363f9f 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -39,15 +39,6 @@
 #define __DQUOT_VERSION__	"dquot_6.5.1"
 #define __DQUOT_NUM_VERSION__	6*10000+5*100+1
 
-/* Size of blocks in which are counted size limits */
-#define QUOTABLOCK_BITS 10
-#define QUOTABLOCK_SIZE (1 << QUOTABLOCK_BITS)
-
-/* Conversion routines from and to quota blocks */
-#define qb2kb(x) ((x) << (QUOTABLOCK_BITS-10))
-#define kb2qb(x) ((x) >> (QUOTABLOCK_BITS-10))
-#define toqb(x) (((x) + QUOTABLOCK_SIZE - 1) >> QUOTABLOCK_BITS)
-
 #define MAXQUOTAS 2
 #define USRQUOTA  0		/* element used for user quotas */
 #define GRPQUOTA  1		/* element used for group quotas */
@@ -80,6 +71,11 @@
 #define Q_GETQUOTA 0x800007	/* get user quota structure */
 #define Q_SETQUOTA 0x800008	/* set user quota structure */
 
+/* Size of block in which space limits are passed through the quota
+ * interface */
+#define QIF_DQBLKSIZE_BITS 10
+#define QIF_DQBLKSIZE (1 << QIF_DQBLKSIZE_BITS)
+
 /*
  * Quota structure used for communication with userspace via quotactl
  * Following flags are used to specify which fields are valid
@@ -187,12 +183,12 @@ extern spinlock_t dq_data_lock;
  * Data for one user/group kept in memory
  */
 struct mem_dqblk {
-	__u32 dqb_bhardlimit;	/* absolute limit on disk blks alloc */
-	__u32 dqb_bsoftlimit;	/* preferred limit on disk blks */
+	qsize_t dqb_bhardlimit;	/* absolute limit on disk blks alloc */
+	qsize_t dqb_bsoftlimit;	/* preferred limit on disk blks */
 	qsize_t dqb_curspace;	/* current used space */
-	__u32 dqb_ihardlimit;	/* absolute limit on allocated inodes */
-	__u32 dqb_isoftlimit;	/* preferred inode limit */
-	__u32 dqb_curinodes;	/* current # allocated inodes */
+	qsize_t dqb_ihardlimit;	/* absolute limit on allocated inodes */
+	qsize_t dqb_isoftlimit;	/* preferred inode limit */
+	qsize_t dqb_curinodes;	/* current # allocated inodes */
 	time_t dqb_btime;	/* time limit for excessive disk use */
 	time_t dqb_itime;	/* time limit for excessive inode use */
 };
@@ -287,9 +283,9 @@ struct dquot_operations {
 	int (*initialize) (struct inode *, int);
 	int (*drop) (struct inode *);
 	int (*alloc_space) (struct inode *, qsize_t, int);
-	int (*alloc_inode) (const struct inode *, unsigned long);
+	int (*alloc_inode) (const struct inode *, qsize_t);
 	int (*free_space) (struct inode *, qsize_t);
-	int (*free_inode) (const struct inode *, unsigned long);
+	int (*free_inode) (const struct inode *, qsize_t);
 	int (*transfer) (struct inode *, struct iattr *);
 	int (*write_dquot) (struct dquot *);		/* Ordinary dquot write */
 	struct dquot *(*alloc_dquot)(struct super_block *, int);	/* Allocate memory for new dquot */
diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h
index a558a4c1d35a..adcc7ba3accb 100644
--- a/include/linux/quotaops.h
+++ b/include/linux/quotaops.h
@@ -26,10 +26,10 @@ int dquot_initialize(struct inode *inode, int type);
 int dquot_drop(struct inode *inode);
 
 int dquot_alloc_space(struct inode *inode, qsize_t number, int prealloc);
-int dquot_alloc_inode(const struct inode *inode, unsigned long number);
+int dquot_alloc_inode(const struct inode *inode, qsize_t number);
 
 int dquot_free_space(struct inode *inode, qsize_t number);
-int dquot_free_inode(const struct inode *inode, unsigned long number);
+int dquot_free_inode(const struct inode *inode, qsize_t number);
 
 int dquot_transfer(struct inode *inode, struct iattr *iattr);
 int dquot_commit(struct dquot *dquot);
-- 
cgit v1.2.3


From e4bc7b4b7ff783779b6928d55a9308910bf180a3 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Wed, 20 Aug 2008 16:21:01 +0200
Subject: quota: Make _SUSPENDED just a flag

Upto now, DQUOT_USR_SUSPENDED behaved like a state - i.e., either quota
was enabled or suspended or none. Now allowed states are 0, ENABLED,
ENABLED | SUSPENDED. This will be useful later when we implement separate
enabling of quota usage tracking and limits enforcement because we need to
keep track of a state which has been suspended.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/dquot.c               | 10 ++++++----
 include/linux/quotaops.h |  6 ++++--
 2 files changed, 10 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/fs/dquot.c b/fs/dquot.c
index c88330602ddd..22340c610e1a 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -1570,18 +1570,20 @@ static inline void reset_enable_flags(struct quota_info *dqopt, int type,
 {
 	switch (type) {
 		case USRQUOTA:
-			dqopt->flags &= ~DQUOT_USR_ENABLED;
 			if (remount)
 				dqopt->flags |= DQUOT_USR_SUSPENDED;
-			else
+			else {
+				dqopt->flags &= ~DQUOT_USR_ENABLED;
 				dqopt->flags &= ~DQUOT_USR_SUSPENDED;
+			}
 			break;
 		case GRPQUOTA:
-			dqopt->flags &= ~DQUOT_GRP_ENABLED;
 			if (remount)
 				dqopt->flags |= DQUOT_GRP_SUSPENDED;
-			else
+			else {
+				dqopt->flags &= ~DQUOT_GRP_ENABLED;
 				dqopt->flags &= ~DQUOT_GRP_SUSPENDED;
+			}
 			break;
 	}
 }
diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h
index adcc7ba3accb..ffd97071cd1e 100644
--- a/include/linux/quotaops.h
+++ b/include/linux/quotaops.h
@@ -67,8 +67,10 @@ static inline struct mem_dqinfo *sb_dqinfo(struct super_block *sb, int type)
 static inline int sb_has_quota_enabled(struct super_block *sb, int type)
 {
 	if (type == USRQUOTA)
-		return sb_dqopt(sb)->flags & DQUOT_USR_ENABLED;
-	return sb_dqopt(sb)->flags & DQUOT_GRP_ENABLED;
+		return (sb_dqopt(sb)->flags & DQUOT_USR_ENABLED)
+			&& !(sb_dqopt(sb)->flags & DQUOT_USR_SUSPENDED);
+	return (sb_dqopt(sb)->flags & DQUOT_GRP_ENABLED)
+		&& !(sb_dqopt(sb)->flags & DQUOT_GROUP_SUSPENDED);
 }
 
 static inline int sb_any_quota_enabled(struct super_block *sb)
-- 
cgit v1.2.3


From f55abc0fb9c3189de3da829adf3220322c0da43e Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Wed, 20 Aug 2008 17:50:32 +0200
Subject: quota: Allow to separately enable quota accounting and enforcing
 limits

Split DQUOT_USR_ENABLED (and DQUOT_GRP_ENABLED) into DQUOT_USR_USAGE_ENABLED
and DQUOT_USR_LIMITS_ENABLED. This way we are able to separately enable /
disable whether we should:
1) ignore quotas completely
2) just keep uptodate information about usage
3) actually enforce quota limits

This is going to be useful when quota is treated as filesystem metadata - we
then want to keep quota information uptodate all the time and just enable /
disable limits enforcement.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/dquot.c               | 222 +++++++++++++++++++++++++++++------------------
 fs/quota.c               |   8 +-
 include/linux/quota.h    |  30 ++++++-
 include/linux/quotaops.h |  91 ++++++++++++++-----
 4 files changed, 239 insertions(+), 112 deletions(-)

(limited to 'include/linux')

diff --git a/fs/dquot.c b/fs/dquot.c
index 22340c610e1a..7569633efe0e 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -489,7 +489,7 @@ int vfs_quota_sync(struct super_block *sb, int type)
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		if (type != -1 && cnt != type)
 			continue;
-		if (!sb_has_quota_enabled(sb, cnt))
+		if (!sb_has_quota_active(sb, cnt))
 			continue;
 		spin_lock(&dq_list_lock);
 		dirty = &dqopt->info[cnt].dqi_dirty_list;
@@ -514,8 +514,8 @@ int vfs_quota_sync(struct super_block *sb, int type)
 	}
 
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
-		if ((cnt == type || type == -1) && sb_has_quota_enabled(sb, cnt)
-			&& info_dirty(&dqopt->info[cnt]))
+		if ((cnt == type || type == -1) && sb_has_quota_active(sb, cnt)
+		    && info_dirty(&dqopt->info[cnt]))
 			sb->dq_op->write_info(sb, cnt);
 	spin_lock(&dq_list_lock);
 	dqstats.syncs++;
@@ -594,7 +594,7 @@ we_slept:
 		/* We have more than one user... nothing to do */
 		atomic_dec(&dquot->dq_count);
 		/* Releasing dquot during quotaoff phase? */
-		if (!sb_has_quota_enabled(dquot->dq_sb, dquot->dq_type) &&
+		if (!sb_has_quota_active(dquot->dq_sb, dquot->dq_type) &&
 		    atomic_read(&dquot->dq_count) == 1)
 			wake_up(&dquot->dq_wait_unused);
 		spin_unlock(&dq_list_lock);
@@ -670,7 +670,7 @@ static struct dquot *dqget(struct super_block *sb, unsigned int id, int type)
 	unsigned int hashent = hashfn(sb, id, type);
 	struct dquot *dquot, *empty = NODQUOT;
 
-        if (!sb_has_quota_enabled(sb, type))
+        if (!sb_has_quota_active(sb, type))
 		return NODQUOT;
 we_slept:
 	spin_lock(&dq_list_lock);
@@ -1041,7 +1041,8 @@ static inline char ignore_hardlimit(struct dquot *dquot)
 static int check_idq(struct dquot *dquot, qsize_t inodes, char *warntype)
 {
 	*warntype = QUOTA_NL_NOWARN;
-	if (test_bit(DQ_FAKE_B, &dquot->dq_flags))
+	if (!sb_has_quota_limits_enabled(dquot->dq_sb, dquot->dq_type) ||
+	    test_bit(DQ_FAKE_B, &dquot->dq_flags))
 		return QUOTA_OK;
 
 	if (dquot->dq_dqb.dqb_ihardlimit &&
@@ -1073,7 +1074,8 @@ static int check_idq(struct dquot *dquot, qsize_t inodes, char *warntype)
 static int check_bdq(struct dquot *dquot, qsize_t space, int prealloc, char *warntype)
 {
 	*warntype = QUOTA_NL_NOWARN;
-	if (test_bit(DQ_FAKE_B, &dquot->dq_flags))
+	if (!sb_has_quota_limits_enabled(dquot->dq_sb, dquot->dq_type) ||
+	    test_bit(DQ_FAKE_B, &dquot->dq_flags))
 		return QUOTA_OK;
 
 	if (dquot->dq_dqb.dqb_bhardlimit &&
@@ -1114,7 +1116,8 @@ static int check_bdq(struct dquot *dquot, qsize_t space, int prealloc, char *war
 static int info_idq_free(struct dquot *dquot, qsize_t inodes)
 {
 	if (test_bit(DQ_FAKE_B, &dquot->dq_flags) ||
-	    dquot->dq_dqb.dqb_curinodes <= dquot->dq_dqb.dqb_isoftlimit)
+	    dquot->dq_dqb.dqb_curinodes <= dquot->dq_dqb.dqb_isoftlimit ||
+	    !sb_has_quota_limits_enabled(dquot->dq_sb, dquot->dq_type))
 		return QUOTA_NL_NOWARN;
 
 	if (dquot->dq_dqb.dqb_curinodes - inodes <= dquot->dq_dqb.dqb_isoftlimit)
@@ -1508,7 +1511,7 @@ warn_put_all:
 /* Wrapper for transferring ownership of an inode */
 int vfs_dq_transfer(struct inode *inode, struct iattr *iattr)
 {
-	if (sb_any_quota_enabled(inode->i_sb) && !IS_NOQUOTA(inode)) {
+	if (sb_any_quota_active(inode->i_sb) && !IS_NOQUOTA(inode)) {
 		vfs_dq_init(inode);
 		if (inode->i_sb->dq_op->transfer(inode, iattr) == NO_QUOTA)
 			return 1;
@@ -1551,53 +1554,22 @@ struct dquot_operations dquot_operations = {
 	.destroy_dquot	= dquot_destroy,
 };
 
-static inline void set_enable_flags(struct quota_info *dqopt, int type)
-{
-	switch (type) {
-		case USRQUOTA:
-			dqopt->flags |= DQUOT_USR_ENABLED;
-			dqopt->flags &= ~DQUOT_USR_SUSPENDED;
-			break;
-		case GRPQUOTA:
-			dqopt->flags |= DQUOT_GRP_ENABLED;
-			dqopt->flags &= ~DQUOT_GRP_SUSPENDED;
-			break;
-	}
-}
-
-static inline void reset_enable_flags(struct quota_info *dqopt, int type,
-				      int remount)
-{
-	switch (type) {
-		case USRQUOTA:
-			if (remount)
-				dqopt->flags |= DQUOT_USR_SUSPENDED;
-			else {
-				dqopt->flags &= ~DQUOT_USR_ENABLED;
-				dqopt->flags &= ~DQUOT_USR_SUSPENDED;
-			}
-			break;
-		case GRPQUOTA:
-			if (remount)
-				dqopt->flags |= DQUOT_GRP_SUSPENDED;
-			else {
-				dqopt->flags &= ~DQUOT_GRP_ENABLED;
-				dqopt->flags &= ~DQUOT_GRP_SUSPENDED;
-			}
-			break;
-	}
-}
-
-
 /*
  * Turn quota off on a device. type == -1 ==> quotaoff for all types (umount)
  */
-int vfs_quota_off(struct super_block *sb, int type, int remount)
+int vfs_quota_disable(struct super_block *sb, int type, unsigned int flags)
 {
 	int cnt, ret = 0;
 	struct quota_info *dqopt = sb_dqopt(sb);
 	struct inode *toputinode[MAXQUOTAS];
 
+	/* Cannot turn off usage accounting without turning off limits, or
+	 * suspend quotas and simultaneously turn quotas off. */
+	if ((flags & DQUOT_USAGE_ENABLED && !(flags & DQUOT_LIMITS_ENABLED))
+	    || (flags & DQUOT_SUSPENDED && flags & (DQUOT_LIMITS_ENABLED |
+	    DQUOT_USAGE_ENABLED)))
+		return -EINVAL;
+
 	/* We need to serialize quota_off() for device */
 	mutex_lock(&dqopt->dqonoff_mutex);
 
@@ -1606,7 +1578,7 @@ int vfs_quota_off(struct super_block *sb, int type, int remount)
 	 * sometimes we are called when fill_super() failed and calling
 	 * sync_fs() in such cases does no good.
 	 */
-	if (!sb_any_quota_enabled(sb) && !sb_any_quota_suspended(sb)) {
+	if (!sb_any_quota_loaded(sb)) {
 		mutex_unlock(&dqopt->dqonoff_mutex);
 		return 0;
 	}
@@ -1614,17 +1586,28 @@ int vfs_quota_off(struct super_block *sb, int type, int remount)
 		toputinode[cnt] = NULL;
 		if (type != -1 && cnt != type)
 			continue;
-		/* If we keep inodes of quota files after remount and quotaoff
-		 * is called, drop kept inodes. */
-		if (!remount && sb_has_quota_suspended(sb, cnt)) {
-			iput(dqopt->files[cnt]);
-			dqopt->files[cnt] = NULL;
-			reset_enable_flags(dqopt, cnt, 0);
+		if (!sb_has_quota_loaded(sb, cnt))
 			continue;
+
+		if (flags & DQUOT_SUSPENDED) {
+			dqopt->flags |=
+				dquot_state_flag(DQUOT_SUSPENDED, cnt);
+		} else {
+			dqopt->flags &= ~dquot_state_flag(flags, cnt);
+			/* Turning off suspended quotas? */
+			if (!sb_has_quota_loaded(sb, cnt) &&
+			    sb_has_quota_suspended(sb, cnt)) {
+				dqopt->flags &=	~dquot_state_flag(
+							DQUOT_SUSPENDED, cnt);
+				iput(dqopt->files[cnt]);
+				dqopt->files[cnt] = NULL;
+				continue;
+			}
 		}
-		if (!sb_has_quota_enabled(sb, cnt))
+
+		/* We still have to keep quota loaded? */
+		if (sb_has_quota_loaded(sb, cnt) && !(flags & DQUOT_SUSPENDED))
 			continue;
-		reset_enable_flags(dqopt, cnt, remount);
 
 		/* Note: these are blocking operations */
 		drop_dquot_ref(sb, cnt);
@@ -1640,7 +1623,7 @@ int vfs_quota_off(struct super_block *sb, int type, int remount)
 		put_quota_format(dqopt->info[cnt].dqi_format);
 
 		toputinode[cnt] = dqopt->files[cnt];
-		if (!remount)
+		if (!sb_has_quota_loaded(sb, cnt))
 			dqopt->files[cnt] = NULL;
 		dqopt->info[cnt].dqi_flags = 0;
 		dqopt->info[cnt].dqi_igrace = 0;
@@ -1663,7 +1646,7 @@ int vfs_quota_off(struct super_block *sb, int type, int remount)
 			mutex_lock(&dqopt->dqonoff_mutex);
 			/* If quota was reenabled in the meantime, we have
 			 * nothing to do */
-			if (!sb_has_quota_enabled(sb, cnt)) {
+			if (!sb_has_quota_loaded(sb, cnt)) {
 				mutex_lock_nested(&toputinode[cnt]->i_mutex, I_MUTEX_QUOTA);
 				toputinode[cnt]->i_flags &= ~(S_IMMUTABLE |
 				  S_NOATIME | S_NOQUOTA);
@@ -1673,10 +1656,13 @@ int vfs_quota_off(struct super_block *sb, int type, int remount)
 			}
 			mutex_unlock(&dqopt->dqonoff_mutex);
 			/* On remount RO, we keep the inode pointer so that we
-			 * can reenable quota on the subsequent remount RW.
-			 * But we have better not keep inode pointer when there
-			 * is pending delete on the quota file... */
-			if (!remount)
+			 * can reenable quota on the subsequent remount RW. We
+			 * have to check 'flags' variable and not use sb_has_
+			 * function because another quotaon / quotaoff could
+			 * change global state before we got here. We refuse
+			 * to suspend quotas when there is pending delete on
+			 * the quota file... */
+			if (!(flags & DQUOT_SUSPENDED))
 				iput(toputinode[cnt]);
 			else if (!toputinode[cnt]->i_nlink)
 				ret = -EBUSY;
@@ -1686,12 +1672,22 @@ int vfs_quota_off(struct super_block *sb, int type, int remount)
 	return ret;
 }
 
+int vfs_quota_off(struct super_block *sb, int type, int remount)
+{
+	return vfs_quota_disable(sb, type, remount ? DQUOT_SUSPENDED :
+				 (DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED));
+}
+
 /*
  *	Turn quotas on on a device
  */
 
-/* Helper function when we already have the inode */
-static int vfs_quota_on_inode(struct inode *inode, int type, int format_id)
+/*
+ * Helper function to turn quotas on when we already have the inode of
+ * quota file and no quota information is loaded.
+ */
+static int vfs_load_quota_inode(struct inode *inode, int type, int format_id,
+	unsigned int flags)
 {
 	struct quota_format_type *fmt = find_quota_format(format_id);
 	struct super_block *sb = inode->i_sb;
@@ -1713,6 +1709,11 @@ static int vfs_quota_on_inode(struct inode *inode, int type, int format_id)
 		error = -EINVAL;
 		goto out_fmt;
 	}
+	/* Usage always has to be set... */
+	if (!(flags & DQUOT_USAGE_ENABLED)) {
+		error = -EINVAL;
+		goto out_fmt;
+	}
 
 	/* As we bypass the pagecache we must now flush the inode so that
 	 * we see all the changes from userspace... */
@@ -1721,8 +1722,7 @@ static int vfs_quota_on_inode(struct inode *inode, int type, int format_id)
 	invalidate_bdev(sb->s_bdev);
 	mutex_lock(&inode->i_mutex);
 	mutex_lock(&dqopt->dqonoff_mutex);
-	if (sb_has_quota_enabled(sb, type) ||
-			sb_has_quota_suspended(sb, type)) {
+	if (sb_has_quota_loaded(sb, type)) {
 		error = -EBUSY;
 		goto out_lock;
 	}
@@ -1754,7 +1754,7 @@ static int vfs_quota_on_inode(struct inode *inode, int type, int format_id)
 	}
 	mutex_unlock(&dqopt->dqio_mutex);
 	mutex_unlock(&inode->i_mutex);
-	set_enable_flags(dqopt, type);
+	dqopt->flags |= dquot_state_flag(flags, type);
 
 	add_dquot_ref(sb, type);
 	mutex_unlock(&dqopt->dqonoff_mutex);
@@ -1787,20 +1787,23 @@ static int vfs_quota_on_remount(struct super_block *sb, int type)
 	struct quota_info *dqopt = sb_dqopt(sb);
 	struct inode *inode;
 	int ret;
+	unsigned int flags;
 
 	mutex_lock(&dqopt->dqonoff_mutex);
 	if (!sb_has_quota_suspended(sb, type)) {
 		mutex_unlock(&dqopt->dqonoff_mutex);
 		return 0;
 	}
-	BUG_ON(sb_has_quota_enabled(sb, type));
-
 	inode = dqopt->files[type];
 	dqopt->files[type] = NULL;
-	reset_enable_flags(dqopt, type, 0);
+	flags = dqopt->flags & dquot_state_flag(DQUOT_USAGE_ENABLED |
+						DQUOT_LIMITS_ENABLED, type);
+	dqopt->flags &= ~dquot_state_flag(DQUOT_STATE_FLAGS, type);
 	mutex_unlock(&dqopt->dqonoff_mutex);
 
-	ret = vfs_quota_on_inode(inode, type, dqopt->info[type].dqi_fmt_id);
+	flags = dquot_generic_flag(flags, type);
+	ret = vfs_load_quota_inode(inode, type, dqopt->info[type].dqi_fmt_id,
+				   flags);
 	iput(inode);
 
 	return ret;
@@ -1816,12 +1819,12 @@ int vfs_quota_on_path(struct super_block *sb, int type, int format_id,
 	if (path->mnt->mnt_sb != sb)
 		error = -EXDEV;
 	else
-		error = vfs_quota_on_inode(path->dentry->d_inode, type,
-					   format_id);
+		error = vfs_load_quota_inode(path->dentry->d_inode, type,
+					     format_id, DQUOT_USAGE_ENABLED |
+					     DQUOT_LIMITS_ENABLED);
 	return error;
 }
 
-/* Actual function called from quotactl() */
 int vfs_quota_on(struct super_block *sb, int type, int format_id, char *name,
 		 int remount)
 {
@@ -1839,6 +1842,50 @@ int vfs_quota_on(struct super_block *sb, int type, int format_id, char *name,
 	return error;
 }
 
+/*
+ * More powerful function for turning on quotas allowing setting
+ * of individual quota flags
+ */
+int vfs_quota_enable(struct inode *inode, int type, int format_id,
+		unsigned int flags)
+{
+	int ret = 0;
+	struct super_block *sb = inode->i_sb;
+	struct quota_info *dqopt = sb_dqopt(sb);
+
+	/* Just unsuspend quotas? */
+	if (flags & DQUOT_SUSPENDED)
+		return vfs_quota_on_remount(sb, type);
+	if (!flags)
+		return 0;
+	/* Just updating flags needed? */
+	if (sb_has_quota_loaded(sb, type)) {
+		mutex_lock(&dqopt->dqonoff_mutex);
+		/* Now do a reliable test... */
+		if (!sb_has_quota_loaded(sb, type)) {
+			mutex_unlock(&dqopt->dqonoff_mutex);
+			goto load_quota;
+		}
+		if (flags & DQUOT_USAGE_ENABLED &&
+		    sb_has_quota_usage_enabled(sb, type)) {
+			ret = -EBUSY;
+			goto out_lock;
+		}
+		if (flags & DQUOT_LIMITS_ENABLED &&
+		    sb_has_quota_limits_enabled(sb, type)) {
+			ret = -EBUSY;
+			goto out_lock;
+		}
+		sb_dqopt(sb)->flags |= dquot_state_flag(flags, type);
+out_lock:
+		mutex_unlock(&dqopt->dqonoff_mutex);
+		return ret;
+	}
+
+load_quota:
+	return vfs_load_quota_inode(inode, type, format_id, flags);
+}
+
 /*
  * This function is used when filesystem needs to initialize quotas
  * during mount time.
@@ -1860,7 +1907,8 @@ int vfs_quota_on_mount(struct super_block *sb, char *qf_name,
 
 	error = security_quota_on(dentry);
 	if (!error)
-		error = vfs_quota_on_inode(dentry->d_inode, type, format_id);
+		error = vfs_load_quota_inode(dentry->d_inode, type, format_id,
+				DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED);
 
 out:
 	dput(dentry);
@@ -1997,12 +2045,14 @@ int vfs_set_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *d
 	int rc;
 
 	mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
-	if (!(dquot = dqget(sb, id, type))) {
-		mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
-		return -ESRCH;
+	dquot = dqget(sb, id, type);
+	if (!dquot) {
+		rc = -ESRCH;
+		goto out;
 	}
 	rc = do_set_dqblk(dquot, di);
 	dqput(dquot);
+out:
 	mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 	return rc;
 }
@@ -2013,7 +2063,7 @@ int vfs_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii)
 	struct mem_dqinfo *mi;
   
 	mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
-	if (!sb_has_quota_enabled(sb, type)) {
+	if (!sb_has_quota_active(sb, type)) {
 		mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 		return -ESRCH;
 	}
@@ -2032,11 +2082,12 @@ int vfs_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii)
 int vfs_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii)
 {
 	struct mem_dqinfo *mi;
+	int err = 0;
 
 	mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
-	if (!sb_has_quota_enabled(sb, type)) {
-		mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
-		return -ESRCH;
+	if (!sb_has_quota_active(sb, type)) {
+		err = -ESRCH;
+		goto out;
 	}
 	mi = sb_dqopt(sb)->info + type;
 	spin_lock(&dq_data_lock);
@@ -2050,8 +2101,9 @@ int vfs_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii)
 	mark_info_dirty(sb, type);
 	/* Force write to disk */
 	sb->dq_op->write_info(sb, type);
+out:
 	mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
-	return 0;
+	return err;
 }
 
 struct quotactl_ops vfs_quotactl_ops = {
@@ -2213,9 +2265,11 @@ EXPORT_SYMBOL(register_quota_format);
 EXPORT_SYMBOL(unregister_quota_format);
 EXPORT_SYMBOL(dqstats);
 EXPORT_SYMBOL(dq_data_lock);
+EXPORT_SYMBOL(vfs_quota_enable);
 EXPORT_SYMBOL(vfs_quota_on);
 EXPORT_SYMBOL(vfs_quota_on_path);
 EXPORT_SYMBOL(vfs_quota_on_mount);
+EXPORT_SYMBOL(vfs_quota_disable);
 EXPORT_SYMBOL(vfs_quota_off);
 EXPORT_SYMBOL(vfs_quota_sync);
 EXPORT_SYMBOL(vfs_get_dqinfo);
diff --git a/fs/quota.c b/fs/quota.c
index b7fe44e01618..8678d9f35ee9 100644
--- a/fs/quota.c
+++ b/fs/quota.c
@@ -73,7 +73,7 @@ static int generic_quotactl_valid(struct super_block *sb, int type, int cmd, qid
 		case Q_SETQUOTA:
 		case Q_GETQUOTA:
 			/* This is just informative test so we are satisfied without a lock */
-			if (!sb_has_quota_enabled(sb, type))
+			if (!sb_has_quota_active(sb, type))
 				return -ESRCH;
 	}
 
@@ -175,7 +175,7 @@ static void quota_sync_sb(struct super_block *sb, int type)
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		if (type != -1 && cnt != type)
 			continue;
-		if (!sb_has_quota_enabled(sb, cnt))
+		if (!sb_has_quota_active(sb, cnt))
 			continue;
 		mutex_lock_nested(&sb_dqopt(sb)->files[cnt]->i_mutex, I_MUTEX_QUOTA);
 		truncate_inode_pages(&sb_dqopt(sb)->files[cnt]->i_data, 0);
@@ -201,7 +201,7 @@ restart:
 		for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 			if (type != -1 && type != cnt)
 				continue;
-			if (!sb_has_quota_enabled(sb, cnt))
+			if (!sb_has_quota_active(sb, cnt))
 				continue;
 			if (!info_dirty(&sb_dqopt(sb)->info[cnt]) &&
 			    list_empty(&sb_dqopt(sb)->info[cnt].dqi_dirty_list))
@@ -245,7 +245,7 @@ static int do_quotactl(struct super_block *sb, int type, int cmd, qid_t id, void
 			__u32 fmt;
 
 			down_read(&sb_dqopt(sb)->dqptr_sem);
-			if (!sb_has_quota_enabled(sb, type)) {
+			if (!sb_has_quota_active(sb, type)) {
 				up_read(&sb_dqopt(sb)->dqptr_sem);
 				return -ESRCH;
 			}
diff --git a/include/linux/quota.h b/include/linux/quota.h
index 9ea468363f9f..93717abcd35b 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -318,12 +318,34 @@ struct quota_format_type {
 	struct quota_format_type *qf_next;
 };
 
-#define DQUOT_USR_ENABLED	0x01		/* User diskquotas enabled */
-#define DQUOT_GRP_ENABLED	0x02		/* Group diskquotas enabled */
-#define DQUOT_USR_SUSPENDED	0x04		/* User diskquotas are off, but
+/* Quota state flags - they actually come in two flavors - for users and groups */
+enum {
+	_DQUOT_USAGE_ENABLED = 0,		/* Track disk usage for users */
+	_DQUOT_LIMITS_ENABLED,			/* Enforce quota limits for users */
+	_DQUOT_SUSPENDED,			/* User diskquotas are off, but
 						 * we have necessary info in
 						 * memory to turn them on */
-#define DQUOT_GRP_SUSPENDED	0x08		/* The same for group quotas */
+	_DQUOT_STATE_FLAGS
+};
+#define DQUOT_USAGE_ENABLED	(1 << _DQUOT_USAGE_ENABLED)
+#define DQUOT_LIMITS_ENABLED	(1 << _DQUOT_LIMITS_ENABLED)
+#define DQUOT_SUSPENDED		(1 << _DQUOT_SUSPENDED)
+#define DQUOT_STATE_FLAGS	(DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED | \
+				 DQUOT_SUSPENDED)
+
+static inline unsigned int dquot_state_flag(unsigned int flags, int type)
+{
+	if (type == USRQUOTA)
+		return flags;
+	return flags << _DQUOT_STATE_FLAGS;
+}
+
+static inline unsigned int dquot_generic_flag(unsigned int flags, int type)
+{
+	if (type == USRQUOTA)
+		return flags;
+	return flags >> _DQUOT_STATE_FLAGS;
+}
 
 struct quota_info {
 	unsigned int flags;			/* Flags for diskquotas on this device */
diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h
index ffd97071cd1e..3b3346fa657c 100644
--- a/include/linux/quotaops.h
+++ b/include/linux/quotaops.h
@@ -40,11 +40,14 @@ int dquot_mark_dquot_dirty(struct dquot *dquot);
 
 int vfs_quota_on(struct super_block *sb, int type, int format_id,
  	char *path, int remount);
+int vfs_quota_enable(struct inode *inode, int type, int format_id,
+	unsigned int flags);
 int vfs_quota_on_path(struct super_block *sb, int type, int format_id,
  	struct path *path);
 int vfs_quota_on_mount(struct super_block *sb, char *qf_name,
  	int format_id, int type);
 int vfs_quota_off(struct super_block *sb, int type, int remount);
+int vfs_quota_disable(struct super_block *sb, int type, unsigned int flags);
 int vfs_quota_sync(struct super_block *sb, int type);
 int vfs_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii);
 int vfs_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii);
@@ -64,26 +67,22 @@ static inline struct mem_dqinfo *sb_dqinfo(struct super_block *sb, int type)
  * Functions for checking status of quota
  */
 
-static inline int sb_has_quota_enabled(struct super_block *sb, int type)
+static inline int sb_has_quota_usage_enabled(struct super_block *sb, int type)
 {
-	if (type == USRQUOTA)
-		return (sb_dqopt(sb)->flags & DQUOT_USR_ENABLED)
-			&& !(sb_dqopt(sb)->flags & DQUOT_USR_SUSPENDED);
-	return (sb_dqopt(sb)->flags & DQUOT_GRP_ENABLED)
-		&& !(sb_dqopt(sb)->flags & DQUOT_GROUP_SUSPENDED);
+	return sb_dqopt(sb)->flags &
+				dquot_state_flag(DQUOT_USAGE_ENABLED, type);
 }
 
-static inline int sb_any_quota_enabled(struct super_block *sb)
+static inline int sb_has_quota_limits_enabled(struct super_block *sb, int type)
 {
-	return sb_has_quota_enabled(sb, USRQUOTA) ||
-		sb_has_quota_enabled(sb, GRPQUOTA);
+	return sb_dqopt(sb)->flags &
+				dquot_state_flag(DQUOT_LIMITS_ENABLED, type);
 }
 
 static inline int sb_has_quota_suspended(struct super_block *sb, int type)
 {
-	if (type == USRQUOTA)
-		return sb_dqopt(sb)->flags & DQUOT_USR_SUSPENDED;
-	return sb_dqopt(sb)->flags & DQUOT_GRP_SUSPENDED;
+	return sb_dqopt(sb)->flags &
+				dquot_state_flag(DQUOT_SUSPENDED, type);
 }
 
 static inline int sb_any_quota_suspended(struct super_block *sb)
@@ -92,6 +91,34 @@ static inline int sb_any_quota_suspended(struct super_block *sb)
 		sb_has_quota_suspended(sb, GRPQUOTA);
 }
 
+/* Does kernel know about any quota information for given sb + type? */
+static inline int sb_has_quota_loaded(struct super_block *sb, int type)
+{
+	/* Currently if anything is on, then quota usage is on as well */
+	return sb_has_quota_usage_enabled(sb, type);
+}
+
+static inline int sb_any_quota_loaded(struct super_block *sb)
+{
+	return sb_has_quota_loaded(sb, USRQUOTA) ||
+		sb_has_quota_loaded(sb, GRPQUOTA);
+}
+
+static inline int sb_has_quota_active(struct super_block *sb, int type)
+{
+	return sb_has_quota_loaded(sb, type) &&
+	       !sb_has_quota_suspended(sb, type);
+}
+
+static inline int sb_any_quota_active(struct super_block *sb)
+{
+	return sb_has_quota_active(sb, USRQUOTA) ||
+	       sb_has_quota_active(sb, GRPQUOTA);
+}
+
+/* For backward compatibility until we remove all users */
+#define sb_any_quota_enabled(sb) sb_any_quota_active(sb)
+
 /*
  * Operations supported for diskquotas.
  */
@@ -106,7 +133,7 @@ extern struct quotactl_ops vfs_quotactl_ops;
 static inline void vfs_dq_init(struct inode *inode)
 {
 	BUG_ON(!inode->i_sb);
-	if (sb_any_quota_enabled(inode->i_sb) && !IS_NOQUOTA(inode))
+	if (sb_any_quota_active(inode->i_sb) && !IS_NOQUOTA(inode))
 		inode->i_sb->dq_op->initialize(inode, -1);
 }
 
@@ -114,7 +141,7 @@ static inline void vfs_dq_init(struct inode *inode)
  * a transaction (deadlocks possible otherwise) */
 static inline int vfs_dq_prealloc_space_nodirty(struct inode *inode, qsize_t nr)
 {
-	if (sb_any_quota_enabled(inode->i_sb)) {
+	if (sb_any_quota_active(inode->i_sb)) {
 		/* Used space is updated in alloc_space() */
 		if (inode->i_sb->dq_op->alloc_space(inode, nr, 1) == NO_QUOTA)
 			return 1;
@@ -134,7 +161,7 @@ static inline int vfs_dq_prealloc_space(struct inode *inode, qsize_t nr)
 
 static inline int vfs_dq_alloc_space_nodirty(struct inode *inode, qsize_t nr)
 {
-	if (sb_any_quota_enabled(inode->i_sb)) {
+	if (sb_any_quota_active(inode->i_sb)) {
 		/* Used space is updated in alloc_space() */
 		if (inode->i_sb->dq_op->alloc_space(inode, nr, 0) == NO_QUOTA)
 			return 1;
@@ -154,7 +181,7 @@ static inline int vfs_dq_alloc_space(struct inode *inode, qsize_t nr)
 
 static inline int vfs_dq_alloc_inode(struct inode *inode)
 {
-	if (sb_any_quota_enabled(inode->i_sb)) {
+	if (sb_any_quota_active(inode->i_sb)) {
 		vfs_dq_init(inode);
 		if (inode->i_sb->dq_op->alloc_inode(inode, 1) == NO_QUOTA)
 			return 1;
@@ -164,7 +191,7 @@ static inline int vfs_dq_alloc_inode(struct inode *inode)
 
 static inline void vfs_dq_free_space_nodirty(struct inode *inode, qsize_t nr)
 {
-	if (sb_any_quota_enabled(inode->i_sb))
+	if (sb_any_quota_active(inode->i_sb))
 		inode->i_sb->dq_op->free_space(inode, nr);
 	else
 		inode_sub_bytes(inode, nr);
@@ -178,7 +205,7 @@ static inline void vfs_dq_free_space(struct inode *inode, qsize_t nr)
 
 static inline void vfs_dq_free_inode(struct inode *inode)
 {
-	if (sb_any_quota_enabled(inode->i_sb))
+	if (sb_any_quota_active(inode->i_sb))
 		inode->i_sb->dq_op->free_inode(inode, 1);
 }
 
@@ -199,12 +226,12 @@ static inline int vfs_dq_off(struct super_block *sb, int remount)
 
 #else
 
-static inline int sb_has_quota_enabled(struct super_block *sb, int type)
+static inline int sb_has_quota_usage_enabled(struct super_block *sb, int type)
 {
 	return 0;
 }
 
-static inline int sb_any_quota_enabled(struct super_block *sb)
+static inline int sb_has_quota_limits_enabled(struct super_block *sb, int type)
 {
 	return 0;
 }
@@ -219,6 +246,30 @@ static inline int sb_any_quota_suspended(struct super_block *sb)
 	return 0;
 }
 
+/* Does kernel know about any quota information for given sb + type? */
+static inline int sb_has_quota_loaded(struct super_block *sb, int type)
+{
+	return 0;
+}
+
+static inline int sb_any_quota_loaded(struct super_block *sb)
+{
+	return 0;
+}
+
+static inline int sb_has_quota_active(struct super_block *sb, int type)
+{
+	return 0;
+}
+
+static inline int sb_any_quota_active(struct super_block *sb)
+{
+	return 0;
+}
+
+/* For backward compatibility until we remove all users */
+#define sb_any_quota_enabled(sb) sb_any_quota_active(sb)
+
 /*
  * NO-OP when quota not configured.
  */
-- 
cgit v1.2.3


From dcb30695f2cac86b71417629a6fe8042b4fe2ab2 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Wed, 20 Aug 2008 18:30:40 +0200
Subject: quota: Remove compatibility function sb_any_quota_enabled()

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 include/linux/quotaops.h | 6 ------
 1 file changed, 6 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h
index 3b3346fa657c..e840ca523175 100644
--- a/include/linux/quotaops.h
+++ b/include/linux/quotaops.h
@@ -116,9 +116,6 @@ static inline int sb_any_quota_active(struct super_block *sb)
 	       sb_has_quota_active(sb, GRPQUOTA);
 }
 
-/* For backward compatibility until we remove all users */
-#define sb_any_quota_enabled(sb) sb_any_quota_active(sb)
-
 /*
  * Operations supported for diskquotas.
  */
@@ -267,9 +264,6 @@ static inline int sb_any_quota_active(struct super_block *sb)
 	return 0;
 }
 
-/* For backward compatibility until we remove all users */
-#define sb_any_quota_enabled(sb) sb_any_quota_active(sb)
-
 /*
  * NO-OP when quota not configured.
  */
-- 
cgit v1.2.3


From ca785ec66b991e9ca74dd9840fc014487ad095e1 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Tue, 30 Sep 2008 17:53:37 +0200
Subject: quota: Introduce DQUOT_QUOTA_SYS_FILE flag

If filesystem can handle quota files as system files hidden from users, we can
skip a lot of cache invalidation, syncing, inode flags setting etc. when
turning quotas on, off and quota_sync. Allow filesystem to indicate that it is
hiding quota files from users by DQUOT_QUOTA_SYS_FILE flag.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/dquot.c            | 45 ++++++++++++++++++++++++++++++---------------
 fs/quota.c            |  3 +++
 include/linux/quota.h |  7 +++++++
 3 files changed, 40 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/fs/dquot.c b/fs/dquot.c
index 7569633efe0e..74185c34a4f0 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -1631,6 +1631,11 @@ int vfs_quota_disable(struct super_block *sb, int type, unsigned int flags)
 		dqopt->ops[cnt] = NULL;
 	}
 	mutex_unlock(&dqopt->dqonoff_mutex);
+
+	/* Skip syncing and setting flags if quota files are hidden */
+	if (dqopt->flags & DQUOT_QUOTA_SYS_FILE)
+		goto put_inodes;
+
 	/* Sync the superblock so that buffers with quota data are written to
 	 * disk (and so userspace sees correct data afterwards). */
 	if (sb->s_op->sync_fs)
@@ -1655,6 +1660,12 @@ int vfs_quota_disable(struct super_block *sb, int type, unsigned int flags)
 				mark_inode_dirty(toputinode[cnt]);
 			}
 			mutex_unlock(&dqopt->dqonoff_mutex);
+		}
+	if (sb->s_bdev)
+		invalidate_bdev(sb->s_bdev);
+put_inodes:
+	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
+		if (toputinode[cnt]) {
 			/* On remount RO, we keep the inode pointer so that we
 			 * can reenable quota on the subsequent remount RW. We
 			 * have to check 'flags' variable and not use sb_has_
@@ -1667,8 +1678,6 @@ int vfs_quota_disable(struct super_block *sb, int type, unsigned int flags)
 			else if (!toputinode[cnt]->i_nlink)
 				ret = -EBUSY;
 		}
-	if (sb->s_bdev)
-		invalidate_bdev(sb->s_bdev);
 	return ret;
 }
 
@@ -1715,25 +1724,31 @@ static int vfs_load_quota_inode(struct inode *inode, int type, int format_id,
 		goto out_fmt;
 	}
 
-	/* As we bypass the pagecache we must now flush the inode so that
-	 * we see all the changes from userspace... */
-	write_inode_now(inode, 1);
-	/* And now flush the block cache so that kernel sees the changes */
-	invalidate_bdev(sb->s_bdev);
+	if (!(dqopt->flags & DQUOT_QUOTA_SYS_FILE)) {
+		/* As we bypass the pagecache we must now flush the inode so
+		 * that we see all the changes from userspace... */
+		write_inode_now(inode, 1);
+		/* And now flush the block cache so that kernel sees the
+		 * changes */
+		invalidate_bdev(sb->s_bdev);
+	}
 	mutex_lock(&inode->i_mutex);
 	mutex_lock(&dqopt->dqonoff_mutex);
 	if (sb_has_quota_loaded(sb, type)) {
 		error = -EBUSY;
 		goto out_lock;
 	}
-	/* We don't want quota and atime on quota files (deadlocks possible)
-	 * Also nobody should write to the file - we use special IO operations
-	 * which ignore the immutable bit. */
-	down_write(&dqopt->dqptr_sem);
-	oldflags = inode->i_flags & (S_NOATIME | S_IMMUTABLE | S_NOQUOTA);
-	inode->i_flags |= S_NOQUOTA | S_NOATIME | S_IMMUTABLE;
-	up_write(&dqopt->dqptr_sem);
-	sb->dq_op->drop(inode);
+
+	if (!(dqopt->flags & DQUOT_QUOTA_SYS_FILE)) {
+		/* We don't want quota and atime on quota files (deadlocks
+		 * possible) Also nobody should write to the file - we use
+		 * special IO operations which ignore the immutable bit. */
+		down_write(&dqopt->dqptr_sem);
+		oldflags = inode->i_flags & (S_NOATIME | S_IMMUTABLE | S_NOQUOTA);
+		inode->i_flags |= S_NOQUOTA | S_NOATIME | S_IMMUTABLE;
+		up_write(&dqopt->dqptr_sem);
+		sb->dq_op->drop(inode);
+	}
 
 	error = -EIO;
 	dqopt->files[type] = igrab(inode);
diff --git a/fs/quota.c b/fs/quota.c
index 8678d9f35ee9..4a8c94f05f76 100644
--- a/fs/quota.c
+++ b/fs/quota.c
@@ -160,6 +160,9 @@ static void quota_sync_sb(struct super_block *sb, int type)
 	int cnt;
 
 	sb->s_qcop->quota_sync(sb, type);
+
+	if (sb_dqopt(sb)->flags & DQUOT_QUOTA_SYS_FILE)
+		return;
 	/* This is not very clever (and fast) but currently I don't know about
 	 * any other simple way of getting quota data to disk and we must get
 	 * them there for userspace to be visible... */
diff --git a/include/linux/quota.h b/include/linux/quota.h
index 93717abcd35b..80b8807b4988 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -332,6 +332,13 @@ enum {
 #define DQUOT_SUSPENDED		(1 << _DQUOT_SUSPENDED)
 #define DQUOT_STATE_FLAGS	(DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED | \
 				 DQUOT_SUSPENDED)
+/* Other quota flags */
+#define DQUOT_QUOTA_SYS_FILE	(1 << 6)	/* Quota file is a special
+						 * system file and user cannot
+						 * touch it. Filesystem is
+						 * responsible for setting
+						 * S_NOQUOTA, S_NOATIME flags
+						 */
 
 static inline unsigned int dquot_state_flag(unsigned int flags, int type)
 {
-- 
cgit v1.2.3


From cf770c137122b78470a67ebd5498947869a09197 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Sun, 21 Sep 2008 23:17:53 +0200
Subject: quota: Move quotaio_v[12].h from include/linux/ to fs/

Since these include files are used only by implementation of quota formats,
there's no need to have them in include/linux/.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/quota_v1.c              |  3 +-
 fs/quota_v2.c              |  7 ++--
 fs/quotaio_v1.h            | 33 +++++++++++++++++++
 fs/quotaio_v2.h            | 79 ++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/Kbuild       |  2 --
 include/linux/quotaio_v1.h | 33 -------------------
 include/linux/quotaio_v2.h | 79 ----------------------------------------------
 7 files changed, 118 insertions(+), 118 deletions(-)
 create mode 100644 fs/quotaio_v1.h
 create mode 100644 fs/quotaio_v2.h
 delete mode 100644 include/linux/quotaio_v1.h
 delete mode 100644 include/linux/quotaio_v2.h

(limited to 'include/linux')

diff --git a/fs/quota_v1.c b/fs/quota_v1.c
index 3e078eee5644..b4af1c69ad16 100644
--- a/fs/quota_v1.c
+++ b/fs/quota_v1.c
@@ -3,13 +3,14 @@
 #include <linux/quota.h>
 #include <linux/quotaops.h>
 #include <linux/dqblk_v1.h>
-#include <linux/quotaio_v1.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/module.h>
 
 #include <asm/byteorder.h>
 
+#include "quotaio_v1.h"
+
 MODULE_AUTHOR("Jan Kara");
 MODULE_DESCRIPTION("Old quota format support");
 MODULE_LICENSE("GPL");
diff --git a/fs/quota_v2.c b/fs/quota_v2.c
index 51c4717f7c6a..a21d1a7c356a 100644
--- a/fs/quota_v2.c
+++ b/fs/quota_v2.c
@@ -6,7 +6,6 @@
 #include <linux/fs.h>
 #include <linux/mount.h>
 #include <linux/dqblk_v2.h>
-#include <linux/quotaio_v2.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/module.h>
@@ -15,6 +14,8 @@
 
 #include <asm/byteorder.h>
 
+#include "quotaio_v2.h"
+
 MODULE_AUTHOR("Jan Kara");
 MODULE_DESCRIPTION("Quota format v2 support");
 MODULE_LICENSE("GPL");
@@ -129,8 +130,8 @@ static void mem2diskdqb(struct v2_disk_dqblk *d, struct mem_dqblk *m, qid_t id)
 	d->dqb_isoftlimit = cpu_to_le32(m->dqb_isoftlimit);
 	d->dqb_curinodes = cpu_to_le32(m->dqb_curinodes);
 	d->dqb_itime = cpu_to_le64(m->dqb_itime);
-	d->dqb_bhardlimit = cpu_to_le32(v2_qbtos(m->dqb_bhardlimit));
-	d->dqb_bsoftlimit = cpu_to_le32(v2_qbtos(m->dqb_bsoftlimit));
+	d->dqb_bhardlimit = cpu_to_le32(v2_stoqb(m->dqb_bhardlimit));
+	d->dqb_bsoftlimit = cpu_to_le32(v2_stoqb(m->dqb_bsoftlimit));
 	d->dqb_curspace = cpu_to_le64(m->dqb_curspace);
 	d->dqb_btime = cpu_to_le64(m->dqb_btime);
 	d->dqb_id = cpu_to_le32(id);
diff --git a/fs/quotaio_v1.h b/fs/quotaio_v1.h
new file mode 100644
index 000000000000..746654b5de70
--- /dev/null
+++ b/fs/quotaio_v1.h
@@ -0,0 +1,33 @@
+#ifndef _LINUX_QUOTAIO_V1_H
+#define _LINUX_QUOTAIO_V1_H
+
+#include <linux/types.h>
+
+/*
+ * The following constants define the amount of time given a user
+ * before the soft limits are treated as hard limits (usually resulting
+ * in an allocation failure). The timer is started when the user crosses
+ * their soft limit, it is reset when they go below their soft limit.
+ */
+#define MAX_IQ_TIME  604800	/* (7*24*60*60) 1 week */
+#define MAX_DQ_TIME  604800	/* (7*24*60*60) 1 week */
+
+/*
+ * The following structure defines the format of the disk quota file
+ * (as it appears on disk) - the file is an array of these structures
+ * indexed by user or group number.
+ */
+struct v1_disk_dqblk {
+	__u32 dqb_bhardlimit;	/* absolute limit on disk blks alloc */
+	__u32 dqb_bsoftlimit;	/* preferred limit on disk blks */
+	__u32 dqb_curblocks;	/* current block count */
+	__u32 dqb_ihardlimit;	/* absolute limit on allocated inodes */
+	__u32 dqb_isoftlimit;	/* preferred inode limit */
+	__u32 dqb_curinodes;	/* current # allocated inodes */
+	time_t dqb_btime;	/* time limit for excessive disk use */
+	time_t dqb_itime;	/* time limit for excessive inode use */
+};
+
+#define v1_dqoff(UID)      ((loff_t)((UID) * sizeof (struct v1_disk_dqblk)))
+
+#endif	/* _LINUX_QUOTAIO_V1_H */
diff --git a/fs/quotaio_v2.h b/fs/quotaio_v2.h
new file mode 100644
index 000000000000..303d7cbe30d4
--- /dev/null
+++ b/fs/quotaio_v2.h
@@ -0,0 +1,79 @@
+/*
+ *	Definitions of structures for vfsv0 quota format
+ */
+
+#ifndef _LINUX_QUOTAIO_V2_H
+#define _LINUX_QUOTAIO_V2_H
+
+#include <linux/types.h>
+#include <linux/quota.h>
+
+/*
+ * Definitions of magics and versions of current quota files
+ */
+#define V2_INITQMAGICS {\
+	0xd9c01f11,	/* USRQUOTA */\
+	0xd9c01927	/* GRPQUOTA */\
+}
+
+#define V2_INITQVERSIONS {\
+	0,		/* USRQUOTA */\
+	0		/* GRPQUOTA */\
+}
+
+/*
+ * The following structure defines the format of the disk quota file
+ * (as it appears on disk) - the file is a radix tree whose leaves point
+ * to blocks of these structures.
+ */
+struct v2_disk_dqblk {
+	__le32 dqb_id;		/* id this quota applies to */
+	__le32 dqb_ihardlimit;	/* absolute limit on allocated inodes */
+	__le32 dqb_isoftlimit;	/* preferred inode limit */
+	__le32 dqb_curinodes;	/* current # allocated inodes */
+	__le32 dqb_bhardlimit;	/* absolute limit on disk space (in QUOTABLOCK_SIZE) */
+	__le32 dqb_bsoftlimit;	/* preferred limit on disk space (in QUOTABLOCK_SIZE) */
+	__le64 dqb_curspace;	/* current space occupied (in bytes) */
+	__le64 dqb_btime;	/* time limit for excessive disk use */
+	__le64 dqb_itime;	/* time limit for excessive inode use */
+};
+
+/*
+ * Here are header structures as written on disk and their in-memory copies
+ */
+/* First generic header */
+struct v2_disk_dqheader {
+	__le32 dqh_magic;	/* Magic number identifying file */
+	__le32 dqh_version;	/* File version */
+};
+
+/* Header with type and version specific information */
+struct v2_disk_dqinfo {
+	__le32 dqi_bgrace;	/* Time before block soft limit becomes hard limit */
+	__le32 dqi_igrace;	/* Time before inode soft limit becomes hard limit */
+	__le32 dqi_flags;	/* Flags for quotafile (DQF_*) */
+	__le32 dqi_blocks;	/* Number of blocks in file */
+	__le32 dqi_free_blk;	/* Number of first free block in the list */
+	__le32 dqi_free_entry;	/* Number of block with at least one free entry */
+};
+
+/*
+ *  Structure of header of block with quota structures. It is padded to 16 bytes so
+ *  there will be space for exactly 21 quota-entries in a block
+ */
+struct v2_disk_dqdbheader {
+	__le32 dqdh_next_free;	/* Number of next block with free entry */
+	__le32 dqdh_prev_free;	/* Number of previous block with free entry */
+	__le16 dqdh_entries;	/* Number of valid entries in block */
+	__le16 dqdh_pad1;
+	__le32 dqdh_pad2;
+};
+
+#define V2_DQINFOOFF	sizeof(struct v2_disk_dqheader)	/* Offset of info header in file */
+#define V2_DQBLKSIZE_BITS	10
+#define V2_DQBLKSIZE	(1 << V2_DQBLKSIZE_BITS)	/* Size of block with quota structures */
+#define V2_DQTREEOFF	1		/* Offset of tree in file in blocks */
+#define V2_DQTREEDEPTH	4		/* Depth of quota tree */
+#define V2_DQSTRINBLK	((V2_DQBLKSIZE - sizeof(struct v2_disk_dqdbheader)) / sizeof(struct v2_disk_dqblk))	/* Number of entries in one blocks */
+
+#endif /* _LINUX_QUOTAIO_V2_H */
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 95ac82340c3b..900a787cbae9 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -134,8 +134,6 @@ header-y += posix_types.h
 header-y += ppdev.h
 header-y += prctl.h
 header-y += qnxtypes.h
-header-y += quotaio_v1.h
-header-y += quotaio_v2.h
 header-y += radeonfb.h
 header-y += raw.h
 header-y += resource.h
diff --git a/include/linux/quotaio_v1.h b/include/linux/quotaio_v1.h
deleted file mode 100644
index 746654b5de70..000000000000
--- a/include/linux/quotaio_v1.h
+++ /dev/null
@@ -1,33 +0,0 @@
-#ifndef _LINUX_QUOTAIO_V1_H
-#define _LINUX_QUOTAIO_V1_H
-
-#include <linux/types.h>
-
-/*
- * The following constants define the amount of time given a user
- * before the soft limits are treated as hard limits (usually resulting
- * in an allocation failure). The timer is started when the user crosses
- * their soft limit, it is reset when they go below their soft limit.
- */
-#define MAX_IQ_TIME  604800	/* (7*24*60*60) 1 week */
-#define MAX_DQ_TIME  604800	/* (7*24*60*60) 1 week */
-
-/*
- * The following structure defines the format of the disk quota file
- * (as it appears on disk) - the file is an array of these structures
- * indexed by user or group number.
- */
-struct v1_disk_dqblk {
-	__u32 dqb_bhardlimit;	/* absolute limit on disk blks alloc */
-	__u32 dqb_bsoftlimit;	/* preferred limit on disk blks */
-	__u32 dqb_curblocks;	/* current block count */
-	__u32 dqb_ihardlimit;	/* absolute limit on allocated inodes */
-	__u32 dqb_isoftlimit;	/* preferred inode limit */
-	__u32 dqb_curinodes;	/* current # allocated inodes */
-	time_t dqb_btime;	/* time limit for excessive disk use */
-	time_t dqb_itime;	/* time limit for excessive inode use */
-};
-
-#define v1_dqoff(UID)      ((loff_t)((UID) * sizeof (struct v1_disk_dqblk)))
-
-#endif	/* _LINUX_QUOTAIO_V1_H */
diff --git a/include/linux/quotaio_v2.h b/include/linux/quotaio_v2.h
deleted file mode 100644
index 303d7cbe30d4..000000000000
--- a/include/linux/quotaio_v2.h
+++ /dev/null
@@ -1,79 +0,0 @@
-/*
- *	Definitions of structures for vfsv0 quota format
- */
-
-#ifndef _LINUX_QUOTAIO_V2_H
-#define _LINUX_QUOTAIO_V2_H
-
-#include <linux/types.h>
-#include <linux/quota.h>
-
-/*
- * Definitions of magics and versions of current quota files
- */
-#define V2_INITQMAGICS {\
-	0xd9c01f11,	/* USRQUOTA */\
-	0xd9c01927	/* GRPQUOTA */\
-}
-
-#define V2_INITQVERSIONS {\
-	0,		/* USRQUOTA */\
-	0		/* GRPQUOTA */\
-}
-
-/*
- * The following structure defines the format of the disk quota file
- * (as it appears on disk) - the file is a radix tree whose leaves point
- * to blocks of these structures.
- */
-struct v2_disk_dqblk {
-	__le32 dqb_id;		/* id this quota applies to */
-	__le32 dqb_ihardlimit;	/* absolute limit on allocated inodes */
-	__le32 dqb_isoftlimit;	/* preferred inode limit */
-	__le32 dqb_curinodes;	/* current # allocated inodes */
-	__le32 dqb_bhardlimit;	/* absolute limit on disk space (in QUOTABLOCK_SIZE) */
-	__le32 dqb_bsoftlimit;	/* preferred limit on disk space (in QUOTABLOCK_SIZE) */
-	__le64 dqb_curspace;	/* current space occupied (in bytes) */
-	__le64 dqb_btime;	/* time limit for excessive disk use */
-	__le64 dqb_itime;	/* time limit for excessive inode use */
-};
-
-/*
- * Here are header structures as written on disk and their in-memory copies
- */
-/* First generic header */
-struct v2_disk_dqheader {
-	__le32 dqh_magic;	/* Magic number identifying file */
-	__le32 dqh_version;	/* File version */
-};
-
-/* Header with type and version specific information */
-struct v2_disk_dqinfo {
-	__le32 dqi_bgrace;	/* Time before block soft limit becomes hard limit */
-	__le32 dqi_igrace;	/* Time before inode soft limit becomes hard limit */
-	__le32 dqi_flags;	/* Flags for quotafile (DQF_*) */
-	__le32 dqi_blocks;	/* Number of blocks in file */
-	__le32 dqi_free_blk;	/* Number of first free block in the list */
-	__le32 dqi_free_entry;	/* Number of block with at least one free entry */
-};
-
-/*
- *  Structure of header of block with quota structures. It is padded to 16 bytes so
- *  there will be space for exactly 21 quota-entries in a block
- */
-struct v2_disk_dqdbheader {
-	__le32 dqdh_next_free;	/* Number of next block with free entry */
-	__le32 dqdh_prev_free;	/* Number of previous block with free entry */
-	__le16 dqdh_entries;	/* Number of valid entries in block */
-	__le16 dqdh_pad1;
-	__le32 dqdh_pad2;
-};
-
-#define V2_DQINFOOFF	sizeof(struct v2_disk_dqheader)	/* Offset of info header in file */
-#define V2_DQBLKSIZE_BITS	10
-#define V2_DQBLKSIZE	(1 << V2_DQBLKSIZE_BITS)	/* Size of block with quota structures */
-#define V2_DQTREEOFF	1		/* Offset of tree in file in blocks */
-#define V2_DQTREEDEPTH	4		/* Depth of quota tree */
-#define V2_DQSTRINBLK	((V2_DQBLKSIZE - sizeof(struct v2_disk_dqdbheader)) / sizeof(struct v2_disk_dqblk))	/* Number of entries in one blocks */
-
-#endif /* _LINUX_QUOTAIO_V2_H */
-- 
cgit v1.2.3


From 1ccd14b9c271c1ac6eec5c5ec5def433100e7248 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Mon, 22 Sep 2008 05:54:49 +0200
Subject: quota: Split off quota tree handling into a separate file

There is going to be a new version of quota format having 64-bit
quota limits and a new quota format for OCFS2. They are both
going to use the same tree structure as VFSv0 quota format. So
split out tree handling into a separate file and make size of
leaf blocks, amount of space usable in each block (needed for
checksumming) and structures contained in them configurable
so that the code can be shared.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/Kconfig                  |   5 +
 fs/Makefile                 |   1 +
 fs/quota_tree.c             | 645 ++++++++++++++++++++++++++++++++++++++++++++
 fs/quota_tree.h             |  25 ++
 fs/quota_v2.c               | 596 ++++------------------------------------
 fs/quotaio_v2.h             |  33 +--
 include/linux/dqblk_qtree.h |  56 ++++
 include/linux/dqblk_v2.h    |  19 +-
 8 files changed, 799 insertions(+), 581 deletions(-)
 create mode 100644 fs/quota_tree.c
 create mode 100644 fs/quota_tree.h
 create mode 100644 include/linux/dqblk_qtree.h

(limited to 'include/linux')

diff --git a/fs/Kconfig b/fs/Kconfig
index b93425ad15de..c1ce3d8831d8 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -302,6 +302,10 @@ config PRINT_QUOTA_WARNING
 	  Note that this behavior is currently deprecated and may go away in
 	  future. Please use notification via netlink socket instead.
 
+# Generic support for tree structured quota files. Seleted when needed.
+config QUOTA_TREE
+	 tristate
+
 config QFMT_V1
 	tristate "Old quota format support"
 	depends on QUOTA
@@ -313,6 +317,7 @@ config QFMT_V1
 config QFMT_V2
 	tristate "Quota format v2 support"
 	depends on QUOTA
+	select QUOTA_TREE
 	help
 	  This quota format allows using quotas with 32-bit UIDs/GIDs. If you
 	  need this functionality say Y here.
diff --git a/fs/Makefile b/fs/Makefile
index e6f423d1d228..c830611550d3 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -54,6 +54,7 @@ obj-$(CONFIG_GENERIC_ACL)	+= generic_acl.o
 obj-$(CONFIG_QUOTA)		+= dquot.o
 obj-$(CONFIG_QFMT_V1)		+= quota_v1.o
 obj-$(CONFIG_QFMT_V2)		+= quota_v2.o
+obj-$(CONFIG_QUOTA_TREE)	+= quota_tree.o
 obj-$(CONFIG_QUOTACTL)		+= quota.o
 
 obj-$(CONFIG_PROC_FS)		+= proc/
diff --git a/fs/quota_tree.c b/fs/quota_tree.c
new file mode 100644
index 000000000000..953404c95b17
--- /dev/null
+++ b/fs/quota_tree.c
@@ -0,0 +1,645 @@
+/*
+ *	vfsv0 quota IO operations on file
+ */
+
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/mount.h>
+#include <linux/dqblk_v2.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/quotaops.h>
+
+#include <asm/byteorder.h>
+
+#include "quota_tree.h"
+
+MODULE_AUTHOR("Jan Kara");
+MODULE_DESCRIPTION("Quota trie support");
+MODULE_LICENSE("GPL");
+
+#define __QUOTA_QT_PARANOIA
+
+typedef char *dqbuf_t;
+
+static int get_index(struct qtree_mem_dqinfo *info, qid_t id, int depth)
+{
+	unsigned int epb = info->dqi_usable_bs >> 2;
+
+	depth = info->dqi_qtree_depth - depth - 1;
+	while (depth--)
+		id /= epb;
+	return id % epb;
+}
+
+/* Number of entries in one blocks */
+static inline int qtree_dqstr_in_blk(struct qtree_mem_dqinfo *info)
+{
+	return (info->dqi_usable_bs - sizeof(struct qt_disk_dqdbheader))
+	       / info->dqi_entry_size;
+}
+
+static dqbuf_t getdqbuf(size_t size)
+{
+	dqbuf_t buf = kmalloc(size, GFP_NOFS);
+	if (!buf)
+		printk(KERN_WARNING "VFS: Not enough memory for quota buffers.\n");
+	return buf;
+}
+
+static inline void freedqbuf(dqbuf_t buf)
+{
+	kfree(buf);
+}
+
+static inline ssize_t read_blk(struct qtree_mem_dqinfo *info, uint blk, dqbuf_t buf)
+{
+	struct super_block *sb = info->dqi_sb;
+
+	memset(buf, 0, info->dqi_usable_bs);
+	return sb->s_op->quota_read(sb, info->dqi_type, (char *)buf,
+	       info->dqi_usable_bs, blk << info->dqi_blocksize_bits);
+}
+
+static inline ssize_t write_blk(struct qtree_mem_dqinfo *info, uint blk, dqbuf_t buf)
+{
+	struct super_block *sb = info->dqi_sb;
+
+	return sb->s_op->quota_write(sb, info->dqi_type, (char *)buf,
+	       info->dqi_usable_bs, blk << info->dqi_blocksize_bits);
+}
+
+/* Remove empty block from list and return it */
+static int get_free_dqblk(struct qtree_mem_dqinfo *info)
+{
+	dqbuf_t buf = getdqbuf(info->dqi_usable_bs);
+	struct qt_disk_dqdbheader *dh = (struct qt_disk_dqdbheader *)buf;
+	int ret, blk;
+
+	if (!buf)
+		return -ENOMEM;
+	if (info->dqi_free_blk) {
+		blk = info->dqi_free_blk;
+		ret = read_blk(info, blk, buf);
+		if (ret < 0)
+			goto out_buf;
+		info->dqi_free_blk = le32_to_cpu(dh->dqdh_next_free);
+	}
+	else {
+		memset(buf, 0, info->dqi_usable_bs);
+		/* Assure block allocation... */
+		ret = write_blk(info, info->dqi_blocks, buf);
+		if (ret < 0)
+			goto out_buf;
+		blk = info->dqi_blocks++;
+	}
+	mark_info_dirty(info->dqi_sb, info->dqi_type);
+	ret = blk;
+out_buf:
+	freedqbuf(buf);
+	return ret;
+}
+
+/* Insert empty block to the list */
+static int put_free_dqblk(struct qtree_mem_dqinfo *info, dqbuf_t buf, uint blk)
+{
+	struct qt_disk_dqdbheader *dh = (struct qt_disk_dqdbheader *)buf;
+	int err;
+
+	dh->dqdh_next_free = cpu_to_le32(info->dqi_free_blk);
+	dh->dqdh_prev_free = cpu_to_le32(0);
+	dh->dqdh_entries = cpu_to_le16(0);
+	err = write_blk(info, blk, buf);
+	if (err < 0)
+		return err;
+	info->dqi_free_blk = blk;
+	mark_info_dirty(info->dqi_sb, info->dqi_type);
+	return 0;
+}
+
+/* Remove given block from the list of blocks with free entries */
+static int remove_free_dqentry(struct qtree_mem_dqinfo *info, dqbuf_t buf, uint blk)
+{
+	dqbuf_t tmpbuf = getdqbuf(info->dqi_usable_bs);
+	struct qt_disk_dqdbheader *dh = (struct qt_disk_dqdbheader *)buf;
+	uint nextblk = le32_to_cpu(dh->dqdh_next_free);
+	uint prevblk = le32_to_cpu(dh->dqdh_prev_free);
+	int err;
+
+	if (!tmpbuf)
+		return -ENOMEM;
+	if (nextblk) {
+		err = read_blk(info, nextblk, tmpbuf);
+		if (err < 0)
+			goto out_buf;
+		((struct qt_disk_dqdbheader *)tmpbuf)->dqdh_prev_free =
+							dh->dqdh_prev_free;
+		err = write_blk(info, nextblk, tmpbuf);
+		if (err < 0)
+			goto out_buf;
+	}
+	if (prevblk) {
+		err = read_blk(info, prevblk, tmpbuf);
+		if (err < 0)
+			goto out_buf;
+		((struct qt_disk_dqdbheader *)tmpbuf)->dqdh_next_free =
+							dh->dqdh_next_free;
+		err = write_blk(info, prevblk, tmpbuf);
+		if (err < 0)
+			goto out_buf;
+	} else {
+		info->dqi_free_entry = nextblk;
+		mark_info_dirty(info->dqi_sb, info->dqi_type);
+	}
+	freedqbuf(tmpbuf);
+	dh->dqdh_next_free = dh->dqdh_prev_free = cpu_to_le32(0);
+	/* No matter whether write succeeds block is out of list */
+	if (write_blk(info, blk, buf) < 0)
+		printk(KERN_ERR "VFS: Can't write block (%u) with free entries.\n", blk);
+	return 0;
+out_buf:
+	freedqbuf(tmpbuf);
+	return err;
+}
+
+/* Insert given block to the beginning of list with free entries */
+static int insert_free_dqentry(struct qtree_mem_dqinfo *info, dqbuf_t buf, uint blk)
+{
+	dqbuf_t tmpbuf = getdqbuf(info->dqi_usable_bs);
+	struct qt_disk_dqdbheader *dh = (struct qt_disk_dqdbheader *)buf;
+	int err;
+
+	if (!tmpbuf)
+		return -ENOMEM;
+	dh->dqdh_next_free = cpu_to_le32(info->dqi_free_entry);
+	dh->dqdh_prev_free = cpu_to_le32(0);
+	err = write_blk(info, blk, buf);
+	if (err < 0)
+		goto out_buf;
+	if (info->dqi_free_entry) {
+		err = read_blk(info, info->dqi_free_entry, tmpbuf);
+		if (err < 0)
+			goto out_buf;
+		((struct qt_disk_dqdbheader *)tmpbuf)->dqdh_prev_free =
+							cpu_to_le32(blk);
+		err = write_blk(info, info->dqi_free_entry, tmpbuf);
+		if (err < 0)
+			goto out_buf;
+	}
+	freedqbuf(tmpbuf);
+	info->dqi_free_entry = blk;
+	mark_info_dirty(info->dqi_sb, info->dqi_type);
+	return 0;
+out_buf:
+	freedqbuf(tmpbuf);
+	return err;
+}
+
+/* Is the entry in the block free? */
+int qtree_entry_unused(struct qtree_mem_dqinfo *info, char *disk)
+{
+	int i;
+
+	for (i = 0; i < info->dqi_entry_size; i++)
+		if (disk[i])
+			return 0;
+	return 1;
+}
+EXPORT_SYMBOL(qtree_entry_unused);
+
+/* Find space for dquot */
+static uint find_free_dqentry(struct qtree_mem_dqinfo *info,
+			      struct dquot *dquot, int *err)
+{
+	uint blk, i;
+	struct qt_disk_dqdbheader *dh;
+	dqbuf_t buf = getdqbuf(info->dqi_usable_bs);
+	char *ddquot;
+
+	*err = 0;
+	if (!buf) {
+		*err = -ENOMEM;
+		return 0;
+	}
+	dh = (struct qt_disk_dqdbheader *)buf;
+	if (info->dqi_free_entry) {
+		blk = info->dqi_free_entry;
+		*err = read_blk(info, blk, buf);
+		if (*err < 0)
+			goto out_buf;
+	} else {
+		blk = get_free_dqblk(info);
+		if ((int)blk < 0) {
+			*err = blk;
+			freedqbuf(buf);
+			return 0;
+		}
+		memset(buf, 0, info->dqi_usable_bs);
+		/* This is enough as block is already zeroed and entry list is empty... */
+		info->dqi_free_entry = blk;
+		mark_info_dirty(dquot->dq_sb, dquot->dq_type);
+	}
+	/* Block will be full? */
+	if (le16_to_cpu(dh->dqdh_entries) + 1 >= qtree_dqstr_in_blk(info)) {
+		*err = remove_free_dqentry(info, buf, blk);
+		if (*err < 0) {
+			printk(KERN_ERR "VFS: find_free_dqentry(): Can't "
+			       "remove block (%u) from entry free list.\n",
+			       blk);
+			goto out_buf;
+		}
+	}
+	le16_add_cpu(&dh->dqdh_entries, 1);
+	/* Find free structure in block */
+	for (i = 0, ddquot = ((char *)buf) + sizeof(struct qt_disk_dqdbheader);
+	     i < qtree_dqstr_in_blk(info) && !qtree_entry_unused(info, ddquot);
+	     i++, ddquot += info->dqi_entry_size);
+#ifdef __QUOTA_QT_PARANOIA
+	if (i == qtree_dqstr_in_blk(info)) {
+		printk(KERN_ERR "VFS: find_free_dqentry(): Data block full "
+				"but it shouldn't.\n");
+		*err = -EIO;
+		goto out_buf;
+	}
+#endif
+	*err = write_blk(info, blk, buf);
+	if (*err < 0) {
+		printk(KERN_ERR "VFS: find_free_dqentry(): Can't write quota "
+				"data block %u.\n", blk);
+		goto out_buf;
+	}
+	dquot->dq_off = (blk << info->dqi_blocksize_bits) +
+			sizeof(struct qt_disk_dqdbheader) +
+			i * info->dqi_entry_size;
+	freedqbuf(buf);
+	return blk;
+out_buf:
+	freedqbuf(buf);
+	return 0;
+}
+
+/* Insert reference to structure into the trie */
+static int do_insert_tree(struct qtree_mem_dqinfo *info, struct dquot *dquot,
+			  uint *treeblk, int depth)
+{
+	dqbuf_t buf = getdqbuf(info->dqi_usable_bs);
+	int ret = 0, newson = 0, newact = 0;
+	__le32 *ref;
+	uint newblk;
+
+	if (!buf)
+		return -ENOMEM;
+	if (!*treeblk) {
+		ret = get_free_dqblk(info);
+		if (ret < 0)
+			goto out_buf;
+		*treeblk = ret;
+		memset(buf, 0, info->dqi_usable_bs);
+		newact = 1;
+	} else {
+		ret = read_blk(info, *treeblk, buf);
+		if (ret < 0) {
+			printk(KERN_ERR "VFS: Can't read tree quota block "
+					"%u.\n", *treeblk);
+			goto out_buf;
+		}
+	}
+	ref = (__le32 *)buf;
+	newblk = le32_to_cpu(ref[get_index(info, dquot->dq_id, depth)]);
+	if (!newblk)
+		newson = 1;
+	if (depth == info->dqi_qtree_depth - 1) {
+#ifdef __QUOTA_QT_PARANOIA
+		if (newblk) {
+			printk(KERN_ERR "VFS: Inserting already present quota "
+					"entry (block %u).\n",
+			       le32_to_cpu(ref[get_index(info,
+						dquot->dq_id, depth)]));
+			ret = -EIO;
+			goto out_buf;
+		}
+#endif
+		newblk = find_free_dqentry(info, dquot, &ret);
+	} else {
+		ret = do_insert_tree(info, dquot, &newblk, depth+1);
+	}
+	if (newson && ret >= 0) {
+		ref[get_index(info, dquot->dq_id, depth)] =
+							cpu_to_le32(newblk);
+		ret = write_blk(info, *treeblk, buf);
+	} else if (newact && ret < 0) {
+		put_free_dqblk(info, buf, *treeblk);
+	}
+out_buf:
+	freedqbuf(buf);
+	return ret;
+}
+
+/* Wrapper for inserting quota structure into tree */
+static inline int dq_insert_tree(struct qtree_mem_dqinfo *info,
+				 struct dquot *dquot)
+{
+	int tmp = QT_TREEOFF;
+	return do_insert_tree(info, dquot, &tmp, 0);
+}
+
+/*
+ *	We don't have to be afraid of deadlocks as we never have quotas on quota files...
+ */
+int qtree_write_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot)
+{
+	int type = dquot->dq_type;
+	struct super_block *sb = dquot->dq_sb;
+	ssize_t ret;
+	dqbuf_t ddquot = getdqbuf(info->dqi_entry_size);
+
+	if (!ddquot)
+		return -ENOMEM;
+
+	/* dq_off is guarded by dqio_mutex */
+	if (!dquot->dq_off) {
+		ret = dq_insert_tree(info, dquot);
+		if (ret < 0) {
+			printk(KERN_ERR "VFS: Error %zd occurred while "
+					"creating quota.\n", ret);
+			freedqbuf(ddquot);
+			return ret;
+		}
+	}
+	spin_lock(&dq_data_lock);
+	info->dqi_ops->mem2disk_dqblk(ddquot, dquot);
+	spin_unlock(&dq_data_lock);
+	ret = sb->s_op->quota_write(sb, type, (char *)ddquot,
+					info->dqi_entry_size, dquot->dq_off);
+	if (ret != info->dqi_entry_size) {
+		printk(KERN_WARNING "VFS: dquota write failed on dev %s\n",
+		       sb->s_id);
+		if (ret >= 0)
+			ret = -ENOSPC;
+	} else {
+		ret = 0;
+	}
+	dqstats.writes++;
+	freedqbuf(ddquot);
+
+	return ret;
+}
+EXPORT_SYMBOL(qtree_write_dquot);
+
+/* Free dquot entry in data block */
+static int free_dqentry(struct qtree_mem_dqinfo *info, struct dquot *dquot,
+			uint blk)
+{
+	struct qt_disk_dqdbheader *dh;
+	dqbuf_t buf = getdqbuf(info->dqi_usable_bs);
+	int ret = 0;
+
+	if (!buf)
+		return -ENOMEM;
+	if (dquot->dq_off >> info->dqi_blocksize_bits != blk) {
+		printk(KERN_ERR "VFS: Quota structure has offset to other "
+		  "block (%u) than it should (%u).\n", blk,
+		  (uint)(dquot->dq_off >> info->dqi_blocksize_bits));
+		goto out_buf;
+	}
+	ret = read_blk(info, blk, buf);
+	if (ret < 0) {
+		printk(KERN_ERR "VFS: Can't read quota data block %u\n", blk);
+		goto out_buf;
+	}
+	dh = (struct qt_disk_dqdbheader *)buf;
+	le16_add_cpu(&dh->dqdh_entries, -1);
+	if (!le16_to_cpu(dh->dqdh_entries)) {	/* Block got free? */
+		ret = remove_free_dqentry(info, buf, blk);
+		if (ret >= 0)
+			ret = put_free_dqblk(info, buf, blk);
+		if (ret < 0) {
+			printk(KERN_ERR "VFS: Can't move quota data block (%u) "
+			  "to free list.\n", blk);
+			goto out_buf;
+		}
+	} else {
+		memset(buf +
+		       (dquot->dq_off & ((1 << info->dqi_blocksize_bits) - 1)),
+		       0, info->dqi_entry_size);
+		if (le16_to_cpu(dh->dqdh_entries) ==
+		    qtree_dqstr_in_blk(info) - 1) {
+			/* Insert will write block itself */
+			ret = insert_free_dqentry(info, buf, blk);
+			if (ret < 0) {
+				printk(KERN_ERR "VFS: Can't insert quota data "
+				       "block (%u) to free entry list.\n", blk);
+				goto out_buf;
+			}
+		} else {
+			ret = write_blk(info, blk, buf);
+			if (ret < 0) {
+				printk(KERN_ERR "VFS: Can't write quota data "
+				  "block %u\n", blk);
+				goto out_buf;
+			}
+		}
+	}
+	dquot->dq_off = 0;	/* Quota is now unattached */
+out_buf:
+	freedqbuf(buf);
+	return ret;
+}
+
+/* Remove reference to dquot from tree */
+static int remove_tree(struct qtree_mem_dqinfo *info, struct dquot *dquot,
+		       uint *blk, int depth)
+{
+	dqbuf_t buf = getdqbuf(info->dqi_usable_bs);
+	int ret = 0;
+	uint newblk;
+	__le32 *ref = (__le32 *)buf;
+
+	if (!buf)
+		return -ENOMEM;
+	ret = read_blk(info, *blk, buf);
+	if (ret < 0) {
+		printk(KERN_ERR "VFS: Can't read quota data block %u\n", *blk);
+		goto out_buf;
+	}
+	newblk = le32_to_cpu(ref[get_index(info, dquot->dq_id, depth)]);
+	if (depth == info->dqi_qtree_depth - 1) {
+		ret = free_dqentry(info, dquot, newblk);
+		newblk = 0;
+	} else {
+		ret = remove_tree(info, dquot, &newblk, depth+1);
+	}
+	if (ret >= 0 && !newblk) {
+		int i;
+		ref[get_index(info, dquot->dq_id, depth)] = cpu_to_le32(0);
+		/* Block got empty? */
+		for (i = 0;
+		     i < (info->dqi_usable_bs >> 2) && !ref[i];
+		     i++);
+		/* Don't put the root block into the free block list */
+		if (i == (info->dqi_usable_bs >> 2)
+		    && *blk != QT_TREEOFF) {
+			put_free_dqblk(info, buf, *blk);
+			*blk = 0;
+		} else {
+			ret = write_blk(info, *blk, buf);
+			if (ret < 0)
+				printk(KERN_ERR "VFS: Can't write quota tree "
+				  "block %u.\n", *blk);
+		}
+	}
+out_buf:
+	freedqbuf(buf);
+	return ret;
+}
+
+/* Delete dquot from tree */
+int qtree_delete_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot)
+{
+	uint tmp = QT_TREEOFF;
+
+	if (!dquot->dq_off)	/* Even not allocated? */
+		return 0;
+	return remove_tree(info, dquot, &tmp, 0);
+}
+EXPORT_SYMBOL(qtree_delete_dquot);
+
+/* Find entry in block */
+static loff_t find_block_dqentry(struct qtree_mem_dqinfo *info,
+				 struct dquot *dquot, uint blk)
+{
+	dqbuf_t buf = getdqbuf(info->dqi_usable_bs);
+	loff_t ret = 0;
+	int i;
+	char *ddquot;
+
+	if (!buf)
+		return -ENOMEM;
+	ret = read_blk(info, blk, buf);
+	if (ret < 0) {
+		printk(KERN_ERR "VFS: Can't read quota tree block %u.\n", blk);
+		goto out_buf;
+	}
+	for (i = 0, ddquot = ((char *)buf) + sizeof(struct qt_disk_dqdbheader);
+	     i < qtree_dqstr_in_blk(info) && !info->dqi_ops->is_id(ddquot, dquot);
+	     i++, ddquot += info->dqi_entry_size);
+	if (i == qtree_dqstr_in_blk(info)) {
+		printk(KERN_ERR "VFS: Quota for id %u referenced "
+		  "but not present.\n", dquot->dq_id);
+		ret = -EIO;
+		goto out_buf;
+	} else {
+		ret = (blk << info->dqi_blocksize_bits) + sizeof(struct
+		  qt_disk_dqdbheader) + i * info->dqi_entry_size;
+	}
+out_buf:
+	freedqbuf(buf);
+	return ret;
+}
+
+/* Find entry for given id in the tree */
+static loff_t find_tree_dqentry(struct qtree_mem_dqinfo *info,
+				struct dquot *dquot, uint blk, int depth)
+{
+	dqbuf_t buf = getdqbuf(info->dqi_usable_bs);
+	loff_t ret = 0;
+	__le32 *ref = (__le32 *)buf;
+
+	if (!buf)
+		return -ENOMEM;
+	ret = read_blk(info, blk, buf);
+	if (ret < 0) {
+		printk(KERN_ERR "VFS: Can't read quota tree block %u.\n", blk);
+		goto out_buf;
+	}
+	ret = 0;
+	blk = le32_to_cpu(ref[get_index(info, dquot->dq_id, depth)]);
+	if (!blk)	/* No reference? */
+		goto out_buf;
+	if (depth < info->dqi_qtree_depth - 1)
+		ret = find_tree_dqentry(info, dquot, blk, depth+1);
+	else
+		ret = find_block_dqentry(info, dquot, blk);
+out_buf:
+	freedqbuf(buf);
+	return ret;
+}
+
+/* Find entry for given id in the tree - wrapper function */
+static inline loff_t find_dqentry(struct qtree_mem_dqinfo *info,
+				  struct dquot *dquot)
+{
+	return find_tree_dqentry(info, dquot, QT_TREEOFF, 0);
+}
+
+int qtree_read_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot)
+{
+	int type = dquot->dq_type;
+	struct super_block *sb = dquot->dq_sb;
+	loff_t offset;
+	dqbuf_t ddquot;
+	int ret = 0;
+
+#ifdef __QUOTA_QT_PARANOIA
+	/* Invalidated quota? */
+	if (!sb_dqopt(dquot->dq_sb)->files[type]) {
+		printk(KERN_ERR "VFS: Quota invalidated while reading!\n");
+		return -EIO;
+	}
+#endif
+	/* Do we know offset of the dquot entry in the quota file? */
+	if (!dquot->dq_off) {
+		offset = find_dqentry(info, dquot);
+		if (offset <= 0) {	/* Entry not present? */
+			if (offset < 0)
+				printk(KERN_ERR "VFS: Can't read quota "
+				  "structure for id %u.\n", dquot->dq_id);
+			dquot->dq_off = 0;
+			set_bit(DQ_FAKE_B, &dquot->dq_flags);
+			memset(&dquot->dq_dqb, 0, sizeof(struct mem_dqblk));
+			ret = offset;
+			goto out;
+		}
+		dquot->dq_off = offset;
+	}
+	ddquot = getdqbuf(info->dqi_entry_size);
+	if (!ddquot)
+		return -ENOMEM;
+	ret = sb->s_op->quota_read(sb, type, (char *)ddquot,
+				   info->dqi_entry_size, dquot->dq_off);
+	if (ret != info->dqi_entry_size) {
+		if (ret >= 0)
+			ret = -EIO;
+		printk(KERN_ERR "VFS: Error while reading quota "
+				"structure for id %u.\n", dquot->dq_id);
+		set_bit(DQ_FAKE_B, &dquot->dq_flags);
+		memset(&dquot->dq_dqb, 0, sizeof(struct mem_dqblk));
+		freedqbuf(ddquot);
+		goto out;
+	}
+	spin_lock(&dq_data_lock);
+	info->dqi_ops->disk2mem_dqblk(dquot, ddquot);
+	if (!dquot->dq_dqb.dqb_bhardlimit &&
+	    !dquot->dq_dqb.dqb_bsoftlimit &&
+	    !dquot->dq_dqb.dqb_ihardlimit &&
+	    !dquot->dq_dqb.dqb_isoftlimit)
+		set_bit(DQ_FAKE_B, &dquot->dq_flags);
+	spin_unlock(&dq_data_lock);
+	freedqbuf(ddquot);
+out:
+	dqstats.reads++;
+	return ret;
+}
+EXPORT_SYMBOL(qtree_read_dquot);
+
+/* Check whether dquot should not be deleted. We know we are
+ * the only one operating on dquot (thanks to dq_lock) */
+int qtree_release_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot)
+{
+	if (test_bit(DQ_FAKE_B, &dquot->dq_flags) && !(dquot->dq_dqb.dqb_curinodes | dquot->dq_dqb.dqb_curspace))
+		return qtree_delete_dquot(info, dquot);
+	return 0;
+}
+EXPORT_SYMBOL(qtree_release_dquot);
diff --git a/fs/quota_tree.h b/fs/quota_tree.h
new file mode 100644
index 000000000000..a1ab8db81a51
--- /dev/null
+++ b/fs/quota_tree.h
@@ -0,0 +1,25 @@
+/*
+ *	Definitions of structures for vfsv0 quota format
+ */
+
+#ifndef _LINUX_QUOTA_TREE_H
+#define _LINUX_QUOTA_TREE_H
+
+#include <linux/types.h>
+#include <linux/quota.h>
+
+/*
+ *  Structure of header of block with quota structures. It is padded to 16 bytes so
+ *  there will be space for exactly 21 quota-entries in a block
+ */
+struct qt_disk_dqdbheader {
+	__le32 dqdh_next_free;	/* Number of next block with free entry */
+	__le32 dqdh_prev_free;	/* Number of previous block with free entry */
+	__le16 dqdh_entries;	/* Number of valid entries in block */
+	__le16 dqdh_pad1;
+	__le32 dqdh_pad2;
+};
+
+#define QT_TREEOFF	1		/* Offset of tree in file in blocks */
+
+#endif /* _LINUX_QUOTAIO_TREE_H */
diff --git a/fs/quota_v2.c b/fs/quota_v2.c
index a21d1a7c356a..a87f1028a425 100644
--- a/fs/quota_v2.c
+++ b/fs/quota_v2.c
@@ -14,6 +14,7 @@
 
 #include <asm/byteorder.h>
 
+#include "quota_tree.h"
 #include "quotaio_v2.h"
 
 MODULE_AUTHOR("Jan Kara");
@@ -22,10 +23,15 @@ MODULE_LICENSE("GPL");
 
 #define __QUOTA_V2_PARANOIA
 
-typedef char *dqbuf_t;
+static void v2_mem2diskdqb(void *dp, struct dquot *dquot);
+static void v2_disk2memdqb(struct dquot *dquot, void *dp);
+static int v2_is_id(void *dp, struct dquot *dquot);
 
-#define GETIDINDEX(id, depth) (((id) >> ((V2_DQTREEDEPTH-(depth)-1)*8)) & 0xff)
-#define GETENTRIES(buf) ((struct v2_disk_dqblk *)(((char *)buf)+sizeof(struct v2_disk_dqdbheader)))
+static struct qtree_fmt_operations v2_qtree_ops = {
+	.mem2disk_dqblk = v2_mem2diskdqb,
+	.disk2mem_dqblk = v2_disk2memdqb,
+	.is_id = v2_is_id,
+};
 
 #define QUOTABLOCK_BITS 10
 #define QUOTABLOCK_SIZE (1 << QUOTABLOCK_BITS)
@@ -64,7 +70,7 @@ static int v2_check_quota_file(struct super_block *sb, int type)
 static int v2_read_file_info(struct super_block *sb, int type)
 {
 	struct v2_disk_dqinfo dinfo;
-	struct mem_dqinfo *info = sb_dqopt(sb)->info+type;
+	struct mem_dqinfo *info = sb_dqinfo(sb, type);
 	ssize_t size;
 
 	size = sb->s_op->quota_read(sb, type, (char *)&dinfo,
@@ -80,9 +86,16 @@ static int v2_read_file_info(struct super_block *sb, int type)
 	info->dqi_bgrace = le32_to_cpu(dinfo.dqi_bgrace);
 	info->dqi_igrace = le32_to_cpu(dinfo.dqi_igrace);
 	info->dqi_flags = le32_to_cpu(dinfo.dqi_flags);
-	info->u.v2_i.dqi_blocks = le32_to_cpu(dinfo.dqi_blocks);
-	info->u.v2_i.dqi_free_blk = le32_to_cpu(dinfo.dqi_free_blk);
-	info->u.v2_i.dqi_free_entry = le32_to_cpu(dinfo.dqi_free_entry);
+	info->u.v2_i.i.dqi_sb = sb;
+	info->u.v2_i.i.dqi_type = type;
+	info->u.v2_i.i.dqi_blocks = le32_to_cpu(dinfo.dqi_blocks);
+	info->u.v2_i.i.dqi_free_blk = le32_to_cpu(dinfo.dqi_free_blk);
+	info->u.v2_i.i.dqi_free_entry = le32_to_cpu(dinfo.dqi_free_entry);
+	info->u.v2_i.i.dqi_blocksize_bits = V2_DQBLKSIZE_BITS;
+	info->u.v2_i.i.dqi_usable_bs = 1 << V2_DQBLKSIZE_BITS;
+	info->u.v2_i.i.dqi_qtree_depth = qtree_depth(&info->u.v2_i.i);
+	info->u.v2_i.i.dqi_entry_size = sizeof(struct v2_disk_dqblk);
+	info->u.v2_i.i.dqi_ops = &v2_qtree_ops;
 	return 0;
 }
 
@@ -90,7 +103,7 @@ static int v2_read_file_info(struct super_block *sb, int type)
 static int v2_write_file_info(struct super_block *sb, int type)
 {
 	struct v2_disk_dqinfo dinfo;
-	struct mem_dqinfo *info = sb_dqopt(sb)->info+type;
+	struct mem_dqinfo *info = sb_dqinfo(sb, type);
 	ssize_t size;
 
 	spin_lock(&dq_data_lock);
@@ -99,9 +112,9 @@ static int v2_write_file_info(struct super_block *sb, int type)
 	dinfo.dqi_igrace = cpu_to_le32(info->dqi_igrace);
 	dinfo.dqi_flags = cpu_to_le32(info->dqi_flags & DQF_MASK);
 	spin_unlock(&dq_data_lock);
-	dinfo.dqi_blocks = cpu_to_le32(info->u.v2_i.dqi_blocks);
-	dinfo.dqi_free_blk = cpu_to_le32(info->u.v2_i.dqi_free_blk);
-	dinfo.dqi_free_entry = cpu_to_le32(info->u.v2_i.dqi_free_entry);
+	dinfo.dqi_blocks = cpu_to_le32(info->u.v2_i.i.dqi_blocks);
+	dinfo.dqi_free_blk = cpu_to_le32(info->u.v2_i.i.dqi_free_blk);
+	dinfo.dqi_free_entry = cpu_to_le32(info->u.v2_i.i.dqi_free_entry);
 	size = sb->s_op->quota_write(sb, type, (char *)&dinfo,
 	       sizeof(struct v2_disk_dqinfo), V2_DQINFOOFF);
 	if (size != sizeof(struct v2_disk_dqinfo)) {
@@ -112,8 +125,11 @@ static int v2_write_file_info(struct super_block *sb, int type)
 	return 0;
 }
 
-static void disk2memdqb(struct mem_dqblk *m, struct v2_disk_dqblk *d)
+static void v2_disk2memdqb(struct dquot *dquot, void *dp)
 {
+	struct v2_disk_dqblk *d = dp, empty;
+	struct mem_dqblk *m = &dquot->dq_dqb;
+
 	m->dqb_ihardlimit = le32_to_cpu(d->dqb_ihardlimit);
 	m->dqb_isoftlimit = le32_to_cpu(d->dqb_isoftlimit);
 	m->dqb_curinodes = le32_to_cpu(d->dqb_curinodes);
@@ -122,10 +138,20 @@ static void disk2memdqb(struct mem_dqblk *m, struct v2_disk_dqblk *d)
 	m->dqb_bsoftlimit = v2_qbtos(le32_to_cpu(d->dqb_bsoftlimit));
 	m->dqb_curspace = le64_to_cpu(d->dqb_curspace);
 	m->dqb_btime = le64_to_cpu(d->dqb_btime);
+	/* We need to escape back all-zero structure */
+	memset(&empty, 0, sizeof(struct v2_disk_dqblk));
+	empty.dqb_itime = cpu_to_le64(1);
+	if (!memcmp(&empty, dp, sizeof(struct v2_disk_dqblk)))
+		m->dqb_itime = 0;
 }
 
-static void mem2diskdqb(struct v2_disk_dqblk *d, struct mem_dqblk *m, qid_t id)
+static void v2_mem2diskdqb(void *dp, struct dquot *dquot)
 {
+	struct v2_disk_dqblk *d = dp;
+	struct mem_dqblk *m = &dquot->dq_dqb;
+	struct qtree_mem_dqinfo *info =
+			&sb_dqinfo(dquot->dq_sb, dquot->dq_type)->u.v2_i.i;
+
 	d->dqb_ihardlimit = cpu_to_le32(m->dqb_ihardlimit);
 	d->dqb_isoftlimit = cpu_to_le32(m->dqb_isoftlimit);
 	d->dqb_curinodes = cpu_to_le32(m->dqb_curinodes);
@@ -134,553 +160,35 @@ static void mem2diskdqb(struct v2_disk_dqblk *d, struct mem_dqblk *m, qid_t id)
 	d->dqb_bsoftlimit = cpu_to_le32(v2_stoqb(m->dqb_bsoftlimit));
 	d->dqb_curspace = cpu_to_le64(m->dqb_curspace);
 	d->dqb_btime = cpu_to_le64(m->dqb_btime);
-	d->dqb_id = cpu_to_le32(id);
-}
-
-static dqbuf_t getdqbuf(void)
-{
-	dqbuf_t buf = kmalloc(V2_DQBLKSIZE, GFP_NOFS);
-	if (!buf)
-		printk(KERN_WARNING "VFS: Not enough memory for quota buffers.\n");
-	return buf;
-}
-
-static inline void freedqbuf(dqbuf_t buf)
-{
-	kfree(buf);
-}
-
-static inline ssize_t read_blk(struct super_block *sb, int type, uint blk, dqbuf_t buf)
-{
-	memset(buf, 0, V2_DQBLKSIZE);
-	return sb->s_op->quota_read(sb, type, (char *)buf,
-	       V2_DQBLKSIZE, blk << V2_DQBLKSIZE_BITS);
-}
-
-static inline ssize_t write_blk(struct super_block *sb, int type, uint blk, dqbuf_t buf)
-{
-	return sb->s_op->quota_write(sb, type, (char *)buf,
-	       V2_DQBLKSIZE, blk << V2_DQBLKSIZE_BITS);
-}
-
-/* Remove empty block from list and return it */
-static int get_free_dqblk(struct super_block *sb, int type)
-{
-	dqbuf_t buf = getdqbuf();
-	struct mem_dqinfo *info = sb_dqinfo(sb, type);
-	struct v2_disk_dqdbheader *dh = (struct v2_disk_dqdbheader *)buf;
-	int ret, blk;
-
-	if (!buf)
-		return -ENOMEM;
-	if (info->u.v2_i.dqi_free_blk) {
-		blk = info->u.v2_i.dqi_free_blk;
-		if ((ret = read_blk(sb, type, blk, buf)) < 0)
-			goto out_buf;
-		info->u.v2_i.dqi_free_blk = le32_to_cpu(dh->dqdh_next_free);
-	}
-	else {
-		memset(buf, 0, V2_DQBLKSIZE);
-		/* Assure block allocation... */
-		if ((ret = write_blk(sb, type, info->u.v2_i.dqi_blocks, buf)) < 0)
-			goto out_buf;
-		blk = info->u.v2_i.dqi_blocks++;
-	}
-	mark_info_dirty(sb, type);
-	ret = blk;
-out_buf:
-	freedqbuf(buf);
-	return ret;
-}
-
-/* Insert empty block to the list */
-static int put_free_dqblk(struct super_block *sb, int type, dqbuf_t buf, uint blk)
-{
-	struct mem_dqinfo *info = sb_dqinfo(sb, type);
-	struct v2_disk_dqdbheader *dh = (struct v2_disk_dqdbheader *)buf;
-	int err;
-
-	dh->dqdh_next_free = cpu_to_le32(info->u.v2_i.dqi_free_blk);
-	dh->dqdh_prev_free = cpu_to_le32(0);
-	dh->dqdh_entries = cpu_to_le16(0);
-	info->u.v2_i.dqi_free_blk = blk;
-	mark_info_dirty(sb, type);
-	/* Some strange block. We had better leave it... */
-	if ((err = write_blk(sb, type, blk, buf)) < 0)
-		return err;
-	return 0;
+	d->dqb_id = cpu_to_le32(dquot->dq_id);
+	if (qtree_entry_unused(info, dp))
+		d->dqb_itime = cpu_to_le64(1);
 }
 
-/* Remove given block from the list of blocks with free entries */
-static int remove_free_dqentry(struct super_block *sb, int type, dqbuf_t buf, uint blk)
+static int v2_is_id(void *dp, struct dquot *dquot)
 {
-	dqbuf_t tmpbuf = getdqbuf();
-	struct mem_dqinfo *info = sb_dqinfo(sb, type);
-	struct v2_disk_dqdbheader *dh = (struct v2_disk_dqdbheader *)buf;
-	uint nextblk = le32_to_cpu(dh->dqdh_next_free), prevblk = le32_to_cpu(dh->dqdh_prev_free);
-	int err;
+	struct v2_disk_dqblk *d = dp;
+	struct qtree_mem_dqinfo *info =
+			&sb_dqinfo(dquot->dq_sb, dquot->dq_type)->u.v2_i.i;
 
-	if (!tmpbuf)
-		return -ENOMEM;
-	if (nextblk) {
-		if ((err = read_blk(sb, type, nextblk, tmpbuf)) < 0)
-			goto out_buf;
-		((struct v2_disk_dqdbheader *)tmpbuf)->dqdh_prev_free = dh->dqdh_prev_free;
-		if ((err = write_blk(sb, type, nextblk, tmpbuf)) < 0)
-			goto out_buf;
-	}
-	if (prevblk) {
-		if ((err = read_blk(sb, type, prevblk, tmpbuf)) < 0)
-			goto out_buf;
-		((struct v2_disk_dqdbheader *)tmpbuf)->dqdh_next_free = dh->dqdh_next_free;
-		if ((err = write_blk(sb, type, prevblk, tmpbuf)) < 0)
-			goto out_buf;
-	}
-	else {
-		info->u.v2_i.dqi_free_entry = nextblk;
-		mark_info_dirty(sb, type);
-	}
-	freedqbuf(tmpbuf);
-	dh->dqdh_next_free = dh->dqdh_prev_free = cpu_to_le32(0);
-	/* No matter whether write succeeds block is out of list */
-	if (write_blk(sb, type, blk, buf) < 0)
-		printk(KERN_ERR "VFS: Can't write block (%u) with free entries.\n", blk);
-	return 0;
-out_buf:
-	freedqbuf(tmpbuf);
-	return err;
-}
-
-/* Insert given block to the beginning of list with free entries */
-static int insert_free_dqentry(struct super_block *sb, int type, dqbuf_t buf, uint blk)
-{
-	dqbuf_t tmpbuf = getdqbuf();
-	struct mem_dqinfo *info = sb_dqinfo(sb, type);
-	struct v2_disk_dqdbheader *dh = (struct v2_disk_dqdbheader *)buf;
-	int err;
-
-	if (!tmpbuf)
-		return -ENOMEM;
-	dh->dqdh_next_free = cpu_to_le32(info->u.v2_i.dqi_free_entry);
-	dh->dqdh_prev_free = cpu_to_le32(0);
-	if ((err = write_blk(sb, type, blk, buf)) < 0)
-		goto out_buf;
-	if (info->u.v2_i.dqi_free_entry) {
-		if ((err = read_blk(sb, type, info->u.v2_i.dqi_free_entry, tmpbuf)) < 0)
-			goto out_buf;
-		((struct v2_disk_dqdbheader *)tmpbuf)->dqdh_prev_free = cpu_to_le32(blk);
-		if ((err = write_blk(sb, type, info->u.v2_i.dqi_free_entry, tmpbuf)) < 0)
-			goto out_buf;
-	}
-	freedqbuf(tmpbuf);
-	info->u.v2_i.dqi_free_entry = blk;
-	mark_info_dirty(sb, type);
-	return 0;
-out_buf:
-	freedqbuf(tmpbuf);
-	return err;
-}
-
-/* Find space for dquot */
-static uint find_free_dqentry(struct dquot *dquot, int *err)
-{
-	struct super_block *sb = dquot->dq_sb;
-	struct mem_dqinfo *info = sb_dqopt(sb)->info+dquot->dq_type;
-	uint blk, i;
-	struct v2_disk_dqdbheader *dh;
-	struct v2_disk_dqblk *ddquot;
-	struct v2_disk_dqblk fakedquot;
-	dqbuf_t buf;
-
-	*err = 0;
-	if (!(buf = getdqbuf())) {
-		*err = -ENOMEM;
+	if (qtree_entry_unused(info, dp))
 		return 0;
-	}
-	dh = (struct v2_disk_dqdbheader *)buf;
-	ddquot = GETENTRIES(buf);
-	if (info->u.v2_i.dqi_free_entry) {
-		blk = info->u.v2_i.dqi_free_entry;
-		if ((*err = read_blk(sb, dquot->dq_type, blk, buf)) < 0)
-			goto out_buf;
-	}
-	else {
-		blk = get_free_dqblk(sb, dquot->dq_type);
-		if ((int)blk < 0) {
-			*err = blk;
-			freedqbuf(buf);
-			return 0;
-		}
-		memset(buf, 0, V2_DQBLKSIZE);
-		/* This is enough as block is already zeroed and entry list is empty... */
-		info->u.v2_i.dqi_free_entry = blk;
-		mark_info_dirty(sb, dquot->dq_type);
-	}
-	if (le16_to_cpu(dh->dqdh_entries)+1 >= V2_DQSTRINBLK)	/* Block will be full? */
-		if ((*err = remove_free_dqentry(sb, dquot->dq_type, buf, blk)) < 0) {
-			printk(KERN_ERR "VFS: find_free_dqentry(): Can't remove block (%u) from entry free list.\n", blk);
-			goto out_buf;
-		}
-	le16_add_cpu(&dh->dqdh_entries, 1);
-	memset(&fakedquot, 0, sizeof(struct v2_disk_dqblk));
-	/* Find free structure in block */
-	for (i = 0; i < V2_DQSTRINBLK && memcmp(&fakedquot, ddquot+i, sizeof(struct v2_disk_dqblk)); i++);
-#ifdef __QUOTA_V2_PARANOIA
-	if (i == V2_DQSTRINBLK) {
-		printk(KERN_ERR "VFS: find_free_dqentry(): Data block full but it shouldn't.\n");
-		*err = -EIO;
-		goto out_buf;
-	}
-#endif
-	if ((*err = write_blk(sb, dquot->dq_type, blk, buf)) < 0) {
-		printk(KERN_ERR "VFS: find_free_dqentry(): Can't write quota data block %u.\n", blk);
-		goto out_buf;
-	}
-	dquot->dq_off = (blk<<V2_DQBLKSIZE_BITS)+sizeof(struct v2_disk_dqdbheader)+i*sizeof(struct v2_disk_dqblk);
-	freedqbuf(buf);
-	return blk;
-out_buf:
-	freedqbuf(buf);
-	return 0;
+	return le32_to_cpu(d->dqb_id) == dquot->dq_id;
 }
 
-/* Insert reference to structure into the trie */
-static int do_insert_tree(struct dquot *dquot, uint *treeblk, int depth)
-{
-	struct super_block *sb = dquot->dq_sb;
-	dqbuf_t buf;
-	int ret = 0, newson = 0, newact = 0;
-	__le32 *ref;
-	uint newblk;
-
-	if (!(buf = getdqbuf()))
-		return -ENOMEM;
-	if (!*treeblk) {
-		ret = get_free_dqblk(sb, dquot->dq_type);
-		if (ret < 0)
-			goto out_buf;
-		*treeblk = ret;
-		memset(buf, 0, V2_DQBLKSIZE);
-		newact = 1;
-	}
-	else {
-		if ((ret = read_blk(sb, dquot->dq_type, *treeblk, buf)) < 0) {
-			printk(KERN_ERR "VFS: Can't read tree quota block %u.\n", *treeblk);
-			goto out_buf;
-		}
-	}
-	ref = (__le32 *)buf;
-	newblk = le32_to_cpu(ref[GETIDINDEX(dquot->dq_id, depth)]);
-	if (!newblk)
-		newson = 1;
-	if (depth == V2_DQTREEDEPTH-1) {
-#ifdef __QUOTA_V2_PARANOIA
-		if (newblk) {
-			printk(KERN_ERR "VFS: Inserting already present quota entry (block %u).\n", le32_to_cpu(ref[GETIDINDEX(dquot->dq_id, depth)]));
-			ret = -EIO;
-			goto out_buf;
-		}
-#endif
-		newblk = find_free_dqentry(dquot, &ret);
-	}
-	else
-		ret = do_insert_tree(dquot, &newblk, depth+1);
-	if (newson && ret >= 0) {
-		ref[GETIDINDEX(dquot->dq_id, depth)] = cpu_to_le32(newblk);
-		ret = write_blk(sb, dquot->dq_type, *treeblk, buf);
-	}
-	else if (newact && ret < 0)
-		put_free_dqblk(sb, dquot->dq_type, buf, *treeblk);
-out_buf:
-	freedqbuf(buf);
-	return ret;
-}
-
-/* Wrapper for inserting quota structure into tree */
-static inline int dq_insert_tree(struct dquot *dquot)
+static int v2_read_dquot(struct dquot *dquot)
 {
-	int tmp = V2_DQTREEOFF;
-	return do_insert_tree(dquot, &tmp, 0);
+	return qtree_read_dquot(&sb_dqinfo(dquot->dq_sb, dquot->dq_type)->u.v2_i.i, dquot);
 }
 
-/*
- *	We don't have to be afraid of deadlocks as we never have quotas on quota files...
- */
 static int v2_write_dquot(struct dquot *dquot)
 {
-	int type = dquot->dq_type;
-	ssize_t ret;
-	struct v2_disk_dqblk ddquot, empty;
-
-	/* dq_off is guarded by dqio_mutex */
-	if (!dquot->dq_off)
-		if ((ret = dq_insert_tree(dquot)) < 0) {
-			printk(KERN_ERR "VFS: Error %zd occurred while creating quota.\n", ret);
-			return ret;
-		}
-	spin_lock(&dq_data_lock);
-	mem2diskdqb(&ddquot, &dquot->dq_dqb, dquot->dq_id);
-	/* Argh... We may need to write structure full of zeroes but that would be
-	 * treated as an empty place by the rest of the code. Format change would
-	 * be definitely cleaner but the problems probably are not worth it */
-	memset(&empty, 0, sizeof(struct v2_disk_dqblk));
-	if (!memcmp(&empty, &ddquot, sizeof(struct v2_disk_dqblk)))
-		ddquot.dqb_itime = cpu_to_le64(1);
-	spin_unlock(&dq_data_lock);
-	ret = dquot->dq_sb->s_op->quota_write(dquot->dq_sb, type,
-	      (char *)&ddquot, sizeof(struct v2_disk_dqblk), dquot->dq_off);
-	if (ret != sizeof(struct v2_disk_dqblk)) {
-		printk(KERN_WARNING "VFS: dquota write failed on dev %s\n", dquot->dq_sb->s_id);
-		if (ret >= 0)
-			ret = -ENOSPC;
-	}
-	else
-		ret = 0;
-	dqstats.writes++;
-
-	return ret;
+	return qtree_write_dquot(&sb_dqinfo(dquot->dq_sb, dquot->dq_type)->u.v2_i.i, dquot);
 }
 
-/* Free dquot entry in data block */
-static int free_dqentry(struct dquot *dquot, uint blk)
-{
-	struct super_block *sb = dquot->dq_sb;
-	int type = dquot->dq_type;
-	struct v2_disk_dqdbheader *dh;
-	dqbuf_t buf = getdqbuf();
-	int ret = 0;
-
-	if (!buf)
-		return -ENOMEM;
-	if (dquot->dq_off >> V2_DQBLKSIZE_BITS != blk) {
-		printk(KERN_ERR "VFS: Quota structure has offset to other "
-		  "block (%u) than it should (%u).\n", blk,
-		  (uint)(dquot->dq_off >> V2_DQBLKSIZE_BITS));
-		goto out_buf;
-	}
-	if ((ret = read_blk(sb, type, blk, buf)) < 0) {
-		printk(KERN_ERR "VFS: Can't read quota data block %u\n", blk);
-		goto out_buf;
-	}
-	dh = (struct v2_disk_dqdbheader *)buf;
-	le16_add_cpu(&dh->dqdh_entries, -1);
-	if (!le16_to_cpu(dh->dqdh_entries)) {	/* Block got free? */
-		if ((ret = remove_free_dqentry(sb, type, buf, blk)) < 0 ||
-		    (ret = put_free_dqblk(sb, type, buf, blk)) < 0) {
-			printk(KERN_ERR "VFS: Can't move quota data block (%u) "
-			  "to free list.\n", blk);
-			goto out_buf;
-		}
-	}
-	else {
-		memset(buf+(dquot->dq_off & ((1 << V2_DQBLKSIZE_BITS)-1)), 0,
-		  sizeof(struct v2_disk_dqblk));
-		if (le16_to_cpu(dh->dqdh_entries) == V2_DQSTRINBLK-1) {
-			/* Insert will write block itself */
-			if ((ret = insert_free_dqentry(sb, type, buf, blk)) < 0) {
-				printk(KERN_ERR "VFS: Can't insert quota data block (%u) to free entry list.\n", blk);
-				goto out_buf;
-			}
-		}
-		else
-			if ((ret = write_blk(sb, type, blk, buf)) < 0) {
-				printk(KERN_ERR "VFS: Can't write quota data "
-				  "block %u\n", blk);
-				goto out_buf;
-			}
-	}
-	dquot->dq_off = 0;	/* Quota is now unattached */
-out_buf:
-	freedqbuf(buf);
-	return ret;
-}
-
-/* Remove reference to dquot from tree */
-static int remove_tree(struct dquot *dquot, uint *blk, int depth)
-{
-	struct super_block *sb = dquot->dq_sb;
-	int type = dquot->dq_type;
-	dqbuf_t buf = getdqbuf();
-	int ret = 0;
-	uint newblk;
-	__le32 *ref = (__le32 *)buf;
-	
-	if (!buf)
-		return -ENOMEM;
-	if ((ret = read_blk(sb, type, *blk, buf)) < 0) {
-		printk(KERN_ERR "VFS: Can't read quota data block %u\n", *blk);
-		goto out_buf;
-	}
-	newblk = le32_to_cpu(ref[GETIDINDEX(dquot->dq_id, depth)]);
-	if (depth == V2_DQTREEDEPTH-1) {
-		ret = free_dqentry(dquot, newblk);
-		newblk = 0;
-	}
-	else
-		ret = remove_tree(dquot, &newblk, depth+1);
-	if (ret >= 0 && !newblk) {
-		int i;
-		ref[GETIDINDEX(dquot->dq_id, depth)] = cpu_to_le32(0);
-		for (i = 0; i < V2_DQBLKSIZE && !buf[i]; i++);	/* Block got empty? */
-		/* Don't put the root block into the free block list */
-		if (i == V2_DQBLKSIZE && *blk != V2_DQTREEOFF) {
-			put_free_dqblk(sb, type, buf, *blk);
-			*blk = 0;
-		}
-		else
-			if ((ret = write_blk(sb, type, *blk, buf)) < 0)
-				printk(KERN_ERR "VFS: Can't write quota tree "
-				  "block %u.\n", *blk);
-	}
-out_buf:
-	freedqbuf(buf);
-	return ret;	
-}
-
-/* Delete dquot from tree */
-static int v2_delete_dquot(struct dquot *dquot)
-{
-	uint tmp = V2_DQTREEOFF;
-
-	if (!dquot->dq_off)	/* Even not allocated? */
-		return 0;
-	return remove_tree(dquot, &tmp, 0);
-}
-
-/* Find entry in block */
-static loff_t find_block_dqentry(struct dquot *dquot, uint blk)
-{
-	dqbuf_t buf = getdqbuf();
-	loff_t ret = 0;
-	int i;
-	struct v2_disk_dqblk *ddquot = GETENTRIES(buf);
-
-	if (!buf)
-		return -ENOMEM;
-	if ((ret = read_blk(dquot->dq_sb, dquot->dq_type, blk, buf)) < 0) {
-		printk(KERN_ERR "VFS: Can't read quota tree block %u.\n", blk);
-		goto out_buf;
-	}
-	if (dquot->dq_id)
-		for (i = 0; i < V2_DQSTRINBLK &&
-		     le32_to_cpu(ddquot[i].dqb_id) != dquot->dq_id; i++);
-	else {	/* ID 0 as a bit more complicated searching... */
-		struct v2_disk_dqblk fakedquot;
-
-		memset(&fakedquot, 0, sizeof(struct v2_disk_dqblk));
-		for (i = 0; i < V2_DQSTRINBLK; i++)
-			if (!le32_to_cpu(ddquot[i].dqb_id) &&
-			    memcmp(&fakedquot, ddquot+i, sizeof(struct v2_disk_dqblk)))
-				break;
-	}
-	if (i == V2_DQSTRINBLK) {
-		printk(KERN_ERR "VFS: Quota for id %u referenced "
-		  "but not present.\n", dquot->dq_id);
-		ret = -EIO;
-		goto out_buf;
-	}
-	else
-		ret = (blk << V2_DQBLKSIZE_BITS) + sizeof(struct
-		  v2_disk_dqdbheader) + i * sizeof(struct v2_disk_dqblk);
-out_buf:
-	freedqbuf(buf);
-	return ret;
-}
-
-/* Find entry for given id in the tree */
-static loff_t find_tree_dqentry(struct dquot *dquot, uint blk, int depth)
-{
-	dqbuf_t buf = getdqbuf();
-	loff_t ret = 0;
-	__le32 *ref = (__le32 *)buf;
-
-	if (!buf)
-		return -ENOMEM;
-	if ((ret = read_blk(dquot->dq_sb, dquot->dq_type, blk, buf)) < 0) {
-		printk(KERN_ERR "VFS: Can't read quota tree block %u.\n", blk);
-		goto out_buf;
-	}
-	ret = 0;
-	blk = le32_to_cpu(ref[GETIDINDEX(dquot->dq_id, depth)]);
-	if (!blk)	/* No reference? */
-		goto out_buf;
-	if (depth < V2_DQTREEDEPTH-1)
-		ret = find_tree_dqentry(dquot, blk, depth+1);
-	else
-		ret = find_block_dqentry(dquot, blk);
-out_buf:
-	freedqbuf(buf);
-	return ret;
-}
-
-/* Find entry for given id in the tree - wrapper function */
-static inline loff_t find_dqentry(struct dquot *dquot)
-{
-	return find_tree_dqentry(dquot, V2_DQTREEOFF, 0);
-}
-
-static int v2_read_dquot(struct dquot *dquot)
-{
-	int type = dquot->dq_type;
-	loff_t offset;
-	struct v2_disk_dqblk ddquot, empty;
-	int ret = 0;
-
-#ifdef __QUOTA_V2_PARANOIA
-	/* Invalidated quota? */
-	if (!dquot->dq_sb || !sb_dqopt(dquot->dq_sb)->files[type]) {
-		printk(KERN_ERR "VFS: Quota invalidated while reading!\n");
-		return -EIO;
-	}
-#endif
-	offset = find_dqentry(dquot);
-	if (offset <= 0) {	/* Entry not present? */
-		if (offset < 0)
-			printk(KERN_ERR "VFS: Can't read quota "
-			  "structure for id %u.\n", dquot->dq_id);
-		dquot->dq_off = 0;
-		set_bit(DQ_FAKE_B, &dquot->dq_flags);
-		memset(&dquot->dq_dqb, 0, sizeof(struct mem_dqblk));
-		ret = offset;
-	}
-	else {
-		dquot->dq_off = offset;
-		if ((ret = dquot->dq_sb->s_op->quota_read(dquot->dq_sb, type,
-		    (char *)&ddquot, sizeof(struct v2_disk_dqblk), offset))
-		    != sizeof(struct v2_disk_dqblk)) {
-			if (ret >= 0)
-				ret = -EIO;
-			printk(KERN_ERR "VFS: Error while reading quota "
-			  "structure for id %u.\n", dquot->dq_id);
-			memset(&ddquot, 0, sizeof(struct v2_disk_dqblk));
-		}
-		else {
-			ret = 0;
-			/* We need to escape back all-zero structure */
-			memset(&empty, 0, sizeof(struct v2_disk_dqblk));
-			empty.dqb_itime = cpu_to_le64(1);
-			if (!memcmp(&empty, &ddquot, sizeof(struct v2_disk_dqblk)))
-				ddquot.dqb_itime = 0;
-		}
-		disk2memdqb(&dquot->dq_dqb, &ddquot);
-		if (!dquot->dq_dqb.dqb_bhardlimit &&
-			!dquot->dq_dqb.dqb_bsoftlimit &&
-			!dquot->dq_dqb.dqb_ihardlimit &&
-			!dquot->dq_dqb.dqb_isoftlimit)
-			set_bit(DQ_FAKE_B, &dquot->dq_flags);
-	}
-	dqstats.reads++;
-
-	return ret;
-}
-
-/* Check whether dquot should not be deleted. We know we are
- * the only one operating on dquot (thanks to dq_lock) */
 static int v2_release_dquot(struct dquot *dquot)
 {
-	if (test_bit(DQ_FAKE_B, &dquot->dq_flags) && !(dquot->dq_dqb.dqb_curinodes | dquot->dq_dqb.dqb_curspace))
-		return v2_delete_dquot(dquot);
-	return 0;
+	return qtree_release_dquot(&sb_dqinfo(dquot->dq_sb, dquot->dq_type)->u.v2_i.i, dquot);
 }
 
 static struct quota_format_ops v2_format_ops = {
diff --git a/fs/quotaio_v2.h b/fs/quotaio_v2.h
index 303d7cbe30d4..530fe580685c 100644
--- a/fs/quotaio_v2.h
+++ b/fs/quotaio_v2.h
@@ -21,6 +21,12 @@
 	0		/* GRPQUOTA */\
 }
 
+/* First generic header */
+struct v2_disk_dqheader {
+	__le32 dqh_magic;	/* Magic number identifying file */
+	__le32 dqh_version;	/* File version */
+};
+
 /*
  * The following structure defines the format of the disk quota file
  * (as it appears on disk) - the file is a radix tree whose leaves point
@@ -38,15 +44,6 @@ struct v2_disk_dqblk {
 	__le64 dqb_itime;	/* time limit for excessive inode use */
 };
 
-/*
- * Here are header structures as written on disk and their in-memory copies
- */
-/* First generic header */
-struct v2_disk_dqheader {
-	__le32 dqh_magic;	/* Magic number identifying file */
-	__le32 dqh_version;	/* File version */
-};
-
 /* Header with type and version specific information */
 struct v2_disk_dqinfo {
 	__le32 dqi_bgrace;	/* Time before block soft limit becomes hard limit */
@@ -57,23 +54,7 @@ struct v2_disk_dqinfo {
 	__le32 dqi_free_entry;	/* Number of block with at least one free entry */
 };
 
-/*
- *  Structure of header of block with quota structures. It is padded to 16 bytes so
- *  there will be space for exactly 21 quota-entries in a block
- */
-struct v2_disk_dqdbheader {
-	__le32 dqdh_next_free;	/* Number of next block with free entry */
-	__le32 dqdh_prev_free;	/* Number of previous block with free entry */
-	__le16 dqdh_entries;	/* Number of valid entries in block */
-	__le16 dqdh_pad1;
-	__le32 dqdh_pad2;
-};
-
 #define V2_DQINFOOFF	sizeof(struct v2_disk_dqheader)	/* Offset of info header in file */
-#define V2_DQBLKSIZE_BITS	10
-#define V2_DQBLKSIZE	(1 << V2_DQBLKSIZE_BITS)	/* Size of block with quota structures */
-#define V2_DQTREEOFF	1		/* Offset of tree in file in blocks */
-#define V2_DQTREEDEPTH	4		/* Depth of quota tree */
-#define V2_DQSTRINBLK	((V2_DQBLKSIZE - sizeof(struct v2_disk_dqdbheader)) / sizeof(struct v2_disk_dqblk))	/* Number of entries in one blocks */
+#define V2_DQBLKSIZE_BITS 10				/* Size of leaf block in tree */
 
 #endif /* _LINUX_QUOTAIO_V2_H */
diff --git a/include/linux/dqblk_qtree.h b/include/linux/dqblk_qtree.h
new file mode 100644
index 000000000000..82a16527b367
--- /dev/null
+++ b/include/linux/dqblk_qtree.h
@@ -0,0 +1,56 @@
+/*
+ *	Definitions of structures and functions for quota formats using trie
+ */
+
+#ifndef _LINUX_DQBLK_QTREE_H
+#define _LINUX_DQBLK_QTREE_H
+
+#include <linux/types.h>
+
+/* Numbers of blocks needed for updates - we count with the smallest
+ * possible block size (1024) */
+#define QTREE_INIT_ALLOC 4
+#define QTREE_INIT_REWRITE 2
+#define QTREE_DEL_ALLOC 0
+#define QTREE_DEL_REWRITE 6
+
+struct dquot;
+
+/* Operations */
+struct qtree_fmt_operations {
+	void (*mem2disk_dqblk)(void *disk, struct dquot *dquot);	/* Convert given entry from in memory format to disk one */
+	void (*disk2mem_dqblk)(struct dquot *dquot, void *disk);	/* Convert given entry from disk format to in memory one */
+	int (*is_id)(void *disk, struct dquot *dquot);	/* Is this structure for given id? */
+};
+
+/* Inmemory copy of version specific information */
+struct qtree_mem_dqinfo {
+	struct super_block *dqi_sb;	/* Sb quota is on */
+	int dqi_type;			/* Quota type */
+	unsigned int dqi_blocks;	/* # of blocks in quota file */
+	unsigned int dqi_free_blk;	/* First block in list of free blocks */
+	unsigned int dqi_free_entry;	/* First block with free entry */
+	unsigned int dqi_blocksize_bits;	/* Block size of quota file */
+	unsigned int dqi_entry_size;	/* Size of quota entry in quota file */
+	unsigned int dqi_usable_bs;	/* Space usable in block for quota data */
+	unsigned int dqi_qtree_depth;	/* Precomputed depth of quota tree */
+	struct qtree_fmt_operations *dqi_ops;	/* Operations for entry manipulation */
+};
+
+int qtree_write_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot);
+int qtree_read_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot);
+int qtree_delete_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot);
+int qtree_release_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot);
+int qtree_entry_unused(struct qtree_mem_dqinfo *info, char *disk);
+static inline int qtree_depth(struct qtree_mem_dqinfo *info)
+{
+	unsigned int epb = info->dqi_usable_bs >> 2;
+	unsigned long long entries = epb;
+	int i;
+
+	for (i = 1; entries < (1ULL << 32); i++)
+		entries *= epb;
+	return i;
+}
+
+#endif /* _LINUX_DQBLK_QTREE_H */
diff --git a/include/linux/dqblk_v2.h b/include/linux/dqblk_v2.h
index 4f853322cb7f..e5e22a787d58 100644
--- a/include/linux/dqblk_v2.h
+++ b/include/linux/dqblk_v2.h
@@ -1,26 +1,23 @@
 /*
- *	Definitions of structures for vfsv0 quota format
+ *  Definitions for vfsv0 quota format
  */
 
 #ifndef _LINUX_DQBLK_V2_H
 #define _LINUX_DQBLK_V2_H
 
-#include <linux/types.h>
+#include <linux/dqblk_qtree.h>
 
-/* id numbers of quota format */
+/* Id number of quota format */
 #define QFMT_VFS_V0 2
 
 /* Numbers of blocks needed for updates */
-#define V2_INIT_ALLOC 4
-#define V2_INIT_REWRITE 2
-#define V2_DEL_ALLOC 0
-#define V2_DEL_REWRITE 6
+#define V2_INIT_ALLOC QTREE_INIT_ALLOC
+#define V2_INIT_REWRITE QTREE_INIT_REWRITE
+#define V2_DEL_ALLOC QTREE_DEL_ALLOC
+#define V2_DEL_REWRITE QTREE_DEL_REWRITE
 
-/* Inmemory copy of version specific information */
 struct v2_mem_dqinfo {
-	unsigned int dqi_blocks;
-	unsigned int dqi_free_blk;
-	unsigned int dqi_free_entry;
+	struct qtree_mem_dqinfo i;
 };
 
 #endif /* _LINUX_DQBLK_V2_H */
-- 
cgit v1.2.3


From e3d4d56b9715e40ded2a84d0d4fa7f3b6c58983c Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Thu, 2 Oct 2008 18:44:14 +0200
Subject: quota: Convert union in mem_dqinfo to a pointer

Coming quota support for OCFS2 is going to need quite a bit
of additional per-sb quota information. Moreover having fs.h
include all the types needed for this structure would be a
pain in the a**. So remove the union from mem_dqinfo and add
a private pointer for filesystem's use.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/quota_v2.c            | 53 +++++++++++++++++++++++++++++++-----------------
 include/linux/dqblk_v1.h |  4 ----
 include/linux/dqblk_v2.h |  4 ----
 include/linux/quota.h    |  5 +----
 4 files changed, 35 insertions(+), 31 deletions(-)

(limited to 'include/linux')

diff --git a/fs/quota_v2.c b/fs/quota_v2.c
index a87f1028a425..b618b563635c 100644
--- a/fs/quota_v2.c
+++ b/fs/quota_v2.c
@@ -71,6 +71,7 @@ static int v2_read_file_info(struct super_block *sb, int type)
 {
 	struct v2_disk_dqinfo dinfo;
 	struct mem_dqinfo *info = sb_dqinfo(sb, type);
+	struct qtree_mem_dqinfo *qinfo;
 	ssize_t size;
 
 	size = sb->s_op->quota_read(sb, type, (char *)&dinfo,
@@ -80,22 +81,29 @@ static int v2_read_file_info(struct super_block *sb, int type)
 			sb->s_id);
 		return -1;
 	}
+	info->dqi_priv = kmalloc(sizeof(struct qtree_mem_dqinfo), GFP_NOFS);
+	if (!info->dqi_priv) {
+		printk(KERN_WARNING
+		       "Not enough memory for quota information structure.\n");
+		return -1;
+	}
+	qinfo = info->dqi_priv;
 	/* limits are stored as unsigned 32-bit data */
 	info->dqi_maxblimit = 0xffffffff;
 	info->dqi_maxilimit = 0xffffffff;
 	info->dqi_bgrace = le32_to_cpu(dinfo.dqi_bgrace);
 	info->dqi_igrace = le32_to_cpu(dinfo.dqi_igrace);
 	info->dqi_flags = le32_to_cpu(dinfo.dqi_flags);
-	info->u.v2_i.i.dqi_sb = sb;
-	info->u.v2_i.i.dqi_type = type;
-	info->u.v2_i.i.dqi_blocks = le32_to_cpu(dinfo.dqi_blocks);
-	info->u.v2_i.i.dqi_free_blk = le32_to_cpu(dinfo.dqi_free_blk);
-	info->u.v2_i.i.dqi_free_entry = le32_to_cpu(dinfo.dqi_free_entry);
-	info->u.v2_i.i.dqi_blocksize_bits = V2_DQBLKSIZE_BITS;
-	info->u.v2_i.i.dqi_usable_bs = 1 << V2_DQBLKSIZE_BITS;
-	info->u.v2_i.i.dqi_qtree_depth = qtree_depth(&info->u.v2_i.i);
-	info->u.v2_i.i.dqi_entry_size = sizeof(struct v2_disk_dqblk);
-	info->u.v2_i.i.dqi_ops = &v2_qtree_ops;
+	qinfo->dqi_sb = sb;
+	qinfo->dqi_type = type;
+	qinfo->dqi_blocks = le32_to_cpu(dinfo.dqi_blocks);
+	qinfo->dqi_free_blk = le32_to_cpu(dinfo.dqi_free_blk);
+	qinfo->dqi_free_entry = le32_to_cpu(dinfo.dqi_free_entry);
+	qinfo->dqi_blocksize_bits = V2_DQBLKSIZE_BITS;
+	qinfo->dqi_usable_bs = 1 << V2_DQBLKSIZE_BITS;
+	qinfo->dqi_qtree_depth = qtree_depth(qinfo);
+	qinfo->dqi_entry_size = sizeof(struct v2_disk_dqblk);
+	qinfo->dqi_ops = &v2_qtree_ops;
 	return 0;
 }
 
@@ -104,6 +112,7 @@ static int v2_write_file_info(struct super_block *sb, int type)
 {
 	struct v2_disk_dqinfo dinfo;
 	struct mem_dqinfo *info = sb_dqinfo(sb, type);
+	struct qtree_mem_dqinfo *qinfo = info->dqi_priv;
 	ssize_t size;
 
 	spin_lock(&dq_data_lock);
@@ -112,9 +121,9 @@ static int v2_write_file_info(struct super_block *sb, int type)
 	dinfo.dqi_igrace = cpu_to_le32(info->dqi_igrace);
 	dinfo.dqi_flags = cpu_to_le32(info->dqi_flags & DQF_MASK);
 	spin_unlock(&dq_data_lock);
-	dinfo.dqi_blocks = cpu_to_le32(info->u.v2_i.i.dqi_blocks);
-	dinfo.dqi_free_blk = cpu_to_le32(info->u.v2_i.i.dqi_free_blk);
-	dinfo.dqi_free_entry = cpu_to_le32(info->u.v2_i.i.dqi_free_entry);
+	dinfo.dqi_blocks = cpu_to_le32(qinfo->dqi_blocks);
+	dinfo.dqi_free_blk = cpu_to_le32(qinfo->dqi_free_blk);
+	dinfo.dqi_free_entry = cpu_to_le32(qinfo->dqi_free_entry);
 	size = sb->s_op->quota_write(sb, type, (char *)&dinfo,
 	       sizeof(struct v2_disk_dqinfo), V2_DQINFOOFF);
 	if (size != sizeof(struct v2_disk_dqinfo)) {
@@ -150,7 +159,7 @@ static void v2_mem2diskdqb(void *dp, struct dquot *dquot)
 	struct v2_disk_dqblk *d = dp;
 	struct mem_dqblk *m = &dquot->dq_dqb;
 	struct qtree_mem_dqinfo *info =
-			&sb_dqinfo(dquot->dq_sb, dquot->dq_type)->u.v2_i.i;
+			sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv;
 
 	d->dqb_ihardlimit = cpu_to_le32(m->dqb_ihardlimit);
 	d->dqb_isoftlimit = cpu_to_le32(m->dqb_isoftlimit);
@@ -169,7 +178,7 @@ static int v2_is_id(void *dp, struct dquot *dquot)
 {
 	struct v2_disk_dqblk *d = dp;
 	struct qtree_mem_dqinfo *info =
-			&sb_dqinfo(dquot->dq_sb, dquot->dq_type)->u.v2_i.i;
+			sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv;
 
 	if (qtree_entry_unused(info, dp))
 		return 0;
@@ -178,24 +187,30 @@ static int v2_is_id(void *dp, struct dquot *dquot)
 
 static int v2_read_dquot(struct dquot *dquot)
 {
-	return qtree_read_dquot(&sb_dqinfo(dquot->dq_sb, dquot->dq_type)->u.v2_i.i, dquot);
+	return qtree_read_dquot(sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv, dquot);
 }
 
 static int v2_write_dquot(struct dquot *dquot)
 {
-	return qtree_write_dquot(&sb_dqinfo(dquot->dq_sb, dquot->dq_type)->u.v2_i.i, dquot);
+	return qtree_write_dquot(sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv, dquot);
 }
 
 static int v2_release_dquot(struct dquot *dquot)
 {
-	return qtree_release_dquot(&sb_dqinfo(dquot->dq_sb, dquot->dq_type)->u.v2_i.i, dquot);
+	return qtree_release_dquot(sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv, dquot);
+}
+
+static int v2_free_file_info(struct super_block *sb, int type)
+{
+	kfree(sb_dqinfo(sb, type)->dqi_priv);
+	return 0;
 }
 
 static struct quota_format_ops v2_format_ops = {
 	.check_quota_file	= v2_check_quota_file,
 	.read_file_info		= v2_read_file_info,
 	.write_file_info	= v2_write_file_info,
-	.free_file_info		= NULL,
+	.free_file_info		= v2_free_file_info,
 	.read_dqblk		= v2_read_dquot,
 	.commit_dqblk		= v2_write_dquot,
 	.release_dqblk		= v2_release_dquot,
diff --git a/include/linux/dqblk_v1.h b/include/linux/dqblk_v1.h
index 57f1250d5a52..9cea901f5bba 100644
--- a/include/linux/dqblk_v1.h
+++ b/include/linux/dqblk_v1.h
@@ -17,8 +17,4 @@
 #define V1_DEL_ALLOC 0
 #define V1_DEL_REWRITE 2
 
-/* Special information about quotafile */
-struct v1_mem_dqinfo {
-};
-
 #endif	/* _LINUX_DQBLK_V1_H */
diff --git a/include/linux/dqblk_v2.h b/include/linux/dqblk_v2.h
index e5e22a787d58..ff8af1b4bda7 100644
--- a/include/linux/dqblk_v2.h
+++ b/include/linux/dqblk_v2.h
@@ -16,8 +16,4 @@
 #define V2_DEL_ALLOC QTREE_DEL_ALLOC
 #define V2_DEL_REWRITE QTREE_DEL_REWRITE
 
-struct v2_mem_dqinfo {
-	struct qtree_mem_dqinfo i;
-};
-
 #endif /* _LINUX_DQBLK_V2_H */
diff --git a/include/linux/quota.h b/include/linux/quota.h
index 80b8807b4988..e51dfdc0aef0 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -208,10 +208,7 @@ struct mem_dqinfo {
 	unsigned int dqi_igrace;
 	qsize_t dqi_maxblimit;
 	qsize_t dqi_maxilimit;
-	union {
-		struct v1_mem_dqinfo v1_i;
-		struct v2_mem_dqinfo v2_i;
-	} u;
+	void *dqi_priv;
 };
 
 struct super_block;
-- 
cgit v1.2.3


From db49d2df489f727096438706a5428115e84a3f0d Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Wed, 1 Oct 2008 18:21:39 +0200
Subject: quota: Allow negative usage of space and inodes

For clustered filesystems, it can happen that space / inode usage goes
negative temporarily (because some node is allocating another node
is freeing and they are not completely in sync). So let quota code
allow this and change qsize_t so a signed type so that we don't
underflow the variables.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/dquot.c            | 6 ++++--
 include/linux/quota.h | 3 ++-
 2 files changed, 6 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/fs/dquot.c b/fs/dquot.c
index 74185c34a4f0..9c78ffe1aad2 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -847,7 +847,8 @@ static inline void dquot_incr_space(struct dquot *dquot, qsize_t number)
 
 static inline void dquot_decr_inodes(struct dquot *dquot, qsize_t number)
 {
-	if (dquot->dq_dqb.dqb_curinodes > number)
+	if (sb_dqopt(dquot->dq_sb)->flags & DQUOT_NEGATIVE_USAGE ||
+	    dquot->dq_dqb.dqb_curinodes >= number)
 		dquot->dq_dqb.dqb_curinodes -= number;
 	else
 		dquot->dq_dqb.dqb_curinodes = 0;
@@ -858,7 +859,8 @@ static inline void dquot_decr_inodes(struct dquot *dquot, qsize_t number)
 
 static inline void dquot_decr_space(struct dquot *dquot, qsize_t number)
 {
-	if (dquot->dq_dqb.dqb_curspace > number)
+	if (sb_dqopt(dquot->dq_sb)->flags & DQUOT_NEGATIVE_USAGE ||
+	    dquot->dq_dqb.dqb_curspace >= number)
 		dquot->dq_dqb.dqb_curspace -= number;
 	else
 		dquot->dq_dqb.dqb_curspace = 0;
diff --git a/include/linux/quota.h b/include/linux/quota.h
index e51dfdc0aef0..75bf761caef2 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -168,7 +168,7 @@ enum {
 #include <asm/atomic.h>
 
 typedef __kernel_uid32_t qid_t; /* Type in which we store ids in memory */
-typedef __u64 qsize_t;          /* Type in which we store sizes */
+typedef long long qsize_t;	/* Type in which we store sizes */
 
 extern spinlock_t dq_data_lock;
 
@@ -336,6 +336,7 @@ enum {
 						 * responsible for setting
 						 * S_NOQUOTA, S_NOATIME flags
 						 */
+#define DQUOT_NEGATIVE_USAGE	(1 << 7)	/* Allow negative quota usage */
 
 static inline unsigned int dquot_state_flag(unsigned int flags, int type)
 {
-- 
cgit v1.2.3


From 4d59bce4f9eaf26d6d9046b56a2f1c0c7f20981d Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Thu, 2 Oct 2008 16:48:10 +0200
Subject: quota: Keep which entries were set by SETQUOTA quotactl

Quota in a clustered environment needs to synchronize quota information
among cluster nodes. This means we have to occasionally update some
information in dquot from disk / network. On the other hand we have to
be careful not to overwrite changes administrator did via SETQUOTA.
So indicate in dquot->dq_flags which entries have been set by SETQUOTA
and quota format can clear these flags when it properly propagated
the changes.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/dquot.c            | 12 ++++++++++--
 include/linux/quota.h | 26 ++++++++++++++++++++------
 2 files changed, 30 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/fs/dquot.c b/fs/dquot.c
index 9c78ffe1aad2..89226726daa8 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -2010,25 +2010,33 @@ static int do_set_dqblk(struct dquot *dquot, struct if_dqblk *di)
 	if (di->dqb_valid & QIF_SPACE) {
 		dm->dqb_curspace = di->dqb_curspace;
 		check_blim = 1;
+		__set_bit(DQ_LASTSET_B + QIF_SPACE_B, &dquot->dq_flags);
 	}
 	if (di->dqb_valid & QIF_BLIMITS) {
 		dm->dqb_bsoftlimit = qbtos(di->dqb_bsoftlimit);
 		dm->dqb_bhardlimit = qbtos(di->dqb_bhardlimit);
 		check_blim = 1;
+		__set_bit(DQ_LASTSET_B + QIF_BLIMITS_B, &dquot->dq_flags);
 	}
 	if (di->dqb_valid & QIF_INODES) {
 		dm->dqb_curinodes = di->dqb_curinodes;
 		check_ilim = 1;
+		__set_bit(DQ_LASTSET_B + QIF_INODES_B, &dquot->dq_flags);
 	}
 	if (di->dqb_valid & QIF_ILIMITS) {
 		dm->dqb_isoftlimit = di->dqb_isoftlimit;
 		dm->dqb_ihardlimit = di->dqb_ihardlimit;
 		check_ilim = 1;
+		__set_bit(DQ_LASTSET_B + QIF_ILIMITS_B, &dquot->dq_flags);
 	}
-	if (di->dqb_valid & QIF_BTIME)
+	if (di->dqb_valid & QIF_BTIME) {
 		dm->dqb_btime = di->dqb_btime;
-	if (di->dqb_valid & QIF_ITIME)
+		__set_bit(DQ_LASTSET_B + QIF_BTIME_B, &dquot->dq_flags);
+	}
+	if (di->dqb_valid & QIF_ITIME) {
 		dm->dqb_itime = di->dqb_itime;
+		__set_bit(DQ_LASTSET_B + QIF_ITIME_B, &dquot->dq_flags);
+	}
 
 	if (check_blim) {
 		if (!dm->dqb_bsoftlimit || dm->dqb_curspace < dm->dqb_bsoftlimit) {
diff --git a/include/linux/quota.h b/include/linux/quota.h
index 75bf761caef2..6d98885c16da 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -80,12 +80,21 @@
  * Quota structure used for communication with userspace via quotactl
  * Following flags are used to specify which fields are valid
  */
-#define QIF_BLIMITS	1
-#define QIF_SPACE	2
-#define QIF_ILIMITS	4
-#define QIF_INODES	8
-#define QIF_BTIME	16
-#define QIF_ITIME	32
+enum {
+	QIF_BLIMITS_B = 0,
+	QIF_SPACE_B,
+	QIF_ILIMITS_B,
+	QIF_INODES_B,
+	QIF_BTIME_B,
+	QIF_ITIME_B,
+};
+
+#define QIF_BLIMITS	(1 << QIF_BLIMITS_B)
+#define QIF_SPACE	(1 << QIF_SPACE_B)
+#define QIF_ILIMITS	(1 << QIF_ILIMITS_B)
+#define QIF_INODES	(1 << QIF_INODES_B)
+#define QIF_BTIME	(1 << QIF_BTIME_B)
+#define QIF_ITIME	(1 << QIF_ITIME_B)
 #define QIF_LIMITS	(QIF_BLIMITS | QIF_ILIMITS)
 #define QIF_USAGE	(QIF_SPACE | QIF_INODES)
 #define QIF_TIMES	(QIF_BTIME | QIF_ITIME)
@@ -242,6 +251,11 @@ extern struct dqstats dqstats;
 #define DQ_FAKE_B	3	/* no limits only usage */
 #define DQ_READ_B	4	/* dquot was read into memory */
 #define DQ_ACTIVE_B	5	/* dquot is active (dquot_release not called) */
+#define DQ_LASTSET_B	6	/* Following 6 bits (see QIF_) are reserved\
+				 * for the mask of entries set via SETQUOTA\
+				 * quotactl. They are set under dq_data_lock\
+				 * and the quota format handling dquot can\
+				 * clear them when it sees fit. */
 
 struct dquot {
 	struct hlist_node dq_hash;	/* Hash list in memory */
-- 
cgit v1.2.3


From 571b46e40bebb0d57130ca24c4a84dfd553adb91 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Thu, 30 Oct 2008 09:17:52 +0100
Subject: quota: Update version number

Increase reported version number of quota support since quota core has changed
significantly. Also remove __DQUOT_NUM_VERSION__ since nobody uses it.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 include/linux/quota.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/quota.h b/include/linux/quota.h
index 6d98885c16da..ec82beb10424 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -36,8 +36,7 @@
 #include <linux/errno.h>
 #include <linux/types.h>
 
-#define __DQUOT_VERSION__	"dquot_6.5.1"
-#define __DQUOT_NUM_VERSION__	6*10000+5*100+1
+#define __DQUOT_VERSION__	"dquot_6.5.2"
 
 #define MAXQUOTAS 2
 #define USRQUOTA  0		/* element used for user quotas */
-- 
cgit v1.2.3


From 3d9ea253a0e73dccaa869888ec2ceb17ea76c810 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 10 Oct 2008 16:12:23 +0200
Subject: quota: Add helpers to allow ocfs2 specific quota initialization,
 freeing and recovery

OCFS2 needs to peek whether quota structure is already in memory so
that it can avoid expensive cluster locking in that case. Similarly
when freeing dquots, it checks whether it is the last quota structure
user or not. Finally, it needs to get reference to dquot structure for
specified id and quota type when recovering quota file after crash.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/dquot.c               | 38 ++++++++++++++++++++++++++++++++------
 include/linux/quotaops.h |  4 ++++
 2 files changed, 36 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/fs/dquot.c b/fs/dquot.c
index 89226726daa8..ae8fd9e645cc 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -211,8 +211,6 @@ static struct hlist_head *dquot_hash;
 
 struct dqstats dqstats;
 
-static void dqput(struct dquot *dquot);
-
 static inline unsigned int
 hashfn(const struct super_block *sb, unsigned int id, int type)
 {
@@ -568,7 +566,7 @@ static struct shrinker dqcache_shrinker = {
  * NOTE: If you change this function please check whether dqput_blocks() works right...
  * MUST be called with either dqptr_sem or dqonoff_mutex held
  */
-static void dqput(struct dquot *dquot)
+void dqput(struct dquot *dquot)
 {
 	int ret;
 
@@ -661,11 +659,29 @@ static struct dquot *get_empty_dquot(struct super_block *sb, int type)
 	return dquot;
 }
 
+/*
+ * Check whether dquot is in memory.
+ * MUST be called with either dqptr_sem or dqonoff_mutex held
+ */
+int dquot_is_cached(struct super_block *sb, unsigned int id, int type)
+{
+	unsigned int hashent = hashfn(sb, id, type);
+	int ret = 0;
+
+        if (!sb_has_quota_active(sb, type))
+		return 0;
+	spin_lock(&dq_list_lock);
+	if (find_dquot(hashent, sb, id, type) != NODQUOT)
+		ret = 1;
+	spin_unlock(&dq_list_lock);
+	return ret;
+}
+
 /*
  * Get reference to dquot
  * MUST be called with either dqptr_sem or dqonoff_mutex held
  */
-static struct dquot *dqget(struct super_block *sb, unsigned int id, int type)
+struct dquot *dqget(struct super_block *sb, unsigned int id, int type)
 {
 	unsigned int hashent = hashfn(sb, id, type);
 	struct dquot *dquot, *empty = NODQUOT;
@@ -1184,17 +1200,23 @@ out_err:
  * 	Release all quotas referenced by inode
  *	Transaction must be started at an entry
  */
-int dquot_drop(struct inode *inode)
+int dquot_drop_locked(struct inode *inode)
 {
 	int cnt;
 
-	down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		if (inode->i_dquot[cnt] != NODQUOT) {
 			dqput(inode->i_dquot[cnt]);
 			inode->i_dquot[cnt] = NODQUOT;
 		}
 	}
+	return 0;
+}
+
+int dquot_drop(struct inode *inode)
+{
+	down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	dquot_drop_locked(inode);
 	up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	return 0;
 }
@@ -2308,7 +2330,11 @@ EXPORT_SYMBOL(dquot_release);
 EXPORT_SYMBOL(dquot_mark_dquot_dirty);
 EXPORT_SYMBOL(dquot_initialize);
 EXPORT_SYMBOL(dquot_drop);
+EXPORT_SYMBOL(dquot_drop_locked);
 EXPORT_SYMBOL(vfs_dq_drop);
+EXPORT_SYMBOL(dqget);
+EXPORT_SYMBOL(dqput);
+EXPORT_SYMBOL(dquot_is_cached);
 EXPORT_SYMBOL(dquot_alloc_space);
 EXPORT_SYMBOL(dquot_alloc_inode);
 EXPORT_SYMBOL(dquot_free_space);
diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h
index e840ca523175..e3a10272d471 100644
--- a/include/linux/quotaops.h
+++ b/include/linux/quotaops.h
@@ -24,6 +24,10 @@ void sync_dquots(struct super_block *sb, int type);
 
 int dquot_initialize(struct inode *inode, int type);
 int dquot_drop(struct inode *inode);
+int dquot_drop_locked(struct inode *inode);
+struct dquot *dqget(struct super_block *sb, unsigned int id, int type);
+void dqput(struct dquot *dquot);
+int dquot_is_cached(struct super_block *sb, unsigned int id, int type);
 
 int dquot_alloc_space(struct inode *inode, qsize_t number, int prealloc);
 int dquot_alloc_inode(const struct inode *inode, qsize_t number);
-- 
cgit v1.2.3


From 12c77527e4138bc3b17d17b0e0c909e4fc84924f Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Mon, 20 Oct 2008 17:05:00 +0200
Subject: quota: Implement function for scanning active dquots

OCFS2 needs to scan all active dquots once in a while and sync quota
information among cluster nodes. Provide a helper function for it so
that it does not have to reimplement internally a list which VFS
already has. Moreover this function is probably going to be useful
for other clustered filesystems if they decide to use VFS quotas.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/dquot.c               | 36 ++++++++++++++++++++++++++++++++++++
 include/linux/quotaops.h |  3 +++
 2 files changed, 39 insertions(+)

(limited to 'include/linux')

diff --git a/fs/dquot.c b/fs/dquot.c
index ae8fd9e645cc..075dc76904e7 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -476,6 +476,41 @@ restart:
 	spin_unlock(&dq_list_lock);
 }
 
+/* Call callback for every active dquot on given filesystem */
+int dquot_scan_active(struct super_block *sb,
+		      int (*fn)(struct dquot *dquot, unsigned long priv),
+		      unsigned long priv)
+{
+	struct dquot *dquot, *old_dquot = NULL;
+	int ret = 0;
+
+	mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
+	spin_lock(&dq_list_lock);
+	list_for_each_entry(dquot, &inuse_list, dq_inuse) {
+		if (!test_bit(DQ_ACTIVE_B, &dquot->dq_flags))
+			continue;
+		if (dquot->dq_sb != sb)
+			continue;
+		/* Now we have active dquot so we can just increase use count */
+		atomic_inc(&dquot->dq_count);
+		dqstats.lookups++;
+		spin_unlock(&dq_list_lock);
+		dqput(old_dquot);
+		old_dquot = dquot;
+		ret = fn(dquot, priv);
+		if (ret < 0)
+			goto out;
+		spin_lock(&dq_list_lock);
+		/* We are safe to continue now because our dquot could not
+		 * be moved out of the inuse list while we hold the reference */
+	}
+	spin_unlock(&dq_list_lock);
+out:
+	dqput(old_dquot);
+	mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
+	return ret;
+}
+
 int vfs_quota_sync(struct super_block *sb, int type)
 {
 	struct list_head *dirty;
@@ -2318,6 +2353,7 @@ EXPORT_SYMBOL(vfs_quota_on_path);
 EXPORT_SYMBOL(vfs_quota_on_mount);
 EXPORT_SYMBOL(vfs_quota_disable);
 EXPORT_SYMBOL(vfs_quota_off);
+EXPORT_SYMBOL(dquot_scan_active);
 EXPORT_SYMBOL(vfs_quota_sync);
 EXPORT_SYMBOL(vfs_get_dqinfo);
 EXPORT_SYMBOL(vfs_set_dqinfo);
diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h
index e3a10272d471..f4913948c305 100644
--- a/include/linux/quotaops.h
+++ b/include/linux/quotaops.h
@@ -28,6 +28,9 @@ int dquot_drop_locked(struct inode *inode);
 struct dquot *dqget(struct super_block *sb, unsigned int id, int type);
 void dqput(struct dquot *dquot);
 int dquot_is_cached(struct super_block *sb, unsigned int id, int type);
+int dquot_scan_active(struct super_block *sb,
+		      int (*fn)(struct dquot *dquot, unsigned long priv),
+		      unsigned long priv);
 
 int dquot_alloc_space(struct inode *inode, qsize_t number, int prealloc);
 int dquot_alloc_inode(const struct inode *inode, qsize_t number);
-- 
cgit v1.2.3


From e97fcd95a4778a8caf1980c6c72fdf68185a0838 Mon Sep 17 00:00:00 2001
From: Mark Fasheh <mfasheh@suse.com>
Date: Tue, 18 Nov 2008 17:15:24 -0800
Subject: jbd2: Add BH_JBDPrivateStart

Add this so that file systems using JBD2 can safely allocate unused b_state
bits.

In this case, we add it so that Ocfs2 can define a single bit for tracking
the validation state of a buffer.

Acked-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 include/linux/jbd2.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index c7d106ef22e2..f36645745489 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -329,6 +329,7 @@ enum jbd_state_bits {
 	BH_State,		/* Pins most journal_head state */
 	BH_JournalHead,		/* Pins bh->b_private and jh->b_bh */
 	BH_Unshadow,		/* Dummy bit, for BJ_Shadow wakeup filtering */
+	BH_JBDPrivateStart,	/* First bit available for private use by FS */
 };
 
 BUFFER_FNS(JBD, jbd)
-- 
cgit v1.2.3


From 5cd9d5bb86daf632a40f90e2321ea9379e42f073 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Tue, 25 Nov 2008 15:31:31 +0100
Subject: quota: Unexport dqblk_v1.h and dqblk_v2.h

Unexport header files dqblk_v[12].h since except for quota format ID they
don't contain information userspace should be interested in. Move ID
definitions to quota.h.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 include/linux/Kbuild     | 2 --
 include/linux/dqblk_v1.h | 3 ---
 include/linux/dqblk_v2.h | 3 ---
 include/linux/quota.h    | 4 ++++
 4 files changed, 4 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 900a787cbae9..39da666067b9 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -56,8 +56,6 @@ header-y += dlm_device.h
 header-y += dlm_netlink.h
 header-y += dm-ioctl.h
 header-y += dn.h
-header-y += dqblk_v1.h
-header-y += dqblk_v2.h
 header-y += dqblk_xfs.h
 header-y += efs_fs_sb.h
 header-y += elf-fdpic.h
diff --git a/include/linux/dqblk_v1.h b/include/linux/dqblk_v1.h
index 9cea901f5bba..3713a7232dd8 100644
--- a/include/linux/dqblk_v1.h
+++ b/include/linux/dqblk_v1.h
@@ -5,9 +5,6 @@
 #ifndef _LINUX_DQBLK_V1_H
 #define _LINUX_DQBLK_V1_H
 
-/* Id of quota format */
-#define QFMT_VFS_OLD 1
-
 /* Root squash turned on */
 #define V1_DQF_RSQUASH 1
 
diff --git a/include/linux/dqblk_v2.h b/include/linux/dqblk_v2.h
index ff8af1b4bda7..18000a542677 100644
--- a/include/linux/dqblk_v2.h
+++ b/include/linux/dqblk_v2.h
@@ -7,9 +7,6 @@
 
 #include <linux/dqblk_qtree.h>
 
-/* Id number of quota format */
-#define QFMT_VFS_V0 2
-
 /* Numbers of blocks needed for updates */
 #define V2_INIT_ALLOC QTREE_INIT_ALLOC
 #define V2_INIT_REWRITE QTREE_INIT_REWRITE
diff --git a/include/linux/quota.h b/include/linux/quota.h
index ec82beb10424..d72d5d84fde5 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -70,6 +70,10 @@
 #define Q_GETQUOTA 0x800007	/* get user quota structure */
 #define Q_SETQUOTA 0x800008	/* set user quota structure */
 
+/* Quota format type IDs */
+#define	QFMT_VFS_OLD 1
+#define	QFMT_VFS_V0 2
+
 /* Size of block in which space limits are passed through the quota
  * interface */
 #define QIF_DQBLKSIZE_BITS 10
-- 
cgit v1.2.3


From 7d9056ba20ebed6e3937a2e23183f6117919cb00 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Tue, 25 Nov 2008 15:31:32 +0100
Subject: quota: Export dquot_alloc() and dquot_destroy() functions

These are default functions for creating and destroying quota structures
and they should be used from filesystems.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/dquot.c               | 6 ++++--
 include/linux/quotaops.h | 2 ++
 2 files changed, 6 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/fs/dquot.c b/fs/dquot.c
index 075dc76904e7..61bfff64e5af 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -413,10 +413,11 @@ out_dqlock:
 	return ret;
 }
 
-static void dquot_destroy(struct dquot *dquot)
+void dquot_destroy(struct dquot *dquot)
 {
 	kmem_cache_free(dquot_cachep, dquot);
 }
+EXPORT_SYMBOL(dquot_destroy);
 
 static inline void do_destroy_dquot(struct dquot *dquot)
 {
@@ -668,10 +669,11 @@ we_slept:
 	spin_unlock(&dq_list_lock);
 }
 
-static struct dquot *dquot_alloc(struct super_block *sb, int type)
+struct dquot *dquot_alloc(struct super_block *sb, int type)
 {
 	return kmem_cache_zalloc(dquot_cachep, GFP_NOFS);
 }
+EXPORT_SYMBOL(dquot_alloc);
 
 static struct dquot *get_empty_dquot(struct super_block *sb, int type)
 {
diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h
index f4913948c305..21b781a3350f 100644
--- a/include/linux/quotaops.h
+++ b/include/linux/quotaops.h
@@ -31,6 +31,8 @@ int dquot_is_cached(struct super_block *sb, unsigned int id, int type);
 int dquot_scan_active(struct super_block *sb,
 		      int (*fn)(struct dquot *dquot, unsigned long priv),
 		      unsigned long priv);
+struct dquot *dquot_alloc(struct super_block *sb, int type);
+void dquot_destroy(struct dquot *dquot);
 
 int dquot_alloc_space(struct inode *inode, qsize_t number, int prealloc);
 int dquot_alloc_inode(const struct inode *inode, qsize_t number);
-- 
cgit v1.2.3


From e06c8227fd94ec181849ba206bf032be31c4295c Mon Sep 17 00:00:00 2001
From: Joel Becker <joel.becker@oracle.com>
Date: Thu, 11 Sep 2008 15:35:47 -0700
Subject: jbd2: Add buffer triggers

Filesystems often to do compute intensive operation on some
metadata.  If this operation is repeated many times, it can be very
expensive.  It would be much nicer if the operation could be performed
once before a buffer goes to disk.

This adds triggers to jbd2 buffer heads.  Just before writing a metadata
buffer to the journal, jbd2 will optionally call a commit trigger associated
with the buffer.  If the journal is aborted, an abort trigger will be
called on any dirty buffers as they are dropped from pending
transactions.

ocfs2 will use this feature.

Initially I tried to come up with a more generic trigger that could be
used for non-buffer-related events like transaction completion.  It
doesn't tie nicely, because the information a buffer trigger needs
(specific to a journal_head) isn't the same as what a transaction
trigger needs (specific to a tranaction_t or perhaps journal_t).  So I
implemented a buffer set, with the understanding that
journal/transaction wide triggers should be implemented separately.

There is only one trigger set allowed per buffer.  I can't think of any
reason to attach more than one set.  Contrast this with a journal or
transaction in which multiple places may want to watch the entire
transaction separately.

The trigger sets are considered static allocation from the jbd2
perspective.  ocfs2 will just have one trigger set per block type,
setting the same set on every bh of the same type.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/jbd2/commit.c             |  9 +++++++++
 fs/jbd2/journal.c            | 19 ++++++++++++++++++
 fs/jbd2/transaction.c        | 47 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/jbd2.h         | 31 +++++++++++++++++++++++++++++
 include/linux/journal-head.h |  8 ++++++++
 5 files changed, 114 insertions(+)

(limited to 'include/linux')

diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index ebc667bc54a8..c8a1bace685a 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -509,6 +509,10 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 		if (is_journal_aborted(journal)) {
 			clear_buffer_jbddirty(jh2bh(jh));
 			JBUFFER_TRACE(jh, "journal is aborting: refile");
+			jbd2_buffer_abort_trigger(jh,
+						  jh->b_frozen_data ?
+						  jh->b_frozen_triggers :
+						  jh->b_triggers);
 			jbd2_journal_refile_buffer(journal, jh);
 			/* If that was the last one, we need to clean up
 			 * any descriptor buffers which may have been
@@ -844,6 +848,9 @@ restart_loop:
 		 * data.
 		 *
 		 * Otherwise, we can just throw away the frozen data now.
+		 *
+		 * We also know that the frozen data has already fired
+		 * its triggers if they exist, so we can clear that too.
 		 */
 		if (jh->b_committed_data) {
 			jbd2_free(jh->b_committed_data, bh->b_size);
@@ -851,10 +858,12 @@ restart_loop:
 			if (jh->b_frozen_data) {
 				jh->b_committed_data = jh->b_frozen_data;
 				jh->b_frozen_data = NULL;
+				jh->b_frozen_triggers = NULL;
 			}
 		} else if (jh->b_frozen_data) {
 			jbd2_free(jh->b_frozen_data, bh->b_size);
 			jh->b_frozen_data = NULL;
+			jh->b_frozen_triggers = NULL;
 		}
 
 		spin_lock(&journal->j_list_lock);
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index e70d657a19f8..f6bff9d6f8df 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -50,6 +50,7 @@ EXPORT_SYMBOL(jbd2_journal_unlock_updates);
 EXPORT_SYMBOL(jbd2_journal_get_write_access);
 EXPORT_SYMBOL(jbd2_journal_get_create_access);
 EXPORT_SYMBOL(jbd2_journal_get_undo_access);
+EXPORT_SYMBOL(jbd2_journal_set_triggers);
 EXPORT_SYMBOL(jbd2_journal_dirty_metadata);
 EXPORT_SYMBOL(jbd2_journal_release_buffer);
 EXPORT_SYMBOL(jbd2_journal_forget);
@@ -290,6 +291,7 @@ int jbd2_journal_write_metadata_buffer(transaction_t *transaction,
 	struct page *new_page;
 	unsigned int new_offset;
 	struct buffer_head *bh_in = jh2bh(jh_in);
+	struct jbd2_buffer_trigger_type *triggers;
 
 	/*
 	 * The buffer really shouldn't be locked: only the current committing
@@ -314,12 +316,22 @@ repeat:
 		done_copy_out = 1;
 		new_page = virt_to_page(jh_in->b_frozen_data);
 		new_offset = offset_in_page(jh_in->b_frozen_data);
+		triggers = jh_in->b_frozen_triggers;
 	} else {
 		new_page = jh2bh(jh_in)->b_page;
 		new_offset = offset_in_page(jh2bh(jh_in)->b_data);
+		triggers = jh_in->b_triggers;
 	}
 
 	mapped_data = kmap_atomic(new_page, KM_USER0);
+	/*
+	 * Fire any commit trigger.  Do this before checking for escaping,
+	 * as the trigger may modify the magic offset.  If a copy-out
+	 * happens afterwards, it will have the correct data in the buffer.
+	 */
+	jbd2_buffer_commit_trigger(jh_in, mapped_data + new_offset,
+				   triggers);
+
 	/*
 	 * Check for escaping
 	 */
@@ -352,6 +364,13 @@ repeat:
 		new_page = virt_to_page(tmp);
 		new_offset = offset_in_page(tmp);
 		done_copy_out = 1;
+
+		/*
+		 * This isn't strictly necessary, as we're using frozen
+		 * data for the escaping, but it keeps consistency with
+		 * b_frozen_data usage.
+		 */
+		jh_in->b_frozen_triggers = jh_in->b_triggers;
 	}
 
 	/*
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index 39b7805a599a..4f925a4f3d05 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -741,6 +741,12 @@ done:
 		source = kmap_atomic(page, KM_USER0);
 		memcpy(jh->b_frozen_data, source+offset, jh2bh(jh)->b_size);
 		kunmap_atomic(source, KM_USER0);
+
+		/*
+		 * Now that the frozen data is saved off, we need to store
+		 * any matching triggers.
+		 */
+		jh->b_frozen_triggers = jh->b_triggers;
 	}
 	jbd_unlock_bh_state(bh);
 
@@ -943,6 +949,47 @@ out:
 	return err;
 }
 
+/**
+ * void jbd2_journal_set_triggers() - Add triggers for commit writeout
+ * @bh: buffer to trigger on
+ * @type: struct jbd2_buffer_trigger_type containing the trigger(s).
+ *
+ * Set any triggers on this journal_head.  This is always safe, because
+ * triggers for a committing buffer will be saved off, and triggers for
+ * a running transaction will match the buffer in that transaction.
+ *
+ * Call with NULL to clear the triggers.
+ */
+void jbd2_journal_set_triggers(struct buffer_head *bh,
+			       struct jbd2_buffer_trigger_type *type)
+{
+	struct journal_head *jh = bh2jh(bh);
+
+	jh->b_triggers = type;
+}
+
+void jbd2_buffer_commit_trigger(struct journal_head *jh, void *mapped_data,
+				struct jbd2_buffer_trigger_type *triggers)
+{
+	struct buffer_head *bh = jh2bh(jh);
+
+	if (!triggers || !triggers->t_commit)
+		return;
+
+	triggers->t_commit(triggers, bh, mapped_data, bh->b_size);
+}
+
+void jbd2_buffer_abort_trigger(struct journal_head *jh,
+			       struct jbd2_buffer_trigger_type *triggers)
+{
+	if (!triggers || !triggers->t_abort)
+		return;
+
+	triggers->t_abort(triggers, jh2bh(jh));
+}
+
+
+
 /**
  * int jbd2_journal_dirty_metadata() -  mark a buffer as containing dirty metadata
  * @handle: transaction to add buffer to.
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index f36645745489..34456476e761 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -1008,6 +1008,35 @@ int __jbd2_journal_clean_checkpoint_list(journal_t *journal);
 int __jbd2_journal_remove_checkpoint(struct journal_head *);
 void __jbd2_journal_insert_checkpoint(struct journal_head *, transaction_t *);
 
+
+/*
+ * Triggers
+ */
+
+struct jbd2_buffer_trigger_type {
+	/*
+	 * Fired just before a buffer is written to the journal.
+	 * mapped_data is a mapped buffer that is the frozen data for
+	 * commit.
+	 */
+	void (*t_commit)(struct jbd2_buffer_trigger_type *type,
+			 struct buffer_head *bh, void *mapped_data,
+			 size_t size);
+
+	/*
+	 * Fired during journal abort for dirty buffers that will not be
+	 * committed.
+	 */
+	void (*t_abort)(struct jbd2_buffer_trigger_type *type,
+			struct buffer_head *bh);
+};
+
+extern void jbd2_buffer_commit_trigger(struct journal_head *jh,
+				       void *mapped_data,
+				       struct jbd2_buffer_trigger_type *triggers);
+extern void jbd2_buffer_abort_trigger(struct journal_head *jh,
+				      struct jbd2_buffer_trigger_type *triggers);
+
 /* Buffer IO */
 extern int
 jbd2_journal_write_metadata_buffer(transaction_t	  *transaction,
@@ -1046,6 +1075,8 @@ extern int	 jbd2_journal_extend (handle_t *, int nblocks);
 extern int	 jbd2_journal_get_write_access(handle_t *, struct buffer_head *);
 extern int	 jbd2_journal_get_create_access (handle_t *, struct buffer_head *);
 extern int	 jbd2_journal_get_undo_access(handle_t *, struct buffer_head *);
+void		 jbd2_journal_set_triggers(struct buffer_head *,
+					   struct jbd2_buffer_trigger_type *type);
 extern int	 jbd2_journal_dirty_metadata (handle_t *, struct buffer_head *);
 extern void	 jbd2_journal_release_buffer (handle_t *, struct buffer_head *);
 extern int	 jbd2_journal_forget (handle_t *, struct buffer_head *);
diff --git a/include/linux/journal-head.h b/include/linux/journal-head.h
index bb70ebb6a2d5..525aac3c97df 100644
--- a/include/linux/journal-head.h
+++ b/include/linux/journal-head.h
@@ -12,6 +12,8 @@
 
 typedef unsigned int		tid_t;		/* Unique transaction ID */
 typedef struct transaction_s	transaction_t;	/* Compound transaction type */
+
+
 struct buffer_head;
 
 struct journal_head {
@@ -87,6 +89,12 @@ struct journal_head {
 	 * [j_list_lock]
 	 */
 	struct journal_head *b_cpnext, *b_cpprev;
+
+	/* Trigger type */
+	struct jbd2_buffer_trigger_type *b_triggers;
+
+	/* Trigger type for the committing transaction's frozen data */
+	struct jbd2_buffer_trigger_type *b_frozen_triggers;
 };
 
 #endif		/* JOURNAL_HEAD_H_INCLUDED */
-- 
cgit v1.2.3


From 4c728ef583b3d82266584da5cb068294c09df31e Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Mon, 22 Dec 2008 21:11:15 +0100
Subject: add a vfs_fsync helper

Fsync currently has a fdatawrite/fdatawait pair around the method call,
and a mutex_lock/unlock of the inode mutex.  All callers of fsync have
to duplicate this, but we have a few and most of them don't quite get
it right.  This patch adds a new vfs_fsync that takes care of this.
It's a little more complicated as usual as ->fsync might get a NULL file
pointer and just a dentry from nfsd, but otherwise gets afile and we
want to take the mapping and file operations from it when it is there.

Notes on the fsync callers:

 - ecryptfs wasn't calling filemap_fdatawrite / filemap_fdatawait on the
   	lower file
 - coda wasn't calling filemap_fdatawrite / filemap_fdatawait on the host
	file, and returning 0 when ->fsync was missing
 - shm wasn't calling either filemap_fdatawrite / filemap_fdatawait nor
   taking i_mutex.  Now given that shared memory doesn't have disk
   backing not doing anything in fsync seems fine and I left it out of
   the vfs_fsync conversion for now, but in that case we might just
   not pass it through to the lower file at all but just call the no-op
   simple_sync_file directly.

[and now actually export vfs_fsync]

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 drivers/usb/gadget/file_storage.c | 18 +--------------
 fs/coda/file.c                    | 12 ++--------
 fs/ecryptfs/file.c                | 15 +++---------
 fs/nfsd/vfs.c                     | 35 +++-------------------------
 fs/sync.c                         | 48 ++++++++++++++++++++++++++++++---------
 include/linux/fs.h                |  2 +-
 mm/msync.c                        |  2 +-
 7 files changed, 48 insertions(+), 84 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/gadget/file_storage.c b/drivers/usb/gadget/file_storage.c
index c4e62a6297d7..2e71368f45b4 100644
--- a/drivers/usb/gadget/file_storage.c
+++ b/drivers/usb/gadget/file_storage.c
@@ -1863,26 +1863,10 @@ static int do_write(struct fsg_dev *fsg)
 static int fsync_sub(struct lun *curlun)
 {
 	struct file	*filp = curlun->filp;
-	struct inode	*inode;
-	int		rc, err;
 
 	if (curlun->ro || !filp)
 		return 0;
-	if (!filp->f_op->fsync)
-		return -EINVAL;
-
-	inode = filp->f_path.dentry->d_inode;
-	mutex_lock(&inode->i_mutex);
-	rc = filemap_fdatawrite(inode->i_mapping);
-	err = filp->f_op->fsync(filp, filp->f_path.dentry, 1);
-	if (!rc)
-		rc = err;
-	err = filemap_fdatawait(inode->i_mapping);
-	if (!rc)
-		rc = err;
-	mutex_unlock(&inode->i_mutex);
-	VLDBG(curlun, "fdatasync -> %d\n", rc);
-	return rc;
+	return vfs_fsync(filp, filp->f_path.dentry, 1);
 }
 
 static void fsync_all(struct fsg_dev *fsg)
diff --git a/fs/coda/file.c b/fs/coda/file.c
index 466303db2df6..6a347fbc998a 100644
--- a/fs/coda/file.c
+++ b/fs/coda/file.c
@@ -201,8 +201,7 @@ int coda_release(struct inode *coda_inode, struct file *coda_file)
 int coda_fsync(struct file *coda_file, struct dentry *coda_dentry, int datasync)
 {
 	struct file *host_file;
-	struct dentry *host_dentry;
-	struct inode *host_inode, *coda_inode = coda_dentry->d_inode;
+	struct inode *coda_inode = coda_dentry->d_inode;
 	struct coda_file_info *cfi;
 	int err = 0;
 
@@ -214,14 +213,7 @@ int coda_fsync(struct file *coda_file, struct dentry *coda_dentry, int datasync)
 	BUG_ON(!cfi || cfi->cfi_magic != CODA_MAGIC);
 	host_file = cfi->cfi_container;
 
-	if (host_file->f_op && host_file->f_op->fsync) {
-		host_dentry = host_file->f_path.dentry;
-		host_inode = host_dentry->d_inode;
-		mutex_lock(&host_inode->i_mutex);
-		err = host_file->f_op->fsync(host_file, host_dentry, datasync);
-		mutex_unlock(&host_inode->i_mutex);
-	}
-
+	err = vfs_fsync(host_file, host_file->f_path.dentry, datasync);
 	if ( !err && !datasync ) {
 		lock_kernel();
 		err = venus_fsync(coda_inode->i_sb, coda_i2f(coda_inode));
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index eb3dc4c7ac06..713834371229 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -275,18 +275,9 @@ static int ecryptfs_release(struct inode *inode, struct file *file)
 static int
 ecryptfs_fsync(struct file *file, struct dentry *dentry, int datasync)
 {
-	struct file *lower_file = ecryptfs_file_to_lower(file);
-	struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry);
-	struct inode *lower_inode = lower_dentry->d_inode;
-	int rc = -EINVAL;
-
-	if (lower_inode->i_fop->fsync) {
-		mutex_lock(&lower_inode->i_mutex);
-		rc = lower_inode->i_fop->fsync(lower_file, lower_dentry,
-					       datasync);
-		mutex_unlock(&lower_inode->i_mutex);
-	}
-	return rc;
+	return vfs_fsync(ecryptfs_file_to_lower(file),
+			 ecryptfs_dentry_to_lower(dentry),
+			 datasync);
 }
 
 static int ecryptfs_fasync(int fd, struct file *file, int flag)
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 5245a3965004..44aa92aba891 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -744,45 +744,16 @@ nfsd_close(struct file *filp)
 	fput(filp);
 }
 
-/*
- * Sync a file
- * As this calls fsync (not fdatasync) there is no need for a write_inode
- * after it.
- */
-static inline int nfsd_dosync(struct file *filp, struct dentry *dp,
-			      const struct file_operations *fop)
-{
-	struct inode *inode = dp->d_inode;
-	int (*fsync) (struct file *, struct dentry *, int);
-	int err;
-
-	err = filemap_fdatawrite(inode->i_mapping);
-	if (err == 0 && fop && (fsync = fop->fsync))
-		err = fsync(filp, dp, 0);
-	if (err == 0)
-		err = filemap_fdatawait(inode->i_mapping);
-
-	return err;
-}
-	
-
 static int
 nfsd_sync(struct file *filp)
 {
-        int err;
-	struct inode *inode = filp->f_path.dentry->d_inode;
-	dprintk("nfsd: sync file %s\n", filp->f_path.dentry->d_name.name);
-	mutex_lock(&inode->i_mutex);
-	err=nfsd_dosync(filp, filp->f_path.dentry, filp->f_op);
-	mutex_unlock(&inode->i_mutex);
-
-	return err;
+	return vfs_fsync(filp, filp->f_path.dentry, 0);
 }
 
 int
-nfsd_sync_dir(struct dentry *dp)
+nfsd_sync_dir(struct dentry *dentry)
 {
-	return nfsd_dosync(NULL, dp, dp->d_inode->i_fop);
+	return vfs_fsync(NULL, dentry, 0);
 }
 
 /*
diff --git a/fs/sync.c b/fs/sync.c
index 2967562d416f..0921d6d4b5e6 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -75,14 +75,39 @@ int file_fsync(struct file *filp, struct dentry *dentry, int datasync)
 	return ret;
 }
 
-long do_fsync(struct file *file, int datasync)
+/**
+ * vfs_fsync - perform a fsync or fdatasync on a file
+ * @file:		file to sync
+ * @dentry:		dentry of @file
+ * @data:		only perform a fdatasync operation
+ *
+ * Write back data and metadata for @file to disk.  If @datasync is
+ * set only metadata needed to access modified file data is written.
+ *
+ * In case this function is called from nfsd @file may be %NULL and
+ * only @dentry is set.  This can only happen when the filesystem
+ * implements the export_operations API.
+ */
+int vfs_fsync(struct file *file, struct dentry *dentry, int datasync)
 {
-	int ret;
-	int err;
-	struct address_space *mapping = file->f_mapping;
+	const struct file_operations *fop;
+	struct address_space *mapping;
+	int err, ret;
+
+	/*
+	 * Get mapping and operations from the file in case we have
+	 * as file, or get the default values for them in case we
+	 * don't have a struct file available.  Damn nfsd..
+	 */
+	if (file) {
+		mapping = file->f_mapping;
+		fop = file->f_op;
+	} else {
+		mapping = dentry->d_inode->i_mapping;
+		fop = dentry->d_inode->i_fop;
+	}
 
-	if (!file->f_op || !file->f_op->fsync) {
-		/* Why?  We can still call filemap_fdatawrite */
+	if (!fop || !fop->fsync) {
 		ret = -EINVAL;
 		goto out;
 	}
@@ -94,7 +119,7 @@ long do_fsync(struct file *file, int datasync)
 	 * livelocks in fsync_buffers_list().
 	 */
 	mutex_lock(&mapping->host->i_mutex);
-	err = file->f_op->fsync(file, file->f_path.dentry, datasync);
+	err = fop->fsync(file, dentry, datasync);
 	if (!ret)
 		ret = err;
 	mutex_unlock(&mapping->host->i_mutex);
@@ -104,15 +129,16 @@ long do_fsync(struct file *file, int datasync)
 out:
 	return ret;
 }
+EXPORT_SYMBOL(vfs_fsync);
 
-static long __do_fsync(unsigned int fd, int datasync)
+static int do_fsync(unsigned int fd, int datasync)
 {
 	struct file *file;
 	int ret = -EBADF;
 
 	file = fget(fd);
 	if (file) {
-		ret = do_fsync(file, datasync);
+		ret = vfs_fsync(file, file->f_path.dentry, datasync);
 		fput(file);
 	}
 	return ret;
@@ -120,12 +146,12 @@ static long __do_fsync(unsigned int fd, int datasync)
 
 asmlinkage long sys_fsync(unsigned int fd)
 {
-	return __do_fsync(fd, 0);
+	return do_fsync(fd, 0);
 }
 
 asmlinkage long sys_fdatasync(unsigned int fd)
 {
-	return __do_fsync(fd, 1);
+	return do_fsync(fd, 1);
 }
 
 /*
diff --git a/include/linux/fs.h b/include/linux/fs.h
index e2170ee21e18..9ad9eac9eb0c 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1827,7 +1827,7 @@ extern int __filemap_fdatawrite_range(struct address_space *mapping,
 extern int filemap_fdatawrite_range(struct address_space *mapping,
 				loff_t start, loff_t end);
 
-extern long do_fsync(struct file *file, int datasync);
+extern int vfs_fsync(struct file *file, struct dentry *dentry, int datasync);
 extern void sync_supers(void);
 extern void sync_filesystems(int wait);
 extern void __fsync_super(struct super_block *sb);
diff --git a/mm/msync.c b/mm/msync.c
index 144a7570535d..07dae08cf31c 100644
--- a/mm/msync.c
+++ b/mm/msync.c
@@ -82,7 +82,7 @@ asmlinkage long sys_msync(unsigned long start, size_t len, int flags)
 				(vma->vm_flags & VM_SHARED)) {
 			get_file(file);
 			up_read(&mm->mmap_sem);
-			error = do_fsync(file, 0);
+			error = vfs_fsync(file, file->f_path.dentry, 0);
 			fput(file);
 			if (error || start >= end)
 				goto out;
-- 
cgit v1.2.3


From 4ae8978cf92a96257cd8998a49e781be83571d64 Mon Sep 17 00:00:00 2001
From: Michael Kerrisk <mtk.manpages@googlemail.com>
Date: Mon, 5 Jan 2009 07:19:16 -0500
Subject: inotify: fix type errors in interfaces

The problems lie in the types used for some inotify interfaces, both at the kernel level and at the glibc level. This mail addresses the kernel problem. I will follow up with some suggestions for glibc changes.

For the sys_inotify_rm_watch() interface, the type of the 'wd' argument is
currently 'u32', it should be '__s32' .  That is Robert's suggestion, and
is consistent with the other declarations of watch descriptors in the
kernel source, in particular, the inotify_event structure in
include/linux/inotify.h:

struct inotify_event {
        __s32           wd;             /* watch descriptor */
        __u32           mask;           /* watch mask */
        __u32           cookie;         /* cookie to synchronize two events */
        __u32           len;            /* length (including nulls) of name */
        char            name[0];        /* stub for possible name */
};

The patch makes the changes needed for inotify_rm_watch().

Signed-off-by: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: Robert Love <rlove@google.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/notify/inotify/inotify_user.c | 2 +-
 include/linux/syscalls.h         | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index 400f8064a548..81b8644b0136 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -704,7 +704,7 @@ fput_and_out:
 	return ret;
 }
 
-asmlinkage long sys_inotify_rm_watch(int fd, u32 wd)
+asmlinkage long sys_inotify_rm_watch(int fd, __s32 wd)
 {
 	struct file *filp;
 	struct inotify_device *dev;
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 04fb47bfb920..18d0a243a7b3 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -549,7 +549,7 @@ asmlinkage long sys_inotify_init(void);
 asmlinkage long sys_inotify_init1(int flags);
 asmlinkage long sys_inotify_add_watch(int fd, const char __user *path,
 					u32 mask);
-asmlinkage long sys_inotify_rm_watch(int fd, u32 wd);
+asmlinkage long sys_inotify_rm_watch(int fd, __s32 wd);
 
 asmlinkage long sys_spu_run(int fd, __u32 __user *unpc,
 				 __u32 __user *ustatus);
-- 
cgit v1.2.3


From 07f2211e4fbce6990722d78c4f04225da9c0e9cf Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Mon, 5 Jan 2009 17:14:31 -0700
Subject: dmaengine: remove dependency on async_tx

async_tx.ko is a consumer of dma channels.  A circular dependency arises
if modules in drivers/dma rely on common code in async_tx.ko.  It
prevents either module from being unloaded.

Move dma_wait_for_async_tx and async_tx_run_dependencies to dmaeninge.o
where they should have been from the beginning.

Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 crypto/async_tx/async_tx.c | 75 -----------------------------------------
 drivers/dma/Kconfig        |  2 --
 drivers/dma/dmaengine.c    | 84 ++++++++++++++++++++++++++++++++++++++++++++++
 drivers/dma/iop-adma.c     |  3 +-
 drivers/dma/mv_xor.c       |  3 +-
 include/linux/async_tx.h   | 15 ---------
 include/linux/dmaengine.h  |  9 +++++
 7 files changed, 95 insertions(+), 96 deletions(-)

(limited to 'include/linux')

diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c
index dcbf1be149f3..8cfac182165d 100644
--- a/crypto/async_tx/async_tx.c
+++ b/crypto/async_tx/async_tx.c
@@ -72,81 +72,6 @@ void async_tx_issue_pending_all(void)
 }
 EXPORT_SYMBOL_GPL(async_tx_issue_pending_all);
 
-/* dma_wait_for_async_tx - spin wait for a transcation to complete
- * @tx: transaction to wait on
- */
-enum dma_status
-dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx)
-{
-	enum dma_status status;
-	struct dma_async_tx_descriptor *iter;
-	struct dma_async_tx_descriptor *parent;
-
-	if (!tx)
-		return DMA_SUCCESS;
-
-	/* poll through the dependency chain, return when tx is complete */
-	do {
-		iter = tx;
-
-		/* find the root of the unsubmitted dependency chain */
-		do {
-			parent = iter->parent;
-			if (!parent)
-				break;
-			else
-				iter = parent;
-		} while (parent);
-
-		/* there is a small window for ->parent == NULL and
-		 * ->cookie == -EBUSY
-		 */
-		while (iter->cookie == -EBUSY)
-			cpu_relax();
-
-		status = dma_sync_wait(iter->chan, iter->cookie);
-	} while (status == DMA_IN_PROGRESS || (iter != tx));
-
-	return status;
-}
-EXPORT_SYMBOL_GPL(dma_wait_for_async_tx);
-
-/* async_tx_run_dependencies - helper routine for dma drivers to process
- *	(start) dependent operations on their target channel
- * @tx: transaction with dependencies
- */
-void async_tx_run_dependencies(struct dma_async_tx_descriptor *tx)
-{
-	struct dma_async_tx_descriptor *dep = tx->next;
-	struct dma_async_tx_descriptor *dep_next;
-	struct dma_chan *chan;
-
-	if (!dep)
-		return;
-
-	chan = dep->chan;
-
-	/* keep submitting up until a channel switch is detected
-	 * in that case we will be called again as a result of
-	 * processing the interrupt from async_tx_channel_switch
-	 */
-	for (; dep; dep = dep_next) {
-		spin_lock_bh(&dep->lock);
-		dep->parent = NULL;
-		dep_next = dep->next;
-		if (dep_next && dep_next->chan == chan)
-			dep->next = NULL; /* ->next will be submitted */
-		else
-			dep_next = NULL; /* submit current dep and terminate */
-		spin_unlock_bh(&dep->lock);
-
-		dep->tx_submit(dep);
-	}
-
-	chan->device->device_issue_pending(chan);
-}
-EXPORT_SYMBOL_GPL(async_tx_run_dependencies);
-
 static void
 free_dma_chan_ref(struct rcu_head *rcu)
 {
diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index 904e57558bb5..e34b06420816 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -33,7 +33,6 @@ config INTEL_IOATDMA
 config INTEL_IOP_ADMA
 	tristate "Intel IOP ADMA support"
 	depends on ARCH_IOP32X || ARCH_IOP33X || ARCH_IOP13XX
-	select ASYNC_CORE
 	select DMA_ENGINE
 	help
 	  Enable support for the Intel(R) IOP Series RAID engines.
@@ -59,7 +58,6 @@ config FSL_DMA
 config MV_XOR
 	bool "Marvell XOR engine support"
 	depends on PLAT_ORION
-	select ASYNC_CORE
 	select DMA_ENGINE
 	---help---
 	  Enable support for the Marvell XOR engine.
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 657996517374..b9008932a8f3 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -626,6 +626,90 @@ void dma_async_tx_descriptor_init(struct dma_async_tx_descriptor *tx,
 }
 EXPORT_SYMBOL(dma_async_tx_descriptor_init);
 
+/* dma_wait_for_async_tx - spin wait for a transaction to complete
+ * @tx: in-flight transaction to wait on
+ *
+ * This routine assumes that tx was obtained from a call to async_memcpy,
+ * async_xor, async_memset, etc which ensures that tx is "in-flight" (prepped
+ * and submitted).  Walking the parent chain is only meant to cover for DMA
+ * drivers that do not implement the DMA_INTERRUPT capability and may race with
+ * the driver's descriptor cleanup routine.
+ */
+enum dma_status
+dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx)
+{
+	enum dma_status status;
+	struct dma_async_tx_descriptor *iter;
+	struct dma_async_tx_descriptor *parent;
+
+	if (!tx)
+		return DMA_SUCCESS;
+
+	WARN_ONCE(tx->parent, "%s: speculatively walking dependency chain for"
+		  " %s\n", __func__, dev_name(&tx->chan->dev));
+
+	/* poll through the dependency chain, return when tx is complete */
+	do {
+		iter = tx;
+
+		/* find the root of the unsubmitted dependency chain */
+		do {
+			parent = iter->parent;
+			if (!parent)
+				break;
+			else
+				iter = parent;
+		} while (parent);
+
+		/* there is a small window for ->parent == NULL and
+		 * ->cookie == -EBUSY
+		 */
+		while (iter->cookie == -EBUSY)
+			cpu_relax();
+
+		status = dma_sync_wait(iter->chan, iter->cookie);
+	} while (status == DMA_IN_PROGRESS || (iter != tx));
+
+	return status;
+}
+EXPORT_SYMBOL_GPL(dma_wait_for_async_tx);
+
+/* dma_run_dependencies - helper routine for dma drivers to process
+ *	(start) dependent operations on their target channel
+ * @tx: transaction with dependencies
+ */
+void dma_run_dependencies(struct dma_async_tx_descriptor *tx)
+{
+	struct dma_async_tx_descriptor *dep = tx->next;
+	struct dma_async_tx_descriptor *dep_next;
+	struct dma_chan *chan;
+
+	if (!dep)
+		return;
+
+	chan = dep->chan;
+
+	/* keep submitting up until a channel switch is detected
+	 * in that case we will be called again as a result of
+	 * processing the interrupt from async_tx_channel_switch
+	 */
+	for (; dep; dep = dep_next) {
+		spin_lock_bh(&dep->lock);
+		dep->parent = NULL;
+		dep_next = dep->next;
+		if (dep_next && dep_next->chan == chan)
+			dep->next = NULL; /* ->next will be submitted */
+		else
+			dep_next = NULL; /* submit current dep and terminate */
+		spin_unlock_bh(&dep->lock);
+
+		dep->tx_submit(dep);
+	}
+
+	chan->device->device_issue_pending(chan);
+}
+EXPORT_SYMBOL_GPL(dma_run_dependencies);
+
 static int __init dma_bus_init(void)
 {
 	mutex_init(&dma_list_mutex);
diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c
index 6be317262200..be9ea9f88805 100644
--- a/drivers/dma/iop-adma.c
+++ b/drivers/dma/iop-adma.c
@@ -24,7 +24,6 @@
 
 #include <linux/init.h>
 #include <linux/module.h>
-#include <linux/async_tx.h>
 #include <linux/delay.h>
 #include <linux/dma-mapping.h>
 #include <linux/spinlock.h>
@@ -116,7 +115,7 @@ iop_adma_run_tx_complete_actions(struct iop_adma_desc_slot *desc,
 	}
 
 	/* run dependent operations */
-	async_tx_run_dependencies(&desc->async_tx);
+	dma_run_dependencies(&desc->async_tx);
 
 	return cookie;
 }
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c
index bcda17426411..3f46df3390c7 100644
--- a/drivers/dma/mv_xor.c
+++ b/drivers/dma/mv_xor.c
@@ -18,7 +18,6 @@
 
 #include <linux/init.h>
 #include <linux/module.h>
-#include <linux/async_tx.h>
 #include <linux/delay.h>
 #include <linux/dma-mapping.h>
 #include <linux/spinlock.h>
@@ -340,7 +339,7 @@ mv_xor_run_tx_complete_actions(struct mv_xor_desc_slot *desc,
 	}
 
 	/* run dependent operations */
-	async_tx_run_dependencies(&desc->async_tx);
+	dma_run_dependencies(&desc->async_tx);
 
 	return cookie;
 }
diff --git a/include/linux/async_tx.h b/include/linux/async_tx.h
index 0f50d4cc4360..1c816775f135 100644
--- a/include/linux/async_tx.h
+++ b/include/linux/async_tx.h
@@ -60,8 +60,6 @@ enum async_tx_flags {
 
 #ifdef CONFIG_DMA_ENGINE
 void async_tx_issue_pending_all(void);
-enum dma_status dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx);
-void async_tx_run_dependencies(struct dma_async_tx_descriptor *tx);
 #ifdef CONFIG_ARCH_HAS_ASYNC_TX_FIND_CHANNEL
 #include <asm/async_tx.h>
 #else
@@ -77,19 +75,6 @@ static inline void async_tx_issue_pending_all(void)
 	do { } while (0);
 }
 
-static inline enum dma_status
-dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx)
-{
-	return DMA_SUCCESS;
-}
-
-static inline void
-async_tx_run_dependencies(struct dma_async_tx_descriptor *tx,
-	struct dma_chan *host_chan)
-{
-	do { } while (0);
-}
-
 static inline struct dma_chan *
 async_tx_find_channel(struct dma_async_tx_descriptor *depend_tx,
 	enum dma_transaction_type tx_type, struct page **dst, int dst_count,
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index adb0b084eb5a..e4ec7e7b8056 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -475,11 +475,20 @@ static inline enum dma_status dma_async_is_complete(dma_cookie_t cookie,
 }
 
 enum dma_status dma_sync_wait(struct dma_chan *chan, dma_cookie_t cookie);
+#ifdef CONFIG_DMA_ENGINE
+enum dma_status dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx);
+#else
+static inline enum dma_status dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx)
+{
+	return DMA_SUCCESS;
+}
+#endif
 
 /* --- DMA device --- */
 
 int dma_async_device_register(struct dma_device *device);
 void dma_async_device_unregister(struct dma_device *device);
+void dma_run_dependencies(struct dma_async_tx_descriptor *tx);
 
 /* --- Helper iov-locking functions --- */
 
-- 
cgit v1.2.3


From e8c82c2e23e3527e0c9dc195e432c16784d270fa Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Tue, 6 Jan 2009 03:05:50 +0100
Subject: mm lockless pagecache barrier fix

An XFS workload showed up a bug in the lockless pagecache patch. Basically it
would go into an "infinite" loop, although it would sometimes be able to break
out of the loop! The reason is a missing compiler barrier in the "increment
reference count unless it was zero" case of the lockless pagecache protocol in
the gang lookup functions.

This would cause the compiler to use a cached value of struct page pointer to
retry the operation with, rather than reload it. So the page might have been
removed from pagecache and freed (refcount==0) but the lookup would not correctly
notice the page is no longer in pagecache, and keep attempting to increment the
refcount and failing, until the page gets reallocated for something else. This
isn't a data corruption because the condition will be detected if the page has
been reallocated. However it can result in a lockup.

Linus points out that ACCESS_ONCE is also required in that pointer load, even
if it's absence is not causing a bug on our particular build. The most general
way to solve this is just to put an rcu_dereference in radix_tree_deref_slot.

Assembly of find_get_pages,
before:
.L220:
        movq    (%rbx), %rax    #* ivtmp.1162, tmp82
        movq    (%rax), %rdi    #, prephitmp.1149
.L218:
        testb   $1, %dil        #, prephitmp.1149
        jne     .L217   #,
        testq   %rdi, %rdi      # prephitmp.1149
        je      .L203   #,
        cmpq    $-1, %rdi       #, prephitmp.1149
        je      .L217   #,
        movl    8(%rdi), %esi   # <variable>._count.counter, c
        testl   %esi, %esi      # c
        je      .L218   #,

after:
.L212:
        movq    (%rbx), %rax    #* ivtmp.1109, tmp81
        movq    (%rax), %rdi    #, ret
        testb   $1, %dil        #, ret
        jne     .L211   #,
        testq   %rdi, %rdi      # ret
        je      .L197   #,
        cmpq    $-1, %rdi       #, ret
        je      .L211   #,
        movl    8(%rdi), %esi   # <variable>._count.counter, c
        testl   %esi, %esi      # c
        je      .L212   #,

(notice the obvious infinite loop in the first example, if page->count remains 0)

Signed-off-by: Nick Piggin <npiggin@suse.de>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/radix-tree.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h
index a916c6660dfa..355f6e80db0d 100644
--- a/include/linux/radix-tree.h
+++ b/include/linux/radix-tree.h
@@ -136,7 +136,7 @@ do {									\
  */
 static inline void *radix_tree_deref_slot(void **pslot)
 {
-	void *ret = *pslot;
+	void *ret = rcu_dereference(*pslot);
 	if (unlikely(radix_tree_is_indirect_ptr(ret)))
 		ret = RADIX_TREE_RETRY;
 	return ret;
-- 
cgit v1.2.3


From 10d3bd09a3c25df114f74f7f86e1b58d070bef32 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Tue, 6 Jan 2009 03:04:58 +0000
Subject: dm: consolidate target deregistration error handling

Change dm_unregister_target to return void and use BUG() for error
reporting.

dm_unregister_target can only fail because of programming bug in the
target driver. It can't fail because of user's behavior or disk errors.

This patch changes unregister_target to return void and use BUG if
someone tries to unregister non-registered target or unregister target
that is in use.

This patch removes code duplication (testing of error codes in all dm
targets) and reports bugs in just one place, in dm_unregister_target. In
some target drivers, these return codes were ignored, which could lead
to a situation where bugs could be missed.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-crypt.c         |  6 +-----
 drivers/md/dm-delay.c         |  6 +-----
 drivers/md/dm-linear.c        |  5 +----
 drivers/md/dm-mpath.c         |  6 +-----
 drivers/md/dm-raid1.c         |  6 +-----
 drivers/md/dm-snap.c          | 11 ++---------
 drivers/md/dm-stripe.c        |  4 +---
 drivers/md/dm-target.c        | 15 +++++++--------
 drivers/md/dm-zero.c          |  5 +----
 include/linux/device-mapper.h |  6 ++++--
 10 files changed, 20 insertions(+), 50 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 3326750ec02c..35bda49796fb 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -1322,11 +1322,7 @@ static int __init dm_crypt_init(void)
 
 static void __exit dm_crypt_exit(void)
 {
-	int r = dm_unregister_target(&crypt_target);
-
-	if (r < 0)
-		DMERR("unregister failed %d", r);
-
+	dm_unregister_target(&crypt_target);
 	kmem_cache_destroy(_crypt_io_pool);
 }
 
diff --git a/drivers/md/dm-delay.c b/drivers/md/dm-delay.c
index 848b381f1173..59ee1b015d2d 100644
--- a/drivers/md/dm-delay.c
+++ b/drivers/md/dm-delay.c
@@ -364,11 +364,7 @@ bad_queue:
 
 static void __exit dm_delay_exit(void)
 {
-	int r = dm_unregister_target(&delay_target);
-
-	if (r < 0)
-		DMERR("unregister failed %d", r);
-
+	dm_unregister_target(&delay_target);
 	kmem_cache_destroy(delayed_cache);
 	destroy_workqueue(kdelayd_wq);
 }
diff --git a/drivers/md/dm-linear.c b/drivers/md/dm-linear.c
index 44042becad8a..79fb53e51c70 100644
--- a/drivers/md/dm-linear.c
+++ b/drivers/md/dm-linear.c
@@ -156,8 +156,5 @@ int __init dm_linear_init(void)
 
 void dm_linear_exit(void)
 {
-	int r = dm_unregister_target(&linear_target);
-
-	if (r < 0)
-		DMERR("unregister failed %d", r);
+	dm_unregister_target(&linear_target);
 }
diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index 3d7f4923cd13..345a26047ae0 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -1495,14 +1495,10 @@ static int __init dm_multipath_init(void)
 
 static void __exit dm_multipath_exit(void)
 {
-	int r;
-
 	destroy_workqueue(kmpath_handlerd);
 	destroy_workqueue(kmultipathd);
 
-	r = dm_unregister_target(&multipath_target);
-	if (r < 0)
-		DMERR("target unregister failed %d", r);
+	dm_unregister_target(&multipath_target);
 	kmem_cache_destroy(_mpio_cache);
 }
 
diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c
index d0fed2b21b08..250f401668d5 100644
--- a/drivers/md/dm-raid1.c
+++ b/drivers/md/dm-raid1.c
@@ -1300,11 +1300,7 @@ static int __init dm_mirror_init(void)
 
 static void __exit dm_mirror_exit(void)
 {
-	int r;
-
-	r = dm_unregister_target(&mirror_target);
-	if (r < 0)
-		DMERR("unregister failed %d", r);
+	dm_unregister_target(&mirror_target);
 }
 
 /* Module hooks */
diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c
index 4ceedd4f22af..a8005b43a06b 100644
--- a/drivers/md/dm-snap.c
+++ b/drivers/md/dm-snap.c
@@ -1470,17 +1470,10 @@ static int __init dm_snapshot_init(void)
 
 static void __exit dm_snapshot_exit(void)
 {
-	int r;
-
 	destroy_workqueue(ksnapd);
 
-	r = dm_unregister_target(&snapshot_target);
-	if (r)
-		DMERR("snapshot unregister failed %d", r);
-
-	r = dm_unregister_target(&origin_target);
-	if (r)
-		DMERR("origin unregister failed %d", r);
+	dm_unregister_target(&snapshot_target);
+	dm_unregister_target(&origin_target);
 
 	exit_origin_hash();
 	kmem_cache_destroy(pending_cache);
diff --git a/drivers/md/dm-stripe.c b/drivers/md/dm-stripe.c
index 9e4ef88d421e..41569bc60abc 100644
--- a/drivers/md/dm-stripe.c
+++ b/drivers/md/dm-stripe.c
@@ -337,9 +337,7 @@ int __init dm_stripe_init(void)
 
 void dm_stripe_exit(void)
 {
-	if (dm_unregister_target(&stripe_target))
-		DMWARN("target unregistration failed");
-
+	dm_unregister_target(&stripe_target);
 	destroy_workqueue(kstriped);
 
 	return;
diff --git a/drivers/md/dm-target.c b/drivers/md/dm-target.c
index 835cf95b857f..7decf10006e4 100644
--- a/drivers/md/dm-target.c
+++ b/drivers/md/dm-target.c
@@ -130,26 +130,26 @@ int dm_register_target(struct target_type *t)
 	return rv;
 }
 
-int dm_unregister_target(struct target_type *t)
+void dm_unregister_target(struct target_type *t)
 {
 	struct tt_internal *ti;
 
 	down_write(&_lock);
 	if (!(ti = __find_target_type(t->name))) {
-		up_write(&_lock);
-		return -EINVAL;
+		DMCRIT("Unregistering unrecognised target: %s", t->name);
+		BUG();
 	}
 
 	if (ti->use) {
-		up_write(&_lock);
-		return -ETXTBSY;
+		DMCRIT("Attempt to unregister target still in use: %s",
+		       t->name);
+		BUG();
 	}
 
 	list_del(&ti->list);
 	kfree(ti);
 
 	up_write(&_lock);
-	return 0;
 }
 
 /*
@@ -187,8 +187,7 @@ int __init dm_target_init(void)
 
 void dm_target_exit(void)
 {
-	if (dm_unregister_target(&error_target))
-		DMWARN("error target unregistration failed");
+	dm_unregister_target(&error_target);
 }
 
 EXPORT_SYMBOL(dm_register_target);
diff --git a/drivers/md/dm-zero.c b/drivers/md/dm-zero.c
index cdbf126ec106..bbc97030c0c2 100644
--- a/drivers/md/dm-zero.c
+++ b/drivers/md/dm-zero.c
@@ -69,10 +69,7 @@ static int __init dm_zero_init(void)
 
 static void __exit dm_zero_exit(void)
 {
-	int r = dm_unregister_target(&zero_target);
-
-	if (r < 0)
-		DMERR("unregister failed %d", r);
+	dm_unregister_target(&zero_target);
 }
 
 module_init(dm_zero_init)
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index c17fd334e574..89ff2df40240 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -157,8 +157,7 @@ struct dm_target {
 };
 
 int dm_register_target(struct target_type *t);
-int dm_unregister_target(struct target_type *t);
-
+void dm_unregister_target(struct target_type *t);
 
 /*-----------------------------------------------------------------
  * Functions for creating and manipulating mapped devices.
@@ -276,6 +275,9 @@ void *dm_vcalloc(unsigned long nmemb, unsigned long elem_size);
  *---------------------------------------------------------------*/
 #define DM_NAME "device-mapper"
 
+#define DMCRIT(f, arg...) \
+	printk(KERN_CRIT DM_NAME ": " DM_MSG_PREFIX ": " f "\n", ## arg)
+
 #define DMERR(f, arg...) \
 	printk(KERN_ERR DM_NAME ": " DM_MSG_PREFIX ": " f "\n", ## arg)
 #define DMERR_LIMIT(f, arg...) \
-- 
cgit v1.2.3


From 7d76345da6ed3927c9cbf5d3f7a7021e8bba7374 Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Tue, 6 Jan 2009 03:05:07 +0000
Subject: dm request: extend target interface

This patch adds the following target interfaces for request-based dm.

  map_rq    : for mapping a request

  rq_end_io : for finishing a request

  busy      : for avoiding performance regression from bio-based dm.
              Target can tell dm core not to map requests now, and
              that may help requests in the block layer queue to be
              bigger by I/O merging.
              In bio-based dm, this behavior is done by device
              drivers managing the block layer queue.
              But in request-based dm, dm core has to do that
              since dm core manages the block layer queue.

Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 include/linux/device-mapper.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index 89ff2df40240..c1ba76c7c0e5 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -45,6 +45,8 @@ typedef void (*dm_dtr_fn) (struct dm_target *ti);
  */
 typedef int (*dm_map_fn) (struct dm_target *ti, struct bio *bio,
 			  union map_info *map_context);
+typedef int (*dm_map_request_fn) (struct dm_target *ti, struct request *clone,
+				  union map_info *map_context);
 
 /*
  * Returns:
@@ -57,6 +59,9 @@ typedef int (*dm_map_fn) (struct dm_target *ti, struct bio *bio,
 typedef int (*dm_endio_fn) (struct dm_target *ti,
 			    struct bio *bio, int error,
 			    union map_info *map_context);
+typedef int (*dm_request_endio_fn) (struct dm_target *ti,
+				    struct request *clone, int error,
+				    union map_info *map_context);
 
 typedef void (*dm_flush_fn) (struct dm_target *ti);
 typedef void (*dm_presuspend_fn) (struct dm_target *ti);
@@ -75,6 +80,13 @@ typedef int (*dm_ioctl_fn) (struct dm_target *ti, unsigned int cmd,
 typedef int (*dm_merge_fn) (struct dm_target *ti, struct bvec_merge_data *bvm,
 			    struct bio_vec *biovec, int max_size);
 
+/*
+ * Returns:
+ *    0: The target can handle the next I/O immediately.
+ *    1: The target can't handle the next I/O immediately.
+ */
+typedef int (*dm_busy_fn) (struct dm_target *ti);
+
 void dm_error(const char *message);
 
 /*
@@ -107,7 +119,9 @@ struct target_type {
 	dm_ctr_fn ctr;
 	dm_dtr_fn dtr;
 	dm_map_fn map;
+	dm_map_request_fn map_rq;
 	dm_endio_fn end_io;
+	dm_request_endio_fn rq_end_io;
 	dm_flush_fn flush;
 	dm_presuspend_fn presuspend;
 	dm_postsuspend_fn postsuspend;
@@ -117,6 +131,7 @@ struct target_type {
 	dm_message_fn message;
 	dm_ioctl_fn ioctl;
 	dm_merge_fn merge;
+	dm_busy_fn busy;
 };
 
 struct io_restrictions {
-- 
cgit v1.2.3


From ab4c1424882be9cd70b89abf2b484add355712fa Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 6 Jan 2009 03:05:09 +0000
Subject: dm: support barriers on simple devices

Implement barrier support for single device DM devices

This patch implements barrier support in DM for the common case of dm linear
just remapping a single underlying device. In this case we can safely
pass the barrier through because there can be no reordering between
devices.

 NB. Any DM device might cease to support barriers if it gets
     reconfigured so code must continue to allow for a possible
     -EOPNOTSUPP on every barrier bio submitted.  - agk

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-linear.c        |  1 +
 drivers/md/dm-table.c         | 19 +++++++++++++++++++
 drivers/md/dm.c               | 15 +++++----------
 drivers/md/dm.h               |  1 +
 include/linux/device-mapper.h |  7 +++++++
 5 files changed, 33 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/dm-linear.c b/drivers/md/dm-linear.c
index 79fb53e51c70..bfa107f59d96 100644
--- a/drivers/md/dm-linear.c
+++ b/drivers/md/dm-linear.c
@@ -142,6 +142,7 @@ static struct target_type linear_target = {
 	.status = linear_status,
 	.ioctl  = linear_ioctl,
 	.merge  = linear_merge,
+	.features = DM_TARGET_SUPPORTS_BARRIERS,
 };
 
 int __init dm_linear_init(void)
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 04e5fd742c2c..ebaaf72cd822 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -38,6 +38,8 @@ struct dm_table {
 	sector_t *highs;
 	struct dm_target *targets;
 
+	unsigned barriers_supported:1;
+
 	/*
 	 * Indicates the rw permissions for the new logical
 	 * device.  This should be a combination of FMODE_READ
@@ -227,6 +229,7 @@ int dm_table_create(struct dm_table **result, fmode_t mode,
 
 	INIT_LIST_HEAD(&t->devices);
 	atomic_set(&t->holders, 1);
+	t->barriers_supported = 1;
 
 	if (!num_targets)
 		num_targets = KEYS_PER_NODE;
@@ -728,6 +731,10 @@ int dm_table_add_target(struct dm_table *t, const char *type,
 	/* FIXME: the plan is to combine high here and then have
 	 * the merge fn apply the target level restrictions. */
 	combine_restrictions_low(&t->limits, &tgt->limits);
+
+	if (!(tgt->type->features & DM_TARGET_SUPPORTS_BARRIERS))
+		t->barriers_supported = 0;
+
 	return 0;
 
  bad:
@@ -772,6 +779,12 @@ int dm_table_complete(struct dm_table *t)
 
 	check_for_valid_limits(&t->limits);
 
+	/*
+	 * We only support barriers if there is exactly one underlying device.
+	 */
+	if (!list_is_singular(&t->devices))
+		t->barriers_supported = 0;
+
 	/* how many indexes will the btree have ? */
 	leaf_nodes = dm_div_up(t->num_targets, KEYS_PER_NODE);
 	t->depth = 1 + int_log(leaf_nodes, CHILDREN_PER_NODE);
@@ -986,6 +999,12 @@ struct mapped_device *dm_table_get_md(struct dm_table *t)
 	return t->md;
 }
 
+int dm_table_barrier_ok(struct dm_table *t)
+{
+	return t->barriers_supported;
+}
+EXPORT_SYMBOL(dm_table_barrier_ok);
+
 EXPORT_SYMBOL(dm_vcalloc);
 EXPORT_SYMBOL(dm_get_device);
 EXPORT_SYMBOL(dm_put_device);
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 4882ce7e88a3..dd953b189f45 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -835,7 +835,11 @@ static int __split_bio(struct mapped_device *md, struct bio *bio)
 	ci.map = dm_get_table(md);
 	if (unlikely(!ci.map))
 		return -EIO;
-
+	if (unlikely(bio_barrier(bio) && !dm_table_barrier_ok(ci.map))) {
+		dm_table_put(ci.map);
+		bio_endio(bio, -EOPNOTSUPP);
+		return 0;
+	}
 	ci.md = md;
 	ci.bio = bio;
 	ci.io = alloc_io(md);
@@ -919,15 +923,6 @@ static int dm_request(struct request_queue *q, struct bio *bio)
 	struct mapped_device *md = q->queuedata;
 	int cpu;
 
-	/*
-	 * There is no use in forwarding any barrier request since we can't
-	 * guarantee it is (or can be) handled by the targets correctly.
-	 */
-	if (unlikely(bio_barrier(bio))) {
-		bio_endio(bio, -EOPNOTSUPP);
-		return 0;
-	}
-
 	down_read(&md->io_lock);
 
 	cpu = part_stat_lock();
diff --git a/drivers/md/dm.h b/drivers/md/dm.h
index 0ade60cdef42..5b5d08ba9e92 100644
--- a/drivers/md/dm.h
+++ b/drivers/md/dm.h
@@ -51,6 +51,7 @@ int dm_table_any_congested(struct dm_table *t, int bdi_bits);
  * To check the return value from dm_table_find_target().
  */
 #define dm_target_is_valid(t) ((t)->table)
+int dm_table_barrier_ok(struct dm_table *t);
 
 /*-----------------------------------------------------------------
  * A registry of target types.
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index c1ba76c7c0e5..8209e08969f9 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -112,7 +112,14 @@ void dm_put_device(struct dm_target *ti, struct dm_dev *d);
 /*
  * Information about a target type
  */
+
+/*
+ * Target features
+ */
+#define DM_TARGET_SUPPORTS_BARRIERS 0x00000001
+
 struct target_type {
+	uint64_t features;
 	const char *name;
 	struct module *module;
 	unsigned version[3];
-- 
cgit v1.2.3


From b3881f74b31b7d47d0f1c4d89ac3e7f0b9c05e3e Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Mon, 5 Jan 2009 22:46:26 -0500
Subject: ext4: Add mount option to set kjournald's I/O priority

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: Jens Axboe <jens.axboe@oracle.com>
---
 Documentation/filesystems/ext4.txt |  7 +++++++
 fs/ext4/super.c                    | 29 +++++++++++++++++++++++++----
 fs/ioprio.c                        |  3 ++-
 include/linux/ioprio.h             |  2 ++
 4 files changed, 36 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt
index 9ec29d86ff8b..8938949b201e 100644
--- a/Documentation/filesystems/ext4.txt
+++ b/Documentation/filesystems/ext4.txt
@@ -308,6 +308,13 @@ min_batch_time=usec	This parameter sets the commit time (as
 			multi-threaded, synchronous workloads on very
 			fast disks, at the cost of increasing latency.
 
+journal_ioprio=prio	The I/O priority (from 0 to 7, where 0 is the
+			highest priorty) which should be used for I/O
+			operations submitted by kjournald2 during a
+			commit operation.  This defaults to 3, which is
+			a slightly higher priority than the default I/O
+			priority.
+
 Data Mode
 =========
 There are 3 different data modes:
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 8036392b2121..8ff8709828fd 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1013,7 +1013,7 @@ enum {
 	Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota,
 	Opt_grpquota, Opt_extents, Opt_noextents, Opt_i_version,
 	Opt_stripe, Opt_delalloc, Opt_nodelalloc,
-	Opt_inode_readahead_blks
+	Opt_inode_readahead_blks, Opt_journal_ioprio
 };
 
 static const match_table_t tokens = {
@@ -1074,6 +1074,7 @@ static const match_table_t tokens = {
 	{Opt_delalloc, "delalloc"},
 	{Opt_nodelalloc, "nodelalloc"},
 	{Opt_inode_readahead_blks, "inode_readahead_blks=%u"},
+	{Opt_journal_ioprio, "journal_ioprio=%u"},
 	{Opt_err, NULL},
 };
 
@@ -1098,8 +1099,11 @@ static ext4_fsblk_t get_sb_block(void **data)
 	return sb_block;
 }
 
+#define DEFAULT_JOURNAL_IOPRIO (IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, 3))
+
 static int parse_options(char *options, struct super_block *sb,
 			 unsigned long *journal_devnum,
+			 unsigned int *journal_ioprio,
 			 ext4_fsblk_t *n_blocks_count, int is_remount)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
@@ -1492,6 +1496,14 @@ set_qf_format:
 				return 0;
 			sbi->s_inode_readahead_blks = option;
 			break;
+		case Opt_journal_ioprio:
+			if (match_int(&args[0], &option))
+				return 0;
+			if (option < 0 || option > 7)
+				break;
+			*journal_ioprio = IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE,
+							    option);
+			break;
 		default:
 			printk(KERN_ERR
 			       "EXT4-fs: Unrecognized mount option \"%s\" "
@@ -2035,6 +2047,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	int features;
 	__u64 blocks_count;
 	int err;
+	unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO;
 
 	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
 	if (!sbi)
@@ -2141,7 +2154,8 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	set_opt(sbi->s_mount_opt, DELALLOC);
 
 
-	if (!parse_options((char *) data, sb, &journal_devnum, NULL, 0))
+	if (!parse_options((char *) data, sb, &journal_devnum,
+			   &journal_ioprio, NULL, 0))
 		goto failed_mount;
 
 	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
@@ -2506,6 +2520,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	default:
 		break;
 	}
+	set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
 
 no_journal:
 
@@ -3127,6 +3142,7 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data)
 	unsigned long old_sb_flags;
 	struct ext4_mount_options old_opts;
 	ext4_group_t g;
+	unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO;
 	int err;
 #ifdef CONFIG_QUOTA
 	int i;
@@ -3145,11 +3161,14 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data)
 	for (i = 0; i < MAXQUOTAS; i++)
 		old_opts.s_qf_names[i] = sbi->s_qf_names[i];
 #endif
+	if (sbi->s_journal && sbi->s_journal->j_task->io_context)
+		journal_ioprio = sbi->s_journal->j_task->io_context->ioprio;
 
 	/*
 	 * Allow the "check" option to be passed as a remount option.
 	 */
-	if (!parse_options(data, sb, NULL, &n_blocks_count, 1)) {
+	if (!parse_options(data, sb, NULL, &journal_ioprio,
+			   &n_blocks_count, 1)) {
 		err = -EINVAL;
 		goto restore_opts;
 	}
@@ -3162,8 +3181,10 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data)
 
 	es = sbi->s_es;
 
-	if (sbi->s_journal)
+	if (sbi->s_journal) {
 		ext4_init_journal_params(sb, sbi->s_journal);
+		set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
+	}
 
 	if ((*flags & MS_RDONLY) != (sb->s_flags & MS_RDONLY) ||
 		n_blocks_count > ext4_blocks_count(es)) {
diff --git a/fs/ioprio.c b/fs/ioprio.c
index 3569e0ad86a2..1a39ac370942 100644
--- a/fs/ioprio.c
+++ b/fs/ioprio.c
@@ -27,7 +27,7 @@
 #include <linux/security.h>
 #include <linux/pid_namespace.h>
 
-static int set_task_ioprio(struct task_struct *task, int ioprio)
+int set_task_ioprio(struct task_struct *task, int ioprio)
 {
 	int err;
 	struct io_context *ioc;
@@ -70,6 +70,7 @@ static int set_task_ioprio(struct task_struct *task, int ioprio)
 	task_unlock(task);
 	return err;
 }
+EXPORT_SYMBOL_GPL(set_task_ioprio);
 
 asmlinkage long sys_ioprio_set(int which, int who, int ioprio)
 {
diff --git a/include/linux/ioprio.h b/include/linux/ioprio.h
index f98a656b17e5..76dad4808847 100644
--- a/include/linux/ioprio.h
+++ b/include/linux/ioprio.h
@@ -86,4 +86,6 @@ static inline int task_nice_ioclass(struct task_struct *task)
  */
 extern int ioprio_best(unsigned short aprio, unsigned short bprio);
 
+extern int set_task_ioprio(struct task_struct *task, int ioprio);
+
 #endif
-- 
cgit v1.2.3


From 835481d9bcd65720b473db6b38746a74a3964218 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Sun, 4 Jan 2009 05:18:06 -0800
Subject: cpumask: convert struct cpufreq_policy to cpumask_var_t

Impact: use new cpumask API to reduce memory usage

This is part of an effort to reduce structure sizes for machines
configured with large NR_CPUS.  cpumask_t gets replaced by
cpumask_var_t, which is either struct cpumask[1] (small NR_CPUS) or
struct cpumask * (large NR_CPUS).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Acked-by: Dave Jones <davej@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c       | 10 +++---
 arch/x86/kernel/cpu/cpufreq/p4-clockmod.c        |  8 ++---
 arch/x86/kernel/cpu/cpufreq/powernow-k8.c        |  6 ++--
 arch/x86/kernel/cpu/cpufreq/powernow-k8.h        |  2 +-
 arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c | 14 ++++----
 arch/x86/kernel/cpu/cpufreq/speedstep-ich.c      | 18 +++++-----
 drivers/cpufreq/cpufreq.c                        | 42 ++++++++++++++++--------
 drivers/cpufreq/cpufreq_conservative.c           |  2 +-
 drivers/cpufreq/cpufreq_ondemand.c               |  4 +--
 include/linux/cpufreq.h                          |  4 +--
 10 files changed, 62 insertions(+), 48 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index 28102ad1a363..0b31939862d6 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -411,7 +411,7 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
 
 #ifdef CONFIG_HOTPLUG_CPU
 	/* cpufreq holds the hotplug lock, so we are safe from here on */
-	cpus_and(online_policy_cpus, cpu_online_map, policy->cpus);
+	cpumask_and(&online_policy_cpus, cpu_online_mask, policy->cpus);
 #else
 	online_policy_cpus = policy->cpus;
 #endif
@@ -626,15 +626,15 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
 	 */
 	if (policy->shared_type == CPUFREQ_SHARED_TYPE_ALL ||
 	    policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) {
-		cpumask_copy(&policy->cpus, perf->shared_cpu_map);
+		cpumask_copy(policy->cpus, perf->shared_cpu_map);
 	}
-	cpumask_copy(&policy->related_cpus, perf->shared_cpu_map);
+	cpumask_copy(policy->related_cpus, perf->shared_cpu_map);
 
 #ifdef CONFIG_SMP
 	dmi_check_system(sw_any_bug_dmi_table);
-	if (bios_with_sw_any_bug && cpus_weight(policy->cpus) == 1) {
+	if (bios_with_sw_any_bug && cpumask_weight(policy->cpus) == 1) {
 		policy->shared_type = CPUFREQ_SHARED_TYPE_ALL;
-		policy->cpus = per_cpu(cpu_core_map, cpu);
+		cpumask_copy(policy->cpus, cpu_core_mask(cpu));
 	}
 #endif
 
diff --git a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
index beea4466b063..b585e04cbc9e 100644
--- a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
+++ b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c
@@ -122,7 +122,7 @@ static int cpufreq_p4_target(struct cpufreq_policy *policy,
 		return 0;
 
 	/* notifiers */
-	for_each_cpu_mask_nr(i, policy->cpus) {
+	for_each_cpu(i, policy->cpus) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
 	}
@@ -130,11 +130,11 @@ static int cpufreq_p4_target(struct cpufreq_policy *policy,
 	/* run on each logical CPU, see section 13.15.3 of IA32 Intel Architecture Software
 	 * Developer's Manual, Volume 3
 	 */
-	for_each_cpu_mask_nr(i, policy->cpus)
+	for_each_cpu(i, policy->cpus)
 		cpufreq_p4_setdc(i, p4clockmod_table[newstate].index);
 
 	/* notifiers */
-	for_each_cpu_mask_nr(i, policy->cpus) {
+	for_each_cpu(i, policy->cpus) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
 	}
@@ -203,7 +203,7 @@ static int cpufreq_p4_cpu_init(struct cpufreq_policy *policy)
 	unsigned int i;
 
 #ifdef CONFIG_SMP
-	policy->cpus = per_cpu(cpu_sibling_map, policy->cpu);
+	cpumask_copy(policy->cpus, &per_cpu(cpu_sibling_map, policy->cpu));
 #endif
 
 	/* Errata workaround */
diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
index c3c9adbaa26f..5c28b37dea11 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
@@ -1199,10 +1199,10 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 	set_cpus_allowed_ptr(current, &oldmask);
 
 	if (cpu_family == CPU_HW_PSTATE)
-		pol->cpus = cpumask_of_cpu(pol->cpu);
+		cpumask_copy(pol->cpus, cpumask_of(pol->cpu));
 	else
-		pol->cpus = per_cpu(cpu_core_map, pol->cpu);
-	data->available_cores = &(pol->cpus);
+		cpumask_copy(pol->cpus, &per_cpu(cpu_core_map, pol->cpu));
+	data->available_cores = pol->cpus;
 
 	/* Take a crude guess here.
 	 * That guess was in microseconds, so multiply with 1000 */
diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h b/arch/x86/kernel/cpu/cpufreq/powernow-k8.h
index 65cfb5d7f77f..8ecc75b6c7c3 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.h
@@ -53,7 +53,7 @@ struct powernow_k8_data {
 	/* we need to keep track of associated cores, but let cpufreq
 	 * handle hotplug events - so just point at cpufreq pol->cpus
 	 * structure */
-	cpumask_t *available_cores;
+	struct cpumask *available_cores;
 };
 
 
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
index d2cc4991cbaa..f08998278a3a 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
@@ -492,8 +492,8 @@ static int centrino_target (struct cpufreq_policy *policy,
 	}
 
 	first_cpu = 1;
-	for_each_cpu_mask_nr(j, policy->cpus) {
-		const cpumask_t *mask;
+	for_each_cpu(j, policy->cpus) {
+		const struct cpumask *mask;
 
 		/* cpufreq holds the hotplug lock, so we are safe here */
 		if (!cpu_online(j))
@@ -504,9 +504,9 @@ static int centrino_target (struct cpufreq_policy *policy,
 		 * Make sure we are running on CPU that wants to change freq
 		 */
 		if (policy->shared_type == CPUFREQ_SHARED_TYPE_ANY)
-			mask = &policy->cpus;
+			mask = policy->cpus;
 		else
-			mask = &cpumask_of_cpu(j);
+			mask = cpumask_of(j);
 
 		set_cpus_allowed_ptr(current, mask);
 		preempt_disable();
@@ -538,7 +538,7 @@ static int centrino_target (struct cpufreq_policy *policy,
 			dprintk("target=%dkHz old=%d new=%d msr=%04x\n",
 				target_freq, freqs.old, freqs.new, msr);
 
-			for_each_cpu_mask_nr(k, policy->cpus) {
+			for_each_cpu(k, policy->cpus) {
 				if (!cpu_online(k))
 					continue;
 				freqs.cpu = k;
@@ -563,7 +563,7 @@ static int centrino_target (struct cpufreq_policy *policy,
 		preempt_enable();
 	}
 
-	for_each_cpu_mask_nr(k, policy->cpus) {
+	for_each_cpu(k, policy->cpus) {
 		if (!cpu_online(k))
 			continue;
 		freqs.cpu = k;
@@ -586,7 +586,7 @@ static int centrino_target (struct cpufreq_policy *policy,
 		tmp = freqs.new;
 		freqs.new = freqs.old;
 		freqs.old = tmp;
-		for_each_cpu_mask_nr(j, policy->cpus) {
+		for_each_cpu(j, policy->cpus) {
 			if (!cpu_online(j))
 				continue;
 			cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
index 04d0376b64b0..dedc1e98f168 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
@@ -229,7 +229,7 @@ static unsigned int speedstep_detect_chipset (void)
 	return 0;
 }
 
-static unsigned int _speedstep_get(const cpumask_t *cpus)
+static unsigned int _speedstep_get(const struct cpumask *cpus)
 {
 	unsigned int speed;
 	cpumask_t cpus_allowed;
@@ -244,7 +244,7 @@ static unsigned int _speedstep_get(const cpumask_t *cpus)
 
 static unsigned int speedstep_get(unsigned int cpu)
 {
-	return _speedstep_get(&cpumask_of_cpu(cpu));
+	return _speedstep_get(cpumask_of(cpu));
 }
 
 /**
@@ -267,7 +267,7 @@ static int speedstep_target (struct cpufreq_policy *policy,
 	if (cpufreq_frequency_table_target(policy, &speedstep_freqs[0], target_freq, relation, &newstate))
 		return -EINVAL;
 
-	freqs.old = _speedstep_get(&policy->cpus);
+	freqs.old = _speedstep_get(policy->cpus);
 	freqs.new = speedstep_freqs[newstate].frequency;
 	freqs.cpu = policy->cpu;
 
@@ -279,20 +279,20 @@ static int speedstep_target (struct cpufreq_policy *policy,
 
 	cpus_allowed = current->cpus_allowed;
 
-	for_each_cpu_mask_nr(i, policy->cpus) {
+	for_each_cpu(i, policy->cpus) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
 	}
 
 	/* switch to physical CPU where state is to be changed */
-	set_cpus_allowed_ptr(current, &policy->cpus);
+	set_cpus_allowed_ptr(current, policy->cpus);
 
 	speedstep_set_state(newstate);
 
 	/* allow to be run on all CPUs */
 	set_cpus_allowed_ptr(current, &cpus_allowed);
 
-	for_each_cpu_mask_nr(i, policy->cpus) {
+	for_each_cpu(i, policy->cpus) {
 		freqs.cpu = i;
 		cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
 	}
@@ -322,11 +322,11 @@ static int speedstep_cpu_init(struct cpufreq_policy *policy)
 
 	/* only run on CPU to be set, or on its sibling */
 #ifdef CONFIG_SMP
-	policy->cpus = per_cpu(cpu_sibling_map, policy->cpu);
+	cpumask_copy(policy->cpus, &per_cpu(cpu_sibling_map, policy->cpu));
 #endif
 
 	cpus_allowed = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &policy->cpus);
+	set_cpus_allowed_ptr(current, policy->cpus);
 
 	/* detect low and high frequency and transition latency */
 	result = speedstep_get_freqs(speedstep_processor,
@@ -339,7 +339,7 @@ static int speedstep_cpu_init(struct cpufreq_policy *policy)
 		return result;
 
 	/* get current speed setting */
-	speed = _speedstep_get(&policy->cpus);
+	speed = _speedstep_get(policy->cpus);
 	if (!speed)
 		return -EIO;
 
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 01dde80597f7..b55cb67435bd 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -584,12 +584,12 @@ out:
 	return i;
 }
 
-static ssize_t show_cpus(cpumask_t mask, char *buf)
+static ssize_t show_cpus(const struct cpumask *mask, char *buf)
 {
 	ssize_t i = 0;
 	unsigned int cpu;
 
-	for_each_cpu_mask_nr(cpu, mask) {
+	for_each_cpu(cpu, mask) {
 		if (i)
 			i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), " ");
 		i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), "%u", cpu);
@@ -606,7 +606,7 @@ static ssize_t show_cpus(cpumask_t mask, char *buf)
  */
 static ssize_t show_related_cpus(struct cpufreq_policy *policy, char *buf)
 {
-	if (cpus_empty(policy->related_cpus))
+	if (cpumask_empty(policy->related_cpus))
 		return show_cpus(policy->cpus, buf);
 	return show_cpus(policy->related_cpus, buf);
 }
@@ -806,9 +806,20 @@ static int cpufreq_add_dev(struct sys_device *sys_dev)
 		ret = -ENOMEM;
 		goto nomem_out;
 	}
+	if (!alloc_cpumask_var(&policy->cpus, GFP_KERNEL)) {
+		kfree(policy);
+		ret = -ENOMEM;
+		goto nomem_out;
+	}
+	if (!alloc_cpumask_var(&policy->related_cpus, GFP_KERNEL)) {
+		free_cpumask_var(policy->cpus);
+		kfree(policy);
+		ret = -ENOMEM;
+		goto nomem_out;
+	}
 
 	policy->cpu = cpu;
-	policy->cpus = cpumask_of_cpu(cpu);
+	cpumask_copy(policy->cpus, cpumask_of(cpu));
 
 	/* Initially set CPU itself as the policy_cpu */
 	per_cpu(policy_cpu, cpu) = cpu;
@@ -843,7 +854,7 @@ static int cpufreq_add_dev(struct sys_device *sys_dev)
 	}
 #endif
 
-	for_each_cpu_mask_nr(j, policy->cpus) {
+	for_each_cpu(j, policy->cpus) {
 		if (cpu == j)
 			continue;
 
@@ -861,7 +872,7 @@ static int cpufreq_add_dev(struct sys_device *sys_dev)
 				goto err_out_driver_exit;
 
 			spin_lock_irqsave(&cpufreq_driver_lock, flags);
-			managed_policy->cpus = policy->cpus;
+			cpumask_copy(managed_policy->cpus, policy->cpus);
 			per_cpu(cpufreq_cpu_data, cpu) = managed_policy;
 			spin_unlock_irqrestore(&cpufreq_driver_lock, flags);
 
@@ -916,14 +927,14 @@ static int cpufreq_add_dev(struct sys_device *sys_dev)
 	}
 
 	spin_lock_irqsave(&cpufreq_driver_lock, flags);
-	for_each_cpu_mask_nr(j, policy->cpus) {
+	for_each_cpu(j, policy->cpus) {
 		per_cpu(cpufreq_cpu_data, j) = policy;
 		per_cpu(policy_cpu, j) = policy->cpu;
 	}
 	spin_unlock_irqrestore(&cpufreq_driver_lock, flags);
 
 	/* symlink affected CPUs */
-	for_each_cpu_mask_nr(j, policy->cpus) {
+	for_each_cpu(j, policy->cpus) {
 		if (j == cpu)
 			continue;
 		if (!cpu_online(j))
@@ -963,7 +974,7 @@ static int cpufreq_add_dev(struct sys_device *sys_dev)
 
 err_out_unregister:
 	spin_lock_irqsave(&cpufreq_driver_lock, flags);
-	for_each_cpu_mask_nr(j, policy->cpus)
+	for_each_cpu(j, policy->cpus)
 		per_cpu(cpufreq_cpu_data, j) = NULL;
 	spin_unlock_irqrestore(&cpufreq_driver_lock, flags);
 
@@ -1024,7 +1035,7 @@ static int __cpufreq_remove_dev(struct sys_device *sys_dev)
 	 */
 	if (unlikely(cpu != data->cpu)) {
 		dprintk("removing link\n");
-		cpu_clear(cpu, data->cpus);
+		cpumask_clear_cpu(cpu, data->cpus);
 		spin_unlock_irqrestore(&cpufreq_driver_lock, flags);
 		sysfs_remove_link(&sys_dev->kobj, "cpufreq");
 		cpufreq_cpu_put(data);
@@ -1045,8 +1056,8 @@ static int __cpufreq_remove_dev(struct sys_device *sys_dev)
 	 * per_cpu(cpufreq_cpu_data) while holding the lock, and remove
 	 * the sysfs links afterwards.
 	 */
-	if (unlikely(cpus_weight(data->cpus) > 1)) {
-		for_each_cpu_mask_nr(j, data->cpus) {
+	if (unlikely(cpumask_weight(data->cpus) > 1)) {
+		for_each_cpu(j, data->cpus) {
 			if (j == cpu)
 				continue;
 			per_cpu(cpufreq_cpu_data, j) = NULL;
@@ -1055,8 +1066,8 @@ static int __cpufreq_remove_dev(struct sys_device *sys_dev)
 
 	spin_unlock_irqrestore(&cpufreq_driver_lock, flags);
 
-	if (unlikely(cpus_weight(data->cpus) > 1)) {
-		for_each_cpu_mask_nr(j, data->cpus) {
+	if (unlikely(cpumask_weight(data->cpus) > 1)) {
+		for_each_cpu(j, data->cpus) {
 			if (j == cpu)
 				continue;
 			dprintk("removing link for cpu %u\n", j);
@@ -1090,7 +1101,10 @@ static int __cpufreq_remove_dev(struct sys_device *sys_dev)
 	if (cpufreq_driver->exit)
 		cpufreq_driver->exit(data);
 
+	free_cpumask_var(data->related_cpus);
+	free_cpumask_var(data->cpus);
 	kfree(data);
+	per_cpu(cpufreq_cpu_data, cpu) = NULL;
 
 	cpufreq_debug_enable_ratelimit();
 	return 0;
diff --git a/drivers/cpufreq/cpufreq_conservative.c b/drivers/cpufreq/cpufreq_conservative.c
index e2657837d954..0320962c4ec5 100644
--- a/drivers/cpufreq/cpufreq_conservative.c
+++ b/drivers/cpufreq/cpufreq_conservative.c
@@ -498,7 +498,7 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy,
 			return rc;
 		}
 
-		for_each_cpu_mask_nr(j, policy->cpus) {
+		for_each_cpu(j, policy->cpus) {
 			struct cpu_dbs_info_s *j_dbs_info;
 			j_dbs_info = &per_cpu(cpu_dbs_info, j);
 			j_dbs_info->cur_policy = policy;
diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c
index 2ab3c12b88af..6a2b036c9389 100644
--- a/drivers/cpufreq/cpufreq_ondemand.c
+++ b/drivers/cpufreq/cpufreq_ondemand.c
@@ -400,7 +400,7 @@ static void dbs_check_cpu(struct cpu_dbs_info_s *this_dbs_info)
 	/* Get Absolute Load - in terms of freq */
 	max_load_freq = 0;
 
-	for_each_cpu_mask_nr(j, policy->cpus) {
+	for_each_cpu(j, policy->cpus) {
 		struct cpu_dbs_info_s *j_dbs_info;
 		cputime64_t cur_wall_time, cur_idle_time;
 		unsigned int idle_time, wall_time;
@@ -568,7 +568,7 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy,
 			return rc;
 		}
 
-		for_each_cpu_mask_nr(j, policy->cpus) {
+		for_each_cpu(j, policy->cpus) {
 			struct cpu_dbs_info_s *j_dbs_info;
 			j_dbs_info = &per_cpu(cpu_dbs_info, j);
 			j_dbs_info->cur_policy = policy;
diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
index 484b3abf61bb..384b38d3e8e2 100644
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -80,8 +80,8 @@ struct cpufreq_real_policy {
 };
 
 struct cpufreq_policy {
-	cpumask_t		cpus;	/* CPUs requiring sw coordination */
-	cpumask_t		related_cpus; /* CPUs with any coordination */
+	cpumask_var_t		cpus;	/* CPUs requiring sw coordination */
+	cpumask_var_t		related_cpus; /* CPUs with any coordination */
 	unsigned int		shared_type; /* ANY or ALL affected CPUs
 						should set cpufreq */
 	unsigned int		cpu;    /* cpu nr of registered CPU */
-- 
cgit v1.2.3


From 025dfdafe77f20b3890981a394774baab7b9c827 Mon Sep 17 00:00:00 2001
From: Frederik Schwarzer <schwarzerf@gmail.com>
Date: Thu, 16 Oct 2008 19:02:37 +0200
Subject: trivial: fix then -> than typos in comments and documentation

- (better, more, bigger ...) then -> (...) than

Signed-off-by: Frederik Schwarzer <schwarzerf@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 Documentation/hwmon/abituguru-datasheet           |  6 +++---
 Documentation/networking/rxrpc.txt                |  2 +-
 Documentation/scsi/ChangeLog.lpfc                 |  2 +-
 arch/blackfin/kernel/kgdb.c                       |  2 +-
 arch/ia64/kernel/kprobes.c                        |  2 +-
 arch/m68k/Kconfig                                 |  2 +-
 arch/mips/pmc-sierra/yosemite/atmel_read_eeprom.c |  2 +-
 arch/powerpc/kernel/kprobes.c                     |  2 +-
 arch/powerpc/oprofile/cell/spu_profiler.c         |  2 +-
 arch/s390/Kconfig                                 |  2 +-
 arch/s390/kernel/kprobes.c                        |  2 +-
 arch/sparc/kernel/kprobes.c                       |  2 +-
 arch/x86/kernel/kprobes.c                         |  2 +-
 arch/x86/kernel/mfgpt_32.c                        |  2 +-
 drivers/hwmon/fschmd.c                            |  2 +-
 drivers/infiniband/hw/mlx4/cq.c                   |  2 +-
 drivers/message/i2o/i2o_scsi.c                    |  2 +-
 drivers/mtd/devices/pmc551.c                      |  2 +-
 drivers/mtd/ubi/eba.c                             |  2 +-
 drivers/mtd/ubi/io.c                              |  2 +-
 drivers/mtd/ubi/scan.c                            |  2 +-
 drivers/mtd/ubi/ubi-media.h                       |  4 ++--
 drivers/mtd/ubi/vtbl.c                            |  2 +-
 drivers/mtd/ubi/wl.c                              |  4 ++--
 drivers/net/bnx2x_link.c                          |  2 +-
 drivers/net/e1000/e1000_hw.c                      |  4 ++--
 drivers/net/slip.h                                |  2 +-
 drivers/net/tehuti.c                              |  4 ++--
 drivers/net/tokenring/smctr.c                     |  2 +-
 drivers/net/wireless/ipw2x00/ipw2100.c            |  2 +-
 drivers/net/wireless/rt2x00/rt2x00crypto.c        |  4 ++--
 drivers/net/wireless/strip.c                      |  2 +-
 drivers/s390/block/dasd_eer.c                     |  4 ++--
 drivers/s390/char/vmlogrdr.c                      |  4 ++--
 drivers/scsi/lpfc/lpfc_hbadisc.c                  |  4 ++--
 drivers/scsi/lpfc/lpfc_sli.c                      | 10 +++++-----
 drivers/serial/crisv10.c                          |  4 ++--
 drivers/video/console/vgacon.c                    |  2 +-
 fs/ocfs2/cluster/heartbeat.c                      |  2 +-
 fs/proc/task_nommu.c                              |  2 +-
 fs/ubifs/Kconfig                                  |  2 +-
 fs/ubifs/budget.c                                 |  4 ++--
 fs/ubifs/gc.c                                     |  2 +-
 fs/ubifs/journal.c                                |  2 +-
 fs/ubifs/shrinker.c                               |  2 +-
 fs/xfs/linux-2.6/xfs_super.c                      |  2 +-
 include/linux/mtd/mtd.h                           |  2 +-
 include/linux/spi/spi.h                           |  4 ++--
 include/mtd/ubi-user.h                            |  2 +-
 kernel/pid.c                                      |  2 +-
 kernel/time/jiffies.c                             |  2 +-
 net/sctp/auth.c                                   |  4 ++--
 net/sctp/sm_statefuns.c                           |  6 +++---
 net/sctp/socket.c                                 |  2 +-
 net/sctp/tsnmap.c                                 |  2 +-
 sound/usb/usx2y/usbusx2y.c                        |  2 +-
 56 files changed, 76 insertions(+), 76 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/hwmon/abituguru-datasheet b/Documentation/hwmon/abituguru-datasheet
index aef5a9b36846..4d184f2db0ea 100644
--- a/Documentation/hwmon/abituguru-datasheet
+++ b/Documentation/hwmon/abituguru-datasheet
@@ -74,7 +74,7 @@ a sensor.
 Notice that some banks have both a read and a write address this is how the
 uGuru determines if a read from or a write to the bank is taking place, thus
 when reading you should always use the read address and when writing the
-write address. The write address is always one (1) more then the read address.
+write address. The write address is always one (1) more than the read address.
 
 
 uGuru ready
@@ -224,7 +224,7 @@ Bit 3: Beep if alarm							(RW)
 Bit 4: 1 if alarm cause measured temp is over the warning threshold	(R)
 Bit 5: 1 if alarm cause measured volt is over the max threshold		(R)
 Bit 6: 1 if alarm cause measured volt is under the min threshold	(R)
-Bit 7: Volt sensor: Shutdown if alarm persist for more then 4 seconds	(RW)
+Bit 7: Volt sensor: Shutdown if alarm persist for more than 4 seconds	(RW)
        Temp sensor: Shutdown if temp is over the shutdown threshold	(RW)
 
 *  This bit is only honored/used by the uGuru if a temp sensor is connected
@@ -293,7 +293,7 @@ Byte 0:
 Alarm behaviour for the selected sensor. A 1 enables the described behaviour.
 Bit 0: Give an alarm if measured rpm is under the min threshold	(RW)
 Bit 3: Beep if alarm						(RW)
-Bit 7: Shutdown if alarm persist for more then 4 seconds	(RW)
+Bit 7: Shutdown if alarm persist for more than 4 seconds	(RW)
 
 Byte 1:
 min threshold (scale as bank 0x26)
diff --git a/Documentation/networking/rxrpc.txt b/Documentation/networking/rxrpc.txt
index c3669a3fb4af..60d05eb77c64 100644
--- a/Documentation/networking/rxrpc.txt
+++ b/Documentation/networking/rxrpc.txt
@@ -540,7 +540,7 @@ A client would issue an operation by:
      MSG_MORE should be set in msghdr::msg_flags on all but the last part of
      the request.  Multiple requests may be made simultaneously.
 
-     If a call is intended to go to a destination other then the default
+     If a call is intended to go to a destination other than the default
      specified through connect(), then msghdr::msg_name should be set on the
      first request message of that call.
 
diff --git a/Documentation/scsi/ChangeLog.lpfc b/Documentation/scsi/ChangeLog.lpfc
index ae3f962a7cfc..ff19a52fe004 100644
--- a/Documentation/scsi/ChangeLog.lpfc
+++ b/Documentation/scsi/ChangeLog.lpfc
@@ -733,7 +733,7 @@ Changes from 20040920 to 20041018
 	  I/O completion path a little more, especially taking care of
 	  fast-pathing the non-error case.  Also removes tons of dead
 	  members and defines from lpfc_scsi.h - e.g. lpfc_target is down
-	  to nothing more then the lpfc_nodelist pointer.
+	  to nothing more than the lpfc_nodelist pointer.
 	* Added binary sysfs file to issue mbox commands
 	* Replaced #if __BIG_ENDIAN with #if __BIG_ENDIAN_BITFIELD for
 	  compatibility with the user space applications.
diff --git a/arch/blackfin/kernel/kgdb.c b/arch/blackfin/kernel/kgdb.c
index b795a207742c..1c5afaeb9504 100644
--- a/arch/blackfin/kernel/kgdb.c
+++ b/arch/blackfin/kernel/kgdb.c
@@ -105,7 +105,7 @@ void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs)
  * Extracts ebp, esp and eip values understandable by gdb from the values
  * saved by switch_to.
  * thread.esp points to ebp. flags and ebp are pushed in switch_to hence esp
- * prior to entering switch_to is 8 greater then the value that is saved.
+ * prior to entering switch_to is 8 greater than the value that is saved.
  * If switch_to changes, change following code appropriately.
  */
 void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p)
diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index f07688da947c..0017b9de2ddf 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -434,7 +434,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 	/*
 	 * It is possible to have multiple instances associated with a given
 	 * task either because an multiple functions in the call path
-	 * have a return probe installed on them, and/or more then one return
+	 * have a return probe installed on them, and/or more than one return
 	 * return probe was registered for a target function.
 	 *
 	 * We can handle this because:
diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig
index c825bde17cb3..fb87c08c6b57 100644
--- a/arch/m68k/Kconfig
+++ b/arch/m68k/Kconfig
@@ -303,7 +303,7 @@ config M68KFPU_EMU_EXTRAPREC
 	  correct rounding, the emulator can (often) do the same but this
 	  extra calculation can cost quite some time, so you can disable
 	  it here. The emulator will then "only" calculate with a 64 bit
-	  mantissa and round slightly incorrect, what is more then enough
+	  mantissa and round slightly incorrect, what is more than enough
 	  for normal usage.
 
 config M68KFPU_EMU_ONLY
diff --git a/arch/mips/pmc-sierra/yosemite/atmel_read_eeprom.c b/arch/mips/pmc-sierra/yosemite/atmel_read_eeprom.c
index 97862f45496d..caf5e9a0acc7 100644
--- a/arch/mips/pmc-sierra/yosemite/atmel_read_eeprom.c
+++ b/arch/mips/pmc-sierra/yosemite/atmel_read_eeprom.c
@@ -148,7 +148,7 @@ int read_eeprom(char *buffer, int eeprom_size, int size)
 	send_byte(W_HEADER);
 	recv_ack();
 
-	/* EEPROM with size of more then 2K need two byte addressing */
+	/* EEPROM with size of more than 2K need two byte addressing */
 	if (eeprom_size > 2048) {
 		send_byte(0x00);
 		recv_ack();
diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c
index de79915452c8..b29005a5a8f5 100644
--- a/arch/powerpc/kernel/kprobes.c
+++ b/arch/powerpc/kernel/kprobes.c
@@ -316,7 +316,7 @@ static int __kprobes trampoline_probe_handler(struct kprobe *p,
 	/*
 	 * It is possible to have multiple instances associated with a given
 	 * task either because an multiple functions in the call path
-	 * have a return probe installed on them, and/or more then one return
+	 * have a return probe installed on them, and/or more than one return
 	 * return probe was registered for a target function.
 	 *
 	 * We can handle this because:
diff --git a/arch/powerpc/oprofile/cell/spu_profiler.c b/arch/powerpc/oprofile/cell/spu_profiler.c
index dd499c3e9da7..83faa958b9d4 100644
--- a/arch/powerpc/oprofile/cell/spu_profiler.c
+++ b/arch/powerpc/oprofile/cell/spu_profiler.c
@@ -49,7 +49,7 @@ void set_spu_profiling_frequency(unsigned int freq_khz, unsigned int cycles_rese
 	 * of precision.  This is close enough for the purpose at hand.
 	 *
 	 * The value of the timeout should be small enough that the hw
-	 * trace buffer will not get more then about 1/3 full for the
+	 * trace buffer will not get more than about 1/3 full for the
 	 * maximum user specified (the LFSR value) hw sampling frequency.
 	 * This is to ensure the trace buffer will never fill even if the
 	 * kernel thread scheduling varies under a heavy system load.
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 19577aeffd7b..a94a3c3ae932 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -299,7 +299,7 @@ config WARN_STACK
 	  This option enables the compiler options -mwarn-framesize and
 	  -mwarn-dynamicstack. If the compiler supports these options it
 	  will generate warnings for function which either use alloca or
-	  create a stack frame bigger then CONFIG_WARN_STACK_SIZE.
+	  create a stack frame bigger than CONFIG_WARN_STACK_SIZE.
 
 	  Say N if you are unsure.
 
diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c
index 569079ec4ff0..267f6698680a 100644
--- a/arch/s390/kernel/kprobes.c
+++ b/arch/s390/kernel/kprobes.c
@@ -381,7 +381,7 @@ static int __kprobes trampoline_probe_handler(struct kprobe *p,
 	/*
 	 * It is possible to have multiple instances associated with a given
 	 * task either because an multiple functions in the call path
-	 * have a return probe installed on them, and/or more then one return
+	 * have a return probe installed on them, and/or more than one return
 	 * return probe was registered for a target function.
 	 *
 	 * We can handle this because:
diff --git a/arch/sparc/kernel/kprobes.c b/arch/sparc/kernel/kprobes.c
index 201a6e547e4a..3bc6527c95af 100644
--- a/arch/sparc/kernel/kprobes.c
+++ b/arch/sparc/kernel/kprobes.c
@@ -517,7 +517,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 	/*
 	 * It is possible to have multiple instances associated with a given
 	 * task either because an multiple functions in the call path
-	 * have a return probe installed on them, and/or more then one return
+	 * have a return probe installed on them, and/or more than one return
 	 * return probe was registered for a target function.
 	 *
 	 * We can handle this because:
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 6c27679ec6aa..a116e6d5726c 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -694,7 +694,7 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs)
 	/*
 	 * It is possible to have multiple instances associated with a given
 	 * task either because multiple functions in the call path have
-	 * return probes installed on them, and/or more then one
+	 * return probes installed on them, and/or more than one
 	 * return probe was registered for a target function.
 	 *
 	 * We can handle this because:
diff --git a/arch/x86/kernel/mfgpt_32.c b/arch/x86/kernel/mfgpt_32.c
index c12314c9e86f..8815f3c7fec7 100644
--- a/arch/x86/kernel/mfgpt_32.c
+++ b/arch/x86/kernel/mfgpt_32.c
@@ -252,7 +252,7 @@ EXPORT_SYMBOL_GPL(geode_mfgpt_alloc_timer);
 /*
  * The MFPGT timers on the CS5536 provide us with suitable timers to use
  * as clock event sources - not as good as a HPET or APIC, but certainly
- * better then the PIT.  This isn't a general purpose MFGPT driver, but
+ * better than the PIT.  This isn't a general purpose MFGPT driver, but
  * a simplified one designed specifically to act as a clock event source.
  * For full details about the MFGPT, please consult the CS5536 data sheet.
  */
diff --git a/drivers/hwmon/fschmd.c b/drivers/hwmon/fschmd.c
index 967170368933..8b2d756595d9 100644
--- a/drivers/hwmon/fschmd.c
+++ b/drivers/hwmon/fschmd.c
@@ -75,7 +75,7 @@ static const u8 FSCHMD_REG_VOLT[3] = { 0x45, 0x42, 0x48 };
 
 /* minimum pwm at which the fan is driven (pwm can by increased depending on
    the temp. Notice that for the scy some fans share there minimum speed.
-   Also notice that with the scy the sensor order is different then with the
+   Also notice that with the scy the sensor order is different than with the
    other chips, this order was in the 2.4 driver and kept for consistency. */
 static const u8 FSCHMD_REG_FAN_MIN[5][6] = {
 	{ 0x55, 0x65 },					/* pos */
diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index a3c5af1d7ec0..de5263beab4a 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -367,7 +367,7 @@ int mlx4_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata)
 		if (err)
 			goto out;
 	} else {
-		/* Can't be smaller then the number of outstanding CQEs */
+		/* Can't be smaller than the number of outstanding CQEs */
 		outst_cqe = mlx4_ib_get_outstanding_cqes(cq);
 		if (entries < outst_cqe + 1) {
 			err = 0;
diff --git a/drivers/message/i2o/i2o_scsi.c b/drivers/message/i2o/i2o_scsi.c
index 1bcdbbb9e7d3..3d45817e6dcd 100644
--- a/drivers/message/i2o/i2o_scsi.c
+++ b/drivers/message/i2o/i2o_scsi.c
@@ -390,7 +390,7 @@ static int i2o_scsi_reply(struct i2o_controller *c, u32 m,
  *	@i2o_dev: the I2O device which was added
  *
  *	If a I2O device is added we catch the notification, because I2O classes
- *	other then SCSI peripheral will not be received through
+ *	other than SCSI peripheral will not be received through
  *	i2o_scsi_probe().
  */
 static void i2o_scsi_notify_device_add(struct i2o_device *i2o_dev)
diff --git a/drivers/mtd/devices/pmc551.c b/drivers/mtd/devices/pmc551.c
index d38bca64bb15..d2fd550f7e09 100644
--- a/drivers/mtd/devices/pmc551.c
+++ b/drivers/mtd/devices/pmc551.c
@@ -34,7 +34,7 @@
  *	aperture size, not the dram size, and the V370PDC supplies no
  *	other method for memory size discovery.  This problem is
  *	mostly only relevant when compiled as a module, as the
- *	unloading of the module with an aperture size smaller then
+ *	unloading of the module with an aperture size smaller than
  *	the ram will cause the driver to detect the onboard memory
  *	size to be equal to the aperture size when the module is
  *	reloaded.  Soooo, to help, the module supports an msize
diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 048a606cebde..25def348e5ba 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -717,7 +717,7 @@ write_error:
  * to the real data size, although the @buf buffer has to contain the
  * alignment. In all other cases, @len has to be aligned.
  *
- * It is prohibited to write more then once to logical eraseblocks of static
+ * It is prohibited to write more than once to logical eraseblocks of static
  * volumes. This function returns zero in case of success and a negative error
  * code in case of failure.
  */
diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c
index a74118c05745..fe81039f2a7c 100644
--- a/drivers/mtd/ubi/io.c
+++ b/drivers/mtd/ubi/io.c
@@ -465,7 +465,7 @@ out:
  * This function synchronously erases physical eraseblock @pnum. If @torture
  * flag is not zero, the physical eraseblock is checked by means of writing
  * different patterns to it and reading them back. If the torturing is enabled,
- * the physical eraseblock is erased more then once.
+ * the physical eraseblock is erased more than once.
  *
  * This function returns the number of erasures made in case of success, %-EIO
  * if the erasure failed or the torturing test failed, and other negative error
diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c
index 41d47e1cf15c..ecde202a5a12 100644
--- a/drivers/mtd/ubi/scan.c
+++ b/drivers/mtd/ubi/scan.c
@@ -478,7 +478,7 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si,
 			return 0;
 		} else {
 			/*
-			 * This logical eraseblock is older then the one found
+			 * This logical eraseblock is older than the one found
 			 * previously.
 			 */
 			if (cmp_res & 4)
diff --git a/drivers/mtd/ubi/ubi-media.h b/drivers/mtd/ubi/ubi-media.h
index 2ad940409053..8419fdccc79c 100644
--- a/drivers/mtd/ubi/ubi-media.h
+++ b/drivers/mtd/ubi/ubi-media.h
@@ -135,7 +135,7 @@ enum {
  * The erase counter header takes 64 bytes and has a plenty of unused space for
  * future usage. The unused fields are zeroed. The @version field is used to
  * indicate the version of UBI implementation which is supposed to be able to
- * work with this UBI image. If @version is greater then the current UBI
+ * work with this UBI image. If @version is greater than the current UBI
  * version, the image is rejected. This may be useful in future if something
  * is changed radically. This field is duplicated in the volume identifier
  * header.
@@ -187,7 +187,7 @@ struct ubi_ec_hdr {
  * (sequence number) is used to distinguish between older and newer versions of
  * logical eraseblocks.
  *
- * There are 2 situations when there may be more then one physical eraseblock
+ * There are 2 situations when there may be more than one physical eraseblock
  * corresponding to the same logical eraseblock, i.e., having the same @vol_id
  * and @lnum values in the volume identifier header. Suppose we have a logical
  * eraseblock L and it is mapped to the physical eraseblock P.
diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c
index 333c8941552f..1afc61e7455d 100644
--- a/drivers/mtd/ubi/vtbl.c
+++ b/drivers/mtd/ubi/vtbl.c
@@ -577,7 +577,7 @@ static int init_volumes(struct ubi_device *ubi, const struct ubi_scan_info *si,
 		if (vtbl[i].flags & UBI_VTBL_AUTORESIZE_FLG) {
 			/* Auto re-size flag may be set only for one volume */
 			if (ubi->autoresize_vol_id != -1) {
-				ubi_err("more then one auto-resize volume (%d "
+				ubi_err("more than one auto-resize volume (%d "
 					"and %d)", ubi->autoresize_vol_id, i);
 				kfree(vol);
 				return -EINVAL;
diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index 14901cb82c18..891534f8210d 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -128,7 +128,7 @@
  * situation when the picked physical eraseblock is constantly erased after the
  * data is written to it. So, we have a constant which limits the highest erase
  * counter of the free physical eraseblock to pick. Namely, the WL sub-system
- * does not pick eraseblocks with erase counter greater then the lowest erase
+ * does not pick eraseblocks with erase counter greater than the lowest erase
  * counter plus %WL_FREE_MAX_DIFF.
  */
 #define WL_FREE_MAX_DIFF (2*UBI_WL_THRESHOLD)
@@ -917,7 +917,7 @@ static int ensure_wear_leveling(struct ubi_device *ubi)
 		/*
 		 * We schedule wear-leveling only if the difference between the
 		 * lowest erase counter of used physical eraseblocks and a high
-		 * erase counter of free physical eraseblocks is greater then
+		 * erase counter of free physical eraseblocks is greater than
 		 * %UBI_WL_THRESHOLD.
 		 */
 		e1 = rb_entry(rb_first(&ubi->used), struct ubi_wl_entry, u.rb);
diff --git a/drivers/net/bnx2x_link.c b/drivers/net/bnx2x_link.c
index 67de94f1f30e..fefa6ab13064 100644
--- a/drivers/net/bnx2x_link.c
+++ b/drivers/net/bnx2x_link.c
@@ -3359,7 +3359,7 @@ static u8 bnx2x_format_ver(u32 num, u8 *str, u16 len)
 	u8 shift = 8*4;
 	u8 digit;
 	if (len < 10) {
-		/* Need more then 10chars for this format */
+		/* Need more than 10chars for this format */
 		*str_ptr = '\0';
 		return -EINVAL;
 	}
diff --git a/drivers/net/e1000/e1000_hw.c b/drivers/net/e1000/e1000_hw.c
index d04eef53571e..e1a3fc1303ee 100644
--- a/drivers/net/e1000/e1000_hw.c
+++ b/drivers/net/e1000/e1000_hw.c
@@ -6758,7 +6758,7 @@ static s32 e1000_get_cable_length(struct e1000_hw *hw, u16 *min_length,
  * returns: - E1000_ERR_XXX
  *            E1000_SUCCESS
  *
- * For phy's older then IGP, this function simply reads the polarity bit in the
+ * For phy's older than IGP, this function simply reads the polarity bit in the
  * Phy Status register.  For IGP phy's, this bit is valid only if link speed is
  * 10 Mbps.  If the link speed is 100 Mbps there is no polarity so this bit will
  * return 0.  If the link speed is 1000 Mbps the polarity status is in the
@@ -6834,7 +6834,7 @@ static s32 e1000_check_polarity(struct e1000_hw *hw,
  * returns: - E1000_ERR_XXX
  *            E1000_SUCCESS
  *
- * For phy's older then IGP, this function reads the Downshift bit in the Phy
+ * For phy's older than IGP, this function reads the Downshift bit in the Phy
  * Specific Status register.  For IGP phy's, it reads the Downgrade bit in the
  * Link Health register.  In IGP this bit is latched high, so the driver must
  * read it immediately after link is established.
diff --git a/drivers/net/slip.h b/drivers/net/slip.h
index 853e0f6ec710..9ea5c11287d2 100644
--- a/drivers/net/slip.h
+++ b/drivers/net/slip.h
@@ -75,7 +75,7 @@ struct slip {
   unsigned long         tx_errors;      /* Planned stuff                */
   unsigned long         rx_dropped;     /* No memory for skb            */
   unsigned long         tx_dropped;     /* When MTU change              */
-  unsigned long         rx_over_errors; /* Frame bigger then SLIP buf.  */
+  unsigned long         rx_over_errors; /* Frame bigger than SLIP buf.  */
 #ifdef SL_INCLUDE_CSLIP
   unsigned long		tx_compressed;
   unsigned long		rx_compressed;
diff --git a/drivers/net/tehuti.c b/drivers/net/tehuti.c
index a10a83a11d9f..a7a4dc4d6313 100644
--- a/drivers/net/tehuti.c
+++ b/drivers/net/tehuti.c
@@ -1004,7 +1004,7 @@ static inline void bdx_rxdb_free_elem(struct rxdb *db, int n)
  * skb for rx. It assumes that Rx is desabled in HW
  * funcs are grouped for better cache usage
  *
- * RxD fifo is smaller then RxF fifo by design. Upon high load, RxD will be
+ * RxD fifo is smaller than RxF fifo by design. Upon high load, RxD will be
  * filled and packets will be dropped by nic without getting into host or
  * cousing interrupt. Anyway, in that condition, host has no chance to proccess
  * all packets, but dropping in nic is cheaper, since it takes 0 cpu cycles
@@ -1826,7 +1826,7 @@ static void bdx_tx_free(struct bdx_priv *priv)
  *
  * Pushes desc to TxD fifo and overlaps it if needed.
  * NOTE: this func does not check for available space. this is responsibility
- *    of the caller. Neither does it check that data size is smaller then
+ *    of the caller. Neither does it check that data size is smaller than
  *    fifo size.
  */
 static void bdx_tx_push_desc(struct bdx_priv *priv, void *data, int size)
diff --git a/drivers/net/tokenring/smctr.c b/drivers/net/tokenring/smctr.c
index a011666342ff..50eb29ce3c87 100644
--- a/drivers/net/tokenring/smctr.c
+++ b/drivers/net/tokenring/smctr.c
@@ -3064,7 +3064,7 @@ static int smctr_load_node_addr(struct net_device *dev)
  * will consequently cause a timeout.
  *
  * NOTE 1: If the monitor_state is MS_BEACON_TEST_STATE, all transmit
- * queues other then the one used for the lobe_media_test should be
+ * queues other than the one used for the lobe_media_test should be
  * disabled.!?
  *
  * NOTE 2: If the monitor_state is MS_BEACON_TEST_STATE and the receive_mask
diff --git a/drivers/net/wireless/ipw2x00/ipw2100.c b/drivers/net/wireless/ipw2x00/ipw2100.c
index 1667065b86a7..753de1a9c4b3 100644
--- a/drivers/net/wireless/ipw2x00/ipw2100.c
+++ b/drivers/net/wireless/ipw2x00/ipw2100.c
@@ -1332,7 +1332,7 @@ static int ipw2100_power_cycle_adapter(struct ipw2100_priv *priv)
 		       IPW_AUX_HOST_RESET_REG_STOP_MASTER);
 
 	/* Step 2. Wait for stop Master Assert
-	 *         (not more then 50us, otherwise ret error */
+	 *         (not more than 50us, otherwise ret error */
 	i = 5;
 	do {
 		udelay(IPW_WAIT_RESET_MASTER_ASSERT_COMPLETE_DELAY);
diff --git a/drivers/net/wireless/rt2x00/rt2x00crypto.c b/drivers/net/wireless/rt2x00/rt2x00crypto.c
index 37ad0d2fb64c..aee9cba13eb3 100644
--- a/drivers/net/wireless/rt2x00/rt2x00crypto.c
+++ b/drivers/net/wireless/rt2x00/rt2x00crypto.c
@@ -184,8 +184,8 @@ void rt2x00crypto_rx_insert_iv(struct sk_buff *skb, unsigned int align,
 	 * Make room for new data, note that we increase both
 	 * headsize and tailsize when required. The tailsize is
 	 * only needed when ICV data needs to be inserted and
-	 * the padding is smaller then the ICV data.
-	 * When alignment requirements is greater then the
+	 * the padding is smaller than the ICV data.
+	 * When alignment requirements is greater than the
 	 * ICV data we must trim the skb to the correct size
 	 * because we need to remove the extra bytes.
 	 */
diff --git a/drivers/net/wireless/strip.c b/drivers/net/wireless/strip.c
index dd0de3a9ed4e..7015f2480550 100644
--- a/drivers/net/wireless/strip.c
+++ b/drivers/net/wireless/strip.c
@@ -236,7 +236,7 @@ struct strip {
 	unsigned long tx_errors;	/* Planned stuff                */
 	unsigned long rx_dropped;	/* No memory for skb            */
 	unsigned long tx_dropped;	/* When MTU change              */
-	unsigned long rx_over_errors;	/* Frame bigger then STRIP buf. */
+	unsigned long rx_over_errors;	/* Frame bigger than STRIP buf. */
 
 	unsigned long pps_timer;	/* Timer to determine pps       */
 	unsigned long rx_pps_count;	/* Counter to determine pps     */
diff --git a/drivers/s390/block/dasd_eer.c b/drivers/s390/block/dasd_eer.c
index 892e2878d61b..f8e05ce98621 100644
--- a/drivers/s390/block/dasd_eer.c
+++ b/drivers/s390/block/dasd_eer.c
@@ -535,8 +535,8 @@ static int dasd_eer_open(struct inode *inp, struct file *filp)
 	    eerb->buffer_page_count > INT_MAX / PAGE_SIZE) {
 		kfree(eerb);
 		MESSAGE(KERN_WARNING, "can't open device since module "
-			"parameter eer_pages is smaller then 1 or"
-			" bigger then %d", (int)(INT_MAX / PAGE_SIZE));
+			"parameter eer_pages is smaller than 1 or"
+			" bigger than %d", (int)(INT_MAX / PAGE_SIZE));
 		unlock_kernel();
 		return -EINVAL;
 	}
diff --git a/drivers/s390/char/vmlogrdr.c b/drivers/s390/char/vmlogrdr.c
index aabbeb909cc6..d8a2289fcb69 100644
--- a/drivers/s390/char/vmlogrdr.c
+++ b/drivers/s390/char/vmlogrdr.c
@@ -427,7 +427,7 @@ static int vmlogrdr_receive_data(struct vmlogrdr_priv_t *priv)
 			buffer = priv->buffer + sizeof(int);
 		}
 		/*
-		 * If the record is bigger then our buffer, we receive only
+		 * If the record is bigger than our buffer, we receive only
 		 * a part of it. We can get the rest later.
 		 */
 		if (iucv_data_count > NET_BUFFER_SIZE)
@@ -437,7 +437,7 @@ static int vmlogrdr_receive_data(struct vmlogrdr_priv_t *priv)
 					  0, buffer, iucv_data_count,
 					  &priv->residual_length);
 		spin_unlock_bh(&priv->priv_lock);
-		/* An rc of 5 indicates that the record was bigger then
+		/* An rc of 5 indicates that the record was bigger than
 		 * the buffer, which is OK for us. A 9 indicates that the
 		 * record was purged befor we could receive it.
 		 */
diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c
index 8c64494444bf..311ed6dea726 100644
--- a/drivers/scsi/lpfc/lpfc_hbadisc.c
+++ b/drivers/scsi/lpfc/lpfc_hbadisc.c
@@ -1964,10 +1964,10 @@ lpfc_set_disctmo(struct lpfc_vport *vport)
 	uint32_t tmo;
 
 	if (vport->port_state == LPFC_LOCAL_CFG_LINK) {
-		/* For FAN, timeout should be greater then edtov */
+		/* For FAN, timeout should be greater than edtov */
 		tmo = (((phba->fc_edtov + 999) / 1000) + 1);
 	} else {
-		/* Normal discovery timeout should be > then ELS/CT timeout
+		/* Normal discovery timeout should be > than ELS/CT timeout
 		 * FC spec states we need 3 * ratov for CT requests
 		 */
 		tmo = ((phba->fc_ratov * 3) + 3);
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 01dfdc8696f8..a36a120561e2 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -420,7 +420,7 @@ lpfc_sli_next_iocb_slot (struct lpfc_hba *phba, struct lpfc_sli_ring *pring)
 		if (unlikely(pring->local_getidx >= max_cmd_idx)) {
 			lpfc_printf_log(phba, KERN_ERR, LOG_SLI,
 					"0315 Ring %d issue: portCmdGet %d "
-					"is bigger then cmd ring %d\n",
+					"is bigger than cmd ring %d\n",
 					pring->ringno,
 					pring->local_getidx, max_cmd_idx);
 
@@ -1628,12 +1628,12 @@ lpfc_sli_rsp_pointers_error(struct lpfc_hba *phba, struct lpfc_sli_ring *pring)
 {
 	struct lpfc_pgp *pgp = &phba->port_gp[pring->ringno];
 	/*
-	 * Ring <ringno> handler: portRspPut <portRspPut> is bigger then
+	 * Ring <ringno> handler: portRspPut <portRspPut> is bigger than
 	 * rsp ring <portRspMax>
 	 */
 	lpfc_printf_log(phba, KERN_ERR, LOG_SLI,
 			"0312 Ring %d handler: portRspPut %d "
-			"is bigger then rsp ring %d\n",
+			"is bigger than rsp ring %d\n",
 			pring->ringno, le32_to_cpu(pgp->rspPutInx),
 			pring->numRiocb);
 
@@ -2083,12 +2083,12 @@ lpfc_sli_handle_slow_ring_event(struct lpfc_hba *phba,
 	portRspPut = le32_to_cpu(pgp->rspPutInx);
 	if (portRspPut >= portRspMax) {
 		/*
-		 * Ring <ringno> handler: portRspPut <portRspPut> is bigger then
+		 * Ring <ringno> handler: portRspPut <portRspPut> is bigger than
 		 * rsp ring <portRspMax>
 		 */
 		lpfc_printf_log(phba, KERN_ERR, LOG_SLI,
 				"0303 Ring %d handler: portRspPut %d "
-				"is bigger then rsp ring %d\n",
+				"is bigger than rsp ring %d\n",
 				pring->ringno, portRspPut, portRspMax);
 
 		phba->link_state = LPFC_HBA_ERROR;
diff --git a/drivers/serial/crisv10.c b/drivers/serial/crisv10.c
index 8b2c619a09f2..e642c22c80e2 100644
--- a/drivers/serial/crisv10.c
+++ b/drivers/serial/crisv10.c
@@ -1203,7 +1203,7 @@ static void e100_disable_txdma_channel(struct e100_serial *info)
 	unsigned long flags;
 
 	/* Disable output DMA channel for the serial port in question
-	 * ( set to something other then serialX)
+	 * ( set to something other than serialX)
 	 */
 	local_irq_save(flags);
 	DFLOW(DEBUG_LOG(info->line, "disable_txdma_channel %i\n", info->line));
@@ -1266,7 +1266,7 @@ static void e100_disable_rxdma_channel(struct e100_serial *info)
 	unsigned long flags;
 
 	/* Disable input DMA channel for the serial port in question
-	 * ( set to something other then serialX)
+	 * ( set to something other than serialX)
 	 */
 	local_irq_save(flags);
 	if (info->line == 0) {
diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c
index e6210725b9ab..d012edda6d11 100644
--- a/drivers/video/console/vgacon.c
+++ b/drivers/video/console/vgacon.c
@@ -1332,7 +1332,7 @@ static void vgacon_save_screen(struct vc_data *c)
 		c->vc_y = screen_info.orig_y;
 	}
 
-	/* We can't copy in more then the size of the video buffer,
+	/* We can't copy in more than the size of the video buffer,
 	 * or we'll be copying in VGA BIOS */
 
 	if (!vga_is_gfx)
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c
index 6ebaa58e2c03..04697ba7f73e 100644
--- a/fs/ocfs2/cluster/heartbeat.c
+++ b/fs/ocfs2/cluster/heartbeat.c
@@ -854,7 +854,7 @@ static int o2hb_thread(void *data)
 
 	while (!kthread_should_stop() && !reg->hr_unclean_stop) {
 		/* We track the time spent inside
-		 * o2hb_do_disk_heartbeat so that we avoid more then
+		 * o2hb_do_disk_heartbeat so that we avoid more than
 		 * hr_timeout_ms between disk writes. On busy systems
 		 * this should result in a heartbeat which is less
 		 * likely to time itself out. */
diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c
index 219bd79ea894..d4a8be32b902 100644
--- a/fs/proc/task_nommu.c
+++ b/fs/proc/task_nommu.c
@@ -9,7 +9,7 @@
 
 /*
  * Logic: we've got two memory sums for each process, "shared", and
- * "non-shared". Shared memory may get counted more then once, for
+ * "non-shared". Shared memory may get counted more than once, for
  * each process that owns it. Non-shared memory is counted
  * accurately.
  */
diff --git a/fs/ubifs/Kconfig b/fs/ubifs/Kconfig
index 91ceeda7e5bf..e35b54d5059d 100644
--- a/fs/ubifs/Kconfig
+++ b/fs/ubifs/Kconfig
@@ -40,7 +40,7 @@ config UBIFS_FS_ZLIB
 	depends on UBIFS_FS
 	default y
 	help
-	  Zlib copresses better then LZO but it is slower. Say 'Y' if unsure.
+	  Zlib compresses better than LZO but it is slower. Say 'Y' if unsure.
 
 # Debugging-related stuff
 config UBIFS_FS_DEBUG
diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c
index 0e5e54d82924..175f9c590b77 100644
--- a/fs/ubifs/budget.c
+++ b/fs/ubifs/budget.c
@@ -142,7 +142,7 @@ static long long get_liability(struct ubifs_info *c)
  *
  * This function is called when an operation cannot be budgeted because there
  * is supposedly no free space. But in most cases there is some free space:
- *   o budgeting is pessimistic, so it always budgets more then it is actually
+ *   o budgeting is pessimistic, so it always budgets more than it is actually
  *     needed, so shrinking the liability is one way to make free space - the
  *     cached data will take less space then it was budgeted for;
  *   o GC may turn some dark space into free space (budgeting treats dark space
@@ -606,7 +606,7 @@ void ubifs_release_budget(struct ubifs_info *c, struct ubifs_budget_req *req)
  * @c: UBIFS file-system description object
  *
  * This function converts budget which was allocated for a new page of data to
- * the budget of changing an existing page of data. The latter is smaller then
+ * the budget of changing an existing page of data. The latter is smaller than
  * the former, so this function only does simple re-calculation and does not
  * involve any write-back.
  */
diff --git a/fs/ubifs/gc.c b/fs/ubifs/gc.c
index 0bef6501d58a..9832f9abe28e 100644
--- a/fs/ubifs/gc.c
+++ b/fs/ubifs/gc.c
@@ -45,7 +45,7 @@
 #define SMALL_NODE_WM  UBIFS_MAX_DENT_NODE_SZ
 
 /*
- * GC may need to move more then one LEB to make progress. The below constants
+ * GC may need to move more than one LEB to make progress. The below constants
  * define "soft" and "hard" limits on the number of LEBs the garbage collector
  * may move.
  */
diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c
index 10ae25b7d1db..9b7c54e0cd2a 100644
--- a/fs/ubifs/journal.c
+++ b/fs/ubifs/journal.c
@@ -191,7 +191,7 @@ again:
 	if (wbuf->lnum != -1 && avail >= len) {
 		/*
 		 * Someone else has switched the journal head and we have
-		 * enough space now. This happens when more then one process is
+		 * enough space now. This happens when more than one process is
 		 * trying to write to the same journal head at the same time.
 		 */
 		dbg_jnl("return LEB %d back, already have LEB %d:%d",
diff --git a/fs/ubifs/shrinker.c b/fs/ubifs/shrinker.c
index f248533841a2..e7bab52a1410 100644
--- a/fs/ubifs/shrinker.c
+++ b/fs/ubifs/shrinker.c
@@ -151,7 +151,7 @@ static int shrink_tnc(struct ubifs_info *c, int nr, int age, int *contention)
  * @contention: if any contention, this is set to %1
  *
  * This function walks the list of mounted UBIFS file-systems and frees clean
- * znodes which are older then @age, until at least @nr znodes are freed.
+ * znodes which are older than @age, until at least @nr znodes are freed.
  * Returns the number of freed znodes.
  */
 static int shrink_tnc_trees(int nr, int age, int *contention)
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 36f6cc703ef2..be846d606ae8 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -1348,7 +1348,7 @@ xfs_finish_flags(
 {
 	int			ronly = (mp->m_flags & XFS_MOUNT_RDONLY);
 
-	/* Fail a mount where the logbuf is smaller then the log stripe */
+	/* Fail a mount where the logbuf is smaller than the log stripe */
 	if (xfs_sb_version_haslogv2(&mp->m_sb)) {
 		if (mp->m_logbsize <= 0 &&
 		    mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE) {
diff --git a/include/linux/mtd/mtd.h b/include/linux/mtd/mtd.h
index eae26bb6430a..64433eb411d7 100644
--- a/include/linux/mtd/mtd.h
+++ b/include/linux/mtd/mtd.h
@@ -83,7 +83,7 @@ typedef enum {
  * @datbuf:	data buffer - if NULL only oob data are read/written
  * @oobbuf:	oob data buffer
  *
- * Note, it is allowed to read more then one OOB area at one go, but not write.
+ * Note, it is allowed to read more than one OOB area at one go, but not write.
  * The interface assumes that the OOB write requests program only one page's
  * OOB area.
  */
diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h
index 82229317753d..68bb1c501d0d 100644
--- a/include/linux/spi/spi.h
+++ b/include/linux/spi/spi.h
@@ -327,9 +327,9 @@ extern struct spi_master *spi_busnum_to_master(u16 busnum);
  * @tx_dma: DMA address of tx_buf, if @spi_message.is_dma_mapped
  * @rx_dma: DMA address of rx_buf, if @spi_message.is_dma_mapped
  * @len: size of rx and tx buffers (in bytes)
- * @speed_hz: Select a speed other then the device default for this
+ * @speed_hz: Select a speed other than the device default for this
  *      transfer. If 0 the default (from @spi_device) is used.
- * @bits_per_word: select a bits_per_word other then the device default
+ * @bits_per_word: select a bits_per_word other than the device default
  *      for this transfer. If 0 the default (from @spi_device) is used.
  * @cs_change: affects chipselect after this transfer completes
  * @delay_usecs: microseconds to delay after this transfer before
diff --git a/include/mtd/ubi-user.h b/include/mtd/ubi-user.h
index ccdc562e444e..2dc2eb2b8e22 100644
--- a/include/mtd/ubi-user.h
+++ b/include/mtd/ubi-user.h
@@ -253,7 +253,7 @@ struct ubi_mkvol_req {
  *
  * Re-sizing is possible for both dynamic and static volumes. But while dynamic
  * volumes may be re-sized arbitrarily, static volumes cannot be made to be
- * smaller then the number of bytes they bear. To arbitrarily shrink a static
+ * smaller than the number of bytes they bear. To arbitrarily shrink a static
  * volume, it must be wiped out first (by means of volume update operation with
  * zero number of bytes).
  */
diff --git a/kernel/pid.c b/kernel/pid.c
index 064e76afa507..af9224cdd6c0 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -475,7 +475,7 @@ pid_t task_session_nr_ns(struct task_struct *tsk, struct pid_namespace *ns)
 EXPORT_SYMBOL(task_session_nr_ns);
 
 /*
- * Used by proc to find the first pid that is greater then or equal to nr.
+ * Used by proc to find the first pid that is greater than or equal to nr.
  *
  * If there is a pid at nr this function is exactly the same as find_pid_ns.
  */
diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c
index 1ca99557e929..06f197560f3b 100644
--- a/kernel/time/jiffies.c
+++ b/kernel/time/jiffies.c
@@ -45,7 +45,7 @@
  *
  * The value 8 is somewhat carefully chosen, as anything
  * larger can result in overflows. NSEC_PER_JIFFY grows as
- * HZ shrinks, so values greater then 8 overflow 32bits when
+ * HZ shrinks, so values greater than 8 overflow 32bits when
  * HZ=100.
  */
 #define JIFFIES_SHIFT	8
diff --git a/net/sctp/auth.c b/net/sctp/auth.c
index 52db5f60daa0..20c576f530fa 100644
--- a/net/sctp/auth.c
+++ b/net/sctp/auth.c
@@ -141,8 +141,8 @@ void sctp_auth_destroy_keys(struct list_head *keys)
 /* Compare two byte vectors as numbers.  Return values
  * are:
  * 	  0 - vectors are equal
- * 	< 0 - vector 1 is smaller then vector2
- * 	> 0 - vector 1 is greater then vector2
+ * 	< 0 - vector 1 is smaller than vector2
+ * 	> 0 - vector 1 is greater than vector2
  *
  * Algorithm is:
  * 	This is performed by selecting the numerically smaller key vector...
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 1c4e5d6c29c0..3a0cd075914f 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -4268,9 +4268,9 @@ nomem:
 
 /*
  * Handle a protocol violation when the chunk length is invalid.
- * "Invalid" length is identified as smaller then the minimal length a
+ * "Invalid" length is identified as smaller than the minimal length a
  * given chunk can be.  For example, a SACK chunk has invalid length
- * if it's length is set to be smaller then the size of sctp_sack_chunk_t.
+ * if its length is set to be smaller than the size of sctp_sack_chunk_t.
  *
  * We inform the other end by sending an ABORT with a Protocol Violation
  * error code.
@@ -4300,7 +4300,7 @@ static sctp_disposition_t sctp_sf_violation_chunklen(
 
 /*
  * Handle a protocol violation when the parameter length is invalid.
- * "Invalid" length is identified as smaller then the minimal length a
+ * "Invalid" length is identified as smaller than the minimal length a
  * given parameter can be.
  */
 static sctp_disposition_t sctp_sf_violation_paramlen(
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index b14a8f33e42d..ff0a8f88de04 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -2717,7 +2717,7 @@ static int sctp_setsockopt_associnfo(struct sock *sk, char __user *optval, int o
 				paths++;
 			}
 
-			/* Only validate asocmaxrxt if we have more then
+			/* Only validate asocmaxrxt if we have more than
 			 * one path/transport.  We do this because path
 			 * retransmissions are only counted when we have more
 			 * then one path.
diff --git a/net/sctp/tsnmap.c b/net/sctp/tsnmap.c
index 35c73e82553a..9bd64565021a 100644
--- a/net/sctp/tsnmap.c
+++ b/net/sctp/tsnmap.c
@@ -227,7 +227,7 @@ void sctp_tsnmap_skip(struct sctp_tsnmap *map, __u32 tsn)
 		 */
 		bitmap_zero(map->tsn_map, map->len);
 	} else {
-		/* If the gap is smaller then the map size,
+		/* If the gap is smaller than the map size,
 		 * shift the map by 'gap' bits and update further.
 		 */
 		bitmap_shift_right(map->tsn_map, map->tsn_map, gap, map->len);
diff --git a/sound/usb/usx2y/usbusx2y.c b/sound/usb/usx2y/usbusx2y.c
index ca26c532e77e..11639bd72a51 100644
--- a/sound/usb/usx2y/usbusx2y.c
+++ b/sound/usb/usx2y/usbusx2y.c
@@ -238,7 +238,7 @@ static void i_usX2Y_In04Int(struct urb *urb)
 					send = 0;
 				for (j = 0; j < URBS_AsyncSeq  &&  !err; ++j)
 					if (0 == usX2Y->AS04.urb[j]->status) {
-						struct us428_p4out *p4out = us428ctls->p4out + send;	// FIXME if more then 1 p4out is new, 1 gets lost.
+						struct us428_p4out *p4out = us428ctls->p4out + send;	// FIXME if more than 1 p4out is new, 1 gets lost.
 						usb_fill_bulk_urb(usX2Y->AS04.urb[j], usX2Y->chip.dev,
 								  usb_sndbulkpipe(usX2Y->chip.dev, 0x04), &p4out->val.vol, 
 								  p4out->type == eLT_Light ? sizeof(struct us428_lights) : 5,
-- 
cgit v1.2.3


From 0211a9c8508b2183e0e539509aad60414f1c3813 Mon Sep 17 00:00:00 2001
From: Frederik Schwarzer <schwarzerf@gmail.com>
Date: Mon, 29 Dec 2008 22:14:56 +0100
Subject: trivial: fix an -> a typos in documentation and comments

It is always "an" if there is a vowel _spoken_ (not written).
So it is:
"an hour" (spoken vowel)
but
"a uniform" (spoken 'j')

Signed-off-by: Frederik Schwarzer <schwarzerf@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
---
 Documentation/dell_rbu.txt              | 4 ++--
 Documentation/laptops/thinkpad-acpi.txt | 2 +-
 Documentation/networking/tuntap.txt     | 2 +-
 arch/m68k/kernel/traps.c                | 2 +-
 drivers/acpi/executer/exprep.c          | 2 +-
 drivers/acpi/executer/exresolv.c        | 2 +-
 drivers/acpi/executer/exstore.c         | 2 +-
 drivers/acpi/resources/rscreate.c       | 2 +-
 drivers/acpi/utilities/utobject.c       | 4 ++--
 drivers/char/epca.c                     | 2 +-
 drivers/cpufreq/Kconfig                 | 4 ++--
 drivers/input/keyboard/atkbd.c          | 2 +-
 drivers/macintosh/Kconfig               | 2 +-
 drivers/misc/phantom.c                  | 2 +-
 fs/ncpfs/ioctl.c                        | 2 +-
 include/acpi/acmacros.h                 | 4 ++--
 include/acpi/actypes.h                  | 2 +-
 include/linux/ncp_fs.h                  | 2 +-
 mm/slub.c                               | 2 +-
 sound/oss/aedsp16.c                     | 2 +-
 20 files changed, 24 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/dell_rbu.txt b/Documentation/dell_rbu.txt
index 2c0d631de0cf..c11b931f8f98 100644
--- a/Documentation/dell_rbu.txt
+++ b/Documentation/dell_rbu.txt
@@ -81,8 +81,8 @@ Until this step is completed the driver cannot be unloaded.
 Also echoing either mono ,packet or init in to image_type will free up the
 memory allocated by the driver.
 
-If an user by accident executes steps 1 and 3 above without executing step 2;
-it will make the /sys/class/firmware/dell_rbu/ entries to disappear.
+If a user by accident executes steps 1 and 3 above without executing step 2;
+it will make the /sys/class/firmware/dell_rbu/ entries disappear.
 The entries can be recreated by doing the following
 echo init > /sys/devices/platform/dell_rbu/image_type
 NOTE: echoing init in image_type does not change it original value.
diff --git a/Documentation/laptops/thinkpad-acpi.txt b/Documentation/laptops/thinkpad-acpi.txt
index 71f0fe1fc1b0..898b4987bb80 100644
--- a/Documentation/laptops/thinkpad-acpi.txt
+++ b/Documentation/laptops/thinkpad-acpi.txt
@@ -1475,7 +1475,7 @@ Sysfs interface changelog:
 
 0x020100:	Marker for thinkpad-acpi with hot key NVRAM polling
 		support.  If you must, use it to know you should not
-		start an userspace NVRAM poller (allows to detect when
+		start a userspace NVRAM poller (allows to detect when
 		NVRAM is compiled out by the user because it is
 		unneeded/undesired in the first place).
 0x020101:	Marker for thinkpad-acpi with hot key NVRAM polling
diff --git a/Documentation/networking/tuntap.txt b/Documentation/networking/tuntap.txt
index 839cbb71388b..c0aab985bad9 100644
--- a/Documentation/networking/tuntap.txt
+++ b/Documentation/networking/tuntap.txt
@@ -118,7 +118,7 @@ As mentioned above, main purpose of TUN/TAP driver is tunneling.
 It is used by VTun (http://vtun.sourceforge.net).
 
 Another interesting application using TUN/TAP is pipsecd
-(http://perso.enst.fr/~beyssac/pipsec/), an userspace IPSec
+(http://perso.enst.fr/~beyssac/pipsec/), a userspace IPSec
 implementation that can use complete kernel routing (unlike FreeS/WAN).
 
 3. How does Virtual network device actually work ? 
diff --git a/arch/m68k/kernel/traps.c b/arch/m68k/kernel/traps.c
index 6d813de2baf1..184acc90808d 100644
--- a/arch/m68k/kernel/traps.c
+++ b/arch/m68k/kernel/traps.c
@@ -401,7 +401,7 @@ static inline void do_040writebacks(struct frame *fp)
  * called from sigreturn(), must ensure userspace code didn't
  * manipulate exception frame to circumvent protection, then complete
  * pending writebacks
- * we just clear TM2 to turn it into an userspace access
+ * we just clear TM2 to turn it into a userspace access
  */
 asmlinkage void berr_040cleanup(struct frame *fp)
 {
diff --git a/drivers/acpi/executer/exprep.c b/drivers/acpi/executer/exprep.c
index 5d438c32989d..a7dc87ecee37 100644
--- a/drivers/acpi/executer/exprep.c
+++ b/drivers/acpi/executer/exprep.c
@@ -404,7 +404,7 @@ acpi_ex_prep_common_field_object(union acpi_operand_object *obj_desc,
  *
  * RETURN:      Status
  *
- * DESCRIPTION: Construct an union acpi_operand_object of type def_field and
+ * DESCRIPTION: Construct a union acpi_operand_object of type def_field and
  *              connect it to the parent Node.
  *
  ******************************************************************************/
diff --git a/drivers/acpi/executer/exresolv.c b/drivers/acpi/executer/exresolv.c
index 89571b92a522..60e8c47128e9 100644
--- a/drivers/acpi/executer/exresolv.c
+++ b/drivers/acpi/executer/exresolv.c
@@ -146,7 +146,7 @@ acpi_ex_resolve_object_to_value(union acpi_operand_object **stack_ptr,
 
 	stack_desc = *stack_ptr;
 
-	/* This is an union acpi_operand_object    */
+	/* This is a union acpi_operand_object    */
 
 	switch (ACPI_GET_OBJECT_TYPE(stack_desc)) {
 	case ACPI_TYPE_LOCAL_REFERENCE:
diff --git a/drivers/acpi/executer/exstore.c b/drivers/acpi/executer/exstore.c
index 3318df4cbd98..1c118ba78adb 100644
--- a/drivers/acpi/executer/exstore.c
+++ b/drivers/acpi/executer/exstore.c
@@ -274,7 +274,7 @@ acpi_ex_do_debug_object(union acpi_operand_object *source_desc,
  *
  * PARAMETERS:  *source_desc        - Value to be stored
  *              *dest_desc          - Where to store it.  Must be an NS node
- *                                    or an union acpi_operand_object of type
+ *                                    or a union acpi_operand_object of type
  *                                    Reference;
  *              walk_state          - Current walk state
  *
diff --git a/drivers/acpi/resources/rscreate.c b/drivers/acpi/resources/rscreate.c
index c0bbfa2c4193..08b8d73e6ee5 100644
--- a/drivers/acpi/resources/rscreate.c
+++ b/drivers/acpi/resources/rscreate.c
@@ -124,7 +124,7 @@ acpi_rs_create_resource_list(union acpi_operand_object *aml_buffer,
  *
  * FUNCTION:    acpi_rs_create_pci_routing_table
  *
- * PARAMETERS:  package_object          - Pointer to an union acpi_operand_object
+ * PARAMETERS:  package_object          - Pointer to a union acpi_operand_object
  *                                        package
  *              output_buffer           - Pointer to the user's buffer
  *
diff --git a/drivers/acpi/utilities/utobject.c b/drivers/acpi/utilities/utobject.c
index c354e7a42bcd..4bef3cfbaccb 100644
--- a/drivers/acpi/utilities/utobject.c
+++ b/drivers/acpi/utilities/utobject.c
@@ -297,7 +297,7 @@ union acpi_operand_object *acpi_ut_create_string_object(acpi_size string_size)
  *
  * RETURN:      TRUE if object is valid, FALSE otherwise
  *
- * DESCRIPTION: Validate a pointer to be an union acpi_operand_object
+ * DESCRIPTION: Validate a pointer to be a union acpi_operand_object
  *
  ******************************************************************************/
 
@@ -389,7 +389,7 @@ void acpi_ut_delete_object_desc(union acpi_operand_object *object)
 {
 	ACPI_FUNCTION_TRACE_PTR(ut_delete_object_desc, object);
 
-	/* Object must be an union acpi_operand_object    */
+	/* Object must be a union acpi_operand_object    */
 
 	if (ACPI_GET_DESCRIPTOR_TYPE(object) != ACPI_DESC_TYPE_OPERAND) {
 		ACPI_ERROR((AE_INFO,
diff --git a/drivers/char/epca.c b/drivers/char/epca.c
index 39ad820b2350..af7c13ca9493 100644
--- a/drivers/char/epca.c
+++ b/drivers/char/epca.c
@@ -769,7 +769,7 @@ static int pc_open(struct tty_struct *tty, struct file *filp)
 	/* Check status of board configured in system.  */
 
 	/*
-	 * I check to see if the epca_setup routine detected an user error. It
+	 * I check to see if the epca_setup routine detected a user error. It
 	 * might be better to put this in pc_init, but for the moment it goes
 	 * here.
 	 */
diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig
index 5f076aef74fa..a8c8d9c19d74 100644
--- a/drivers/cpufreq/Kconfig
+++ b/drivers/cpufreq/Kconfig
@@ -83,7 +83,7 @@ config CPU_FREQ_DEFAULT_GOV_USERSPACE
 	select CPU_FREQ_GOV_USERSPACE
 	help
 	  Use the CPUFreq governor 'userspace' as default. This allows
-	  you to set the CPU frequency manually or when an userspace 
+	  you to set the CPU frequency manually or when a userspace 
 	  program shall be able to set the CPU dynamically without having
 	  to enable the userspace governor manually.
 
@@ -138,7 +138,7 @@ config CPU_FREQ_GOV_USERSPACE
 	tristate "'userspace' governor for userspace frequency scaling"
 	help
 	  Enable this cpufreq governor when you either want to set the
-	  CPU frequency manually or when an userspace program shall
+	  CPU frequency manually or when a userspace program shall
 	  be able to set the CPU dynamically, like on LART 
 	  <http://www.lartmaker.nl/>.
 
diff --git a/drivers/input/keyboard/atkbd.c b/drivers/input/keyboard/atkbd.c
index 379b7ff354ec..b9e6bef594ac 100644
--- a/drivers/input/keyboard/atkbd.c
+++ b/drivers/input/keyboard/atkbd.c
@@ -65,7 +65,7 @@ MODULE_PARM_DESC(extra, "Enable extra LEDs and keys on IBM RapidAcces, EzKey and
 
 /*
  * Scancode to keycode tables. These are just the default setting, and
- * are loadable via an userland utility.
+ * are loadable via a userland utility.
  */
 
 static const unsigned short atkbd_set2_keycode[512] = {
diff --git a/drivers/macintosh/Kconfig b/drivers/macintosh/Kconfig
index b52659620d50..173cf55c64d0 100644
--- a/drivers/macintosh/Kconfig
+++ b/drivers/macintosh/Kconfig
@@ -138,7 +138,7 @@ config PMAC_BACKLIGHT
 	  Say Y here to enable Macintosh specific extensions of the generic
 	  backlight code. With this enabled, the brightness keys on older
 	  PowerBooks will be enabled so you can change the screen brightness.
-	  Newer models should use an userspace daemon like pbbuttonsd.
+	  Newer models should use a userspace daemon like pbbuttonsd.
 
 config PMAC_BACKLIGHT_LEGACY
 	bool "Provide legacy ioctl's on /dev/pmu for the backlight"
diff --git a/drivers/misc/phantom.c b/drivers/misc/phantom.c
index abdebe347383..fa57b67593ae 100644
--- a/drivers/misc/phantom.c
+++ b/drivers/misc/phantom.c
@@ -6,7 +6,7 @@
  *  the Free Software Foundation; either version 2 of the License, or
  *  (at your option) any later version.
  *
- *  You need an userspace library to cooperate with this driver. It (and other
+ *  You need a userspace library to cooperate with this driver. It (and other
  *  info) may be obtained here:
  *  http://www.fi.muni.cz/~xslaby/phantom.html
  *  or alternatively, you might use OpenHaptics provided by Sensable.
diff --git a/fs/ncpfs/ioctl.c b/fs/ncpfs/ioctl.c
index 6d04e050c74e..f54360f50a9c 100644
--- a/fs/ncpfs/ioctl.c
+++ b/fs/ncpfs/ioctl.c
@@ -98,7 +98,7 @@ struct compat_ncp_objectname_ioctl
 {
 	s32		auth_type;
 	u32		object_name_len;
-	compat_caddr_t	object_name;	/* an userspace data, in most cases user name */
+	compat_caddr_t	object_name;	/* a userspace data, in most cases user name */
 };
 
 struct compat_ncp_fs_info_v2 {
diff --git a/include/acpi/acmacros.h b/include/acpi/acmacros.h
index a597207e2835..1954c9d1d012 100644
--- a/include/acpi/acmacros.h
+++ b/include/acpi/acmacros.h
@@ -333,8 +333,8 @@ struct acpi_integer_overlay {
 #define ACPI_INSERT_BITS(target, mask, source)          target = ((target & (~(mask))) | (source & mask))
 
 /*
- * An struct acpi_namespace_node can appear in some contexts
- * where a pointer to an union acpi_operand_object can also
+ * A struct acpi_namespace_node can appear in some contexts
+ * where a pointer to a union acpi_operand_object can also
  * appear. This macro is used to distinguish them.
  *
  * The "Descriptor" field is the first field in both structures.
diff --git a/include/acpi/actypes.h b/include/acpi/actypes.h
index 7220361790b3..8222e8de0d1c 100644
--- a/include/acpi/actypes.h
+++ b/include/acpi/actypes.h
@@ -467,7 +467,7 @@ typedef u32 acpi_object_type;
 
 /*
  * These are special object types that never appear in
- * a Namespace node, only in an union acpi_operand_object
+ * a Namespace node, only in a union acpi_operand_object
  */
 #define ACPI_TYPE_LOCAL_EXTRA           0x1C
 #define ACPI_TYPE_LOCAL_DATA            0x1D
diff --git a/include/linux/ncp_fs.h b/include/linux/ncp_fs.h
index 9f2d76347f19..f69e66d151cc 100644
--- a/include/linux/ncp_fs.h
+++ b/include/linux/ncp_fs.h
@@ -87,7 +87,7 @@ struct ncp_objectname_ioctl
 #define NCP_AUTH_NDS	0x32
 	int		auth_type;
 	size_t		object_name_len;
-	void __user *	object_name;	/* an userspace data, in most cases user name */
+	void __user *	object_name;	/* a userspace data, in most cases user name */
 };
 
 struct ncp_privatedata_ioctl
diff --git a/mm/slub.c b/mm/slub.c
index f0e2892fe403..6392ae5cc6b1 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -2254,7 +2254,7 @@ static int calculate_sizes(struct kmem_cache *s, int forced_order)
 		 * Add some empty padding so that we can catch
 		 * overwrites from earlier objects rather than let
 		 * tracking information or the free pointer be
-		 * corrupted if an user writes before the start
+		 * corrupted if a user writes before the start
 		 * of the object.
 		 */
 		size += sizeof(void *);
diff --git a/sound/oss/aedsp16.c b/sound/oss/aedsp16.c
index a0274f3dac08..3ee9900ffd7b 100644
--- a/sound/oss/aedsp16.c
+++ b/sound/oss/aedsp16.c
@@ -157,7 +157,7 @@
 
    Started Fri Mar 17 16:13:18 MET 1995
 
-   v0.1 (ALPHA, was an user-level program called AudioExcelDSP16.c)
+   v0.1 (ALPHA, was a user-level program called AudioExcelDSP16.c)
    - Initial code.
    v0.2 (ALPHA)
    - Cleanups.
-- 
cgit v1.2.3


From bd53cbcce501b61921a1af2dddfa87c5b9923dfd Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:48 +0100
Subject: ide: add ->cur_port to struct ide_host and use it for serialized
 hosts

* Pass 'ide_hwif_t *' instead of 'ide_hwgroup_t *' to unexpected_intr().

* Cache pointer to the port currently being serviced in ->cur_port
  and use it instead of hwif->hwgroup on serialized hosts.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c | 25 +++++++++++++++----------
 include/linux/ide.h  |  1 +
 2 files changed, 16 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 1d63159fc97d..43bf43d802c3 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -726,7 +726,7 @@ void do_ide_request(struct request_queue *q)
 	if (!ide_lock_hwgroup(hwgroup)) {
 		ide_hwif_t *prev_port;
 repeat:
-		prev_port = hwgroup->hwif;
+		prev_port = hwif->host->cur_port;
 		hwgroup->rq = NULL;
 
 		if (drive->dev_flags & IDE_DFLAG_SLEEPING) {
@@ -736,15 +736,17 @@ repeat:
 			}
 		}
 
-		if (hwif != prev_port) {
+		if ((hwif->host->host_flags & IDE_HFLAG_SERIALIZE) &&
+		    hwif != prev_port) {
 			/*
 			 * set nIEN for previous port, drives in the
 			 * quirk_list may not like intr setups/cleanups
 			 */
-			if (hwgroup->drive->quirk_list == 0)
+			if (prev_port && hwgroup->drive->quirk_list == 0)
 				prev_port->tp_ops->set_irq(prev_port, 0);
+
+			hwif->host->cur_port = hwif;
 		}
-		hwgroup->hwif = hwif;
 		hwgroup->drive = drive;
 		drive->dev_flags &= ~(IDE_DFLAG_SLEEPING | IDE_DFLAG_PARKED);
 
@@ -976,7 +978,7 @@ void ide_timer_expiry (unsigned long data)
 /**
  *	unexpected_intr		-	handle an unexpected IDE interrupt
  *	@irq: interrupt line
- *	@hwgroup: hwgroup being processed
+ *	@hwif: port being processed
  *
  *	There's nothing really useful we can do with an unexpected interrupt,
  *	other than reading the status register (to clear it), and logging it.
@@ -1005,11 +1007,11 @@ void ide_timer_expiry (unsigned long data)
  *	is doing the current command, but we don't know which hwif burped
  *	mysteriously.
  */
- 
-static void unexpected_intr (int irq, ide_hwgroup_t *hwgroup)
+
+static void unexpected_intr(int irq, ide_hwif_t *hwif)
 {
+	ide_hwgroup_t *hwgroup = hwif->hwgroup;
 	u8 stat;
-	ide_hwif_t *hwif = hwgroup->hwif;
 
 	/*
 	 * handle the unexpected interrupt
@@ -1044,7 +1046,7 @@ static void unexpected_intr (int irq, ide_hwgroup_t *hwgroup)
  *	not need to override it. If you do be aware it is subtle in
  *	places
  *
- *	hwgroup->hwif is the interface in the group currently performing
+ *	hwif is the interface in the group currently performing
  *	a command. hwgroup->drive is the drive and hwgroup->handler is
  *	the IRQ handler to call. As we issue a command the handlers
  *	step through multiple states, reassigning the handler to the
@@ -1070,6 +1072,9 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 	irqreturn_t irq_ret = IRQ_NONE;
 	int plug_device = 0;
 
+	if (hwif->host->host_flags & IDE_HFLAG_SERIALIZE)
+		hwif = hwif->host->cur_port;
+
 	spin_lock_irqsave(&hwgroup->lock, flags);
 
 	if (!ide_ack_intr(hwif))
@@ -1099,7 +1104,7 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 			 * Probably not a shared PCI interrupt,
 			 * so we can safely try to do something about it:
 			 */
-			unexpected_intr(irq, hwgroup);
+			unexpected_intr(irq, hwif);
 #ifdef CONFIG_BLK_DEV_IDEPCI
 		} else {
 			/*
diff --git a/include/linux/ide.h b/include/linux/ide.h
index db5ef8ae1ab9..3de13df8bcef 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -852,6 +852,7 @@ struct ide_host {
 	unsigned int	(*init_chipset)(struct pci_dev *);
 	unsigned long	host_flags;
 	void		*host_priv;
+	ide_hwif_t	*cur_port;	/* for hosts requiring serialization */
 };
 
 /*
-- 
cgit v1.2.3


From ae86afaee6a1c77c7a06d81dcc3bf872204d3bec Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:48 +0100
Subject: ide: use per-port IRQ handlers

Use hwif instead of hwgroup as {request,free}_irq()'s cookie,
teach ide_intr() to return early for non-active serialized ports,
modify unexpected_intr() accordingly and then use per-port IRQ
handlers instead of per-hwgroup ones.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c    | 58 ++++++++++++++++++++-----------------------------
 drivers/ide/ide-probe.c | 21 +++++++-----------
 drivers/ide/ide.c       | 17 +--------------
 include/linux/ide.h     |  4 ++--
 4 files changed, 34 insertions(+), 66 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 43bf43d802c3..1fc739f44154 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -723,7 +723,7 @@ void do_ide_request(struct request_queue *q)
 	spin_unlock_irq(q->queue_lock);
 	spin_lock_irq(&hwgroup->lock);
 
-	if (!ide_lock_hwgroup(hwgroup)) {
+	if (!ide_lock_hwgroup(hwgroup, hwif)) {
 		ide_hwif_t *prev_port;
 repeat:
 		prev_port = hwif->host->cur_port;
@@ -1002,44 +1002,30 @@ void ide_timer_expiry (unsigned long data)
  *	before completing the issuance of any new drive command, so we will not
  *	be accidentally invoked as a result of any valid command completion
  *	interrupt.
- *
- *	Note that we must walk the entire hwgroup here. We know which hwif
- *	is doing the current command, but we don't know which hwif burped
- *	mysteriously.
  */
 
 static void unexpected_intr(int irq, ide_hwif_t *hwif)
 {
-	ide_hwgroup_t *hwgroup = hwif->hwgroup;
-	u8 stat;
-
-	/*
-	 * handle the unexpected interrupt
-	 */
-	do {
-		if (hwif->irq == irq) {
-			stat = hwif->tp_ops->read_status(hwif);
-
-			if (!OK_STAT(stat, ATA_DRDY, BAD_STAT)) {
-				/* Try to not flood the console with msgs */
-				static unsigned long last_msgtime, count;
-				++count;
-				if (time_after(jiffies, last_msgtime + HZ)) {
-					last_msgtime = jiffies;
-					printk(KERN_ERR "%s%s: unexpected interrupt, "
-						"status=0x%02x, count=%ld\n",
-						hwif->name,
-						(hwif->next==hwgroup->hwif) ? "" : "(?)", stat, count);
-				}
-			}
+	u8 stat = hwif->tp_ops->read_status(hwif);
+
+	if (!OK_STAT(stat, ATA_DRDY, BAD_STAT)) {
+		/* Try to not flood the console with msgs */
+		static unsigned long last_msgtime, count;
+		++count;
+
+		if (time_after(jiffies, last_msgtime + HZ)) {
+			last_msgtime = jiffies;
+			printk(KERN_ERR "%s: unexpected interrupt, "
+				"status=0x%02x, count=%ld\n",
+				hwif->name, stat, count);
 		}
-	} while ((hwif = hwif->next) != hwgroup->hwif);
+	}
 }
 
 /**
  *	ide_intr	-	default IDE interrupt handler
  *	@irq: interrupt number
- *	@dev_id: hwif group
+ *	@dev_id: hwif
  *	@regs: unused weirdness from the kernel irq layer
  *
  *	This is the default IRQ handler for the IDE layer. You should
@@ -1063,17 +1049,19 @@ static void unexpected_intr(int irq, ide_hwif_t *hwif)
  
 irqreturn_t ide_intr (int irq, void *dev_id)
 {
-	unsigned long flags;
-	ide_hwgroup_t *hwgroup = (ide_hwgroup_t *)dev_id;
-	ide_hwif_t *hwif = hwgroup->hwif;
+	ide_hwif_t *hwif = (ide_hwif_t *)dev_id;
+	ide_hwgroup_t *hwgroup = hwif->hwgroup;
 	ide_drive_t *uninitialized_var(drive);
 	ide_handler_t *handler;
+	unsigned long flags;
 	ide_startstop_t startstop;
 	irqreturn_t irq_ret = IRQ_NONE;
 	int plug_device = 0;
 
-	if (hwif->host->host_flags & IDE_HFLAG_SERIALIZE)
-		hwif = hwif->host->cur_port;
+	if (hwif->host->host_flags & IDE_HFLAG_SERIALIZE) {
+		if (hwif != hwif->host->cur_port)
+			goto out_early;
+	}
 
 	spin_lock_irqsave(&hwgroup->lock, flags);
 
@@ -1172,7 +1160,7 @@ out_handled:
 	irq_ret = IRQ_HANDLED;
 out:
 	spin_unlock_irqrestore(&hwgroup->lock, flags);
-
+out_early:
 	if (plug_device)
 		ide_plug_device(drive);
 
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index c5adb7b9c5b5..2752509531b7 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1022,6 +1022,7 @@ static int init_irq (ide_hwif_t *hwif)
 	unsigned int index;
 	ide_hwgroup_t *hwgroup;
 	ide_hwif_t *match = NULL;
+	int sa = 0;
 
 	mutex_lock(&ide_cfg_mtx);
 	hwif->hwgroup = NULL;
@@ -1076,24 +1077,18 @@ static int init_irq (ide_hwif_t *hwif)
 
 	ide_ports[hwif->index] = hwif;
 
-	/*
-	 * Allocate the irq, if not already obtained for another hwif
-	 */
-	if (!match || match->irq != hwif->irq) {
-		int sa = 0;
 #if defined(__mc68000__)
-		sa = IRQF_SHARED;
+	sa = IRQF_SHARED;
 #endif /* __mc68000__ */
 
-		if (hwif->chipset == ide_pci)
-			sa = IRQF_SHARED;
+	if (hwif->chipset == ide_pci)
+		sa = IRQF_SHARED;
 
-		if (io_ports->ctl_addr)
-			hwif->tp_ops->set_irq(hwif, 1);
+	if (io_ports->ctl_addr)
+		hwif->tp_ops->set_irq(hwif, 1);
 
-		if (request_irq(hwif->irq,&ide_intr,sa,hwif->name,hwgroup))
-	       		goto out_unlink;
-	}
+	if (request_irq(hwif->irq, &ide_intr, sa, hwif->name, hwif))
+		goto out_unlink;
 
 	if (!hwif->rqsize) {
 		if ((hwif->host_flags & IDE_HFLAG_NO_LBA48) ||
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 46a2d4ca812b..5bc2e4782a55 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -175,10 +175,6 @@ EXPORT_SYMBOL_GPL(ide_port_unregister_devices);
 
 void ide_unregister(ide_hwif_t *hwif)
 {
-	ide_hwif_t *g;
-	ide_hwgroup_t *hwgroup;
-	int irq_count = 0;
-
 	BUG_ON(in_interrupt());
 	BUG_ON(irqs_disabled());
 
@@ -191,18 +187,7 @@ void ide_unregister(ide_hwif_t *hwif)
 
 	ide_proc_unregister_port(hwif);
 
-	hwgroup = hwif->hwgroup;
-	/*
-	 * free the irq if we were the only hwif using it
-	 */
-	g = hwgroup->hwif;
-	do {
-		if (g->irq == hwif->irq)
-			++irq_count;
-		g = g->next;
-	} while (g != hwgroup->hwif);
-	if (irq_count == 1)
-		free_irq(hwif->irq, hwgroup);
+	free_irq(hwif->irq, hwif);
 
 	ide_remove_port_from_hwgroup(hwif);
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 3de13df8bcef..f5382ad0bd4c 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1274,14 +1274,14 @@ extern void ide_stall_queue(ide_drive_t *drive, unsigned long timeout);
 extern void ide_timer_expiry(unsigned long);
 extern irqreturn_t ide_intr(int irq, void *dev_id);
 
-static inline int ide_lock_hwgroup(ide_hwgroup_t *hwgroup)
+static inline int ide_lock_hwgroup(ide_hwgroup_t *hwgroup, ide_hwif_t *hwif)
 {
 	if (hwgroup->busy)
 		return 1;
 
 	hwgroup->busy = 1;
 	/* for atari only */
-	ide_get_lock(ide_intr, hwgroup);
+	ide_get_lock(ide_intr, hwif);
 
 	return 0;
 }
-- 
cgit v1.2.3


From efe0397eef544ac4bcca23d39aa8d5db154952e0 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:49 +0100
Subject: ide: remove hwgroup->hwif and {drive,hwif}->next

* Add 'int port_count' field to ide_hwgroup_t to keep the track
  of the number of ports in the hwgroup.  Then update init_irq()
  and ide_remove_port_from_hwgroup() to use it.

* Remove no longer needed hwgroup->hwif, {drive,hwif}->next,
  ide_add_drive_to_hwgroup() and ide_remove_drive_from_hwgroup()
  (hwgroup->drive now only denotes the currently active device
   in the hwgroup).

* Update locking documentation in <linux/ide.h>.

While at it:

* Rename ->drive field in ide_hwgroup_t to ->cur_dev.

* Use __func__ in ide_timer_expiry().

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c    | 12 +++----
 drivers/ide/ide-probe.c | 90 ++++---------------------------------------------
 include/linux/ide.h     | 13 ++-----
 3 files changed, 15 insertions(+), 100 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 1fc739f44154..4ce793c05629 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -742,12 +742,12 @@ repeat:
 			 * set nIEN for previous port, drives in the
 			 * quirk_list may not like intr setups/cleanups
 			 */
-			if (prev_port && hwgroup->drive->quirk_list == 0)
+			if (prev_port && hwgroup->cur_dev->quirk_list == 0)
 				prev_port->tp_ops->set_irq(prev_port, 0);
 
 			hwif->host->cur_port = hwif;
 		}
-		hwgroup->drive = drive;
+		hwgroup->cur_dev = drive;
 		drive->dev_flags &= ~(IDE_DFLAG_SLEEPING | IDE_DFLAG_PARKED);
 
 		spin_unlock_irq(&hwgroup->lock);
@@ -913,9 +913,9 @@ void ide_timer_expiry (unsigned long data)
 		 * Either way, we don't really want to complain about anything.
 		 */
 	} else {
-		drive = hwgroup->drive;
+		drive = hwgroup->cur_dev;
 		if (!drive) {
-			printk(KERN_ERR "ide_timer_expiry: hwgroup->drive was NULL\n");
+			printk(KERN_ERR "%s: ->cur_dev was NULL\n", __func__);
 			hwgroup->handler = NULL;
 		} else {
 			ide_hwif_t *hwif;
@@ -1033,7 +1033,7 @@ static void unexpected_intr(int irq, ide_hwif_t *hwif)
  *	places
  *
  *	hwif is the interface in the group currently performing
- *	a command. hwgroup->drive is the drive and hwgroup->handler is
+ *	a command. hwgroup->cur_dev is the drive and hwgroup->handler is
  *	the IRQ handler to call. As we issue a command the handlers
  *	step through multiple states, reassigning the handler to the
  *	next step in the process. Unlike a smart SCSI controller IDE
@@ -1105,7 +1105,7 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 		goto out;
 	}
 
-	drive = hwgroup->drive;
+	drive = hwgroup->cur_dev;
 	if (!drive) {
 		/*
 		 * This should NEVER happen, and there isn't much
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 2752509531b7..68f3c87b8284 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -918,27 +918,9 @@ static int ide_init_queue(ide_drive_t *drive)
 	return 0;
 }
 
-static void ide_add_drive_to_hwgroup(ide_drive_t *drive)
-{
-	ide_hwgroup_t *hwgroup = drive->hwif->hwgroup;
-
-	spin_lock_irq(&hwgroup->lock);
-	if (!hwgroup->drive) {
-		/* first drive for hwgroup. */
-		drive->next = drive;
-		hwgroup->drive = drive;
-		hwgroup->hwif = HWIF(hwgroup->drive);
-	} else {
-		drive->next = hwgroup->drive->next;
-		hwgroup->drive->next = drive;
-	}
-	spin_unlock_irq(&hwgroup->lock);
-}
-
 /*
  * For any present drive:
  * - allocate the block device queue
- * - link drive into the hwgroup
  */
 static int ide_port_setup_devices(ide_hwif_t *hwif)
 {
@@ -961,8 +943,6 @@ static int ide_port_setup_devices(ide_hwif_t *hwif)
 		}
 
 		j++;
-
-		ide_add_drive_to_hwgroup(drive);
 	}
 	mutex_unlock(&ide_cfg_mtx);
 
@@ -978,33 +958,9 @@ void ide_remove_port_from_hwgroup(ide_hwif_t *hwif)
 	ide_ports[hwif->index] = NULL;
 
 	spin_lock_irq(&hwgroup->lock);
-	/*
-	 * Remove us from the hwgroup, and free
-	 * the hwgroup if we were the only member
-	 */
-	if (hwif->next == hwif) {
-		BUG_ON(hwgroup->hwif != hwif);
+	/* Free the hwgroup if we were the only member. */
+	if (--hwgroup->port_count == 0)
 		kfree(hwgroup);
-	} else {
-		/* There is another interface in hwgroup.
-		 * Unlink us, and set hwgroup->drive and ->hwif to
-		 * something sane.
-		 */
-		ide_hwif_t *g = hwgroup->hwif;
-
-		while (g->next != hwif)
-			g = g->next;
-		g->next = hwif->next;
-		if (hwgroup->hwif == hwif) {
-			/* Chose a random hwif for hwgroup->hwif.
-			 * It's guaranteed that there are no drives
-			 * left in the hwgroup.
-			 */
-			BUG_ON(hwgroup->drive != NULL);
-			hwgroup->hwif = g;
-		}
-		BUG_ON(hwgroup->hwif == hwif);
-	}
 	spin_unlock_irq(&hwgroup->lock);
 }
 
@@ -1044,20 +1000,9 @@ static int init_irq (ide_hwif_t *hwif)
 	if (match) {
 		hwgroup = match->hwgroup;
 		hwif->hwgroup = hwgroup;
-		/*
-		 * Link us into the hwgroup.
-		 * This must be done early, do ensure that unexpected_intr
-		 * can find the hwif and prevent irq storms.
-		 * No drives are attached to the new hwif, choose_drive
-		 * can't do anything stupid (yet).
-		 * Add ourself as the 2nd entry to the hwgroup->hwif
-		 * linked list, the first entry is the hwif that owns
-		 * hwgroup->handler - do not change that.
-		 */
+
 		spin_lock_irq(&hwgroup->lock);
-		hwif->next = hwgroup->hwif->next;
-		hwgroup->hwif->next = hwif;
-		BUG_ON(hwif->next == hwif);
+		hwgroup->port_count++;
 		spin_unlock_irq(&hwgroup->lock);
 	} else {
 		hwgroup = kmalloc_node(sizeof(*hwgroup), GFP_KERNEL|__GFP_ZERO,
@@ -1068,7 +1013,8 @@ static int init_irq (ide_hwif_t *hwif)
 		spin_lock_init(&hwgroup->lock);
 
 		hwif->hwgroup = hwgroup;
-		hwgroup->hwif = hwif->next = hwif;
+
+		hwgroup->port_count = 1;
 
 		init_timer(&hwgroup->timer);
 		hwgroup->timer.function = &ide_timer_expiry;
@@ -1191,29 +1137,6 @@ void ide_init_disk(struct gendisk *disk, ide_drive_t *drive)
 
 EXPORT_SYMBOL_GPL(ide_init_disk);
 
-static void ide_remove_drive_from_hwgroup(ide_drive_t *drive)
-{
-	ide_hwgroup_t *hwgroup = drive->hwif->hwgroup;
-
-	if (drive == drive->next) {
-		/* special case: last drive from hwgroup. */
-		BUG_ON(hwgroup->drive != drive);
-		hwgroup->drive = NULL;
-	} else {
-		ide_drive_t *walk;
-
-		walk = hwgroup->drive;
-		while (walk->next != drive)
-			walk = walk->next;
-		walk->next = drive->next;
-		if (hwgroup->drive == drive) {
-			hwgroup->drive = drive->next;
-			hwgroup->hwif = hwgroup->drive->hwif;
-		}
-	}
-	BUG_ON(hwgroup->drive == drive);
-}
-
 static void drive_release_dev (struct device *dev)
 {
 	ide_drive_t *drive = container_of(dev, ide_drive_t, gendev);
@@ -1222,7 +1145,6 @@ static void drive_release_dev (struct device *dev)
 	ide_proc_unregister_device(drive);
 
 	spin_lock_irq(&hwgroup->lock);
-	ide_remove_drive_from_hwgroup(drive);
 	kfree(drive->id);
 	drive->id = NULL;
 	drive->dev_flags &= ~IDE_DFLAG_PRESENT;
diff --git a/include/linux/ide.h b/include/linux/ide.h
index f5382ad0bd4c..8b74ccdd221c 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -588,7 +588,6 @@ struct ide_drive_s {
 	struct request_queue	*queue;	/* request queue */
 
 	struct request		*rq;	/* current request */
-	struct ide_drive_s 	*next;	/* circular list of hwgroup drives */
 	void		*driver_data;	/* extra driver data */
 	u16			*id;	/* identification info */
 #ifdef CONFIG_IDE_PROC_FS
@@ -750,7 +749,6 @@ struct ide_dma_ops {
 struct ide_host;
 
 typedef struct hwif_s {
-	struct hwif_s *next;		/* for linked-list in ide_hwgroup_t */
 	struct hwif_s *mate;		/* other hwif from same PCI chip */
 	struct hwgroup_s *hwgroup;	/* actually (ide_hwgroup_t *) */
 	struct proc_dir_entry *proc;	/* /proc/ide/ directory entry */
@@ -874,9 +872,7 @@ typedef struct hwgroup_s {
 	unsigned int polling	: 1;
 
 		/* current drive */
-	ide_drive_t *drive;
-		/* ptr to current hwif in linked-list */
-	ide_hwif_t *hwif;
+	ide_drive_t *cur_dev;
 
 		/* current request */
 	struct request *rq;
@@ -892,6 +888,8 @@ typedef struct hwgroup_s {
 	int req_gen_timer;
 
 	spinlock_t lock;
+
+	int port_count;
 } ide_hwgroup_t;
 
 typedef struct ide_driver_s ide_driver_t;
@@ -1622,12 +1620,7 @@ extern struct mutex ide_cfg_mtx;
 /*
  * Structure locking:
  *
- * ide_cfg_mtx and hwgroup->lock together protect changes to
- * ide_hwif_t->next
- * ide_drive_t->next
- *
  * ide_hwgroup_t->busy: hwgroup->lock
- * ide_hwgroup_t->hwif: hwgroup->lock
  * ide_hwif_t->{hwgroup,mate}: constant, no locking
  * ide_drive_t->hwif: constant, no locking
  */
-- 
cgit v1.2.3


From 5b31f855f10d0053e738baa6d91fb6a3fad35119 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:49 +0100
Subject: ide: use lock bitops for ports serialization (v2)

* Add ->host_busy field to struct ide_host and use it's first bit
  together with lock bitops to provide new ports serialization method.

* Convert core IDE code to use new ide_[un]lock_host() helpers.

  This removes the need for taking hwgroup->lock if host is already
  busy on serialized hosts and makes it possible to merge ide_hwgroup_t
  into ide_hwif_t (done in the later patch).

* Remove no longer needed ide_hwgroup_t.busy and ide_[un]lock_hwgroup().

* Update do_ide_request() documentation.

v2:
* ide_release_lock() should be called inside IDE_HFLAG_SERIALIZE check.

* Add ide_hwif_t.busy flag and ide_[un]lock_port() for serializing
  devices on a port.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-io.c | 105 ++++++++++++++++++++++++++++++---------------------
 include/linux/ide.h  |  35 +++--------------
 2 files changed, 69 insertions(+), 71 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 4ce793c05629..8f371821c614 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -666,45 +666,54 @@ void ide_stall_queue (ide_drive_t *drive, unsigned long timeout)
 }
 EXPORT_SYMBOL(ide_stall_queue);
 
+static inline int ide_lock_port(ide_hwif_t *hwif)
+{
+	if (hwif->busy)
+		return 1;
+
+	hwif->busy = 1;
+
+	return 0;
+}
+
+static inline void ide_unlock_port(ide_hwif_t *hwif)
+{
+	hwif->busy = 0;
+}
+
+static inline int ide_lock_host(struct ide_host *host, ide_hwif_t *hwif)
+{
+	int rc = 0;
+
+	if (host->host_flags & IDE_HFLAG_SERIALIZE) {
+		rc = test_and_set_bit_lock(IDE_HOST_BUSY, &host->host_busy);
+		if (rc == 0) {
+			/* for atari only */
+			ide_get_lock(ide_intr, hwif);
+		}
+	}
+	return rc;
+}
+
+static inline void ide_unlock_host(struct ide_host *host)
+{
+	if (host->host_flags & IDE_HFLAG_SERIALIZE) {
+		/* for atari only */
+		ide_release_lock();
+		clear_bit_unlock(IDE_HOST_BUSY, &host->host_busy);
+	}
+}
+
 /*
  * Issue a new request to a drive from hwgroup
- *
- * A hwgroup is a serialized group of IDE interfaces.  Usually there is
- * exactly one hwif (interface) per hwgroup, but buggy controllers (eg. CMD640)
- * may have both interfaces in a single hwgroup to "serialize" access.
- * Or possibly multiple ISA interfaces can share a common IRQ by being grouped
- * together into one hwgroup for serialized access.
- *
- * Note also that several hwgroups can end up sharing a single IRQ,
- * possibly along with many other devices.  This is especially common in
- * PCI-based systems with off-board IDE controller cards.
- *
- * The IDE driver uses a per-hwgroup lock to protect the hwgroup->busy flag.
- *
- * The first thread into the driver for a particular hwgroup sets the
- * hwgroup->busy flag to indicate that this hwgroup is now active,
- * and then initiates processing of the top request from the request queue.
- *
- * Other threads attempting entry notice the busy setting, and will simply
- * queue their new requests and exit immediately.  Note that hwgroup->busy
- * remains set even when the driver is merely awaiting the next interrupt.
- * Thus, the meaning is "this hwgroup is busy processing a request".
- *
- * When processing of a request completes, the completing thread or IRQ-handler
- * will start the next request from the queue.  If no more work remains,
- * the driver will clear the hwgroup->busy flag and exit.
- *
- * The per-hwgroup spinlock is used to protect all access to the
- * hwgroup->busy flag, but is otherwise not needed for most processing in
- * the driver.  This makes the driver much more friendlier to shared IRQs
- * than previous designs, while remaining 100% (?) SMP safe and capable.
  */
 void do_ide_request(struct request_queue *q)
 {
 	ide_drive_t	*drive = q->queuedata;
 	ide_hwif_t	*hwif = drive->hwif;
+	struct ide_host *host = hwif->host;
 	ide_hwgroup_t	*hwgroup = hwif->hwgroup;
-	struct request	*rq;
+	struct request	*rq = NULL;
 	ide_startstop_t	startstop;
 
 	/*
@@ -721,9 +730,13 @@ void do_ide_request(struct request_queue *q)
 		blk_remove_plug(q);
 
 	spin_unlock_irq(q->queue_lock);
+
+	if (ide_lock_host(host, hwif))
+		goto plug_device_2;
+
 	spin_lock_irq(&hwgroup->lock);
 
-	if (!ide_lock_hwgroup(hwgroup, hwif)) {
+	if (!ide_lock_port(hwif)) {
 		ide_hwif_t *prev_port;
 repeat:
 		prev_port = hwif->host->cur_port;
@@ -731,7 +744,7 @@ repeat:
 
 		if (drive->dev_flags & IDE_DFLAG_SLEEPING) {
 			if (time_before(drive->sleep, jiffies)) {
-				ide_unlock_hwgroup(hwgroup);
+				ide_unlock_port(hwif);
 				goto plug_device;
 			}
 		}
@@ -761,7 +774,7 @@ repeat:
 		spin_lock_irq(&hwgroup->lock);
 
 		if (!rq) {
-			ide_unlock_hwgroup(hwgroup);
+			ide_unlock_port(hwif);
 			goto out;
 		}
 
@@ -782,7 +795,7 @@ repeat:
 		    blk_pm_request(rq) == 0 &&
 		    (rq->cmd_flags & REQ_PREEMPT) == 0) {
 			/* there should be no pending command at this point */
-			ide_unlock_hwgroup(hwgroup);
+			ide_unlock_port(hwif);
 			goto plug_device;
 		}
 
@@ -798,11 +811,15 @@ repeat:
 		goto plug_device;
 out:
 	spin_unlock_irq(&hwgroup->lock);
+	if (rq == NULL)
+		ide_unlock_host(host);
 	spin_lock_irq(q->queue_lock);
 	return;
 
 plug_device:
 	spin_unlock_irq(&hwgroup->lock);
+	ide_unlock_host(host);
+plug_device_2:
 	spin_lock_irq(q->queue_lock);
 
 	if (!elv_queue_empty(q))
@@ -844,9 +861,9 @@ static ide_startstop_t ide_dma_timeout_retry(ide_drive_t *drive, int error)
 	ide_dma_off_quietly(drive);
 
 	/*
-	 * un-busy drive etc (hwgroup->busy is cleared on return) and
-	 * make sure request is sane
+	 * un-busy drive etc and make sure request is sane
 	 */
+
 	rq = HWGROUP(drive)->rq;
 
 	if (!rq)
@@ -895,6 +912,7 @@ static void ide_plug_device(ide_drive_t *drive)
 void ide_timer_expiry (unsigned long data)
 {
 	ide_hwgroup_t	*hwgroup = (ide_hwgroup_t *) data;
+	ide_hwif_t	*uninitialized_var(hwif);
 	ide_drive_t	*uninitialized_var(drive);
 	ide_handler_t	*handler;
 	ide_expiry_t	*expiry;
@@ -918,7 +936,6 @@ void ide_timer_expiry (unsigned long data)
 			printk(KERN_ERR "%s: ->cur_dev was NULL\n", __func__);
 			hwgroup->handler = NULL;
 		} else {
-			ide_hwif_t *hwif;
 			ide_startstop_t startstop = ide_stopped;
 
 			if ((expiry = hwgroup->expiry) != NULL) {
@@ -964,15 +981,17 @@ void ide_timer_expiry (unsigned long data)
 			spin_lock_irq(&hwgroup->lock);
 			enable_irq(hwif->irq);
 			if (startstop == ide_stopped) {
-				ide_unlock_hwgroup(hwgroup);
+				ide_unlock_port(hwif);
 				plug_device = 1;
 			}
 		}
 	}
 	spin_unlock_irqrestore(&hwgroup->lock, flags);
 
-	if (plug_device)
+	if (plug_device) {
+		ide_unlock_host(hwif->host);
 		ide_plug_device(drive);
+	}
 }
 
 /**
@@ -1150,7 +1169,7 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 	 */
 	if (startstop == ide_stopped) {
 		if (hwgroup->handler == NULL) {	/* paranoia */
-			ide_unlock_hwgroup(hwgroup);
+			ide_unlock_port(hwif);
 			plug_device = 1;
 		} else
 			printk(KERN_ERR "%s: %s: huh? expected NULL handler "
@@ -1161,8 +1180,10 @@ out_handled:
 out:
 	spin_unlock_irqrestore(&hwgroup->lock, flags);
 out_early:
-	if (plug_device)
+	if (plug_device) {
+		ide_unlock_host(hwif->host);
 		ide_plug_device(drive);
+	}
 
 	return irq_ret;
 }
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 8b74ccdd221c..00df155b5a02 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -828,6 +828,7 @@ typedef struct hwif_s {
 
 	unsigned	present    : 1;	/* this interface exists */
 	unsigned	sg_mapped  : 1;	/* sg_table and sg_nents are ready */
+	unsigned	busy	   : 1; /* serializes devices on a port */
 
 	struct device		gendev;
 	struct device		*portdev;
@@ -851,8 +852,13 @@ struct ide_host {
 	unsigned long	host_flags;
 	void		*host_priv;
 	ide_hwif_t	*cur_port;	/* for hosts requiring serialization */
+
+	/* used for hosts requiring serialization */
+	volatile long	host_busy;
 };
 
+#define IDE_HOST_BUSY 0
+
 /*
  *  internal ide interrupt handler type
  */
@@ -866,8 +872,6 @@ typedef struct hwgroup_s {
 		/* irq handler, if active */
 	ide_startstop_t	(*handler)(ide_drive_t *);
 
-		/* BOOL: protects all fields below */
-	volatile int busy;
 		/* BOOL: polling active & poll_timeout field valid */
 	unsigned int polling	: 1;
 
@@ -1271,26 +1275,6 @@ extern void ide_stall_queue(ide_drive_t *drive, unsigned long timeout);
 
 extern void ide_timer_expiry(unsigned long);
 extern irqreturn_t ide_intr(int irq, void *dev_id);
-
-static inline int ide_lock_hwgroup(ide_hwgroup_t *hwgroup, ide_hwif_t *hwif)
-{
-	if (hwgroup->busy)
-		return 1;
-
-	hwgroup->busy = 1;
-	/* for atari only */
-	ide_get_lock(ide_intr, hwif);
-
-	return 0;
-}
-
-static inline void ide_unlock_hwgroup(ide_hwgroup_t *hwgroup)
-{
-	/* for atari only */
-	ide_release_lock();
-	hwgroup->busy = 0;
-}
-
 extern void do_ide_request(struct request_queue *);
 
 void ide_init_disk(struct gendisk *, ide_drive_t *);
@@ -1617,13 +1601,6 @@ static inline void ide_set_max_pio(ide_drive_t *drive)
 
 extern spinlock_t ide_lock;
 extern struct mutex ide_cfg_mtx;
-/*
- * Structure locking:
- *
- * ide_hwgroup_t->busy: hwgroup->lock
- * ide_hwif_t->{hwgroup,mate}: constant, no locking
- * ide_drive_t->hwif: constant, no locking
- */
 
 #define local_irq_set(flags)	do { local_save_flags((flags)); local_irq_enable_in_hardirq(); } while (0)
 
-- 
cgit v1.2.3


From b65fac32cfe3b2f98cd472fef400bd1c1340de23 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:50 +0100
Subject: ide: merge ide_hwgroup_t with ide_hwif_t (v2)

* Merge ide_hwgroup_t with ide_hwif_t.

* Cleanup init_irq() accordingly, then remove no longer needed
  ide_remove_port_from_hwgroup() and ide_ports[].

* Remove now unused HWGROUP() macro.

While at it:

* ide_dump_ata_error() fixups

v2:
* Fix ->quirk_list check in do_ide_request()
  (s/hwif->cur_dev/prev_port->cur_dev).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/alim15x3.c     |   2 +-
 drivers/ide/au1xxx-ide.c   |   6 +--
 drivers/ide/icside.c       |   2 +-
 drivers/ide/ide-atapi.c    |   8 ++--
 drivers/ide/ide-cd.c       |  16 +++----
 drivers/ide/ide-dma-sff.c  |   4 +-
 drivers/ide/ide-dma.c      |   2 +-
 drivers/ide/ide-floppy.c   |   2 +-
 drivers/ide/ide-io.c       | 117 ++++++++++++++++++++++-----------------------
 drivers/ide/ide-iops.c     |  69 +++++++++++++-------------
 drivers/ide/ide-lib.c      |   9 ++--
 drivers/ide/ide-park.c     |  16 +++----
 drivers/ide/ide-pm.c       |   2 +-
 drivers/ide/ide-probe.c    |  87 +++++----------------------------
 drivers/ide/ide-tape.c     |   9 ++--
 drivers/ide/ide-taskfile.c |   4 +-
 drivers/ide/ide.c          |   7 +--
 drivers/ide/pdc202xx_old.c |   2 +-
 drivers/ide/pmac.c         |   2 +-
 drivers/ide/scc_pata.c     |   4 +-
 drivers/ide/sgiioc4.c      |   2 +-
 drivers/ide/tc86c001.c     |  10 ++--
 drivers/ide/trm290.c       |   2 +-
 drivers/ide/tx4939ide.c    |   2 +-
 drivers/ide/umc8672.c      |  13 +++--
 include/linux/ide.h        |  55 ++++++++++-----------
 26 files changed, 188 insertions(+), 266 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/alim15x3.c b/drivers/ide/alim15x3.c
index 45d2356bb725..290204e89eaf 100644
--- a/drivers/ide/alim15x3.c
+++ b/drivers/ide/alim15x3.c
@@ -198,7 +198,7 @@ static void ali_set_dma_mode(ide_drive_t *drive, const u8 speed)
 static int ali15x3_dma_setup(ide_drive_t *drive)
 {
 	if (m5229_revision < 0xC2 && drive->media != ide_disk) {
-		if (rq_data_dir(drive->hwif->hwgroup->rq))
+		if (rq_data_dir(drive->hwif->rq))
 			return 1;	/* try PIO instead of DMA */
 	}
 	return ide_dma_setup(drive);
diff --git a/drivers/ide/au1xxx-ide.c b/drivers/ide/au1xxx-ide.c
index 0ec8fd1e4dcb..db412bc772d1 100644
--- a/drivers/ide/au1xxx-ide.c
+++ b/drivers/ide/au1xxx-ide.c
@@ -213,7 +213,7 @@ static int auide_build_dmatable(ide_drive_t *drive)
 {
 	int i, iswrite, count = 0;
 	ide_hwif_t *hwif = HWIF(drive);
-	struct request *rq = HWGROUP(drive)->rq;
+	struct request *rq = hwif->rq;
 	_auide_hwif *ahwif = &auide_hwif;
 	struct scatterlist *sg;
 
@@ -309,8 +309,8 @@ static void auide_dma_exec_cmd(ide_drive_t *drive, u8 command)
 }
 
 static int auide_dma_setup(ide_drive_t *drive)
-{       	
-	struct request *rq = HWGROUP(drive)->rq;
+{
+	struct request *rq = drive->hwif->rq;
 
 	if (!auide_build_dmatable(drive)) {
 		ide_map_sg(drive, rq);
diff --git a/drivers/ide/icside.c b/drivers/ide/icside.c
index 81f70caeb40f..72d95c99f147 100644
--- a/drivers/ide/icside.c
+++ b/drivers/ide/icside.c
@@ -312,7 +312,7 @@ static int icside_dma_setup(ide_drive_t *drive)
 	ide_hwif_t *hwif = HWIF(drive);
 	struct expansion_card *ec = ECARD_DEV(hwif->dev);
 	struct icside_state *state = ecard_get_drvdata(ec);
-	struct request *rq = hwif->hwgroup->rq;
+	struct request *rq = hwif->rq;
 	unsigned int dma_mode;
 
 	if (rq_data_dir(rq))
diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index e8688c0f8645..d6a50d67b7db 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -243,7 +243,7 @@ EXPORT_SYMBOL_GPL(ide_retry_pc);
 
 int ide_cd_expiry(ide_drive_t *drive)
 {
-	struct request *rq = HWGROUP(drive)->rq;
+	struct request *rq = drive->hwif->rq;
 	unsigned long wait = 0;
 
 	debug_log("%s: rq->cmd[0]: 0x%x\n", __func__, rq->cmd[0]);
@@ -294,7 +294,7 @@ static ide_startstop_t ide_pc_intr(ide_drive_t *drive)
 {
 	struct ide_atapi_pc *pc = drive->pc;
 	ide_hwif_t *hwif = drive->hwif;
-	struct request *rq = hwif->hwgroup->rq;
+	struct request *rq = hwif->rq;
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	xfer_func_t *xferfunc;
 	unsigned int timeout, temp;
@@ -491,7 +491,7 @@ static ide_startstop_t ide_transfer_pc(ide_drive_t *drive)
 {
 	struct ide_atapi_pc *uninitialized_var(pc);
 	ide_hwif_t *hwif = drive->hwif;
-	struct request *rq = hwif->hwgroup->rq;
+	struct request *rq = hwif->rq;
 	ide_expiry_t *expiry;
 	unsigned int timeout;
 	int cmd_len;
@@ -580,7 +580,7 @@ ide_startstop_t ide_issue_pc(ide_drive_t *drive)
 
 	if (dev_is_idecd(drive)) {
 		tf_flags = IDE_TFLAG_OUT_NSECT | IDE_TFLAG_OUT_LBAL;
-		bcount = ide_cd_get_xferlen(hwif->hwgroup->rq);
+		bcount = ide_cd_get_xferlen(hwif->rq);
 		expiry = ide_cd_expiry;
 		timeout = ATAPI_WAIT_PC;
 
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 1a7410f88249..c35b64d495f5 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -239,7 +239,7 @@ static void cdrom_queue_request_sense(ide_drive_t *drive, void *sense,
 
 static void cdrom_end_request(ide_drive_t *drive, int uptodate)
 {
-	struct request *rq = HWGROUP(drive)->rq;
+	struct request *rq = drive->hwif->rq;
 	int nsectors = rq->hard_cur_sectors;
 
 	ide_debug_log(IDE_DBG_FUNC, "Call %s, cmd: 0x%x, uptodate: 0x%x, "
@@ -306,8 +306,7 @@ static void ide_dump_status_no_sense(ide_drive_t *drive, const char *msg, u8 st)
 static int cdrom_decode_status(ide_drive_t *drive, int good_stat, int *stat_ret)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	ide_hwgroup_t *hwgroup = hwif->hwgroup;
-	struct request *rq = hwgroup->rq;
+	struct request *rq = hwif->rq;
 	int stat, err, sense_key;
 
 	/* check for errors */
@@ -502,7 +501,7 @@ end_request:
 		blkdev_dequeue_request(rq);
 		spin_unlock_irqrestore(q->queue_lock, flags);
 
-		hwgroup->rq = NULL;
+		hwif->rq = NULL;
 
 		cdrom_queue_request_sense(drive, rq->sense, rq);
 	} else
@@ -525,7 +524,7 @@ static ide_startstop_t cdrom_newpc_intr(ide_drive_t *);
 static ide_startstop_t cdrom_start_packet_command(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	struct request *rq = hwif->hwgroup->rq;
+	struct request *rq = hwif->rq;
 	int xferlen;
 
 	xferlen = ide_cd_get_xferlen(rq);
@@ -567,7 +566,7 @@ static ide_startstop_t cdrom_start_packet_command(ide_drive_t *drive)
 static ide_startstop_t cdrom_transfer_packet_command(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	struct request *rq = hwif->hwgroup->rq;
+	struct request *rq = hwif->rq;
 	int cmd_len;
 	ide_startstop_t startstop;
 
@@ -854,8 +853,7 @@ static int cdrom_newpc_intr_dummy_cb(struct request *rq)
 static ide_startstop_t cdrom_newpc_intr(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	ide_hwgroup_t *hwgroup = hwif->hwgroup;
-	struct request *rq = hwgroup->rq;
+	struct request *rq = hwif->rq;
 	xfer_func_t *xferfunc;
 	ide_expiry_t *expiry = NULL;
 	int dma_error = 0, dma, stat, thislen, uptodate = 0;
@@ -1061,7 +1059,7 @@ end_request:
 		if (blk_end_request(rq, 0, dlen))
 			BUG();
 
-		hwgroup->rq = NULL;
+		hwif->rq = NULL;
 	} else {
 		if (!uptodate)
 			rq->cmd_flags |= REQ_FAILED;
diff --git a/drivers/ide/ide-dma-sff.c b/drivers/ide/ide-dma-sff.c
index f6d2d44d8a9a..623a82d1535d 100644
--- a/drivers/ide/ide-dma-sff.c
+++ b/drivers/ide/ide-dma-sff.c
@@ -175,7 +175,7 @@ EXPORT_SYMBOL_GPL(ide_build_dmatable);
 int ide_dma_setup(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	struct request *rq = hwif->hwgroup->rq;
+	struct request *rq = hwif->rq;
 	unsigned int reading = rq_data_dir(rq) ? 0 : ATA_DMA_WR;
 	u8 mmio = (hwif->host_flags & IDE_HFLAG_MMIO) ? 1 : 0;
 	u8 dma_stat;
@@ -240,7 +240,7 @@ static int dma_timer_expiry(ide_drive_t *drive)
 	if ((dma_stat & 0x18) == 0x18)	/* BUSY Stupid Early Timer !! */
 		return WAIT_CMD;
 
-	hwif->hwgroup->expiry = NULL;	/* one free ride for now */
+	hwif->expiry = NULL;	/* one free ride for now */
 
 	if (dma_stat & ATA_DMA_ERR)	/* ERROR */
 		return -1;
diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index fffd11717b2d..72ebab0bc755 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -96,7 +96,7 @@ ide_startstop_t ide_dma_intr(ide_drive_t *drive)
 
 	if (OK_STAT(stat, DRIVE_READY, drive->bad_wstat | ATA_DRQ)) {
 		if (!dma_stat) {
-			struct request *rq = hwif->hwgroup->rq;
+			struct request *rq = hwif->rq;
 
 			task_end_request(drive, rq, stat);
 			return ide_stopped;
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 0a48e2dc53a2..3eab1c6c9b31 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -71,7 +71,7 @@
 static int ide_floppy_end_request(ide_drive_t *drive, int uptodate, int nsecs)
 {
 	struct ide_disk_obj *floppy = drive->driver_data;
-	struct request *rq = HWGROUP(drive)->rq;
+	struct request *rq = drive->hwif->rq;
 	int error;
 
 	ide_debug_log(IDE_DBG_FUNC, "Call %s\n", __func__);
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 8f371821c614..6ff82d7055b9 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -88,7 +88,7 @@ static int __ide_end_request(ide_drive_t *drive, struct request *rq,
 		ret = 0;
 
 	if (ret == 0 && dequeue)
-		drive->hwif->hwgroup->rq = NULL;
+		drive->hwif->rq = NULL;
 
 	return ret;
 }
@@ -107,7 +107,7 @@ static int __ide_end_request(ide_drive_t *drive, struct request *rq,
 int ide_end_request (ide_drive_t *drive, int uptodate, int nr_sectors)
 {
 	unsigned int nr_bytes = nr_sectors << 9;
-	struct request *rq = drive->hwif->hwgroup->rq;
+	struct request *rq = drive->hwif->rq;
 
 	if (!nr_bytes) {
 		if (blk_pc_request(rq))
@@ -160,8 +160,8 @@ EXPORT_SYMBOL_GPL(ide_end_dequeued_request);
  
 void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 {
-	ide_hwgroup_t *hwgroup = drive->hwif->hwgroup;
-	struct request *rq = hwgroup->rq;
+	ide_hwif_t *hwif = drive->hwif;
+	struct request *rq = hwif->rq;
 
 	if (rq->cmd_type == REQ_TYPE_ATA_TASKFILE) {
 		ide_task_t *task = (ide_task_t *)rq->special;
@@ -186,7 +186,7 @@ void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 		return;
 	}
 
-	hwgroup->rq = NULL;
+	hwif->rq = NULL;
 
 	rq->errors = err;
 
@@ -321,7 +321,8 @@ ide_startstop_t ide_error (ide_drive_t *drive, const char *msg, u8 stat)
 
 	err = ide_dump_status(drive, msg, stat);
 
-	if ((rq = HWGROUP(drive)->rq) == NULL)
+	rq = drive->hwif->rq;
+	if (rq == NULL)
 		return ide_stopped;
 
 	/* retry only "normal" I/O: */
@@ -654,7 +655,7 @@ kill_rq:
  *	@timeout: time to stall for (jiffies)
  *
  *	ide_stall_queue() can be used by a drive to give excess bandwidth back
- *	to the hwgroup by sleeping for timeout jiffies.
+ *	to the port by sleeping for timeout jiffies.
  */
  
 void ide_stall_queue (ide_drive_t *drive, unsigned long timeout)
@@ -705,14 +706,13 @@ static inline void ide_unlock_host(struct ide_host *host)
 }
 
 /*
- * Issue a new request to a drive from hwgroup
+ * Issue a new request to a device.
  */
 void do_ide_request(struct request_queue *q)
 {
 	ide_drive_t	*drive = q->queuedata;
 	ide_hwif_t	*hwif = drive->hwif;
 	struct ide_host *host = hwif->host;
-	ide_hwgroup_t	*hwgroup = hwif->hwgroup;
 	struct request	*rq = NULL;
 	ide_startstop_t	startstop;
 
@@ -734,13 +734,13 @@ void do_ide_request(struct request_queue *q)
 	if (ide_lock_host(host, hwif))
 		goto plug_device_2;
 
-	spin_lock_irq(&hwgroup->lock);
+	spin_lock_irq(&hwif->lock);
 
 	if (!ide_lock_port(hwif)) {
 		ide_hwif_t *prev_port;
 repeat:
 		prev_port = hwif->host->cur_port;
-		hwgroup->rq = NULL;
+		hwif->rq = NULL;
 
 		if (drive->dev_flags & IDE_DFLAG_SLEEPING) {
 			if (time_before(drive->sleep, jiffies)) {
@@ -755,15 +755,15 @@ repeat:
 			 * set nIEN for previous port, drives in the
 			 * quirk_list may not like intr setups/cleanups
 			 */
-			if (prev_port && hwgroup->cur_dev->quirk_list == 0)
+			if (prev_port && prev_port->cur_dev->quirk_list == 0)
 				prev_port->tp_ops->set_irq(prev_port, 0);
 
 			hwif->host->cur_port = hwif;
 		}
-		hwgroup->cur_dev = drive;
+		hwif->cur_dev = drive;
 		drive->dev_flags &= ~(IDE_DFLAG_SLEEPING | IDE_DFLAG_PARKED);
 
-		spin_unlock_irq(&hwgroup->lock);
+		spin_unlock_irq(&hwif->lock);
 		spin_lock_irq(q->queue_lock);
 		/*
 		 * we know that the queue isn't empty, but this can happen
@@ -771,7 +771,7 @@ repeat:
 		 */
 		rq = elv_next_request(drive->queue);
 		spin_unlock_irq(q->queue_lock);
-		spin_lock_irq(&hwgroup->lock);
+		spin_lock_irq(&hwif->lock);
 
 		if (!rq) {
 			ide_unlock_port(hwif);
@@ -799,25 +799,25 @@ repeat:
 			goto plug_device;
 		}
 
-		hwgroup->rq = rq;
+		hwif->rq = rq;
 
-		spin_unlock_irq(&hwgroup->lock);
+		spin_unlock_irq(&hwif->lock);
 		startstop = start_request(drive, rq);
-		spin_lock_irq(&hwgroup->lock);
+		spin_lock_irq(&hwif->lock);
 
 		if (startstop == ide_stopped)
 			goto repeat;
 	} else
 		goto plug_device;
 out:
-	spin_unlock_irq(&hwgroup->lock);
+	spin_unlock_irq(&hwif->lock);
 	if (rq == NULL)
 		ide_unlock_host(host);
 	spin_lock_irq(q->queue_lock);
 	return;
 
 plug_device:
-	spin_unlock_irq(&hwgroup->lock);
+	spin_unlock_irq(&hwif->lock);
 	ide_unlock_host(host);
 plug_device_2:
 	spin_lock_irq(q->queue_lock);
@@ -827,7 +827,7 @@ plug_device_2:
 }
 
 /*
- * un-busy the hwgroup etc, and clear any pending DMA status. we want to
+ * un-busy the port etc, and clear any pending DMA status. we want to
  * retry the current request in pio mode instead of risking tossing it
  * all away
  */
@@ -864,12 +864,11 @@ static ide_startstop_t ide_dma_timeout_retry(ide_drive_t *drive, int error)
 	 * un-busy drive etc and make sure request is sane
 	 */
 
-	rq = HWGROUP(drive)->rq;
-
+	rq = hwif->rq;
 	if (!rq)
 		goto out;
 
-	HWGROUP(drive)->rq = NULL;
+	hwif->rq = NULL;
 
 	rq->errors = 0;
 
@@ -897,7 +896,7 @@ static void ide_plug_device(ide_drive_t *drive)
 
 /**
  *	ide_timer_expiry	-	handle lack of an IDE interrupt
- *	@data: timer callback magic (hwgroup)
+ *	@data: timer callback magic (hwif)
  *
  *	An IDE command has timed out before the expected drive return
  *	occurred. At this point we attempt to clean up the current
@@ -911,19 +910,18 @@ static void ide_plug_device(ide_drive_t *drive)
  
 void ide_timer_expiry (unsigned long data)
 {
-	ide_hwgroup_t	*hwgroup = (ide_hwgroup_t *) data;
-	ide_hwif_t	*uninitialized_var(hwif);
+	ide_hwif_t	*hwif = (ide_hwif_t *)data;
 	ide_drive_t	*uninitialized_var(drive);
 	ide_handler_t	*handler;
-	ide_expiry_t	*expiry;
 	unsigned long	flags;
 	unsigned long	wait = -1;
 	int		plug_device = 0;
 
-	spin_lock_irqsave(&hwgroup->lock, flags);
+	spin_lock_irqsave(&hwif->lock, flags);
+
+	handler = hwif->handler;
 
-	if (((handler = hwgroup->handler) == NULL) ||
-	    (hwgroup->req_gen != hwgroup->req_gen_timer)) {
+	if (handler == NULL || hwif->req_gen != hwif->req_gen_timer) {
 		/*
 		 * Either a marginal timeout occurred
 		 * (got the interrupt just as timer expired),
@@ -931,38 +929,39 @@ void ide_timer_expiry (unsigned long data)
 		 * Either way, we don't really want to complain about anything.
 		 */
 	} else {
-		drive = hwgroup->cur_dev;
+		drive = hwif->cur_dev;
 		if (!drive) {
 			printk(KERN_ERR "%s: ->cur_dev was NULL\n", __func__);
-			hwgroup->handler = NULL;
+			hwif->handler = NULL;
 		} else {
+			ide_expiry_t *expiry = hwif->expiry;
 			ide_startstop_t startstop = ide_stopped;
 
-			if ((expiry = hwgroup->expiry) != NULL) {
+			if (expiry) {
 				/* continue */
 				if ((wait = expiry(drive)) > 0) {
 					/* reset timer */
-					hwgroup->timer.expires  = jiffies + wait;
-					hwgroup->req_gen_timer = hwgroup->req_gen;
-					add_timer(&hwgroup->timer);
-					spin_unlock_irqrestore(&hwgroup->lock, flags);
+					hwif->timer.expires  = jiffies + wait;
+					hwif->req_gen_timer = hwif->req_gen;
+					add_timer(&hwif->timer);
+					spin_unlock_irqrestore(&hwif->lock, flags);
 					return;
 				}
 			}
-			hwgroup->handler = NULL;
+			hwif->handler = NULL;
 			/*
 			 * We need to simulate a real interrupt when invoking
 			 * the handler() function, which means we need to
 			 * globally mask the specific IRQ:
 			 */
-			spin_unlock(&hwgroup->lock);
+			spin_unlock(&hwif->lock);
 			hwif  = HWIF(drive);
 			/* disable_irq_nosync ?? */
 			disable_irq(hwif->irq);
 			/* local CPU only,
 			 * as if we were handling an interrupt */
 			local_irq_disable();
-			if (hwgroup->polling) {
+			if (hwif->polling) {
 				startstop = handler(drive);
 			} else if (drive_is_ready(drive)) {
 				if (drive->waiting_for_dma)
@@ -978,7 +977,7 @@ void ide_timer_expiry (unsigned long data)
 					ide_error(drive, "irq timeout",
 						  hwif->tp_ops->read_status(hwif));
 			}
-			spin_lock_irq(&hwgroup->lock);
+			spin_lock_irq(&hwif->lock);
 			enable_irq(hwif->irq);
 			if (startstop == ide_stopped) {
 				ide_unlock_port(hwif);
@@ -986,7 +985,7 @@ void ide_timer_expiry (unsigned long data)
 			}
 		}
 	}
-	spin_unlock_irqrestore(&hwgroup->lock, flags);
+	spin_unlock_irqrestore(&hwif->lock, flags);
 
 	if (plug_device) {
 		ide_unlock_host(hwif->host);
@@ -1052,7 +1051,7 @@ static void unexpected_intr(int irq, ide_hwif_t *hwif)
  *	places
  *
  *	hwif is the interface in the group currently performing
- *	a command. hwgroup->cur_dev is the drive and hwgroup->handler is
+ *	a command. hwif->cur_dev is the drive and hwif->handler is
  *	the IRQ handler to call. As we issue a command the handlers
  *	step through multiple states, reassigning the handler to the
  *	next step in the process. Unlike a smart SCSI controller IDE
@@ -1063,13 +1062,12 @@ static void unexpected_intr(int irq, ide_hwif_t *hwif)
  *
  *	The handler eventually returns ide_stopped to indicate the
  *	request completed. At this point we issue the next request
- *	on the hwgroup and the process begins again.
+ *	on the port and the process begins again.
  */
- 
+
 irqreturn_t ide_intr (int irq, void *dev_id)
 {
 	ide_hwif_t *hwif = (ide_hwif_t *)dev_id;
-	ide_hwgroup_t *hwgroup = hwif->hwgroup;
 	ide_drive_t *uninitialized_var(drive);
 	ide_handler_t *handler;
 	unsigned long flags;
@@ -1082,12 +1080,14 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 			goto out_early;
 	}
 
-	spin_lock_irqsave(&hwgroup->lock, flags);
+	spin_lock_irqsave(&hwif->lock, flags);
 
 	if (!ide_ack_intr(hwif))
 		goto out;
 
-	if ((handler = hwgroup->handler) == NULL || hwgroup->polling) {
+	handler = hwif->handler;
+
+	if (handler == NULL || hwif->polling) {
 		/*
 		 * Not expecting an interrupt from this drive.
 		 * That means this could be:
@@ -1124,7 +1124,7 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 		goto out;
 	}
 
-	drive = hwgroup->cur_dev;
+	drive = hwif->cur_dev;
 	if (!drive) {
 		/*
 		 * This should NEVER happen, and there isn't much
@@ -1145,10 +1145,10 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 		 */
 		goto out;
 
-	hwgroup->handler = NULL;
-	hwgroup->req_gen++;
-	del_timer(&hwgroup->timer);
-	spin_unlock(&hwgroup->lock);
+	hwif->handler = NULL;
+	hwif->req_gen++;
+	del_timer(&hwif->timer);
+	spin_unlock(&hwif->lock);
 
 	if (hwif->port_ops && hwif->port_ops->clear_irq)
 		hwif->port_ops->clear_irq(drive);
@@ -1159,7 +1159,7 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 	/* service this interrupt, may set handler for next interrupt */
 	startstop = handler(drive);
 
-	spin_lock_irq(&hwgroup->lock);
+	spin_lock_irq(&hwif->lock);
 	/*
 	 * Note that handler() may have set things up for another
 	 * interrupt to occur soon, but it cannot happen until
@@ -1168,7 +1168,7 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 	 * won't allow another of the same (on any CPU) until we return.
 	 */
 	if (startstop == ide_stopped) {
-		if (hwgroup->handler == NULL) {	/* paranoia */
+		if (hwif->handler == NULL) {	/* paranoia */
 			ide_unlock_port(hwif);
 			plug_device = 1;
 		} else
@@ -1178,7 +1178,7 @@ irqreturn_t ide_intr (int irq, void *dev_id)
 out_handled:
 	irq_ret = IRQ_HANDLED;
 out:
-	spin_unlock_irqrestore(&hwgroup->lock, flags);
+	spin_unlock_irqrestore(&hwif->lock, flags);
 out_early:
 	if (plug_device) {
 		ide_unlock_host(hwif->host);
@@ -1205,11 +1205,10 @@ out_early:
 
 void ide_do_drive_cmd(ide_drive_t *drive, struct request *rq)
 {
-	ide_hwgroup_t *hwgroup = drive->hwif->hwgroup;
 	struct request_queue *q = drive->queue;
 	unsigned long flags;
 
-	hwgroup->rq = NULL;
+	drive->hwif->rq = NULL;
 
 	spin_lock_irqsave(q->queue_lock, flags);
 	__elv_add_request(q, rq, ELEVATOR_INSERT_FRONT, 0);
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index ad8bd6539283..b92304d0e79a 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -822,25 +822,25 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 static void __ide_set_handler (ide_drive_t *drive, ide_handler_t *handler,
 		      unsigned int timeout, ide_expiry_t *expiry)
 {
-	ide_hwgroup_t *hwgroup = HWGROUP(drive);
-
-	BUG_ON(hwgroup->handler);
-	hwgroup->handler	= handler;
-	hwgroup->expiry		= expiry;
-	hwgroup->timer.expires	= jiffies + timeout;
-	hwgroup->req_gen_timer	= hwgroup->req_gen;
-	add_timer(&hwgroup->timer);
+	ide_hwif_t *hwif = drive->hwif;
+
+	BUG_ON(hwif->handler);
+	hwif->handler		= handler;
+	hwif->expiry		= expiry;
+	hwif->timer.expires	= jiffies + timeout;
+	hwif->req_gen_timer	= hwif->req_gen;
+	add_timer(&hwif->timer);
 }
 
 void ide_set_handler (ide_drive_t *drive, ide_handler_t *handler,
 		      unsigned int timeout, ide_expiry_t *expiry)
 {
-	ide_hwgroup_t *hwgroup = drive->hwif->hwgroup;
+	ide_hwif_t *hwif = drive->hwif;
 	unsigned long flags;
 
-	spin_lock_irqsave(&hwgroup->lock, flags);
+	spin_lock_irqsave(&hwif->lock, flags);
 	__ide_set_handler(drive, handler, timeout, expiry);
-	spin_unlock_irqrestore(&hwgroup->lock, flags);
+	spin_unlock_irqrestore(&hwif->lock, flags);
 }
 
 EXPORT_SYMBOL(ide_set_handler);
@@ -863,10 +863,9 @@ void ide_execute_command(ide_drive_t *drive, u8 cmd, ide_handler_t *handler,
 			 unsigned timeout, ide_expiry_t *expiry)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	ide_hwgroup_t *hwgroup = hwif->hwgroup;
 	unsigned long flags;
 
-	spin_lock_irqsave(&hwgroup->lock, flags);
+	spin_lock_irqsave(&hwif->lock, flags);
 	__ide_set_handler(drive, handler, timeout, expiry);
 	hwif->tp_ops->exec_command(hwif, cmd);
 	/*
@@ -876,26 +875,25 @@ void ide_execute_command(ide_drive_t *drive, u8 cmd, ide_handler_t *handler,
 	 * FIXME: we could skip this delay with care on non shared devices
 	 */
 	ndelay(400);
-	spin_unlock_irqrestore(&hwgroup->lock, flags);
+	spin_unlock_irqrestore(&hwif->lock, flags);
 }
 EXPORT_SYMBOL(ide_execute_command);
 
 void ide_execute_pkt_cmd(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	ide_hwgroup_t *hwgroup = hwif->hwgroup;
 	unsigned long flags;
 
-	spin_lock_irqsave(&hwgroup->lock, flags);
+	spin_lock_irqsave(&hwif->lock, flags);
 	hwif->tp_ops->exec_command(hwif, ATA_CMD_PACKET);
 	ndelay(400);
-	spin_unlock_irqrestore(&hwgroup->lock, flags);
+	spin_unlock_irqrestore(&hwif->lock, flags);
 }
 EXPORT_SYMBOL_GPL(ide_execute_pkt_cmd);
 
 static inline void ide_complete_drive_reset(ide_drive_t *drive, int err)
 {
-	struct request *rq = drive->hwif->hwgroup->rq;
+	struct request *rq = drive->hwif->rq;
 
 	if (rq && blk_special_request(rq) && rq->cmd[0] == REQ_DRIVE_RESET)
 		ide_end_request(drive, err ? err : 1, 0);
@@ -913,7 +911,6 @@ static ide_startstop_t do_reset1 (ide_drive_t *, int);
 static ide_startstop_t atapi_reset_pollfunc (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	ide_hwgroup_t *hwgroup = hwif->hwgroup;
 	u8 stat;
 
 	SELECT_DRIVE(drive);
@@ -923,20 +920,20 @@ static ide_startstop_t atapi_reset_pollfunc (ide_drive_t *drive)
 	if (OK_STAT(stat, 0, ATA_BUSY))
 		printk("%s: ATAPI reset complete\n", drive->name);
 	else {
-		if (time_before(jiffies, hwgroup->poll_timeout)) {
+		if (time_before(jiffies, hwif->poll_timeout)) {
 			ide_set_handler(drive, &atapi_reset_pollfunc, HZ/20, NULL);
 			/* continue polling */
 			return ide_started;
 		}
 		/* end of polling */
-		hwgroup->polling = 0;
+		hwif->polling = 0;
 		printk("%s: ATAPI reset timed-out, status=0x%02x\n",
 				drive->name, stat);
 		/* do it the old fashioned way */
 		return do_reset1(drive, 1);
 	}
 	/* done polling */
-	hwgroup->polling = 0;
+	hwif->polling = 0;
 	ide_complete_drive_reset(drive, 0);
 	return ide_stopped;
 }
@@ -968,7 +965,6 @@ static void ide_reset_report_error(ide_hwif_t *hwif, u8 err)
  */
 static ide_startstop_t reset_pollfunc (ide_drive_t *drive)
 {
-	ide_hwgroup_t *hwgroup	= HWGROUP(drive);
 	ide_hwif_t *hwif	= HWIF(drive);
 	const struct ide_port_ops *port_ops = hwif->port_ops;
 	u8 tmp;
@@ -986,7 +982,7 @@ static ide_startstop_t reset_pollfunc (ide_drive_t *drive)
 	tmp = hwif->tp_ops->read_status(hwif);
 
 	if (!OK_STAT(tmp, 0, ATA_BUSY)) {
-		if (time_before(jiffies, hwgroup->poll_timeout)) {
+		if (time_before(jiffies, hwif->poll_timeout)) {
 			ide_set_handler(drive, &reset_pollfunc, HZ/20, NULL);
 			/* continue polling */
 			return ide_started;
@@ -1007,7 +1003,7 @@ static ide_startstop_t reset_pollfunc (ide_drive_t *drive)
 		}
 	}
 out:
-	hwgroup->polling = 0;	/* done polling */
+	hwif->polling = 0;	/* done polling */
 	ide_complete_drive_reset(drive, err);
 	return ide_stopped;
 }
@@ -1081,7 +1077,6 @@ static void pre_reset(ide_drive_t *drive)
 static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	ide_hwgroup_t *hwgroup = hwif->hwgroup;
 	struct ide_io_ports *io_ports = &hwif->io_ports;
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	const struct ide_port_ops *port_ops;
@@ -1089,10 +1084,10 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	unsigned int unit;
 	DEFINE_WAIT(wait);
 
-	spin_lock_irqsave(&hwgroup->lock, flags);
+	spin_lock_irqsave(&hwif->lock, flags);
 
 	/* We must not reset with running handlers */
-	BUG_ON(hwgroup->handler != NULL);
+	BUG_ON(hwif->handler != NULL);
 
 	/* For an ATAPI device, first try an ATAPI SRST. */
 	if (drive->media != ide_disk && !do_not_try_atapi) {
@@ -1101,10 +1096,10 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 		udelay (20);
 		tp_ops->exec_command(hwif, ATA_CMD_DEV_RESET);
 		ndelay(400);
-		hwgroup->poll_timeout = jiffies + WAIT_WORSTCASE;
-		hwgroup->polling = 1;
+		hwif->poll_timeout = jiffies + WAIT_WORSTCASE;
+		hwif->polling = 1;
 		__ide_set_handler(drive, &atapi_reset_pollfunc, HZ/20, NULL);
-		spin_unlock_irqrestore(&hwgroup->lock, flags);
+		spin_unlock_irqrestore(&hwif->lock, flags);
 		return ide_started;
 	}
 
@@ -1127,9 +1122,9 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 		if (time_before_eq(timeout, now))
 			break;
 
-		spin_unlock_irqrestore(&hwgroup->lock, flags);
+		spin_unlock_irqrestore(&hwif->lock, flags);
 		timeout = schedule_timeout_uninterruptible(timeout - now);
-		spin_lock_irqsave(&hwgroup->lock, flags);
+		spin_lock_irqsave(&hwif->lock, flags);
 	} while (timeout);
 	finish_wait(&ide_park_wq, &wait);
 
@@ -1141,7 +1136,7 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 		pre_reset(&hwif->drives[unit]);
 
 	if (io_ports->ctl_addr == 0) {
-		spin_unlock_irqrestore(&hwgroup->lock, flags);
+		spin_unlock_irqrestore(&hwif->lock, flags);
 		ide_complete_drive_reset(drive, -ENXIO);
 		return ide_stopped;
 	}
@@ -1164,8 +1159,8 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	tp_ops->set_irq(hwif, drive->quirk_list == 2);
 	/* more than enough time */
 	udelay(10);
-	hwgroup->poll_timeout = jiffies + WAIT_WORSTCASE;
-	hwgroup->polling = 1;
+	hwif->poll_timeout = jiffies + WAIT_WORSTCASE;
+	hwif->polling = 1;
 	__ide_set_handler(drive, &reset_pollfunc, HZ/20, NULL);
 
 	/*
@@ -1177,7 +1172,7 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	if (port_ops && port_ops->resetproc)
 		port_ops->resetproc(drive);
 
-	spin_unlock_irqrestore(&hwgroup->lock, flags);
+	spin_unlock_irqrestore(&hwif->lock, flags);
 	return ide_started;
 }
 
diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index 9f6e33d8a8b2..09526a0de734 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -273,7 +273,7 @@ int ide_set_xfer_rate(ide_drive_t *drive, u8 rate)
 
 static void ide_dump_opcode(ide_drive_t *drive)
 {
-	struct request *rq = drive->hwif->hwgroup->rq;
+	struct request *rq = drive->hwif->rq;
 	ide_task_t *task = NULL;
 
 	if (!rq)
@@ -346,10 +346,13 @@ static void ide_dump_ata_error(ide_drive_t *drive, u8 err)
 	printk(KERN_CONT "}");
 	if ((err & (ATA_BBK | ATA_ABORTED)) == ATA_BBK ||
 	    (err & (ATA_UNC | ATA_IDNF | ATA_AMNF))) {
+		struct request *rq = drive->hwif->rq;
+
 		ide_dump_sector(drive);
-		if (HWGROUP(drive) && HWGROUP(drive)->rq)
+
+		if (rq)
 			printk(KERN_CONT ", sector=%llu",
-			       (unsigned long long)HWGROUP(drive)->rq->sector);
+			       (unsigned long long)rq->sector);
 	}
 	printk(KERN_CONT "\n");
 }
diff --git a/drivers/ide/ide-park.c b/drivers/ide/ide-park.c
index 678454ac2483..c875a957596c 100644
--- a/drivers/ide/ide-park.c
+++ b/drivers/ide/ide-park.c
@@ -7,22 +7,22 @@ DECLARE_WAIT_QUEUE_HEAD(ide_park_wq);
 
 static void issue_park_cmd(ide_drive_t *drive, unsigned long timeout)
 {
-	ide_hwgroup_t *hwgroup = drive->hwif->hwgroup;
+	ide_hwif_t *hwif = drive->hwif;
 	struct request_queue *q = drive->queue;
 	struct request *rq;
 	int rc;
 
 	timeout += jiffies;
-	spin_lock_irq(&hwgroup->lock);
+	spin_lock_irq(&hwif->lock);
 	if (drive->dev_flags & IDE_DFLAG_PARKED) {
 		int reset_timer = time_before(timeout, drive->sleep);
 		int start_queue = 0;
 
 		drive->sleep = timeout;
 		wake_up_all(&ide_park_wq);
-		if (reset_timer && del_timer(&hwgroup->timer))
+		if (reset_timer && del_timer(&hwif->timer))
 			start_queue = 1;
-		spin_unlock_irq(&hwgroup->lock);
+		spin_unlock_irq(&hwif->lock);
 
 		if (start_queue) {
 			spin_lock_irq(q->queue_lock);
@@ -31,7 +31,7 @@ static void issue_park_cmd(ide_drive_t *drive, unsigned long timeout)
 		}
 		return;
 	}
-	spin_unlock_irq(&hwgroup->lock);
+	spin_unlock_irq(&hwif->lock);
 
 	rq = blk_get_request(q, READ, __GFP_WAIT);
 	rq->cmd[0] = REQ_PARK_HEADS;
@@ -64,21 +64,21 @@ ssize_t ide_park_show(struct device *dev, struct device_attribute *attr,
 		      char *buf)
 {
 	ide_drive_t *drive = to_ide_device(dev);
-	ide_hwgroup_t *hwgroup = drive->hwif->hwgroup;
+	ide_hwif_t *hwif = drive->hwif;
 	unsigned long now;
 	unsigned int msecs;
 
 	if (drive->dev_flags & IDE_DFLAG_NO_UNLOAD)
 		return -EOPNOTSUPP;
 
-	spin_lock_irq(&hwgroup->lock);
+	spin_lock_irq(&hwif->lock);
 	now = jiffies;
 	if (drive->dev_flags & IDE_DFLAG_PARKED &&
 	    time_after(drive->sleep, now))
 		msecs = jiffies_to_msecs(drive->sleep - now);
 	else
 		msecs = 0;
-	spin_unlock_irq(&hwgroup->lock);
+	spin_unlock_irq(&hwif->lock);
 
 	return snprintf(buf, 20, "%u\n", msecs);
 }
diff --git a/drivers/ide/ide-pm.c b/drivers/ide/ide-pm.c
index 8282c6086e6a..abb84a2dd821 100644
--- a/drivers/ide/ide-pm.c
+++ b/drivers/ide/ide-pm.c
@@ -194,7 +194,7 @@ void ide_complete_pm_request(ide_drive_t *drive, struct request *rq)
 	}
 	spin_unlock_irqrestore(q->queue_lock, flags);
 
-	drive->hwif->hwgroup->rq = NULL;
+	drive->hwif->rq = NULL;
 
 	if (blk_end_request(rq, 0, 0))
 		BUG();
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 68f3c87b8284..e953b70706c2 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -949,79 +949,20 @@ static int ide_port_setup_devices(ide_hwif_t *hwif)
 	return j;
 }
 
-static ide_hwif_t *ide_ports[MAX_HWIFS];
-
-void ide_remove_port_from_hwgroup(ide_hwif_t *hwif)
-{
-	ide_hwgroup_t *hwgroup = hwif->hwgroup;
-
-	ide_ports[hwif->index] = NULL;
-
-	spin_lock_irq(&hwgroup->lock);
-	/* Free the hwgroup if we were the only member. */
-	if (--hwgroup->port_count == 0)
-		kfree(hwgroup);
-	spin_unlock_irq(&hwgroup->lock);
-}
-
 /*
- * This routine sets up the irq for an ide interface, and creates a new
- * hwgroup for the irq/hwif if none was previously assigned.
- *
- * Much of the code is for correctly detecting/handling irq sharing
- * and irq serialization situations.  This is somewhat complex because
- * it handles static as well as dynamic (PCMCIA) IDE interfaces.
+ * This routine sets up the IRQ for an IDE interface.
  */
 static int init_irq (ide_hwif_t *hwif)
 {
 	struct ide_io_ports *io_ports = &hwif->io_ports;
-	unsigned int index;
-	ide_hwgroup_t *hwgroup;
-	ide_hwif_t *match = NULL;
 	int sa = 0;
 
 	mutex_lock(&ide_cfg_mtx);
-	hwif->hwgroup = NULL;
-
-	for (index = 0; index < MAX_HWIFS; index++) {
-		ide_hwif_t *h = ide_ports[index];
+	spin_lock_init(&hwif->lock);
 
-		if (h && h->hwgroup) {  /* scan only initialized ports */
-			if (hwif->host->host_flags & IDE_HFLAG_SERIALIZE) {
-				if (hwif->host == h->host)
-					match = h;
-			}
-		}
-	}
-
-	/*
-	 * If we are still without a hwgroup, then form a new one
-	 */
-	if (match) {
-		hwgroup = match->hwgroup;
-		hwif->hwgroup = hwgroup;
-
-		spin_lock_irq(&hwgroup->lock);
-		hwgroup->port_count++;
-		spin_unlock_irq(&hwgroup->lock);
-	} else {
-		hwgroup = kmalloc_node(sizeof(*hwgroup), GFP_KERNEL|__GFP_ZERO,
-				       hwif_to_node(hwif));
-		if (hwgroup == NULL)
-			goto out_up;
-
-		spin_lock_init(&hwgroup->lock);
-
-		hwif->hwgroup = hwgroup;
-
-		hwgroup->port_count = 1;
-
-		init_timer(&hwgroup->timer);
-		hwgroup->timer.function = &ide_timer_expiry;
-		hwgroup->timer.data = (unsigned long) hwgroup;
-	}
-
-	ide_ports[hwif->index] = hwif;
+	init_timer(&hwif->timer);
+	hwif->timer.function = &ide_timer_expiry;
+	hwif->timer.data = (unsigned long)hwif;
 
 #if defined(__mc68000__)
 	sa = IRQF_SHARED;
@@ -1034,7 +975,7 @@ static int init_irq (ide_hwif_t *hwif)
 		hwif->tp_ops->set_irq(hwif, 1);
 
 	if (request_irq(hwif->irq, &ide_intr, sa, hwif->name, hwif))
-		goto out_unlink;
+		goto out_up;
 
 	if (!hwif->rqsize) {
 		if ((hwif->host_flags & IDE_HFLAG_NO_LBA48) ||
@@ -1052,14 +993,12 @@ static int init_irq (ide_hwif_t *hwif)
 	printk(KERN_INFO "%s at 0x%08lx on irq %d", hwif->name,
 		io_ports->data_addr, hwif->irq);
 #endif /* __mc68000__ */
-	if (match)
-		printk(KERN_CONT " (serialized with %s)", match->name);
+	if (hwif->host->host_flags & IDE_HFLAG_SERIALIZE)
+		printk(KERN_CONT " (serialized)");
 	printk(KERN_CONT "\n");
 
 	mutex_unlock(&ide_cfg_mtx);
 	return 0;
-out_unlink:
-	ide_remove_port_from_hwgroup(hwif);
 out_up:
 	mutex_unlock(&ide_cfg_mtx);
 	return 1;
@@ -1140,20 +1079,20 @@ EXPORT_SYMBOL_GPL(ide_init_disk);
 static void drive_release_dev (struct device *dev)
 {
 	ide_drive_t *drive = container_of(dev, ide_drive_t, gendev);
-	ide_hwgroup_t *hwgroup = drive->hwif->hwgroup;
+	ide_hwif_t *hwif = drive->hwif;
 
 	ide_proc_unregister_device(drive);
 
-	spin_lock_irq(&hwgroup->lock);
+	spin_lock_irq(&hwif->lock);
 	kfree(drive->id);
 	drive->id = NULL;
 	drive->dev_flags &= ~IDE_DFLAG_PRESENT;
 	/* Messed up locking ... */
-	spin_unlock_irq(&hwgroup->lock);
+	spin_unlock_irq(&hwif->lock);
 	blk_cleanup_queue(drive->queue);
-	spin_lock_irq(&hwgroup->lock);
+	spin_lock_irq(&hwif->lock);
 	drive->queue = NULL;
-	spin_unlock_irq(&hwgroup->lock);
+	spin_unlock_irq(&hwif->lock);
 
 	complete(&drive->gendev_rel_comp);
 }
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 5d2aa22cd6e4..e39f2f46d982 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -479,7 +479,7 @@ static void ide_tape_kfree_buffer(idetape_tape_t *tape)
 
 static int idetape_end_request(ide_drive_t *drive, int uptodate, int nr_sects)
 {
-	struct request *rq = HWGROUP(drive)->rq;
+	struct request *rq = drive->hwif->rq;
 	idetape_tape_t *tape = drive->driver_data;
 	unsigned long flags;
 	int error;
@@ -531,7 +531,7 @@ static void ide_tape_callback(ide_drive_t *drive, int dsc)
 			printk(KERN_ERR "ide-tape: Error in REQUEST SENSE "
 					"itself - Aborting request!\n");
 	} else if (pc->c[0] == READ_6 || pc->c[0] == WRITE_6) {
-		struct request *rq = drive->hwif->hwgroup->rq;
+		struct request *rq = drive->hwif->rq;
 		int blocks = pc->xferred / tape->blk_size;
 
 		tape->avg_size += blocks * tape->blk_size;
@@ -576,7 +576,7 @@ static void ide_tape_callback(ide_drive_t *drive, int dsc)
 
 /*
  * Postpone the current request so that ide.c will be able to service requests
- * from another device on the same hwgroup while we are polling for DSC.
+ * from another device on the same port while we are polling for DSC.
  */
 static void idetape_postpone_request(ide_drive_t *drive)
 {
@@ -584,7 +584,8 @@ static void idetape_postpone_request(ide_drive_t *drive)
 
 	debug_log(DBG_PROCS, "Enter %s\n", __func__);
 
-	tape->postponed_rq = HWGROUP(drive)->rq;
+	tape->postponed_rq = drive->hwif->rq;
+
 	ide_stall_queue(drive, tape->dsc_poll_freq);
 }
 
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index bf4fb9d8d176..693e8d15fb78 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -361,7 +361,7 @@ static ide_startstop_t task_in_unexpected(ide_drive_t *drive, struct request *rq
 static ide_startstop_t task_in_intr(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	struct request *rq = hwif->hwgroup->rq;
+	struct request *rq = hwif->rq;
 	u8 stat = hwif->tp_ops->read_status(hwif);
 
 	/* Error? */
@@ -395,7 +395,7 @@ static ide_startstop_t task_in_intr(ide_drive_t *drive)
 static ide_startstop_t task_out_intr (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	struct request *rq = HWGROUP(drive)->rq;
+	struct request *rq = hwif->rq;
 	u8 stat = hwif->tp_ops->read_status(hwif);
 
 	if (!OK_STAT(stat, DRIVE_READY, drive->bad_wstat))
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 5bc2e4782a55..9e403282dfa7 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -189,8 +189,6 @@ void ide_unregister(ide_hwif_t *hwif)
 
 	free_irq(hwif->irq, hwif);
 
-	ide_remove_port_from_hwgroup(hwif);
-
 	device_unregister(hwif->portdev);
 	device_unregister(&hwif->gendev);
 	wait_for_completion(&hwif->gendev_rel_comp);
@@ -315,7 +313,6 @@ static int set_pio_mode_abuse(ide_hwif_t *hwif, u8 req_pio)
 static int set_pio_mode(ide_drive_t *drive, int arg)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	ide_hwgroup_t *hwgroup = hwif->hwgroup;
 	const struct ide_port_ops *port_ops = hwif->port_ops;
 
 	if (arg < 0 || arg > 255)
@@ -330,9 +327,9 @@ static int set_pio_mode(ide_drive_t *drive, int arg)
 			unsigned long flags;
 
 			/* take lock for IDE_DFLAG_[NO_]UNMASK/[NO_]IO_32BIT */
-			spin_lock_irqsave(&hwgroup->lock, flags);
+			spin_lock_irqsave(&hwif->lock, flags);
 			port_ops->set_pio_mode(drive, arg);
-			spin_unlock_irqrestore(&hwgroup->lock, flags);
+			spin_unlock_irqrestore(&hwif->lock, flags);
 		} else
 			port_ops->set_pio_mode(drive, arg);
 	} else {
diff --git a/drivers/ide/pdc202xx_old.c b/drivers/ide/pdc202xx_old.c
index 624e62e5cc9a..072ef70bf061 100644
--- a/drivers/ide/pdc202xx_old.c
+++ b/drivers/ide/pdc202xx_old.c
@@ -169,8 +169,8 @@ static void pdc202xx_dma_start(ide_drive_t *drive)
 	if (drive->current_speed > XFER_UDMA_2)
 		pdc_old_enable_66MHz_clock(drive->hwif);
 	if (drive->media != ide_disk || (drive->dev_flags & IDE_DFLAG_LBA48)) {
-		struct request *rq	= HWGROUP(drive)->rq;
 		ide_hwif_t *hwif	= HWIF(drive);
+		struct request *rq	= hwif->rq;
 		unsigned long high_16	= hwif->extra_base - 16;
 		unsigned long atapi_reg	= high_16 + (hwif->channel ? 0x24 : 0x20);
 		u32 word_count	= 0;
diff --git a/drivers/ide/pmac.c b/drivers/ide/pmac.c
index 7c481bb56fab..899b96baf215 100644
--- a/drivers/ide/pmac.c
+++ b/drivers/ide/pmac.c
@@ -1516,7 +1516,7 @@ pmac_ide_dma_setup(ide_drive_t *drive)
 	ide_hwif_t *hwif = HWIF(drive);
 	pmac_ide_hwif_t *pmif =
 		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
-	struct request *rq = HWGROUP(drive)->rq;
+	struct request *rq = hwif->rq;
 	u8 unit = drive->dn & 1, ata4 = (pmif->kind == controller_kl_ata4);
 
 	if (!pmac_ide_build_dmatable(drive, rq)) {
diff --git a/drivers/ide/scc_pata.c b/drivers/ide/scc_pata.c
index 0f48f9dacfa5..e966113fd569 100644
--- a/drivers/ide/scc_pata.c
+++ b/drivers/ide/scc_pata.c
@@ -316,7 +316,7 @@ static void scc_dma_host_set(ide_drive_t *drive, int on)
 static int scc_dma_setup(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	struct request *rq = HWGROUP(drive)->rq;
+	struct request *rq = hwif->rq;
 	unsigned int reading;
 	u8 dma_stat;
 
@@ -405,7 +405,7 @@ static int scc_dma_end(ide_drive_t *drive)
 			       drive->name);
 			data_loss = 1;
 			if (retry++) {
-				struct request *rq = HWGROUP(drive)->rq;
+				struct request *rq = hwif->rq;
 				int unit;
 				/* ERROR_RESET and drive->crc_count are needed
 				 * to reduce DMA transfer mode in retry process.
diff --git a/drivers/ide/sgiioc4.c b/drivers/ide/sgiioc4.c
index a687a7dfea6f..c68b71b1087b 100644
--- a/drivers/ide/sgiioc4.c
+++ b/drivers/ide/sgiioc4.c
@@ -492,7 +492,7 @@ use_pio_instead:
 
 static int sgiioc4_dma_setup(ide_drive_t *drive)
 {
-	struct request *rq = HWGROUP(drive)->rq;
+	struct request *rq = drive->hwif->rq;
 	unsigned int count = 0;
 	int ddir;
 
diff --git a/drivers/ide/tc86c001.c b/drivers/ide/tc86c001.c
index 93e2cce4b296..accb379bcad6 100644
--- a/drivers/ide/tc86c001.c
+++ b/drivers/ide/tc86c001.c
@@ -64,11 +64,10 @@ static int tc86c001_timer_expiry(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	ide_expiry_t *expiry	= ide_get_hwifdata(hwif);
-	ide_hwgroup_t *hwgroup	= HWGROUP(drive);
 	u8 dma_stat		= inb(hwif->dma_base + ATA_DMA_STATUS);
 
 	/* Restore a higher level driver's expiry handler first. */
-	hwgroup->expiry	= expiry;
+	hwif->expiry = expiry;
 
 	if ((dma_stat & 5) == 1) {	/* DMA active and no interrupt */
 		unsigned long sc_base	= hwif->config_data;
@@ -111,10 +110,9 @@ static int tc86c001_timer_expiry(ide_drive_t *drive)
 static void tc86c001_dma_start(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
-	ide_hwgroup_t *hwgroup	= HWGROUP(drive);
 	unsigned long sc_base	= hwif->config_data;
 	unsigned long twcr_port	= sc_base + (drive->dn ? 0x06 : 0x04);
-	unsigned long nsectors	= hwgroup->rq->nr_sectors;
+	unsigned long nsectors	= hwif->rq->nr_sectors;
 
 	/*
 	 * We have to manually load the sector count and size into
@@ -125,8 +123,8 @@ static void tc86c001_dma_start(ide_drive_t *drive)
 	outw(SECTOR_SIZE / 2, twcr_port); /* Transfer Word Count 1/2 */
 
 	/* Install our timeout expiry hook, saving the current handler... */
-	ide_set_hwifdata(hwif, hwgroup->expiry);
-	hwgroup->expiry = &tc86c001_timer_expiry;
+	ide_set_hwifdata(hwif, hwif->expiry);
+	hwif->expiry = &tc86c001_timer_expiry;
 
 	ide_dma_start(drive);
 }
diff --git a/drivers/ide/trm290.c b/drivers/ide/trm290.c
index 2a5ea90cf8b8..79a03df118d8 100644
--- a/drivers/ide/trm290.c
+++ b/drivers/ide/trm290.c
@@ -184,7 +184,7 @@ static void trm290_dma_exec_cmd(ide_drive_t *drive, u8 command)
 static int trm290_dma_setup(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	struct request *rq = hwif->hwgroup->rq;
+	struct request *rq = hwif->rq;
 	unsigned int count, rw;
 
 	if (rq_data_dir(rq)) {
diff --git a/drivers/ide/tx4939ide.c b/drivers/ide/tx4939ide.c
index 4a8c5a21bd4c..1ac27ac7283b 100644
--- a/drivers/ide/tx4939ide.c
+++ b/drivers/ide/tx4939ide.c
@@ -293,7 +293,7 @@ static int tx4939ide_dma_setup(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	void __iomem *base = TX4939IDE_BASE(hwif);
-	struct request *rq = hwif->hwgroup->rq;
+	struct request *rq = hwif->rq;
 	u8 reading;
 	int nent;
 
diff --git a/drivers/ide/umc8672.c b/drivers/ide/umc8672.c
index e29978cf6197..0608d41fb6d0 100644
--- a/drivers/ide/umc8672.c
+++ b/drivers/ide/umc8672.c
@@ -106,22 +106,21 @@ static void umc_set_speeds(u8 speeds[])
 
 static void umc_set_pio_mode(ide_drive_t *drive, const u8 pio)
 {
-	ide_hwif_t *hwif = drive->hwif;
-	ide_hwgroup_t *mate_hwgroup = hwif->mate ? hwif->mate->hwgroup : NULL;
+	ide_hwif_t *hwif = drive->hwif, *mate = hwif->mate;
 	unsigned long uninitialized_var(flags);
 
 	printk("%s: setting umc8672 to PIO mode%d (speed %d)\n",
 		drive->name, pio, pio_to_umc[pio]);
-	if (mate_hwgroup)
-		spin_lock_irqsave(&mate_hwgroup->lock, flags);
-	if (mate_hwgroup && mate_hwgroup->handler) {
+	if (mate)
+		spin_lock_irqsave(&mate->lock, flags);
+	if (mate && mate->handler) {
 		printk(KERN_ERR "umc8672: other interface is busy: exiting tune_umc()\n");
 	} else {
 		current_speeds[drive->name[2] - 'a'] = pio_to_umc[pio];
 		umc_set_speeds(current_speeds);
 	}
-	if (mate_hwgroup)
-		spin_unlock_irqrestore(&mate_hwgroup->lock, flags);
+	if (mate)
+		spin_unlock_irqrestore(&mate->lock, flags);
 }
 
 static const struct ide_port_ops umc8672_port_ops = {
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 00df155b5a02..f27f130ba000 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -42,7 +42,6 @@ typedef unsigned char	byte;	/* used everywhere */
 #define ERROR_RECAL	1	/* Recalibrate every 2nd retry */
 
 #define HWIF(drive)		((ide_hwif_t *)((drive)->hwif))
-#define HWGROUP(drive)		((ide_hwgroup_t *)(HWIF(drive)->hwgroup))
 
 /*
  * Definitions for accessing IDE controller registers
@@ -750,7 +749,6 @@ struct ide_host;
 
 typedef struct hwif_s {
 	struct hwif_s *mate;		/* other hwif from same PCI chip */
-	struct hwgroup_s *hwgroup;	/* actually (ide_hwgroup_t *) */
 	struct proc_dir_entry *proc;	/* /proc/ide/ directory entry */
 
 	struct ide_host *host;
@@ -840,6 +838,30 @@ typedef struct hwif_s {
 #ifdef CONFIG_BLK_DEV_IDEACPI
 	struct ide_acpi_hwif_link *acpidata;
 #endif
+
+	/* IRQ handler, if active */
+	ide_startstop_t	(*handler)(ide_drive_t *);
+
+	/* BOOL: polling active & poll_timeout field valid */
+	unsigned int polling : 1;
+
+	/* current drive */
+	ide_drive_t *cur_dev;
+
+	/* current request */
+	struct request *rq;
+
+	/* failsafe timer */
+	struct timer_list timer;
+	/* timeout value during long polls */
+	unsigned long poll_timeout;
+	/* queried upon timeouts */
+	int (*expiry)(ide_drive_t *);
+
+	int req_gen;
+	int req_gen_timer;
+
+	spinlock_t lock;
 } ____cacheline_internodealigned_in_smp ide_hwif_t;
 
 #define MAX_HOST_PORTS 4
@@ -868,34 +890,6 @@ typedef int (ide_expiry_t)(ide_drive_t *);
 /* used by ide-cd, ide-floppy, etc. */
 typedef void (xfer_func_t)(ide_drive_t *, struct request *rq, void *, unsigned);
 
-typedef struct hwgroup_s {
-		/* irq handler, if active */
-	ide_startstop_t	(*handler)(ide_drive_t *);
-
-		/* BOOL: polling active & poll_timeout field valid */
-	unsigned int polling	: 1;
-
-		/* current drive */
-	ide_drive_t *cur_dev;
-
-		/* current request */
-	struct request *rq;
-
-		/* failsafe timer */
-	struct timer_list timer;
-		/* timeout value during long polls */
-	unsigned long poll_timeout;
-		/* queried upon timeouts */
-	int (*expiry)(ide_drive_t *);
-
-	int req_gen;
-	int req_gen_timer;
-
-	spinlock_t lock;
-
-	int port_count;
-} ide_hwgroup_t;
-
 typedef struct ide_driver_s ide_driver_t;
 
 extern struct mutex ide_setting_mtx;
@@ -1512,7 +1506,6 @@ static inline void ide_acpi_port_init_devices(ide_hwif_t *hwif) { ; }
 static inline void ide_acpi_set_state(ide_hwif_t *hwif, int on) {}
 #endif
 
-void ide_remove_port_from_hwgroup(ide_hwif_t *);
 void ide_unregister(ide_hwif_t *);
 
 void ide_register_region(struct gendisk *);
-- 
cgit v1.2.3


From b40d1b88f1001f0224c63fa2c008914514bcef33 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:51 +0100
Subject: ide: move ide_init_port_data() and friends to ide-probe.c

* Move IDE_DEFAULT_MAX_FAILURES to <linux/ide.h>.

* Move ide_cfg_mtx, ide_hwif_to_major[], ide_port_init_devices_data(),
  ide_init_port_data(), ide_init_port_hw() and ide_unregister() to
  ide-probe.c from ide.c.

* Make ide_unregister(), ide_init_port_data(), ide_init_port_hw()
  and ide_cfg_mtx static.

While at it:

* Remove stale ide_init_port_data() documentation and ide_lock extern.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-probe.c | 135 ++++++++++++++++++++++++++++++++++++++++++++
 drivers/ide/ide.c       | 146 ------------------------------------------------
 include/linux/ide.h     |   9 +--
 3 files changed, 136 insertions(+), 154 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index e953b70706c2..09ea50118e63 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -918,6 +918,8 @@ static int ide_init_queue(ide_drive_t *drive)
 	return 0;
 }
 
+static DEFINE_MUTEX(ide_cfg_mtx);
+
 /*
  * For any present drive:
  * - allocate the block device queue
@@ -1273,6 +1275,69 @@ static void ide_port_cable_detect(ide_hwif_t *hwif)
 	}
 }
 
+static const u8 ide_hwif_to_major[] =
+	{ IDE0_MAJOR, IDE1_MAJOR, IDE2_MAJOR, IDE3_MAJOR, IDE4_MAJOR,
+	  IDE5_MAJOR, IDE6_MAJOR, IDE7_MAJOR, IDE8_MAJOR, IDE9_MAJOR };
+
+static void ide_port_init_devices_data(ide_hwif_t *hwif)
+{
+	int unit;
+
+	for (unit = 0; unit < MAX_DRIVES; ++unit) {
+		ide_drive_t *drive = &hwif->drives[unit];
+		u8 j = (hwif->index * MAX_DRIVES) + unit;
+
+		memset(drive, 0, sizeof(*drive));
+
+		drive->media			= ide_disk;
+		drive->select			= (unit << 4) | ATA_DEVICE_OBS;
+		drive->hwif			= hwif;
+		drive->ready_stat		= ATA_DRDY;
+		drive->bad_wstat		= BAD_W_STAT;
+		drive->special.b.recalibrate	= 1;
+		drive->special.b.set_geometry	= 1;
+		drive->name[0]			= 'h';
+		drive->name[1]			= 'd';
+		drive->name[2]			= 'a' + j;
+		drive->max_failures		= IDE_DEFAULT_MAX_FAILURES;
+
+		INIT_LIST_HEAD(&drive->list);
+		init_completion(&drive->gendev_rel_comp);
+	}
+}
+
+static void ide_init_port_data(ide_hwif_t *hwif, unsigned int index)
+{
+	/* bulk initialize hwif & drive info with zeros */
+	memset(hwif, 0, sizeof(ide_hwif_t));
+
+	/* fill in any non-zero initial values */
+	hwif->index	= index;
+	hwif->major	= ide_hwif_to_major[index];
+
+	hwif->name[0]	= 'i';
+	hwif->name[1]	= 'd';
+	hwif->name[2]	= 'e';
+	hwif->name[3]	= '0' + index;
+
+	init_completion(&hwif->gendev_rel_comp);
+
+	hwif->tp_ops = &default_tp_ops;
+
+	ide_port_init_devices_data(hwif);
+}
+
+static void ide_init_port_hw(ide_hwif_t *hwif, hw_regs_t *hw)
+{
+	memcpy(&hwif->io_ports, &hw->io_ports, sizeof(hwif->io_ports));
+	hwif->irq = hw->irq;
+	hwif->chipset = hw->chipset;
+	hwif->dev = hw->dev;
+	hwif->gendev.parent = hw->parent ? hw->parent : hw->dev;
+	hwif->ack_intr = hw->ack_intr;
+	hwif->config_data = hw->config;
+}
+
 static unsigned int ide_indexes;
 
 /**
@@ -1503,6 +1568,76 @@ int ide_host_add(const struct ide_port_info *d, hw_regs_t **hws,
 }
 EXPORT_SYMBOL_GPL(ide_host_add);
 
+static void __ide_port_unregister_devices(ide_hwif_t *hwif)
+{
+	int i;
+
+	for (i = 0; i < MAX_DRIVES; i++) {
+		ide_drive_t *drive = &hwif->drives[i];
+
+		if (drive->dev_flags & IDE_DFLAG_PRESENT) {
+			device_unregister(&drive->gendev);
+			wait_for_completion(&drive->gendev_rel_comp);
+		}
+	}
+}
+
+void ide_port_unregister_devices(ide_hwif_t *hwif)
+{
+	mutex_lock(&ide_cfg_mtx);
+	__ide_port_unregister_devices(hwif);
+	hwif->present = 0;
+	ide_port_init_devices_data(hwif);
+	mutex_unlock(&ide_cfg_mtx);
+}
+EXPORT_SYMBOL_GPL(ide_port_unregister_devices);
+
+/**
+ *	ide_unregister		-	free an IDE interface
+ *	@hwif: IDE interface
+ *
+ *	Perform the final unregister of an IDE interface.
+ *
+ *	Locking:
+ *	The caller must not hold the IDE locks.
+ *
+ *	It is up to the caller to be sure there is no pending I/O here,
+ *	and that the interface will not be reopened (present/vanishing
+ *	locking isn't yet done BTW).
+ */
+
+static void ide_unregister(ide_hwif_t *hwif)
+{
+	BUG_ON(in_interrupt());
+	BUG_ON(irqs_disabled());
+
+	mutex_lock(&ide_cfg_mtx);
+
+	if (hwif->present) {
+		__ide_port_unregister_devices(hwif);
+		hwif->present = 0;
+	}
+
+	ide_proc_unregister_port(hwif);
+
+	free_irq(hwif->irq, hwif);
+
+	device_unregister(hwif->portdev);
+	device_unregister(&hwif->gendev);
+	wait_for_completion(&hwif->gendev_rel_comp);
+
+	/*
+	 * Remove us from the kernel's knowledge
+	 */
+	blk_unregister_region(MKDEV(hwif->major, 0), MAX_DRIVES<<PARTN_BITS);
+	kfree(hwif->sg_table);
+	unregister_blkdev(hwif->major, hwif->name);
+
+	ide_release_dma_engine(hwif);
+
+	mutex_unlock(&ide_cfg_mtx);
+}
+
 void ide_host_free(struct ide_host *host)
 {
 	ide_hwif_t *hwif;
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 6538b63158bf..c1bb0f6784a9 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -60,154 +60,8 @@
 #include <linux/completion.h>
 #include <linux/device.h>
 
-
-/* default maximum number of failures */
-#define IDE_DEFAULT_MAX_FAILURES 	1
-
 struct class *ide_port_class;
 
-static const u8 ide_hwif_to_major[] = { IDE0_MAJOR, IDE1_MAJOR,
-					IDE2_MAJOR, IDE3_MAJOR,
-					IDE4_MAJOR, IDE5_MAJOR,
-					IDE6_MAJOR, IDE7_MAJOR,
-					IDE8_MAJOR, IDE9_MAJOR };
-
-DEFINE_MUTEX(ide_cfg_mtx);
-
-static void ide_port_init_devices_data(ide_hwif_t *);
-
-/*
- * Do not even *think* about calling this!
- */
-void ide_init_port_data(ide_hwif_t *hwif, unsigned int index)
-{
-	/* bulk initialize hwif & drive info with zeros */
-	memset(hwif, 0, sizeof(ide_hwif_t));
-
-	/* fill in any non-zero initial values */
-	hwif->index	= index;
-	hwif->major	= ide_hwif_to_major[index];
-
-	hwif->name[0]	= 'i';
-	hwif->name[1]	= 'd';
-	hwif->name[2]	= 'e';
-	hwif->name[3]	= '0' + index;
-
-	init_completion(&hwif->gendev_rel_comp);
-
-	hwif->tp_ops = &default_tp_ops;
-
-	ide_port_init_devices_data(hwif);
-}
-
-static void ide_port_init_devices_data(ide_hwif_t *hwif)
-{
-	int unit;
-
-	for (unit = 0; unit < MAX_DRIVES; ++unit) {
-		ide_drive_t *drive = &hwif->drives[unit];
-		u8 j = (hwif->index * MAX_DRIVES) + unit;
-
-		memset(drive, 0, sizeof(*drive));
-
-		drive->media			= ide_disk;
-		drive->select			= (unit << 4) | ATA_DEVICE_OBS;
-		drive->hwif			= hwif;
-		drive->ready_stat		= ATA_DRDY;
-		drive->bad_wstat		= BAD_W_STAT;
-		drive->special.b.recalibrate	= 1;
-		drive->special.b.set_geometry	= 1;
-		drive->name[0]			= 'h';
-		drive->name[1]			= 'd';
-		drive->name[2]			= 'a' + j;
-		drive->max_failures		= IDE_DEFAULT_MAX_FAILURES;
-
-		INIT_LIST_HEAD(&drive->list);
-		init_completion(&drive->gendev_rel_comp);
-	}
-}
-
-static void __ide_port_unregister_devices(ide_hwif_t *hwif)
-{
-	int i;
-
-	for (i = 0; i < MAX_DRIVES; i++) {
-		ide_drive_t *drive = &hwif->drives[i];
-
-		if (drive->dev_flags & IDE_DFLAG_PRESENT) {
-			device_unregister(&drive->gendev);
-			wait_for_completion(&drive->gendev_rel_comp);
-		}
-	}
-}
-
-void ide_port_unregister_devices(ide_hwif_t *hwif)
-{
-	mutex_lock(&ide_cfg_mtx);
-	__ide_port_unregister_devices(hwif);
-	hwif->present = 0;
-	ide_port_init_devices_data(hwif);
-	mutex_unlock(&ide_cfg_mtx);
-}
-EXPORT_SYMBOL_GPL(ide_port_unregister_devices);
-
-/**
- *	ide_unregister		-	free an IDE interface
- *	@hwif: IDE interface
- *
- *	Perform the final unregister of an IDE interface.
- *
- *	Locking:
- *	The caller must not hold the IDE locks.
- *
- *	It is up to the caller to be sure there is no pending I/O here,
- *	and that the interface will not be reopened (present/vanishing
- *	locking isn't yet done BTW).
- */
-
-void ide_unregister(ide_hwif_t *hwif)
-{
-	BUG_ON(in_interrupt());
-	BUG_ON(irqs_disabled());
-
-	mutex_lock(&ide_cfg_mtx);
-
-	if (hwif->present) {
-		__ide_port_unregister_devices(hwif);
-		hwif->present = 0;
-	}
-
-	ide_proc_unregister_port(hwif);
-
-	free_irq(hwif->irq, hwif);
-
-	device_unregister(hwif->portdev);
-	device_unregister(&hwif->gendev);
-	wait_for_completion(&hwif->gendev_rel_comp);
-
-	/*
-	 * Remove us from the kernel's knowledge
-	 */
-	blk_unregister_region(MKDEV(hwif->major, 0), MAX_DRIVES<<PARTN_BITS);
-	kfree(hwif->sg_table);
-	unregister_blkdev(hwif->major, hwif->name);
-
-	ide_release_dma_engine(hwif);
-
-	mutex_unlock(&ide_cfg_mtx);
-}
-
-void ide_init_port_hw(ide_hwif_t *hwif, hw_regs_t *hw)
-{
-	memcpy(&hwif->io_ports, &hw->io_ports, sizeof(hwif->io_ports));
-	hwif->irq = hw->irq;
-	hwif->chipset = hw->chipset;
-	hwif->dev = hw->dev;
-	hwif->gendev.parent = hw->parent ? hw->parent : hw->dev;
-	hwif->ack_intr = hw->ack_intr;
-	hwif->config_data = hw->config;
-}
-
 /*
  *	Locks for IDE setting functionality
  */
diff --git a/include/linux/ide.h b/include/linux/ide.h
index f27f130ba000..ee2f461882ad 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -37,6 +37,7 @@ typedef unsigned char	byte;	/* used everywhere */
 /*
  * Probably not wise to fiddle with these
  */
+#define IDE_DEFAULT_MAX_FAILURES	1
 #define ERROR_MAX	8	/* Max read/write errors per sector */
 #define ERROR_RESET	3	/* Reset controller every 4th retry */
 #define ERROR_RECAL	1	/* Recalibrate every 2nd retry */
@@ -184,9 +185,6 @@ typedef struct hw_regs_s {
 	unsigned long	config;
 } hw_regs_t;
 
-void ide_init_port_data(struct hwif_s *, unsigned int);
-void ide_init_port_hw(struct hwif_s *, hw_regs_t *);
-
 static inline void ide_std_init_ports(hw_regs_t *hw,
 				      unsigned long io_addr,
 				      unsigned long ctl_addr)
@@ -1506,8 +1504,6 @@ static inline void ide_acpi_port_init_devices(ide_hwif_t *hwif) { ; }
 static inline void ide_acpi_set_state(ide_hwif_t *hwif, int on) {}
 #endif
 
-void ide_unregister(ide_hwif_t *);
-
 void ide_register_region(struct gendisk *);
 void ide_unregister_region(struct gendisk *);
 
@@ -1592,9 +1588,6 @@ static inline void ide_set_max_pio(ide_drive_t *drive)
 	ide_set_pio(drive, 255);
 }
 
-extern spinlock_t ide_lock;
-extern struct mutex ide_cfg_mtx;
-
 #define local_irq_set(flags)	do { local_save_flags((flags)); local_irq_enable_in_hardirq(); } while (0)
 
 char *ide_media_string(ide_drive_t *);
-- 
cgit v1.2.3


From 898ec223fea2a2df88035e58dbf50f493577e225 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:52 +0100
Subject: ide: remove HWIF() macro

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/aec62xx.c      |  4 ++--
 drivers/ide/alim15x3.c     |  4 ++--
 drivers/ide/au1xxx-ide.c   |  4 ++--
 drivers/ide/cmd64x.c       | 14 +++++++-------
 drivers/ide/cs5520.c       |  2 +-
 drivers/ide/cy82c693.c     |  2 +-
 drivers/ide/hpt366.c       | 18 +++++++++---------
 drivers/ide/icside.c       | 10 +++++-----
 drivers/ide/ide-acpi.c     |  4 ++--
 drivers/ide/ide-disk.c     |  4 ++--
 drivers/ide/ide-io.c       |  8 ++++----
 drivers/ide/ide-iops.c     |  4 ++--
 drivers/ide/ide-pm.c       |  4 ++--
 drivers/ide/ide-probe.c    | 12 ++++++------
 drivers/ide/ide-taskfile.c |  2 +-
 drivers/ide/it8213.c       |  4 ++--
 drivers/ide/ns87415.c      |  6 +++---
 drivers/ide/pdc202xx_new.c |  4 ++--
 drivers/ide/pdc202xx_old.c | 10 +++++-----
 drivers/ide/piix.c         |  6 +++---
 drivers/ide/pmac.c         |  4 ++--
 drivers/ide/qd65xx.c       |  7 ++++---
 drivers/ide/sc1200.c       |  6 +++---
 drivers/ide/scc_pata.c     |  8 ++++----
 drivers/ide/serverworks.c  |  2 +-
 drivers/ide/sgiioc4.c      | 12 ++++++------
 drivers/ide/siimage.c      | 10 +++++-----
 drivers/ide/sis5513.c      |  2 +-
 drivers/ide/sl82c105.c     |  4 ++--
 drivers/ide/slc90e66.c     |  4 ++--
 drivers/ide/tc86c001.c     |  6 +++---
 drivers/ide/triflex.c      |  2 +-
 drivers/ide/trm290.c       |  8 ++++----
 drivers/ide/via82cxxx.c    |  2 +-
 include/linux/ide.h        |  2 --
 35 files changed, 102 insertions(+), 103 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/aec62xx.c b/drivers/ide/aec62xx.c
index 4142c698e0d3..4485b9c6f0e6 100644
--- a/drivers/ide/aec62xx.c
+++ b/drivers/ide/aec62xx.c
@@ -83,7 +83,7 @@ static u8 pci_bus_clock_list_ultra (u8 speed, struct chipset_bus_clock_list_entr
 
 static void aec6210_set_mode(ide_drive_t *drive, const u8 speed)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	struct ide_host *host	= pci_get_drvdata(dev);
 	struct chipset_bus_clock_list_entry *bus_clock = host->host_priv;
@@ -111,7 +111,7 @@ static void aec6210_set_mode(ide_drive_t *drive, const u8 speed)
 
 static void aec6260_set_mode(ide_drive_t *drive, const u8 speed)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	struct ide_host *host	= pci_get_drvdata(dev);
 	struct chipset_bus_clock_list_entry *bus_clock = host->host_priv;
diff --git a/drivers/ide/alim15x3.c b/drivers/ide/alim15x3.c
index 290204e89eaf..8d217430c997 100644
--- a/drivers/ide/alim15x3.c
+++ b/drivers/ide/alim15x3.c
@@ -68,7 +68,7 @@ static struct pci_dev *isa_dev;
 
 static void ali_set_pio_mode(ide_drive_t *drive, const u8 pio)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
 	struct ide_timing *t = ide_timing_find_mode(XFER_PIO_0 + pio);
 	int s_time = t->setup, a_time = t->active, c_time = t->cycle;
@@ -150,7 +150,7 @@ static u8 ali_udma_filter(ide_drive_t *drive)
 
 static void ali_set_dma_mode(ide_drive_t *drive, const u8 speed)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u8 speed1		= speed;
 	u8 unit			= drive->dn & 1;
diff --git a/drivers/ide/au1xxx-ide.c b/drivers/ide/au1xxx-ide.c
index db412bc772d1..11f2c8f3db48 100644
--- a/drivers/ide/au1xxx-ide.c
+++ b/drivers/ide/au1xxx-ide.c
@@ -212,7 +212,7 @@ static void auide_set_dma_mode(ide_drive_t *drive, const u8 speed)
 static int auide_build_dmatable(ide_drive_t *drive)
 {
 	int i, iswrite, count = 0;
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct request *rq = hwif->rq;
 	_auide_hwif *ahwif = &auide_hwif;
 	struct scatterlist *sg;
@@ -286,7 +286,7 @@ static int auide_build_dmatable(ide_drive_t *drive)
 
 static int auide_dma_end(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 
 	if (hwif->sg_nents) {
 		ide_destroy_dmatable(drive);
diff --git a/drivers/ide/cmd64x.c b/drivers/ide/cmd64x.c
index 3623bf013bcf..d1fc198719b2 100644
--- a/drivers/ide/cmd64x.c
+++ b/drivers/ide/cmd64x.c
@@ -115,7 +115,7 @@ static void program_cycle_times (ide_drive_t *drive, int cycle_time, int active_
  */
 static void cmd64x_tune_pio(ide_drive_t *drive, const u8 pio)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	struct ide_timing *t	= ide_timing_find_mode(XFER_PIO_0 + pio);
 	unsigned int cycle_time;
@@ -180,7 +180,7 @@ static void cmd64x_set_pio_mode(ide_drive_t *drive, const u8 pio)
 
 static void cmd64x_set_dma_mode(ide_drive_t *drive, const u8 speed)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u8 unit			= drive->dn & 0x01;
 	u8 regU = 0, pciU	= hwif->channel ? UDIDETCR1 : UDIDETCR0;
@@ -226,7 +226,7 @@ static void cmd64x_set_dma_mode(ide_drive_t *drive, const u8 speed)
 
 static int cmd648_dma_end(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	unsigned long base	= hwif->dma_base - (hwif->channel * 8);
 	int err			= ide_dma_end(drive);
 	u8  irq_mask		= hwif->channel ? MRDMODE_INTR_CH1 :
@@ -242,7 +242,7 @@ static int cmd648_dma_end(ide_drive_t *drive)
 
 static int cmd64x_dma_end(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	int irq_reg		= hwif->channel ? ARTTIM23 : CFR;
 	u8  irq_mask		= hwif->channel ? ARTTIM23_INTR_CH1 :
@@ -259,7 +259,7 @@ static int cmd64x_dma_end(ide_drive_t *drive)
 
 static int cmd648_dma_test_irq(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	unsigned long base	= hwif->dma_base - (hwif->channel * 8);
 	u8 irq_mask		= hwif->channel ? MRDMODE_INTR_CH1 :
 						  MRDMODE_INTR_CH0;
@@ -282,7 +282,7 @@ static int cmd648_dma_test_irq(ide_drive_t *drive)
 
 static int cmd64x_dma_test_irq(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	int irq_reg		= hwif->channel ? ARTTIM23 : CFR;
 	u8  irq_mask		= hwif->channel ? ARTTIM23_INTR_CH1 :
@@ -313,7 +313,7 @@ static int cmd64x_dma_test_irq(ide_drive_t *drive)
 
 static int cmd646_1_dma_end(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	u8 dma_stat = 0, dma_cmd = 0;
 
 	drive->waiting_for_dma = 0;
diff --git a/drivers/ide/cs5520.c b/drivers/ide/cs5520.c
index 5efb467f8fa0..d003bec56ff9 100644
--- a/drivers/ide/cs5520.c
+++ b/drivers/ide/cs5520.c
@@ -59,7 +59,7 @@ static struct pio_clocks cs5520_pio_clocks[]={
 
 static void cs5520_set_pio_mode(ide_drive_t *drive, const u8 pio)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct pci_dev *pdev = to_pci_dev(hwif->dev);
 	int controller = drive->dn > 1 ? 1 : 0;
 
diff --git a/drivers/ide/cy82c693.c b/drivers/ide/cy82c693.c
index d37baf8ecc5f..74fc5401f407 100644
--- a/drivers/ide/cy82c693.c
+++ b/drivers/ide/cy82c693.c
@@ -203,7 +203,7 @@ static void cy82c693_set_dma_mode(ide_drive_t *drive, const u8 mode)
 
 static void cy82c693_set_pio_mode(ide_drive_t *drive, const u8 pio)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
 	pio_clocks_t pclk;
 	unsigned int addrCtrl;
diff --git a/drivers/ide/hpt366.c b/drivers/ide/hpt366.c
index b18e10d99d2e..a18a02a96977 100644
--- a/drivers/ide/hpt366.c
+++ b/drivers/ide/hpt366.c
@@ -626,7 +626,7 @@ static struct hpt_info *hpt3xx_get_info(struct device *dev)
 
 static u8 hpt3xx_udma_filter(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct hpt_info *info	= hpt3xx_get_info(hwif->dev);
 	u8 mask 		= hwif->ultra_mask;
 
@@ -665,7 +665,7 @@ static u8 hpt3xx_udma_filter(ide_drive_t *drive)
 
 static u8 hpt3xx_mdma_filter(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct hpt_info *info	= hpt3xx_get_info(hwif->dev);
 
 	switch (info->chip_type) {
@@ -743,7 +743,7 @@ static void hpt3xx_quirkproc(ide_drive_t *drive)
 
 static void hpt3xx_maskproc(ide_drive_t *drive, int mask)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev	*dev	= to_pci_dev(hwif->dev);
 	struct hpt_info *info	= hpt3xx_get_info(hwif->dev);
 
@@ -788,7 +788,7 @@ static void hpt366_dma_lost_irq(ide_drive_t *drive)
 
 static void hpt370_clear_engine(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
 
 	pci_write_config_byte(dev, hwif->select_data, 0x37);
@@ -797,7 +797,7 @@ static void hpt370_clear_engine(ide_drive_t *drive)
 
 static void hpt370_irq_timeout(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u16 bfifo		= 0;
 	u8  dma_cmd;
@@ -822,7 +822,7 @@ static void hpt370_dma_start(ide_drive_t *drive)
 
 static int hpt370_dma_end(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	u8  dma_stat		= inb(hwif->dma_base + ATA_DMA_STATUS);
 
 	if (dma_stat & 0x01) {
@@ -844,7 +844,7 @@ static void hpt370_dma_timeout(ide_drive_t *drive)
 /* returns 1 if DMA IRQ issued, 0 otherwise */
 static int hpt374_dma_test_irq(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u16 bfifo		= 0;
 	u8  dma_stat;
@@ -865,7 +865,7 @@ static int hpt374_dma_test_irq(ide_drive_t *drive)
 
 static int hpt374_dma_end(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u8 mcr	= 0, mcr_addr	= hwif->select_data;
 	u8 bwsr = 0, mask	= hwif->channel ? 0x02 : 0x01;
@@ -927,7 +927,7 @@ static void hpt3xxn_set_clock(ide_hwif_t *hwif, u8 mode)
 
 static void hpt3xxn_rw_disk(ide_drive_t *drive, struct request *rq)
 {
-	hpt3xxn_set_clock(HWIF(drive), rq_data_dir(rq) ? 0x23 : 0x21);
+	hpt3xxn_set_clock(drive->hwif, rq_data_dir(rq) ? 0x23 : 0x21);
 }
 
 /**
diff --git a/drivers/ide/icside.c b/drivers/ide/icside.c
index 72d95c99f147..97a35c667aee 100644
--- a/drivers/ide/icside.c
+++ b/drivers/ide/icside.c
@@ -166,7 +166,7 @@ static const expansioncard_ops_t icside_ops_arcin_v6 = {
  */
 static void icside_maskproc(ide_drive_t *drive, int mask)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct expansion_card *ec = ECARD_DEV(hwif->dev);
 	struct icside_state *state = ecard_get_drvdata(ec);
 	unsigned long flags;
@@ -284,7 +284,7 @@ static void icside_dma_host_set(ide_drive_t *drive, int on)
 
 static int icside_dma_end(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct expansion_card *ec = ECARD_DEV(hwif->dev);
 
 	drive->waiting_for_dma = 0;
@@ -299,7 +299,7 @@ static int icside_dma_end(ide_drive_t *drive)
 
 static void icside_dma_start(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct expansion_card *ec = ECARD_DEV(hwif->dev);
 
 	/* We can not enable DMA on both channels simultaneously. */
@@ -309,7 +309,7 @@ static void icside_dma_start(ide_drive_t *drive)
 
 static int icside_dma_setup(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct expansion_card *ec = ECARD_DEV(hwif->dev);
 	struct icside_state *state = ecard_get_drvdata(ec);
 	struct request *rq = hwif->rq;
@@ -362,7 +362,7 @@ static void icside_dma_exec_cmd(ide_drive_t *drive, u8 cmd)
 
 static int icside_dma_test_irq(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct expansion_card *ec = ECARD_DEV(hwif->dev);
 	struct icside_state *state = ecard_get_drvdata(ec);
 
diff --git a/drivers/ide/ide-acpi.c b/drivers/ide/ide-acpi.c
index fd4a36433050..f89b6ecf7d1a 100644
--- a/drivers/ide/ide-acpi.c
+++ b/drivers/ide/ide-acpi.c
@@ -218,7 +218,7 @@ static acpi_handle ide_acpi_hwif_get_handle(ide_hwif_t *hwif)
  */
 static acpi_handle ide_acpi_drive_get_handle(ide_drive_t *drive)
 {
-	ide_hwif_t	*hwif = HWIF(drive);
+	ide_hwif_t	*hwif = drive->hwif;
 	int		 port;
 	acpi_handle	 drive_handle;
 
@@ -263,7 +263,7 @@ static int do_drive_get_GTF(ide_drive_t *drive,
 	acpi_status			status;
 	struct acpi_buffer		output;
 	union acpi_object 		*out_obj;
-	ide_hwif_t			*hwif = HWIF(drive);
+	ide_hwif_t			*hwif = drive->hwif;
 	struct device			*dev = hwif->gendev.parent;
 	int				err = -ENODEV;
 	int				port;
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index eb9fac4d0f0c..4088a622873e 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -89,7 +89,7 @@ static void ide_tf_set_cmd(ide_drive_t *drive, ide_task_t *task, u8 dma)
 static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 					sector_t block)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	u16 nsectors		= (u16)rq->nr_sectors;
 	u8 lba48		= !!(drive->dev_flags & IDE_DFLAG_LBA48);
 	u8 dma			= !!(drive->dev_flags & IDE_DFLAG_USING_DMA);
@@ -187,7 +187,7 @@ static ide_startstop_t __ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 static ide_startstop_t ide_do_rw_disk(ide_drive_t *drive, struct request *rq,
 				      sector_t block)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 
 	BUG_ON(drive->dev_flags & IDE_DFLAG_BLOCKED);
 
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 6ff82d7055b9..0e2b95ec08a5 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -463,7 +463,7 @@ EXPORT_SYMBOL_GPL(ide_init_sg_cmd);
 static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
 		struct request *rq)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	ide_task_t *task = rq->special;
 
 	if (task) {
@@ -587,7 +587,7 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 
 #ifdef DEBUG
 	printk("%s: start_request: current=0x%08lx\n",
-		HWIF(drive)->name, (unsigned long) rq);
+		drive->hwif->name, (unsigned long) rq);
 #endif
 
 	/* bail early if we've exceeded max_failures */
@@ -833,7 +833,7 @@ plug_device_2:
  */
 static ide_startstop_t ide_dma_timeout_retry(ide_drive_t *drive, int error)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct request *rq;
 	ide_startstop_t ret = ide_stopped;
 
@@ -955,7 +955,7 @@ void ide_timer_expiry (unsigned long data)
 			 * globally mask the specific IRQ:
 			 */
 			spin_unlock(&hwif->lock);
-			hwif  = HWIF(drive);
+			hwif = drive->hwif;
 			/* disable_irq_nosync ?? */
 			disable_irq(hwif->irq);
 			/* local CPU only,
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index b92304d0e79a..386080ee4608 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -451,7 +451,7 @@ EXPORT_SYMBOL(ide_fixstring);
  */
 int drive_is_ready (ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	u8 stat			= 0;
 
 	if (drive->waiting_for_dma)
@@ -965,7 +965,7 @@ static void ide_reset_report_error(ide_hwif_t *hwif, u8 err)
  */
 static ide_startstop_t reset_pollfunc (ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	const struct ide_port_ops *port_ops = hwif->port_ops;
 	u8 tmp;
 	int err = 0;
diff --git a/drivers/ide/ide-pm.c b/drivers/ide/ide-pm.c
index abb84a2dd821..0c206c68bbb1 100644
--- a/drivers/ide/ide-pm.c
+++ b/drivers/ide/ide-pm.c
@@ -5,7 +5,7 @@
 int generic_ide_suspend(struct device *dev, pm_message_t mesg)
 {
 	ide_drive_t *drive = dev->driver_data, *pair = ide_get_pair_dev(drive);
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct request *rq;
 	struct request_pm_state rqpm;
 	ide_task_t args;
@@ -39,7 +39,7 @@ int generic_ide_suspend(struct device *dev, pm_message_t mesg)
 int generic_ide_resume(struct device *dev)
 {
 	ide_drive_t *drive = dev->driver_data, *pair = ide_get_pair_dev(drive);
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct request *rq;
 	struct request_pm_state rqpm;
 	ide_task_t args;
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 09ea50118e63..9dfa99f062aa 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -189,7 +189,7 @@ static void ide_classify_atapi_dev(ide_drive_t *drive)
 
 static void do_identify(ide_drive_t *drive, u8 cmd)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	u16 *id = drive->id;
 	char *m = (char *)&id[ATA_ID_PROD];
 	unsigned long flags;
@@ -266,7 +266,7 @@ err_misc:
 
 static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct ide_io_ports *io_ports = &hwif->io_ports;
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	int use_altstatus = 0, rc;
@@ -341,7 +341,7 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
  
 static int try_to_identify (ide_drive_t *drive, u8 cmd)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	int retval;
 	int autoprobe = 0;
@@ -438,7 +438,7 @@ static u8 ide_read_device(ide_drive_t *drive)
 
 static int do_probe (ide_drive_t *drive, u8 cmd)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	int rc;
 	u8 present = !!(drive->dev_flags & IDE_DFLAG_PRESENT), stat;
@@ -522,7 +522,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
  */
 static void enable_nest (ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	u8 stat;
 
@@ -869,7 +869,7 @@ static void ide_port_tune_devices(ide_hwif_t *hwif)
 static int ide_init_queue(ide_drive_t *drive)
 {
 	struct request_queue *q;
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	int max_sectors = 256;
 	int max_sg_entries = PRD_ENTRIES;
 
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 693e8d15fb78..55d451ce9b4f 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -58,7 +58,7 @@ static ide_startstop_t task_in_intr(ide_drive_t *);
 
 ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct ide_taskfile *tf = &task->tf;
 	ide_handler_t *handler = NULL;
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
diff --git a/drivers/ide/it8213.c b/drivers/ide/it8213.c
index 7c2feeb3c5ec..d7969b6d139e 100644
--- a/drivers/ide/it8213.c
+++ b/drivers/ide/it8213.c
@@ -25,7 +25,7 @@
 
 static void it8213_set_pio_mode(ide_drive_t *drive, const u8 pio)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	int is_slave		= drive->dn & 1;
 	int master_port		= 0x40;
@@ -82,7 +82,7 @@ static void it8213_set_pio_mode(ide_drive_t *drive, const u8 pio)
 
 static void it8213_set_dma_mode(ide_drive_t *drive, const u8 speed)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u8 maslave		= 0x40;
 	int a_speed		= 3 << (drive->dn * 4);
diff --git a/drivers/ide/ns87415.c b/drivers/ide/ns87415.c
index 13789060f407..aceb2fcbe1d1 100644
--- a/drivers/ide/ns87415.c
+++ b/drivers/ide/ns87415.c
@@ -138,12 +138,12 @@ static unsigned int ns87415_count = 0, ns87415_control[MAX_HWIFS] = { 0 };
 
 /*
  * This routine either enables/disables (according to IDE_DFLAG_PRESENT)
- * the IRQ associated with the port (HWIF(drive)),
+ * the IRQ associated with the port,
  * and selects either PIO or DMA handshaking for the next I/O operation.
  */
 static void ns87415_prepare_drive (ide_drive_t *drive, unsigned int use_dma)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
 	unsigned int bit, other, new, *old = (unsigned int *) hwif->select_data;
 	unsigned long flags;
@@ -197,7 +197,7 @@ static void ns87415_selectproc (ide_drive_t *drive)
 
 static int ns87415_dma_end(ide_drive_t *drive)
 {
-	ide_hwif_t      *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	u8 dma_stat = 0, dma_cmd = 0;
 
 	drive->waiting_for_dma = 0;
diff --git a/drivers/ide/pdc202xx_new.c b/drivers/ide/pdc202xx_new.c
index 211ae46e3e0c..f21290c4b447 100644
--- a/drivers/ide/pdc202xx_new.c
+++ b/drivers/ide/pdc202xx_new.c
@@ -143,7 +143,7 @@ static struct udma_timing {
 
 static void pdcnew_set_dma_mode(ide_drive_t *drive, const u8 speed)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u8 adj			= (drive->dn & 1) ? 0x08 : 0x00;
 
@@ -219,7 +219,7 @@ static void pdcnew_reset(ide_drive_t *drive)
 	 * Deleted this because it is redundant from the caller.
 	 */
 	printk(KERN_WARNING "pdc202xx_new: %s channel reset.\n",
-		HWIF(drive)->channel ? "Secondary" : "Primary");
+		drive->hwif->channel ? "Secondary" : "Primary");
 }
 
 /**
diff --git a/drivers/ide/pdc202xx_old.c b/drivers/ide/pdc202xx_old.c
index 072ef70bf061..e8e6b29d9e41 100644
--- a/drivers/ide/pdc202xx_old.c
+++ b/drivers/ide/pdc202xx_old.c
@@ -39,7 +39,7 @@ static void pdc_old_disable_66MHz_clock(ide_hwif_t *);
 
 static void pdc202xx_set_mode(ide_drive_t *drive, const u8 speed)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u8 drive_pci		= 0x60 + (drive->dn << 2);
 
@@ -169,7 +169,7 @@ static void pdc202xx_dma_start(ide_drive_t *drive)
 	if (drive->current_speed > XFER_UDMA_2)
 		pdc_old_enable_66MHz_clock(drive->hwif);
 	if (drive->media != ide_disk || (drive->dev_flags & IDE_DFLAG_LBA48)) {
-		ide_hwif_t *hwif	= HWIF(drive);
+		ide_hwif_t *hwif	= drive->hwif;
 		struct request *rq	= hwif->rq;
 		unsigned long high_16	= hwif->extra_base - 16;
 		unsigned long atapi_reg	= high_16 + (hwif->channel ? 0x24 : 0x20);
@@ -189,7 +189,7 @@ static void pdc202xx_dma_start(ide_drive_t *drive)
 static int pdc202xx_dma_end(ide_drive_t *drive)
 {
 	if (drive->media != ide_disk || (drive->dev_flags & IDE_DFLAG_LBA48)) {
-		ide_hwif_t *hwif	= HWIF(drive);
+		ide_hwif_t *hwif	= drive->hwif;
 		unsigned long high_16	= hwif->extra_base - 16;
 		unsigned long atapi_reg	= high_16 + (hwif->channel ? 0x24 : 0x20);
 		u8 clock		= 0;
@@ -205,7 +205,7 @@ static int pdc202xx_dma_end(ide_drive_t *drive)
 
 static int pdc202xx_dma_test_irq(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	unsigned long high_16	= hwif->extra_base - 16;
 	u8 dma_stat		= inb(hwif->dma_base + ATA_DMA_STATUS);
 	u8 sc1d			= inb(high_16 + 0x001d);
@@ -243,7 +243,7 @@ static void pdc202xx_reset_host (ide_hwif_t *hwif)
 
 static void pdc202xx_reset (ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	ide_hwif_t *mate	= hwif->mate;
 
 	pdc202xx_reset_host(hwif);
diff --git a/drivers/ide/piix.c b/drivers/ide/piix.c
index 61d2d920a5cd..e57b6c3ac78e 100644
--- a/drivers/ide/piix.c
+++ b/drivers/ide/piix.c
@@ -67,7 +67,7 @@ static int no_piix_dma;
 
 static void piix_set_pio_mode(ide_drive_t *drive, const u8 pio)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	int is_slave		= drive->dn & 1;
 	int master_port		= hwif->channel ? 0x42 : 0x40;
@@ -136,7 +136,7 @@ static void piix_set_pio_mode(ide_drive_t *drive, const u8 pio)
 
 static void piix_set_dma_mode(ide_drive_t *drive, const u8 speed)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u8 maslave		= hwif->channel ? 0x42 : 0x40;
 	int a_speed		= 3 << (drive->dn * 4);
@@ -224,7 +224,7 @@ static unsigned int init_chipset_ich(struct pci_dev *dev)
  */
 static void ich_clear_irq(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	u8 dma_stat;
 
 	/*
diff --git a/drivers/ide/pmac.c b/drivers/ide/pmac.c
index 899b96baf215..ee52a21af1be 100644
--- a/drivers/ide/pmac.c
+++ b/drivers/ide/pmac.c
@@ -1513,7 +1513,7 @@ use_pio_instead:
 static int
 pmac_ide_dma_setup(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	pmac_ide_hwif_t *pmif =
 		(pmac_ide_hwif_t *)dev_get_drvdata(hwif->gendev.parent);
 	struct request *rq = hwif->rq;
@@ -1637,7 +1637,7 @@ pmac_ide_dma_test_irq (ide_drive_t *drive)
 			break;
 		if (++timeout > 100) {
 			printk(KERN_WARNING "ide%d, ide_dma_test_irq \
-			timeout flushing channel\n", HWIF(drive)->index);
+			timeout flushing channel\n", hwif->index);
 			break;
 		}
 	}	
diff --git a/drivers/ide/qd65xx.c b/drivers/ide/qd65xx.c
index bc27c7aba936..5b2e3af43c4b 100644
--- a/drivers/ide/qd65xx.c
+++ b/drivers/ide/qd65xx.c
@@ -202,7 +202,8 @@ static void qd6500_set_pio_mode(ide_drive_t *drive, const u8 pio)
 		recovery_time = drive->id[ATA_ID_EIDE_PIO] - 120;
 	}
 
-	qd_set_timing(drive, qd6500_compute_timing(HWIF(drive), active_time, recovery_time));
+	qd_set_timing(drive, qd6500_compute_timing(drive->hwif,
+				active_time, recovery_time));
 }
 
 static void qd6580_set_pio_mode(ide_drive_t *drive, const u8 pio)
@@ -245,11 +246,11 @@ static void qd6580_set_pio_mode(ide_drive_t *drive, const u8 pio)
 		printk(KERN_INFO "%s: PIO mode%d\n", drive->name,pio);
 	}
 
-	if (!HWIF(drive)->channel && drive->media != ide_disk) {
+	if (!hwif->channel && drive->media != ide_disk) {
 		outb(0x5f, QD_CONTROL_PORT);
 		printk(KERN_WARNING "%s: ATAPI: disabled read-ahead FIFO "
 			"and post-write buffer on %s.\n",
-			drive->name, HWIF(drive)->name);
+			drive->name, hwif->name);
 	}
 
 	qd_set_timing(drive, qd6580_compute_timing(active_time, recovery_time));
diff --git a/drivers/ide/sc1200.c b/drivers/ide/sc1200.c
index ec7f766ef5e4..6f68fe984bfb 100644
--- a/drivers/ide/sc1200.c
+++ b/drivers/ide/sc1200.c
@@ -125,7 +125,7 @@ out:
 
 static void sc1200_set_dma_mode(ide_drive_t *drive, const u8 mode)
 {
-	ide_hwif_t		*hwif = HWIF(drive);
+	ide_hwif_t		*hwif = drive->hwif;
 	struct pci_dev		*dev = to_pci_dev(hwif->dev);
 	unsigned int		reg, timings;
 	unsigned short		pci_clock;
@@ -170,7 +170,7 @@ static void sc1200_set_dma_mode(ide_drive_t *drive, const u8 mode)
  */
 static int sc1200_dma_end(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	unsigned long dma_base = hwif->dma_base;
 	byte dma_stat;
 
@@ -199,7 +199,7 @@ static int sc1200_dma_end(ide_drive_t *drive)
 
 static void sc1200_set_pio_mode(ide_drive_t *drive, const u8 pio)
 {
-	ide_hwif_t	*hwif = HWIF(drive);
+	ide_hwif_t	*hwif = drive->hwif;
 	int		mode = -1;
 
 	/*
diff --git a/drivers/ide/scc_pata.c b/drivers/ide/scc_pata.c
index e966113fd569..90574ab76345 100644
--- a/drivers/ide/scc_pata.c
+++ b/drivers/ide/scc_pata.c
@@ -217,7 +217,7 @@ scc_ide_outsl(unsigned long port, void *addr, u32 count)
 
 static void scc_set_pio_mode(ide_drive_t *drive, const u8 pio)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct scc_ports *ports = ide_get_hwifdata(hwif);
 	unsigned long ctl_base = ports->ctl;
 	unsigned long cckctrl_port = ctl_base + 0xff0;
@@ -249,7 +249,7 @@ static void scc_set_pio_mode(ide_drive_t *drive, const u8 pio)
 
 static void scc_set_dma_mode(ide_drive_t *drive, const u8 speed)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct scc_ports *ports = ide_get_hwifdata(hwif);
 	unsigned long ctl_base = ports->ctl;
 	unsigned long cckctrl_port = ctl_base + 0xff0;
@@ -387,7 +387,7 @@ static int __scc_dma_end(ide_drive_t *drive)
 
 static int scc_dma_end(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	void __iomem *dma_base = (void __iomem *)hwif->dma_base;
 	unsigned long intsts_port = hwif->dma_base + 0x014;
 	u32 reg;
@@ -496,7 +496,7 @@ static int scc_dma_end(ide_drive_t *drive)
 /* returns 1 if dma irq issued, 0 otherwise */
 static int scc_dma_test_irq(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	u32 int_stat = in_be32((void __iomem *)hwif->dma_base + 0x014);
 
 	/* SCC errata A252,A308 workaround: Step4 */
diff --git a/drivers/ide/serverworks.c b/drivers/ide/serverworks.c
index 437bc919dafd..382102ba467b 100644
--- a/drivers/ide/serverworks.c
+++ b/drivers/ide/serverworks.c
@@ -151,7 +151,7 @@ static void svwks_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	static const u8 dma_modes[]		= { 0x77, 0x21, 0x20 };
 	static const u8 drive_pci2[]		= { 0x45, 0x44, 0x47, 0x46 };
 
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u8 unit			= drive->dn & 1;
 
diff --git a/drivers/ide/sgiioc4.c b/drivers/ide/sgiioc4.c
index c68b71b1087b..8e1ffd57a86d 100644
--- a/drivers/ide/sgiioc4.c
+++ b/drivers/ide/sgiioc4.c
@@ -123,7 +123,7 @@ static int
 sgiioc4_clearirq(ide_drive_t * drive)
 {
 	u32 intr_reg;
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct ide_io_ports *io_ports = &hwif->io_ports;
 	unsigned long other_ir = io_ports->irq_addr + (IOC4_INTR_REG << 2);
 
@@ -181,7 +181,7 @@ sgiioc4_clearirq(ide_drive_t * drive)
 
 static void sgiioc4_dma_start(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	unsigned long ioc4_dma_addr = hwif->dma_base + IOC4_DMA_CTRL * 4;
 	unsigned int reg = readl((void __iomem *)ioc4_dma_addr);
 	unsigned int temp_reg = reg | IOC4_S_DMA_START;
@@ -209,7 +209,7 @@ sgiioc4_ide_dma_stop(ide_hwif_t *hwif, u64 dma_base)
 static int sgiioc4_dma_end(ide_drive_t *drive)
 {
 	u32 ioc4_dma, bc_dev, bc_mem, num, valid = 0, cnt = 0;
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	unsigned long dma_base = hwif->dma_base;
 	int dma_stat = 0;
 	unsigned long *ending_dma = ide_get_hwifdata(hwif);
@@ -271,7 +271,7 @@ static void sgiioc4_set_dma_mode(ide_drive_t *drive, const u8 speed)
 /* returns 1 if dma irq issued, 0 otherwise */
 static int sgiioc4_dma_test_irq(ide_drive_t *drive)
 {
-	return sgiioc4_checkirq(HWIF(drive));
+	return sgiioc4_checkirq(drive->hwif);
 }
 
 static void sgiioc4_dma_host_set(ide_drive_t *drive, int on)
@@ -367,7 +367,7 @@ static void
 sgiioc4_configure_for_dma(int dma_direction, ide_drive_t * drive)
 {
 	u32 ioc4_dma;
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	unsigned long dma_base = hwif->dma_base;
 	unsigned long ioc4_dma_addr = dma_base + IOC4_DMA_CTRL * 4;
 	u32 dma_addr, ending_dma_addr;
@@ -427,7 +427,7 @@ sgiioc4_configure_for_dma(int dma_direction, ide_drive_t * drive)
 static unsigned int
 sgiioc4_build_dma_table(ide_drive_t * drive, struct request *rq, int ddir)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	unsigned int *table = hwif->dmatable_cpu;
 	unsigned int count = 0, i = 1;
 	struct scatterlist *sg;
diff --git a/drivers/ide/siimage.c b/drivers/ide/siimage.c
index 7d622d20bc4c..652b3a04620f 100644
--- a/drivers/ide/siimage.c
+++ b/drivers/ide/siimage.c
@@ -114,7 +114,7 @@ static unsigned long siimage_selreg(ide_hwif_t *hwif, int r)
 
 static inline unsigned long siimage_seldev(ide_drive_t *drive, int r)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	unsigned long base	= (unsigned long)hwif->hwif_data;
 	u8 unit			= drive->dn & 1;
 
@@ -243,7 +243,7 @@ static void sil_set_pio_mode(ide_drive_t *drive, u8 pio)
 	static const u16 tf_speed[]   = { 0x328a, 0x2283, 0x1281, 0x10c3, 0x10c1 };
 	static const u16 data_speed[] = { 0x328a, 0x2283, 0x1104, 0x10c3, 0x10c1 };
 
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	ide_drive_t *pair	= ide_get_pair_dev(drive);
 	u32 speedt		= 0;
@@ -300,7 +300,7 @@ static void sil_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	static const u8 ultra5[] = { 0x0C, 0x07, 0x05, 0x04, 0x02, 0x01 };
 	static const u16 dma[]	 = { 0x2208, 0x10C2, 0x10C1 };
 
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	unsigned long base	= (unsigned long)hwif->hwif_data;
 	u16 ultra = 0, multi	= 0;
@@ -340,7 +340,7 @@ static void sil_set_dma_mode(ide_drive_t *drive, const u8 speed)
 /* returns 1 if dma irq issued, 0 otherwise */
 static int siimage_io_dma_test_irq(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u8 dma_altstat		= 0;
 	unsigned long addr	= siimage_selreg(hwif, 1);
@@ -367,7 +367,7 @@ static int siimage_io_dma_test_irq(ide_drive_t *drive)
 
 static int siimage_mmio_dma_test_irq(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	unsigned long addr	= siimage_selreg(hwif, 0x1);
 	void __iomem *sata_error_addr
 		= (void __iomem *)hwif->sata_scr[SATA_ERROR_OFFSET];
diff --git a/drivers/ide/sis5513.c b/drivers/ide/sis5513.c
index ad32e18c5ba3..9ec1a4a4432c 100644
--- a/drivers/ide/sis5513.c
+++ b/drivers/ide/sis5513.c
@@ -274,7 +274,7 @@ static void sis_program_timings(ide_drive_t *drive, const u8 mode)
 
 static void config_drive_art_rwp(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u8 reg4bh		= 0;
 	u8 rw_prefetch		= 0;
diff --git a/drivers/ide/sl82c105.c b/drivers/ide/sl82c105.c
index 84dc33602ff8..1ded01d81ab3 100644
--- a/drivers/ide/sl82c105.c
+++ b/drivers/ide/sl82c105.c
@@ -140,7 +140,7 @@ static inline void sl82c105_reset_host(struct pci_dev *dev)
  */
 static void sl82c105_dma_lost_irq(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u32 val, mask		= hwif->channel ? CTRL_IDE_IRQB : CTRL_IDE_IRQA;
 	u8 dma_cmd;
@@ -177,7 +177,7 @@ static void sl82c105_dma_lost_irq(ide_drive_t *drive)
  */
 static void sl82c105_dma_start(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	int reg 		= 0x44 + drive->dn * 4;
 
diff --git a/drivers/ide/slc90e66.c b/drivers/ide/slc90e66.c
index 0f759e4ed779..40b4b94a4288 100644
--- a/drivers/ide/slc90e66.c
+++ b/drivers/ide/slc90e66.c
@@ -20,7 +20,7 @@ static DEFINE_SPINLOCK(slc90e66_lock);
 
 static void slc90e66_set_pio_mode(ide_drive_t *drive, const u8 pio)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	int is_slave		= drive->dn & 1;
 	int master_port		= hwif->channel ? 0x42 : 0x40;
@@ -73,7 +73,7 @@ static void slc90e66_set_pio_mode(ide_drive_t *drive, const u8 pio)
 
 static void slc90e66_set_dma_mode(ide_drive_t *drive, const u8 speed)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
 	u8 maslave		= hwif->channel ? 0x42 : 0x40;
 	int sitre = 0, a_speed	= 7 << (drive->dn * 4);
diff --git a/drivers/ide/tc86c001.c b/drivers/ide/tc86c001.c
index accb379bcad6..d2c00fb928e4 100644
--- a/drivers/ide/tc86c001.c
+++ b/drivers/ide/tc86c001.c
@@ -15,7 +15,7 @@
 
 static void tc86c001_set_mode(ide_drive_t *drive, const u8 speed)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	unsigned long scr_port	= hwif->config_data + (drive->dn ? 0x02 : 0x00);
 	u16 mode, scr		= inw(scr_port);
 
@@ -62,7 +62,7 @@ static void tc86c001_set_pio_mode(ide_drive_t *drive, const u8 pio)
  */
 static int tc86c001_timer_expiry(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	ide_expiry_t *expiry	= ide_get_hwifdata(hwif);
 	u8 dma_stat		= inb(hwif->dma_base + ATA_DMA_STATUS);
 
@@ -109,7 +109,7 @@ static int tc86c001_timer_expiry(ide_drive_t *drive)
 
 static void tc86c001_dma_start(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_hwif_t *hwif	= drive->hwif;
 	unsigned long sc_base	= hwif->config_data;
 	unsigned long twcr_port	= sc_base + (drive->dn ? 0x06 : 0x04);
 	unsigned long nsectors	= hwif->rq->nr_sectors;
diff --git a/drivers/ide/triflex.c b/drivers/ide/triflex.c
index b6ff40336aa9..8773c3ba7462 100644
--- a/drivers/ide/triflex.c
+++ b/drivers/ide/triflex.c
@@ -36,7 +36,7 @@
 
 static void triflex_set_mode(ide_drive_t *drive, const u8 speed)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
 	u32 triflex_timings = 0;
 	u16 timing = 0;
diff --git a/drivers/ide/trm290.c b/drivers/ide/trm290.c
index 79a03df118d8..b6a1285a4021 100644
--- a/drivers/ide/trm290.c
+++ b/drivers/ide/trm290.c
@@ -144,7 +144,7 @@
 
 static void trm290_prepare_drive (ide_drive_t *drive, unsigned int use_dma)
 {
-	ide_hwif_t *hwif = HWIF(drive);
+	ide_hwif_t *hwif = drive->hwif;
 	u16 reg = 0;
 	unsigned long flags;
 
@@ -222,15 +222,15 @@ static int trm290_dma_end(ide_drive_t *drive)
 	drive->waiting_for_dma = 0;
 	/* purge DMA mappings */
 	ide_destroy_dmatable(drive);
-	status = inw(HWIF(drive)->dma_base + 2);
+	status = inw(drive->hwif->dma_base + 2);
+
 	return status != 0x00ff;
 }
 
 static int trm290_dma_test_irq(ide_drive_t *drive)
 {
-	u16 status;
+	u16 status = inw(drive->hwif->dma_base + 2);
 
-	status = inw(HWIF(drive)->dma_base + 2);
 	return status == 0x00ff;
 }
 
diff --git a/drivers/ide/via82cxxx.c b/drivers/ide/via82cxxx.c
index 2a812d3207e9..fecc0e03c3fc 100644
--- a/drivers/ide/via82cxxx.c
+++ b/drivers/ide/via82cxxx.c
@@ -178,7 +178,7 @@ static void via_set_drive(ide_drive_t *drive, const u8 speed)
 		ide_timing_merge(&p, &t, &t, IDE_TIMING_8BIT);
 	}
 
-	via_set_speed(HWIF(drive), drive->dn, &t);
+	via_set_speed(hwif, drive->dn, &t);
 }
 
 /**
diff --git a/include/linux/ide.h b/include/linux/ide.h
index ee2f461882ad..58b9c99482cd 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -42,8 +42,6 @@ typedef unsigned char	byte;	/* used everywhere */
 #define ERROR_RESET	3	/* Reset controller every 4th retry */
 #define ERROR_RECAL	1	/* Recalibrate every 2nd retry */
 
-#define HWIF(drive)		((ide_hwif_t *)((drive)->hwif))
-
 /*
  * Definitions for accessing IDE controller registers
  */
-- 
cgit v1.2.3


From 54cc1428cfa619e16d75baae8cb041a2eff015f0 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:52 +0100
Subject: ide: remove local_irq_set() macro

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-iops.c  | 3 ++-
 drivers/ide/ide-probe.c | 3 ++-
 include/linux/ide.h     | 2 --
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 386080ee4608..44ad03bf9cb4 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -503,7 +503,8 @@ static int __ide_wait_stat(ide_drive_t *drive, u8 good, u8 bad, unsigned long ti
 	stat = tp_ops->read_status(hwif);
 
 	if (stat & ATA_BUSY) {
-		local_irq_set(flags);
+		local_irq_save(flags);
+		local_irq_enable_in_hardirq();
 		timeout += jiffies;
 		while ((stat = tp_ops->read_status(hwif)) & ATA_BUSY) {
 			if (time_after(jiffies, timeout)) {
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 9dfa99f062aa..2fca32ed576f 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -796,7 +796,8 @@ static int ide_probe_port(ide_hwif_t *hwif)
 	if (irqd)
 		disable_irq(hwif->irq);
 
-	local_irq_set(flags);
+	local_irq_save(flags);
+	local_irq_enable_in_hardirq();
 
 	if (ide_port_wait_ready(hwif) == -EBUSY)
 		printk(KERN_DEBUG "%s: Wait for ready failed before probe !\n", hwif->name);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 58b9c99482cd..82d500c5a847 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1586,8 +1586,6 @@ static inline void ide_set_max_pio(ide_drive_t *drive)
 	ide_set_pio(drive, 255);
 }
 
-#define local_irq_set(flags)	do { local_save_flags((flags)); local_irq_enable_in_hardirq(); } while (0)
-
 char *ide_media_string(ide_drive_t *);
 
 extern struct device_attribute ide_dev_attrs[];
-- 
cgit v1.2.3


From c0ae50234771684ae0cbac5dfb70e1a09c22aa89 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:52 +0100
Subject: ide: remove ide_pci_enablebit_t typedef

Remove needless parens while at it.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/setup-pci.c | 2 +-
 include/linux/ide.h     | 7 ++++---
 2 files changed, 5 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index 9f1f9163a136..09f820af379d 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -471,7 +471,7 @@ void ide_pci_setup_ports(struct pci_dev *dev, const struct ide_port_info *d,
 	 */
 
 	for (port = 0; port < channels; ++port) {
-		const ide_pci_enablebit_t *e = &(d->enablebits[port]);
+		const struct ide_pci_enablebit *e = &d->enablebits[port];
 
 		if (e->reg && (pci_read_config_byte(dev, e->reg, &tmp) ||
 		    (tmp & e->mask) != e->val)) {
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 82d500c5a847..cca6cfb299bc 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1300,11 +1300,11 @@ static inline int ide_hwif_setup_dma(ide_hwif_t *hwif,
 }
 #endif
 
-typedef struct ide_pci_enablebit_s {
+struct ide_pci_enablebit {
 	u8	reg;	/* byte pci reg holding the enable-bit */
 	u8	mask;	/* mask to isolate the enable-bit */
 	u8	val;	/* value of masked reg when "enabled" */
-} ide_pci_enablebit_t;
+};
 
 enum {
 	/* Uses ISA control ports not PCI ones. */
@@ -1393,7 +1393,8 @@ struct ide_port_info {
 	const struct ide_port_ops	*port_ops;
 	const struct ide_dma_ops	*dma_ops;
 
-	ide_pci_enablebit_t	enablebits[2];
+	struct ide_pci_enablebit	enablebits[2];
+
 	hwif_chipset_t		chipset;
 
 	u16			max_sectors;	/* if < than the default one */
-- 
cgit v1.2.3


From 9892ec5497af1ec38c46974b5a370ba11c636b19 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:53 +0100
Subject: ide: remove 'byte' typedef

Just use u8 instead, also s/__u8/u8/ in ide-cd.h while at it.

Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-cd.h | 26 +++++++++++++-------------
 drivers/ide/it821x.c |  4 ++--
 drivers/ide/qd65xx.h |  4 ++--
 drivers/ide/sc1200.c |  2 +-
 include/linux/ide.h  |  4 +---
 5 files changed, 19 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-cd.h b/drivers/ide/ide-cd.h
index bf676b262181..d5ae036af00c 100644
--- a/drivers/ide/ide-cd.h
+++ b/drivers/ide/ide-cd.h
@@ -33,33 +33,33 @@
 
 /* Structure of a MSF cdrom address. */
 struct atapi_msf {
-	byte reserved;
-	byte minute;
-	byte second;
-	byte frame;
+	u8 reserved;
+	u8 minute;
+	u8 second;
+	u8 frame;
 };
 
 /* Space to hold the disk TOC. */
 #define MAX_TRACKS 99
 struct atapi_toc_header {
 	unsigned short toc_length;
-	byte first_track;
-	byte last_track;
+	u8 first_track;
+	u8 last_track;
 };
 
 struct atapi_toc_entry {
-	byte reserved1;
+	u8 reserved1;
 #if defined(__BIG_ENDIAN_BITFIELD)
-	__u8 adr     : 4;
-	__u8 control : 4;
+	u8 adr     : 4;
+	u8 control : 4;
 #elif defined(__LITTLE_ENDIAN_BITFIELD)
-	__u8 control : 4;
-	__u8 adr     : 4;
+	u8 control : 4;
+	u8 adr     : 4;
 #else
 #error "Please fix <asm/byteorder.h>"
 #endif
-	byte track;
-	byte reserved2;
+	u8 track;
+	u8 reserved2;
 	union {
 		unsigned lba;
 		struct atapi_msf msf;
diff --git a/drivers/ide/it821x.c b/drivers/ide/it821x.c
index ef004089761b..a34f3e19b4a4 100644
--- a/drivers/ide/it821x.c
+++ b/drivers/ide/it821x.c
@@ -279,7 +279,7 @@ static void it821x_set_pio_mode(ide_drive_t *drive, const u8 pio)
  *	the shared MWDMA/PIO timing register.
  */
 
-static void it821x_tune_mwdma (ide_drive_t *drive, byte mode_wanted)
+static void it821x_tune_mwdma(ide_drive_t *drive, u8 mode_wanted)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
@@ -316,7 +316,7 @@ static void it821x_tune_mwdma (ide_drive_t *drive, byte mode_wanted)
  *	controller when doing UDMA modes in pass through.
  */
 
-static void it821x_tune_udma (ide_drive_t *drive, byte mode_wanted)
+static void it821x_tune_udma(ide_drive_t *drive, u8 mode_wanted)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
diff --git a/drivers/ide/qd65xx.h b/drivers/ide/qd65xx.h
index c83dea85e621..6636f9665d16 100644
--- a/drivers/ide/qd65xx.h
+++ b/drivers/ide/qd65xx.h
@@ -31,8 +31,8 @@
 
 #define QD_CONFIG(hwif)		((hwif)->config_data & 0x00ff)
 
-#define QD_TIMING(drive)	(byte)(((drive)->drive_data) & 0x00ff)
-#define QD_TIMREG(drive)	(byte)((((drive)->drive_data) & 0xff00) >> 8)
+#define QD_TIMING(drive)	(u8)(((drive)->drive_data) & 0x00ff)
+#define QD_TIMREG(drive)	(u8)((((drive)->drive_data) & 0xff00) >> 8)
 
 #define QD6500_DEF_DATA		((QD_TIM1_PORT<<8) | (QD_ID3 ? 0x0c : 0x08))
 #define QD6580_DEF_DATA		((QD_TIM1_PORT<<8) | (QD_ID3 ? 0x0a : 0x00))
diff --git a/drivers/ide/sc1200.c b/drivers/ide/sc1200.c
index 6f68fe984bfb..1cf477aaae36 100644
--- a/drivers/ide/sc1200.c
+++ b/drivers/ide/sc1200.c
@@ -172,7 +172,7 @@ static int sc1200_dma_end(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	unsigned long dma_base = hwif->dma_base;
-	byte dma_stat;
+	u8 dma_stat;
 
 	dma_stat = inb(dma_base+2);		/* get DMA status */
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index cca6cfb299bc..545a67f1f6b5 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -32,8 +32,6 @@
 # define SUPPORT_VLB_SYNC 1
 #endif
 
-typedef unsigned char	byte;	/* used everywhere */
-
 /*
  * Probably not wise to fiddle with these
  */
@@ -1161,7 +1159,7 @@ void ide_pad_transfer(ide_drive_t *, int, int);
 
 ide_startstop_t __ide_error(ide_drive_t *, struct request *, u8, u8);
 
-ide_startstop_t ide_error (ide_drive_t *drive, const char *msg, byte stat);
+ide_startstop_t ide_error(ide_drive_t *, const char *, u8);
 
 void ide_fix_driveid(u16 *);
 
-- 
cgit v1.2.3


From 7f3c868ba78e486bd9d7569f884dd46d8f59bb18 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:53 +0100
Subject: ide: remove ide_driver_t typedef

While at it:
- s/struct ide_driver_s/struct ide_driver/
- use to_ide_driver() macro in ide-proc.c

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-cd.c       |  2 +-
 drivers/ide/ide-cd.h       |  8 ++++----
 drivers/ide/ide-gd.c       |  2 +-
 drivers/ide/ide-gd.h       | 10 +++++-----
 drivers/ide/ide-io.c       | 12 ++++++------
 drivers/ide/ide-pm.c       |  2 +-
 drivers/ide/ide-proc.c     | 16 ++++++++--------
 drivers/ide/ide-tape.c     | 10 +++++-----
 drivers/ide/ide-taskfile.c |  8 ++++----
 drivers/ide/ide.c          |  6 +++---
 include/linux/ide.h        | 18 +++++++++---------
 11 files changed, 47 insertions(+), 47 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index c35b64d495f5..bc982dfacc36 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -1914,7 +1914,7 @@ static void ide_cd_release(struct kref *kref)
 
 static int ide_cd_probe(ide_drive_t *);
 
-static ide_driver_t ide_cdrom_driver = {
+static struct ide_driver ide_cdrom_driver = {
 	.gen_driver = {
 		.owner		= THIS_MODULE,
 		.name		= "ide-cdrom",
diff --git a/drivers/ide/ide-cd.h b/drivers/ide/ide-cd.h
index d5ae036af00c..ac40d6cb90a2 100644
--- a/drivers/ide/ide-cd.h
+++ b/drivers/ide/ide-cd.h
@@ -77,10 +77,10 @@ struct atapi_toc {
 
 /* Extra per-device info for cdrom drives. */
 struct cdrom_info {
-	ide_drive_t	*drive;
-	ide_driver_t	*driver;
-	struct gendisk	*disk;
-	struct kref	kref;
+	ide_drive_t		*drive;
+	struct ide_driver	*driver;
+	struct gendisk		*disk;
+	struct kref		kref;
 
 	/* Buffer for table of contents.  NULL if we haven't allocated
 	   a TOC buffer for this device yet. */
diff --git a/drivers/ide/ide-gd.c b/drivers/ide/ide-gd.c
index b8078b3231f7..cae82b3c432a 100644
--- a/drivers/ide/ide-gd.c
+++ b/drivers/ide/ide-gd.c
@@ -149,7 +149,7 @@ static int ide_gd_end_request(ide_drive_t *drive, int uptodate, int nrsecs)
 	return drive->disk_ops->end_request(drive, uptodate, nrsecs);
 }
 
-static ide_driver_t ide_gd_driver = {
+static struct ide_driver ide_gd_driver = {
 	.gen_driver = {
 		.owner		= THIS_MODULE,
 		.name		= "ide-gd",
diff --git a/drivers/ide/ide-gd.h b/drivers/ide/ide-gd.h
index 7d3d101713e0..a86779f0756b 100644
--- a/drivers/ide/ide-gd.h
+++ b/drivers/ide/ide-gd.h
@@ -14,11 +14,11 @@
 #endif
 
 struct ide_disk_obj {
-	ide_drive_t	*drive;
-	ide_driver_t	*driver;
-	struct gendisk	*disk;
-	struct kref	kref;
-	unsigned int	openers;	/* protected by BKL for now */
+	ide_drive_t		*drive;
+	struct ide_driver	*driver;
+	struct gendisk		*disk;
+	struct kref		kref;
+	unsigned int		openers;	/* protected by BKL for now */
 
 	/* Last failed packet command */
 	struct ide_atapi_pc *failed_pc;
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 0e2b95ec08a5..e788b3622f2e 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -199,9 +199,9 @@ EXPORT_SYMBOL(ide_end_drive_cmd);
 static void ide_kill_rq(ide_drive_t *drive, struct request *rq)
 {
 	if (rq->rq_disk) {
-		ide_driver_t *drv;
+		struct ide_driver *drv;
 
-		drv = *(ide_driver_t **)rq->rq_disk->private_data;
+		drv = *(struct ide_driver **)rq->rq_disk->private_data;
 		drv->end_request(drive, 0, 0);
 	} else
 		ide_end_request(drive, 0, 0);
@@ -333,9 +333,9 @@ ide_startstop_t ide_error (ide_drive_t *drive, const char *msg, u8 stat)
 	}
 
 	if (rq->rq_disk) {
-		ide_driver_t *drv;
+		struct ide_driver *drv;
 
-		drv = *(ide_driver_t **)rq->rq_disk->private_data;
+		drv = *(struct ide_driver **)rq->rq_disk->private_data;
 		return drv->error(drive, rq, stat, err);
 	} else
 		return __ide_error(drive, rq, stat, err);
@@ -606,7 +606,7 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 		return startstop;
 	}
 	if (!drive->special.all) {
-		ide_driver_t *drv;
+		struct ide_driver *drv;
 
 		/*
 		 * We reset the drive so we need to issue a SETFEATURES.
@@ -639,7 +639,7 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 			 */
 			return ide_special_rq(drive, rq);
 
-		drv = *(ide_driver_t **)rq->rq_disk->private_data;
+		drv = *(struct ide_driver **)rq->rq_disk->private_data;
 
 		return drv->do_request(drive, rq, rq->sector);
 	}
diff --git a/drivers/ide/ide-pm.c b/drivers/ide/ide-pm.c
index 0c206c68bbb1..4b3bf6a06b70 100644
--- a/drivers/ide/ide-pm.c
+++ b/drivers/ide/ide-pm.c
@@ -67,7 +67,7 @@ int generic_ide_resume(struct device *dev)
 	blk_put_request(rq);
 
 	if (err == 0 && dev->driver) {
-		ide_driver_t *drv = to_ide_driver(dev->driver);
+		struct ide_driver *drv = to_ide_driver(dev->driver);
 
 		if (drv->resume)
 			drv->resume(drive);
diff --git a/drivers/ide/ide-proc.c b/drivers/ide/ide-proc.c
index a14e2938e4f3..d985a9ec6bef 100644
--- a/drivers/ide/ide-proc.c
+++ b/drivers/ide/ide-proc.c
@@ -439,13 +439,13 @@ static int proc_ide_read_dmodel
 static int proc_ide_read_driver
 	(char *page, char **start, off_t off, int count, int *eof, void *data)
 {
-	ide_drive_t	*drive = (ide_drive_t *) data;
-	struct device	*dev = &drive->gendev;
-	ide_driver_t	*ide_drv;
-	int		len;
+	ide_drive_t		*drive = (ide_drive_t *)data;
+	struct device		*dev = &drive->gendev;
+	struct ide_driver	*ide_drv;
+	int			len;
 
 	if (dev->driver) {
-		ide_drv = container_of(dev->driver, ide_driver_t, gen_driver);
+		ide_drv = to_ide_driver(dev->driver);
 		len = sprintf(page, "%s version %s\n",
 				dev->driver->name, ide_drv->version);
 	} else
@@ -555,7 +555,7 @@ static void ide_remove_proc_entries(struct proc_dir_entry *dir, ide_proc_entry_t
 	}
 }
 
-void ide_proc_register_driver(ide_drive_t *drive, ide_driver_t *driver)
+void ide_proc_register_driver(ide_drive_t *drive, struct ide_driver *driver)
 {
 	mutex_lock(&ide_setting_mtx);
 	drive->settings = driver->proc_devsets(drive);
@@ -577,7 +577,7 @@ EXPORT_SYMBOL(ide_proc_register_driver);
  *	Takes ide_setting_mtx.
  */
 
-void ide_proc_unregister_driver(ide_drive_t *drive, ide_driver_t *driver)
+void ide_proc_unregister_driver(ide_drive_t *drive, struct ide_driver *driver)
 {
 	ide_remove_proc_entries(drive->proc, driver->proc_entries(drive));
 
@@ -653,7 +653,7 @@ void ide_proc_unregister_port(ide_hwif_t *hwif)
 
 static int proc_print_driver(struct device_driver *drv, void *data)
 {
-	ide_driver_t *ide_drv = container_of(drv, ide_driver_t, gen_driver);
+	struct ide_driver *ide_drv = to_ide_driver(drv);
 	struct seq_file *s = data;
 
 	seq_printf(s, "%s version %s\n", drv->name, ide_drv->version);
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index e39f2f46d982..fd865039bfdb 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -166,10 +166,10 @@ struct idetape_bh {
  * to an interrupt or a timer event is stored in the struct defined below.
  */
 typedef struct ide_tape_obj {
-	ide_drive_t	*drive;
-	ide_driver_t	*driver;
-	struct gendisk	*disk;
-	struct kref	kref;
+	ide_drive_t		*drive;
+	struct ide_driver	*driver;
+	struct gendisk		*disk;
+	struct kref		kref;
 
 	/*
 	 *	failed_pc points to the last failed packet command, or contains
@@ -2313,7 +2313,7 @@ static const struct ide_proc_devset *ide_tape_proc_devsets(ide_drive_t *drive)
 
 static int ide_tape_probe(ide_drive_t *);
 
-static ide_driver_t idetape_driver = {
+static struct ide_driver idetape_driver = {
 	.gen_driver = {
 		.owner		= THIS_MODULE,
 		.name		= "ide-tape",
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 55d451ce9b4f..16138bce84a7 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -309,9 +309,9 @@ static ide_startstop_t task_error(ide_drive_t *drive, struct request *rq,
 		}
 
 		if (sectors > 0) {
-			ide_driver_t *drv;
+			struct ide_driver *drv;
 
-			drv = *(ide_driver_t **)rq->rq_disk->private_data;
+			drv = *(struct ide_driver **)rq->rq_disk->private_data;
 			drv->end_request(drive, 1, sectors);
 		}
 	}
@@ -328,9 +328,9 @@ void task_end_request(ide_drive_t *drive, struct request *rq, u8 stat)
 	}
 
 	if (rq->rq_disk) {
-		ide_driver_t *drv;
+		struct ide_driver *drv;
 
-		drv = *(ide_driver_t **)rq->rq_disk->private_data;;
+		drv = *(struct ide_driver **)rq->rq_disk->private_data;;
 		drv->end_request(drive, 1, rq->nr_sectors);
 	} else
 		ide_end_request(drive, 1, rq->nr_sectors);
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index c1bb0f6784a9..6971c285a212 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -281,7 +281,7 @@ static int ide_uevent(struct device *dev, struct kobj_uevent_env *env)
 static int generic_ide_probe(struct device *dev)
 {
 	ide_drive_t *drive = to_ide_device(dev);
-	ide_driver_t *drv = to_ide_driver(dev->driver);
+	struct ide_driver *drv = to_ide_driver(dev->driver);
 
 	return drv->probe ? drv->probe(drive) : -ENODEV;
 }
@@ -289,7 +289,7 @@ static int generic_ide_probe(struct device *dev)
 static int generic_ide_remove(struct device *dev)
 {
 	ide_drive_t *drive = to_ide_device(dev);
-	ide_driver_t *drv = to_ide_driver(dev->driver);
+	struct ide_driver *drv = to_ide_driver(dev->driver);
 
 	if (drv->remove)
 		drv->remove(drive);
@@ -300,7 +300,7 @@ static int generic_ide_remove(struct device *dev)
 static void generic_ide_shutdown(struct device *dev)
 {
 	ide_drive_t *drive = to_ide_device(dev);
-	ide_driver_t *drv = to_ide_driver(dev->driver);
+	struct ide_driver *drv = to_ide_driver(dev->driver);
 
 	if (dev->driver && drv->shutdown)
 		drv->shutdown(drive);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 545a67f1f6b5..fcbcfa2cbe75 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -437,7 +437,7 @@ struct ide_atapi_pc {
 };
 
 struct ide_devset;
-struct ide_driver_s;
+struct ide_driver;
 
 #ifdef CONFIG_BLK_DEV_IDEACPI
 struct ide_acpi_drive_link;
@@ -884,8 +884,6 @@ typedef int (ide_expiry_t)(ide_drive_t *);
 /* used by ide-cd, ide-floppy, etc. */
 typedef void (xfer_func_t)(ide_drive_t *, struct request *rq, void *, unsigned);
 
-typedef struct ide_driver_s ide_driver_t;
-
 extern struct mutex ide_setting_mtx;
 
 /*
@@ -1011,8 +1009,8 @@ void ide_proc_register_port(ide_hwif_t *);
 void ide_proc_port_register_devices(ide_hwif_t *);
 void ide_proc_unregister_device(ide_drive_t *);
 void ide_proc_unregister_port(ide_hwif_t *);
-void ide_proc_register_driver(ide_drive_t *, ide_driver_t *);
-void ide_proc_unregister_driver(ide_drive_t *, ide_driver_t *);
+void ide_proc_register_driver(ide_drive_t *, struct ide_driver *);
+void ide_proc_unregister_driver(ide_drive_t *, struct ide_driver *);
 
 read_proc_t proc_ide_read_capacity;
 read_proc_t proc_ide_read_geometry;
@@ -1039,8 +1037,10 @@ static inline void ide_proc_register_port(ide_hwif_t *hwif) { ; }
 static inline void ide_proc_port_register_devices(ide_hwif_t *hwif) { ; }
 static inline void ide_proc_unregister_device(ide_drive_t *drive) { ; }
 static inline void ide_proc_unregister_port(ide_hwif_t *hwif) { ; }
-static inline void ide_proc_register_driver(ide_drive_t *drive, ide_driver_t *driver) { ; }
-static inline void ide_proc_unregister_driver(ide_drive_t *drive, ide_driver_t *driver) { ; }
+static inline void ide_proc_register_driver(ide_drive_t *drive,
+					    struct ide_driver *driver) { ; }
+static inline void ide_proc_unregister_driver(ide_drive_t *drive,
+					      struct ide_driver *driver) { ; }
 #define PROC_IDE_READ_RETURN(page,start,off,count,eof,len) return 0;
 #endif
 
@@ -1109,7 +1109,7 @@ void ide_check_pm_state(ide_drive_t *, struct request *);
  * The gendriver.owner field should be set to the module owner of this driver.
  * The gendriver.name field should be set to the name of this driver
  */
-struct ide_driver_s {
+struct ide_driver {
 	const char			*version;
 	ide_startstop_t	(*do_request)(ide_drive_t *, struct request *, sector_t);
 	int		(*end_request)(ide_drive_t *, int, int);
@@ -1125,7 +1125,7 @@ struct ide_driver_s {
 #endif
 };
 
-#define to_ide_driver(drv) container_of(drv, ide_driver_t, gen_driver)
+#define to_ide_driver(drv) container_of(drv, struct ide_driver, gen_driver)
 
 int ide_device_get(ide_drive_t *);
 void ide_device_put(ide_drive_t *);
-- 
cgit v1.2.3


From 627e05daa10896a8f012fa78e8434c07e9e55ea7 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:54 +0100
Subject: ide: remove ->error method from struct ide_driver

* Remove (now superfluous) ->error method from struct ide_driver.

* Unexport __ide_error() and make it static.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-cd.c   |  1 -
 drivers/ide/ide-gd.c   |  1 -
 drivers/ide/ide-io.c   | 13 ++-----------
 drivers/ide/ide-tape.c |  1 -
 include/linux/ide.h    |  3 ---
 5 files changed, 2 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index bc982dfacc36..6c7dd8fd8638 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -1925,7 +1925,6 @@ static struct ide_driver ide_cdrom_driver = {
 	.version		= IDECD_VERSION,
 	.do_request		= ide_cd_do_request,
 	.end_request		= ide_end_request,
-	.error			= __ide_error,
 #ifdef CONFIG_IDE_PROC_FS
 	.proc_entries		= ide_cd_proc_entries,
 	.proc_devsets		= ide_cd_proc_devsets,
diff --git a/drivers/ide/ide-gd.c b/drivers/ide/ide-gd.c
index cae82b3c432a..7857b209c6df 100644
--- a/drivers/ide/ide-gd.c
+++ b/drivers/ide/ide-gd.c
@@ -162,7 +162,6 @@ static struct ide_driver ide_gd_driver = {
 	.version		= IDE_GD_VERSION,
 	.do_request		= ide_gd_do_request,
 	.end_request		= ide_gd_end_request,
-	.error			= __ide_error,
 #ifdef CONFIG_IDE_PROC_FS
 	.proc_entries		= ide_disk_proc_entries,
 	.proc_devsets		= ide_disk_proc_devsets,
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index e788b3622f2e..9d363a1b5573 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -291,7 +291,7 @@ static ide_startstop_t ide_atapi_error(ide_drive_t *drive, struct request *rq, u
 	return ide_stopped;
 }
 
-ide_startstop_t
+static ide_startstop_t
 __ide_error(ide_drive_t *drive, struct request *rq, u8 stat, u8 err)
 {
 	if (drive->media == ide_disk)
@@ -299,8 +299,6 @@ __ide_error(ide_drive_t *drive, struct request *rq, u8 stat, u8 err)
 	return ide_atapi_error(drive, rq, stat, err);
 }
 
-EXPORT_SYMBOL_GPL(__ide_error);
-
 /**
  *	ide_error	-	handle an error on the IDE
  *	@drive: drive the error occurred on
@@ -332,15 +330,8 @@ ide_startstop_t ide_error (ide_drive_t *drive, const char *msg, u8 stat)
 		return ide_stopped;
 	}
 
-	if (rq->rq_disk) {
-		struct ide_driver *drv;
-
-		drv = *(struct ide_driver **)rq->rq_disk->private_data;
-		return drv->error(drive, rq, stat, err);
-	} else
-		return __ide_error(drive, rq, stat, err);
+	return __ide_error(drive, rq, stat, err);
 }
-
 EXPORT_SYMBOL_GPL(ide_error);
 
 static void ide_tf_set_specify_cmd(ide_drive_t *drive, struct ide_taskfile *tf)
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index fd865039bfdb..d7ecd3c79757 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -2324,7 +2324,6 @@ static struct ide_driver idetape_driver = {
 	.version		= IDETAPE_VERSION,
 	.do_request		= idetape_do_request,
 	.end_request		= idetape_end_request,
-	.error			= __ide_error,
 #ifdef CONFIG_IDE_PROC_FS
 	.proc_entries		= ide_tape_proc_entries,
 	.proc_devsets		= ide_tape_proc_devsets,
diff --git a/include/linux/ide.h b/include/linux/ide.h
index fcbcfa2cbe75..9f6fe1fe7a6c 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1113,7 +1113,6 @@ struct ide_driver {
 	const char			*version;
 	ide_startstop_t	(*do_request)(ide_drive_t *, struct request *, sector_t);
 	int		(*end_request)(ide_drive_t *, int, int);
-	ide_startstop_t	(*error)(ide_drive_t *, struct request *rq, u8, u8);
 	struct device_driver	gen_driver;
 	int		(*probe)(ide_drive_t *);
 	void		(*remove)(ide_drive_t *);
@@ -1157,8 +1156,6 @@ void ide_execute_pkt_cmd(ide_drive_t *);
 
 void ide_pad_transfer(ide_drive_t *, int, int);
 
-ide_startstop_t __ide_error(ide_drive_t *, struct request *, u8, u8);
-
 ide_startstop_t ide_error(ide_drive_t *, const char *, u8);
 
 void ide_fix_driveid(u16 *);
-- 
cgit v1.2.3


From 5e7f3a46690f7f6c9f2781c700ab4370874aa0e8 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:56 +0100
Subject: ide: dynamic allocation of device structures

Allocate device structures dynamically instead of having them embedded
in ide_hwif_t:

* Remove needless zeroing of port structure from ide_init_port_data().

* Add ide_hwif_t.devices[MAX_DRIVES] (table of pointers to the devices).

* Add ide_port_{alloc,free}_devices() helpers and use them respectively
  in ide_{host,free}_alloc().

* Convert all users of ->drives[] to use ->devices[] instead.

While at it:

* Use drive->dn for the slave device check in scc_pata.c.

As a nice side-effect this patch cuts ~1kB (x86-32) from the resulting
code size:

   text    data     bss     dec     hex filename
  53963    1244     237   55444    d894 drivers/ide/ide-core.o.before
  52981    1244     237   54462    d4be drivers/ide/ide-core.o.after

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-acpi.c  | 10 +++----
 drivers/ide/ide-iops.c  |  4 +--
 drivers/ide/ide-probe.c | 71 +++++++++++++++++++++++++++++++++++--------------
 drivers/ide/ide-proc.c  |  2 +-
 drivers/ide/ide.c       |  2 +-
 drivers/ide/scc_pata.c  |  5 ++--
 include/linux/ide.h     |  4 +--
 7 files changed, 65 insertions(+), 33 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-acpi.c b/drivers/ide/ide-acpi.c
index f89b6ecf7d1a..fd155b8a256c 100644
--- a/drivers/ide/ide-acpi.c
+++ b/drivers/ide/ide-acpi.c
@@ -656,7 +656,7 @@ void ide_acpi_set_state(ide_hwif_t *hwif, int on)
 	if (on)
 		acpi_bus_set_power(hwif->acpidata->obj_handle, ACPI_STATE_D0);
 	for (unit = 0; unit < MAX_DRIVES; ++unit) {
-		ide_drive_t *drive = &hwif->drives[unit];
+		ide_drive_t *drive = hwif->devices[unit];
 
 		if (!drive->acpidata->obj_handle)
 			drive->acpidata->obj_handle = ide_acpi_drive_get_handle(drive);
@@ -711,14 +711,14 @@ void ide_acpi_port_init_devices(ide_hwif_t *hwif)
 	 * for both drives, regardless whether they are connected
 	 * or not.
 	 */
-	hwif->drives[0].acpidata = &hwif->acpidata->master;
-	hwif->drives[1].acpidata = &hwif->acpidata->slave;
+	hwif->devices[0]->acpidata = &hwif->acpidata->master;
+	hwif->devices[1]->acpidata = &hwif->acpidata->slave;
 
 	/*
 	 * Send IDENTIFY for each drive
 	 */
 	for (i = 0; i < MAX_DRIVES; i++) {
-		drive = &hwif->drives[i];
+		drive = hwif->devices[i];
 
 		memset(drive->acpidata, 0, sizeof(*drive->acpidata));
 
@@ -745,7 +745,7 @@ void ide_acpi_port_init_devices(ide_hwif_t *hwif)
 	ide_acpi_push_timing(hwif);
 
 	for (i = 0; i < MAX_DRIVES; i++) {
-		drive = &hwif->drives[i];
+		drive = hwif->devices[i];
 
 		if (drive->dev_flags & IDE_DFLAG_PRESENT)
 			/* Execute ACPI startup code */
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 1bcb9484f49e..26b58d15c4e6 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -1111,7 +1111,7 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 		prepare_to_wait(&ide_park_wq, &wait, TASK_UNINTERRUPTIBLE);
 		timeout = jiffies;
 		for (unit = 0; unit < MAX_DRIVES; unit++) {
-			ide_drive_t *tdrive = &hwif->drives[unit];
+			ide_drive_t *tdrive = hwif->devices[unit];
 
 			if (tdrive->dev_flags & IDE_DFLAG_PRESENT &&
 			    tdrive->dev_flags & IDE_DFLAG_PARKED &&
@@ -1134,7 +1134,7 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	 * for any of the drives on this interface.
 	 */
 	for (unit = 0; unit < MAX_DRIVES; ++unit)
-		pre_reset(&hwif->drives[unit]);
+		pre_reset(hwif->devices[unit]);
 
 	if (io_ports->ctl_addr == 0) {
 		spin_unlock_irqrestore(&hwif->lock, flags);
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 00c7e5a67bd1..006e601cafb8 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -463,7 +463,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 	if (ide_read_device(drive) != drive->select && present == 0) {
 		if (drive->dn & 1) {
 			/* exit with drive0 selected */
-			SELECT_DRIVE(&hwif->drives[0]);
+			SELECT_DRIVE(hwif->devices[0]);
 			/* allow ATA_BUSY to assert & clear */
 			msleep(50);
 		}
@@ -509,7 +509,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 	}
 	if (drive->dn & 1) {
 		/* exit with drive0 selected */
-		SELECT_DRIVE(&hwif->drives[0]);
+		SELECT_DRIVE(hwif->devices[0]);
 		msleep(50);
 		/* ensure drive irq is clear */
 		(void)tp_ops->read_status(hwif);
@@ -715,7 +715,7 @@ static int ide_port_wait_ready(ide_hwif_t *hwif)
 
 	/* Now make sure both master & slave are ready */
 	for (unit = 0; unit < MAX_DRIVES; unit++) {
-		ide_drive_t *drive = &hwif->drives[unit];
+		ide_drive_t *drive = hwif->devices[unit];
 
 		/* Ignore disks that we will not probe for later. */
 		if ((drive->dev_flags & IDE_DFLAG_NOPROBE) == 0 ||
@@ -733,7 +733,7 @@ static int ide_port_wait_ready(ide_hwif_t *hwif)
 out:
 	/* Exit function with master reselected (let's be sane) */
 	if (unit)
-		SELECT_DRIVE(&hwif->drives[0]);
+		SELECT_DRIVE(hwif->devices[0]);
 
 	return rc;
 }
@@ -749,7 +749,7 @@ out:
 
 void ide_undecoded_slave(ide_drive_t *dev1)
 {
-	ide_drive_t *dev0 = &dev1->hwif->drives[0];
+	ide_drive_t *dev0 = dev1->hwif->devices[0];
 
 	if ((dev1->dn & 1) == 0 || (dev0->dev_flags & IDE_DFLAG_PRESENT) == 0)
 		return;
@@ -784,8 +784,8 @@ static int ide_probe_port(ide_hwif_t *hwif)
 
 	BUG_ON(hwif->present);
 
-	if ((hwif->drives[0].dev_flags & IDE_DFLAG_NOPROBE) &&
-	    (hwif->drives[1].dev_flags & IDE_DFLAG_NOPROBE))
+	if ((hwif->devices[0]->dev_flags & IDE_DFLAG_NOPROBE) &&
+	    (hwif->devices[1]->dev_flags & IDE_DFLAG_NOPROBE))
 		return -EACCES;
 
 	/*
@@ -807,7 +807,7 @@ static int ide_probe_port(ide_hwif_t *hwif)
 	 * but a lot of cdrom drives are configured as single slaves.
 	 */
 	for (unit = 0; unit < MAX_DRIVES; ++unit) {
-		ide_drive_t *drive = &hwif->drives[unit];
+		ide_drive_t *drive = hwif->devices[unit];
 
 		(void) probe_for_drive(drive);
 		if (drive->dev_flags & IDE_DFLAG_PRESENT)
@@ -832,7 +832,7 @@ static void ide_port_tune_devices(ide_hwif_t *hwif)
 	int unit;
 
 	for (unit = 0; unit < MAX_DRIVES; unit++) {
-		ide_drive_t *drive = &hwif->drives[unit];
+		ide_drive_t *drive = hwif->devices[unit];
 
 		if (drive->dev_flags & IDE_DFLAG_PRESENT) {
 			if (port_ops && port_ops->quirkproc)
@@ -841,7 +841,7 @@ static void ide_port_tune_devices(ide_hwif_t *hwif)
 	}
 
 	for (unit = 0; unit < MAX_DRIVES; ++unit) {
-		ide_drive_t *drive = &hwif->drives[unit];
+		ide_drive_t *drive = hwif->devices[unit];
 
 		if (drive->dev_flags & IDE_DFLAG_PRESENT) {
 			ide_set_max_pio(drive);
@@ -854,7 +854,7 @@ static void ide_port_tune_devices(ide_hwif_t *hwif)
 	}
 
 	for (unit = 0; unit < MAX_DRIVES; ++unit) {
-		ide_drive_t *drive = &hwif->drives[unit];
+		ide_drive_t *drive = hwif->devices[unit];
 
 		if ((hwif->host_flags & IDE_HFLAG_NO_IO_32BIT) ||
 		    drive->id[ATA_ID_DWORD_IO])
@@ -931,7 +931,7 @@ static int ide_port_setup_devices(ide_hwif_t *hwif)
 
 	mutex_lock(&ide_cfg_mtx);
 	for (i = 0; i < MAX_DRIVES; i++) {
-		ide_drive_t *drive = &hwif->drives[i];
+		ide_drive_t *drive = hwif->devices[i];
 
 		if ((drive->dev_flags & IDE_DFLAG_PRESENT) == 0)
 			continue;
@@ -1017,7 +1017,7 @@ static struct kobject *ata_probe(dev_t dev, int *part, void *data)
 {
 	ide_hwif_t *hwif = data;
 	int unit = *part >> PARTN_BITS;
-	ide_drive_t *drive = &hwif->drives[unit];
+	ide_drive_t *drive = hwif->devices[unit];
 
 	if ((drive->dev_flags & IDE_DFLAG_PRESENT) == 0)
 		return NULL;
@@ -1164,7 +1164,7 @@ static void hwif_register_devices(ide_hwif_t *hwif)
 	unsigned int i;
 
 	for (i = 0; i < MAX_DRIVES; i++) {
-		ide_drive_t *drive = &hwif->drives[i];
+		ide_drive_t *drive = hwif->devices[i];
 		struct device *dev = &drive->gendev;
 		int ret;
 
@@ -1190,7 +1190,7 @@ static void ide_port_init_devices(ide_hwif_t *hwif)
 	int i;
 
 	for (i = 0; i < MAX_DRIVES; i++) {
-		ide_drive_t *drive = &hwif->drives[i];
+		ide_drive_t *drive = hwif->devices[i];
 
 		drive->dn = i + hwif->channel * 2;
 
@@ -1285,7 +1285,7 @@ static void ide_port_init_devices_data(ide_hwif_t *hwif)
 	int unit;
 
 	for (unit = 0; unit < MAX_DRIVES; ++unit) {
-		ide_drive_t *drive = &hwif->drives[unit];
+		ide_drive_t *drive = hwif->devices[unit];
 		u8 j = (hwif->index * MAX_DRIVES) + unit;
 
 		memset(drive, 0, sizeof(*drive));
@@ -1309,9 +1309,6 @@ static void ide_port_init_devices_data(ide_hwif_t *hwif)
 
 static void ide_init_port_data(ide_hwif_t *hwif, unsigned int index)
 {
-	/* bulk initialize hwif & drive info with zeros */
-	memset(hwif, 0, sizeof(ide_hwif_t));
-
 	/* fill in any non-zero initial values */
 	hwif->index	= index;
 	hwif->major	= ide_hwif_to_major[index];
@@ -1388,6 +1385,34 @@ static void ide_free_port_slot(int idx)
 	mutex_unlock(&ide_cfg_mtx);
 }
 
+static void ide_port_free_devices(ide_hwif_t *hwif)
+{
+	int i;
+
+	for (i = 0; i < MAX_DRIVES; i++)
+		kfree(hwif->devices[i]);
+}
+
+static int ide_port_alloc_devices(ide_hwif_t *hwif, int node)
+{
+	int i;
+
+	for (i = 0; i < MAX_DRIVES; i++) {
+		ide_drive_t *drive;
+
+		drive = kzalloc_node(sizeof(*drive), GFP_KERNEL, node);
+		if (drive == NULL)
+			goto out_nomem;
+
+		hwif->devices[i] = drive;
+	}
+	return 0;
+
+out_nomem:
+	ide_port_free_devices(hwif);
+	return -ENOMEM;
+}
+
 struct ide_host *ide_host_alloc(const struct ide_port_info *d, hw_regs_t **hws)
 {
 	struct ide_host *host;
@@ -1410,6 +1435,11 @@ struct ide_host *ide_host_alloc(const struct ide_port_info *d, hw_regs_t **hws)
 		if (hwif == NULL)
 			continue;
 
+		if (ide_port_alloc_devices(hwif, node) < 0) {
+			kfree(hwif);
+			continue;
+		}
+
 		idx = ide_find_port_slot(d);
 		if (idx < 0) {
 			printk(KERN_ERR "%s: no free slot for interface\n",
@@ -1575,7 +1605,7 @@ static void __ide_port_unregister_devices(ide_hwif_t *hwif)
 	int i;
 
 	for (i = 0; i < MAX_DRIVES; i++) {
-		ide_drive_t *drive = &hwif->drives[i];
+		ide_drive_t *drive = hwif->devices[i];
 
 		if (drive->dev_flags & IDE_DFLAG_PRESENT) {
 			device_unregister(&drive->gendev);
@@ -1651,6 +1681,7 @@ void ide_host_free(struct ide_host *host)
 		if (hwif == NULL)
 			continue;
 
+		ide_port_free_devices(hwif);
 		ide_free_port_slot(hwif->index);
 		kfree(hwif);
 	}
diff --git a/drivers/ide/ide-proc.c b/drivers/ide/ide-proc.c
index d985a9ec6bef..1dc827fa7061 100644
--- a/drivers/ide/ide-proc.c
+++ b/drivers/ide/ide-proc.c
@@ -599,7 +599,7 @@ void ide_proc_port_register_devices(ide_hwif_t *hwif)
 	char name[64];
 
 	for (d = 0; d < MAX_DRIVES; d++) {
-		ide_drive_t *drive = &hwif->drives[d];
+		ide_drive_t *drive = hwif->devices[d];
 
 		if ((drive->dev_flags & IDE_DFLAG_PRESENT) == 0 || drive->proc)
 			continue;
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 6971c285a212..8a6f893d127e 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -497,7 +497,7 @@ void ide_port_apply_params(ide_hwif_t *hwif)
 	}
 
 	for (i = 0; i < MAX_DRIVES; i++)
-		ide_dev_apply_params(&hwif->drives[i], i);
+		ide_dev_apply_params(hwif->devices[i], i);
 }
 
 /*
diff --git a/drivers/ide/scc_pata.c b/drivers/ide/scc_pata.c
index 90574ab76345..841164d7d3b9 100644
--- a/drivers/ide/scc_pata.c
+++ b/drivers/ide/scc_pata.c
@@ -259,7 +259,7 @@ static void scc_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	unsigned long scrcst_port = ctl_base + 0x014;
 	unsigned long udenvt_port = ctl_base + 0x018;
 	unsigned long tdvhsel_port   = ctl_base + 0x020;
-	int is_slave = (&hwif->drives[1] == drive);
+	int is_slave = drive->dn & 1;
 	int offset, idx;
 	unsigned long reg;
 	unsigned long jcactsel;
@@ -413,7 +413,8 @@ static int scc_dma_end(ide_drive_t *drive)
 				if (rq)
 					rq->errors |= ERROR_RESET;
 				for (unit = 0; unit < MAX_DRIVES; unit++) {
-					ide_drive_t *drive = &hwif->drives[unit];
+					ide_drive_t *drive = hwif->devices[unit];
+
 					drive->crc_count++;
 				}
 			}
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 9f6fe1fe7a6c..f00086b10be3 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -753,7 +753,7 @@ typedef struct hwif_s {
 
 	unsigned long	sata_scr[SATA_NR_PORTS];
 
-	ide_drive_t	drives[MAX_DRIVES];	/* drive info */
+	ide_drive_t	*devices[MAX_DRIVES];
 
 	u8 major;	/* our major number */
 	u8 index;	/* 0 for ide0; 1 for ide1; ... */
@@ -1600,7 +1600,7 @@ static inline int hwif_to_node(ide_hwif_t *hwif)
 
 static inline ide_drive_t *ide_get_pair_dev(ide_drive_t *drive)
 {
-	ide_drive_t *peer = &drive->hwif->drives[(drive->dn ^ 1) & 1];
+	ide_drive_t *peer = drive->hwif->devices[(drive->dn ^ 1) & 1];
 
 	return (peer->dev_flags & IDE_DFLAG_PRESENT) ? peer : NULL;
 }
-- 
cgit v1.2.3


From 2bd24a1cfc99d242c2cff9a6b74ca49fcaac3fb6 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:56 +0100
Subject: ide: add port and host iterators

Add ide_port_for_each_dev() / ide_host_for_each_port() iterators
and update IDE code to use them.

While at it:
- s/unit/i/ variable in ide_port_wait_ready(), ide_probe_port(),
  ide_port_tune_devices(), ide_port_init_devices_data(), do_reset1(),
  ide_acpi_set_state() and scc_dma_end()
- s/d/i/ variable in ide_proc_port_register_devices()

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-acpi.c  | 14 +++-----
 drivers/ide/ide-iops.c  | 11 +++---
 drivers/ide/ide-probe.c | 96 ++++++++++++++++++++-----------------------------
 drivers/ide/ide-proc.c  |  7 ++--
 drivers/ide/ide.c       |  5 +--
 drivers/ide/scc_pata.c  |  8 ++---
 include/linux/ide.h     | 11 ++++--
 7 files changed, 67 insertions(+), 85 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-acpi.c b/drivers/ide/ide-acpi.c
index fd155b8a256c..2f9e941968d6 100644
--- a/drivers/ide/ide-acpi.c
+++ b/drivers/ide/ide-acpi.c
@@ -641,7 +641,8 @@ void ide_acpi_push_timing(ide_hwif_t *hwif)
  */
 void ide_acpi_set_state(ide_hwif_t *hwif, int on)
 {
-	int unit;
+	ide_drive_t *drive;
+	int i;
 
 	if (ide_noacpi || ide_noacpi_psx)
 		return;
@@ -655,9 +656,8 @@ void ide_acpi_set_state(ide_hwif_t *hwif, int on)
 	/* channel first and then drives for power on and verse versa for power off */
 	if (on)
 		acpi_bus_set_power(hwif->acpidata->obj_handle, ACPI_STATE_D0);
-	for (unit = 0; unit < MAX_DRIVES; ++unit) {
-		ide_drive_t *drive = hwif->devices[unit];
 
+	ide_port_for_each_dev(i, drive, hwif) {
 		if (!drive->acpidata->obj_handle)
 			drive->acpidata->obj_handle = ide_acpi_drive_get_handle(drive);
 
@@ -717,9 +717,7 @@ void ide_acpi_port_init_devices(ide_hwif_t *hwif)
 	/*
 	 * Send IDENTIFY for each drive
 	 */
-	for (i = 0; i < MAX_DRIVES; i++) {
-		drive = hwif->devices[i];
-
+	ide_port_for_each_dev(i, drive, hwif) {
 		memset(drive->acpidata, 0, sizeof(*drive->acpidata));
 
 		if ((drive->dev_flags & IDE_DFLAG_PRESENT) == 0)
@@ -744,9 +742,7 @@ void ide_acpi_port_init_devices(ide_hwif_t *hwif)
 	ide_acpi_get_timing(hwif);
 	ide_acpi_push_timing(hwif);
 
-	for (i = 0; i < MAX_DRIVES; i++) {
-		drive = hwif->devices[i];
-
+	ide_port_for_each_dev(i, drive, hwif) {
 		if (drive->dev_flags & IDE_DFLAG_PRESENT)
 			/* Execute ACPI startup code */
 			ide_acpi_exec_tfs(drive);
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 26b58d15c4e6..1a22c31ce7ac 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -1081,8 +1081,9 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	struct ide_io_ports *io_ports = &hwif->io_ports;
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	const struct ide_port_ops *port_ops;
+	ide_drive_t *tdrive;
 	unsigned long flags, timeout;
-	unsigned int unit;
+	int i;
 	DEFINE_WAIT(wait);
 
 	spin_lock_irqsave(&hwif->lock, flags);
@@ -1110,9 +1111,7 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 
 		prepare_to_wait(&ide_park_wq, &wait, TASK_UNINTERRUPTIBLE);
 		timeout = jiffies;
-		for (unit = 0; unit < MAX_DRIVES; unit++) {
-			ide_drive_t *tdrive = hwif->devices[unit];
-
+		ide_port_for_each_dev(i, tdrive, hwif) {
 			if (tdrive->dev_flags & IDE_DFLAG_PRESENT &&
 			    tdrive->dev_flags & IDE_DFLAG_PARKED &&
 			    time_after(tdrive->sleep, timeout))
@@ -1133,8 +1132,8 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 	 * First, reset any device state data we were maintaining
 	 * for any of the drives on this interface.
 	 */
-	for (unit = 0; unit < MAX_DRIVES; ++unit)
-		pre_reset(hwif->devices[unit]);
+	ide_port_for_each_dev(i, tdrive, hwif)
+		pre_reset(tdrive);
 
 	if (io_ports->ctl_addr == 0) {
 		spin_unlock_irqrestore(&hwif->lock, flags);
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 006e601cafb8..e688ca1c967c 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -697,7 +697,8 @@ out:
 
 static int ide_port_wait_ready(ide_hwif_t *hwif)
 {
-	int unit, rc;
+	ide_drive_t *drive;
+	int i, rc;
 
 	printk(KERN_DEBUG "Probing IDE interface %s...\n", hwif->name);
 
@@ -714,9 +715,7 @@ static int ide_port_wait_ready(ide_hwif_t *hwif)
 		return rc;
 
 	/* Now make sure both master & slave are ready */
-	for (unit = 0; unit < MAX_DRIVES; unit++) {
-		ide_drive_t *drive = hwif->devices[unit];
-
+	ide_port_for_each_dev(i, drive, hwif) {
 		/* Ignore disks that we will not probe for later. */
 		if ((drive->dev_flags & IDE_DFLAG_NOPROBE) == 0 ||
 		    (drive->dev_flags & IDE_DFLAG_PRESENT)) {
@@ -732,7 +731,7 @@ static int ide_port_wait_ready(ide_hwif_t *hwif)
 	}
 out:
 	/* Exit function with master reselected (let's be sane) */
-	if (unit)
+	if (i)
 		SELECT_DRIVE(hwif->devices[0]);
 
 	return rc;
@@ -778,9 +777,10 @@ EXPORT_SYMBOL_GPL(ide_undecoded_slave);
 
 static int ide_probe_port(ide_hwif_t *hwif)
 {
+	ide_drive_t *drive;
 	unsigned long flags;
 	unsigned int irqd;
-	int unit, rc = -ENODEV;
+	int i, rc = -ENODEV;
 
 	BUG_ON(hwif->present);
 
@@ -806,9 +806,7 @@ static int ide_probe_port(ide_hwif_t *hwif)
 	 * Second drive should only exist if first drive was found,
 	 * but a lot of cdrom drives are configured as single slaves.
 	 */
-	for (unit = 0; unit < MAX_DRIVES; ++unit) {
-		ide_drive_t *drive = hwif->devices[unit];
-
+	ide_port_for_each_dev(i, drive, hwif) {
 		(void) probe_for_drive(drive);
 		if (drive->dev_flags & IDE_DFLAG_PRESENT)
 			rc = 0;
@@ -829,20 +827,17 @@ static int ide_probe_port(ide_hwif_t *hwif)
 static void ide_port_tune_devices(ide_hwif_t *hwif)
 {
 	const struct ide_port_ops *port_ops = hwif->port_ops;
-	int unit;
-
-	for (unit = 0; unit < MAX_DRIVES; unit++) {
-		ide_drive_t *drive = hwif->devices[unit];
+	ide_drive_t *drive;
+	int i;
 
+	ide_port_for_each_dev(i, drive, hwif) {
 		if (drive->dev_flags & IDE_DFLAG_PRESENT) {
 			if (port_ops && port_ops->quirkproc)
 				port_ops->quirkproc(drive);
 		}
 	}
 
-	for (unit = 0; unit < MAX_DRIVES; ++unit) {
-		ide_drive_t *drive = hwif->devices[unit];
-
+	ide_port_for_each_dev(i, drive, hwif) {
 		if (drive->dev_flags & IDE_DFLAG_PRESENT) {
 			ide_set_max_pio(drive);
 
@@ -853,9 +848,7 @@ static void ide_port_tune_devices(ide_hwif_t *hwif)
 		}
 	}
 
-	for (unit = 0; unit < MAX_DRIVES; ++unit) {
-		ide_drive_t *drive = hwif->devices[unit];
-
+	ide_port_for_each_dev(i, drive, hwif) {
 		if ((hwif->host_flags & IDE_HFLAG_NO_IO_32BIT) ||
 		    drive->id[ATA_ID_DWORD_IO])
 			drive->dev_flags |= IDE_DFLAG_NO_IO_32BIT;
@@ -927,12 +920,11 @@ static DEFINE_MUTEX(ide_cfg_mtx);
  */
 static int ide_port_setup_devices(ide_hwif_t *hwif)
 {
+	ide_drive_t *drive;
 	int i, j = 0;
 
 	mutex_lock(&ide_cfg_mtx);
-	for (i = 0; i < MAX_DRIVES; i++) {
-		ide_drive_t *drive = hwif->devices[i];
-
+	ide_port_for_each_dev(i, drive, hwif) {
 		if ((drive->dev_flags & IDE_DFLAG_PRESENT) == 0)
 			continue;
 
@@ -1161,10 +1153,10 @@ out:
 
 static void hwif_register_devices(ide_hwif_t *hwif)
 {
+	ide_drive_t *drive;
 	unsigned int i;
 
-	for (i = 0; i < MAX_DRIVES; i++) {
-		ide_drive_t *drive = hwif->devices[i];
+	ide_port_for_each_dev(i, drive, hwif) {
 		struct device *dev = &drive->gendev;
 		int ret;
 
@@ -1187,11 +1179,10 @@ static void hwif_register_devices(ide_hwif_t *hwif)
 static void ide_port_init_devices(ide_hwif_t *hwif)
 {
 	const struct ide_port_ops *port_ops = hwif->port_ops;
+	ide_drive_t *drive;
 	int i;
 
-	for (i = 0; i < MAX_DRIVES; i++) {
-		ide_drive_t *drive = hwif->devices[i];
-
+	ide_port_for_each_dev(i, drive, hwif) {
 		drive->dn = i + hwif->channel * 2;
 
 		if (hwif->host_flags & IDE_HFLAG_IO_32BIT)
@@ -1282,16 +1273,16 @@ static const u8 ide_hwif_to_major[] =
 
 static void ide_port_init_devices_data(ide_hwif_t *hwif)
 {
-	int unit;
+	ide_drive_t *drive;
+	int i;
 
-	for (unit = 0; unit < MAX_DRIVES; ++unit) {
-		ide_drive_t *drive = hwif->devices[unit];
-		u8 j = (hwif->index * MAX_DRIVES) + unit;
+	ide_port_for_each_dev(i, drive, hwif) {
+		u8 j = (hwif->index * MAX_DRIVES) + i;
 
 		memset(drive, 0, sizeof(*drive));
 
 		drive->media			= ide_disk;
-		drive->select			= (unit << 4) | ATA_DEVICE_OBS;
+		drive->select			= (i << 4) | ATA_DEVICE_OBS;
 		drive->hwif			= hwif;
 		drive->ready_stat		= ATA_DRDY;
 		drive->bad_wstat		= BAD_W_STAT;
@@ -1387,10 +1378,11 @@ static void ide_free_port_slot(int idx)
 
 static void ide_port_free_devices(ide_hwif_t *hwif)
 {
+	ide_drive_t *drive;
 	int i;
 
-	for (i = 0; i < MAX_DRIVES; i++)
-		kfree(hwif->devices[i]);
+	ide_port_for_each_dev(i, drive, hwif)
+		kfree(drive);
 }
 
 static int ide_port_alloc_devices(ide_hwif_t *hwif, int node)
@@ -1478,9 +1470,7 @@ int ide_host_register(struct ide_host *host, const struct ide_port_info *d,
 	ide_hwif_t *hwif, *mate = NULL;
 	int i, j = 0;
 
-	for (i = 0; i < MAX_HOST_PORTS; i++) {
-		hwif = host->ports[i];
-
+	ide_host_for_each_port(i, hwif, host) {
 		if (hwif == NULL) {
 			mate = NULL;
 			continue;
@@ -1506,9 +1496,7 @@ int ide_host_register(struct ide_host *host, const struct ide_port_info *d,
 		ide_port_init_devices(hwif);
 	}
 
-	for (i = 0; i < MAX_HOST_PORTS; i++) {
-		hwif = host->ports[i];
-
+	ide_host_for_each_port(i, hwif, host) {
 		if (hwif == NULL)
 			continue;
 
@@ -1523,9 +1511,7 @@ int ide_host_register(struct ide_host *host, const struct ide_port_info *d,
 			ide_port_tune_devices(hwif);
 	}
 
-	for (i = 0; i < MAX_HOST_PORTS; i++) {
-		hwif = host->ports[i];
-
+	ide_host_for_each_port(i, hwif, host) {
 		if (hwif == NULL)
 			continue;
 
@@ -1550,9 +1536,7 @@ int ide_host_register(struct ide_host *host, const struct ide_port_info *d,
 			ide_acpi_port_init_devices(hwif);
 	}
 
-	for (i = 0; i < MAX_HOST_PORTS; i++) {
-		hwif = host->ports[i];
-
+	ide_host_for_each_port(i, hwif, host) {
 		if (hwif == NULL)
 			continue;
 
@@ -1560,9 +1544,7 @@ int ide_host_register(struct ide_host *host, const struct ide_port_info *d,
 			hwif_register_devices(hwif);
 	}
 
-	for (i = 0; i < MAX_HOST_PORTS; i++) {
-		hwif = host->ports[i];
-
+	ide_host_for_each_port(i, hwif, host) {
 		if (hwif == NULL)
 			continue;
 
@@ -1602,11 +1584,10 @@ EXPORT_SYMBOL_GPL(ide_host_add);
 
 static void __ide_port_unregister_devices(ide_hwif_t *hwif)
 {
+	ide_drive_t *drive;
 	int i;
 
-	for (i = 0; i < MAX_DRIVES; i++) {
-		ide_drive_t *drive = hwif->devices[i];
-
+	ide_port_for_each_dev(i, drive, hwif) {
 		if (drive->dev_flags & IDE_DFLAG_PRESENT) {
 			device_unregister(&drive->gendev);
 			wait_for_completion(&drive->gendev_rel_comp);
@@ -1675,9 +1656,7 @@ void ide_host_free(struct ide_host *host)
 	ide_hwif_t *hwif;
 	int i;
 
-	for (i = 0; i < MAX_HOST_PORTS; i++) {
-		hwif = host->ports[i];
-
+	ide_host_for_each_port(i, hwif, host) {
 		if (hwif == NULL)
 			continue;
 
@@ -1692,11 +1671,12 @@ EXPORT_SYMBOL_GPL(ide_host_free);
 
 void ide_host_remove(struct ide_host *host)
 {
+	ide_hwif_t *hwif;
 	int i;
 
-	for (i = 0; i < MAX_HOST_PORTS; i++) {
-		if (host->ports[i])
-			ide_unregister(host->ports[i]);
+	ide_host_for_each_port(i, hwif, host) {
+		if (hwif)
+			ide_unregister(hwif);
 	}
 
 	ide_host_free(host);
diff --git a/drivers/ide/ide-proc.c b/drivers/ide/ide-proc.c
index 1dc827fa7061..1d8978b3314a 100644
--- a/drivers/ide/ide-proc.c
+++ b/drivers/ide/ide-proc.c
@@ -593,14 +593,13 @@ EXPORT_SYMBOL(ide_proc_unregister_driver);
 
 void ide_proc_port_register_devices(ide_hwif_t *hwif)
 {
-	int	d;
 	struct proc_dir_entry *ent;
 	struct proc_dir_entry *parent = hwif->proc;
+	ide_drive_t *drive;
 	char name[64];
+	int i;
 
-	for (d = 0; d < MAX_DRIVES; d++) {
-		ide_drive_t *drive = hwif->devices[d];
-
+	ide_port_for_each_dev(i, drive, hwif) {
 		if ((drive->dev_flags & IDE_DFLAG_PRESENT) == 0 || drive->proc)
 			continue;
 
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 8a6f893d127e..258805da15c3 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -488,6 +488,7 @@ MODULE_PARM_DESC(ignore_cable, "ignore cable detection");
 
 void ide_port_apply_params(ide_hwif_t *hwif)
 {
+	ide_drive_t *drive;
 	int i;
 
 	if (ide_ignore_cable & (1 << hwif->index)) {
@@ -496,8 +497,8 @@ void ide_port_apply_params(ide_hwif_t *hwif)
 		hwif->cbl = ATA_CBL_PATA40_SHORT;
 	}
 
-	for (i = 0; i < MAX_DRIVES; i++)
-		ide_dev_apply_params(hwif->devices[i], i);
+	ide_port_for_each_dev(i, drive, hwif)
+		ide_dev_apply_params(drive, i);
 }
 
 /*
diff --git a/drivers/ide/scc_pata.c b/drivers/ide/scc_pata.c
index 841164d7d3b9..5d53850c79d7 100644
--- a/drivers/ide/scc_pata.c
+++ b/drivers/ide/scc_pata.c
@@ -406,17 +406,17 @@ static int scc_dma_end(ide_drive_t *drive)
 			data_loss = 1;
 			if (retry++) {
 				struct request *rq = hwif->rq;
-				int unit;
+				ide_drive_t *drive;
+				int i;
+
 				/* ERROR_RESET and drive->crc_count are needed
 				 * to reduce DMA transfer mode in retry process.
 				 */
 				if (rq)
 					rq->errors |= ERROR_RESET;
-				for (unit = 0; unit < MAX_DRIVES; unit++) {
-					ide_drive_t *drive = hwif->devices[unit];
 
+				ide_port_for_each_dev(i, drive, hwif)
 					drive->crc_count++;
-				}
 			}
 		}
 	}
diff --git a/include/linux/ide.h b/include/linux/ide.h
index f00086b10be3..4cecd923fc79 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -753,7 +753,7 @@ typedef struct hwif_s {
 
 	unsigned long	sata_scr[SATA_NR_PORTS];
 
-	ide_drive_t	*devices[MAX_DRIVES];
+	ide_drive_t	*devices[MAX_DRIVES + 1];
 
 	u8 major;	/* our major number */
 	u8 index;	/* 0 for ide0; 1 for ide1; ... */
@@ -861,7 +861,7 @@ typedef struct hwif_s {
 #define MAX_HOST_PORTS 4
 
 struct ide_host {
-	ide_hwif_t	*ports[MAX_HOST_PORTS];
+	ide_hwif_t	*ports[MAX_HOST_PORTS + 1];
 	unsigned int	n_ports;
 	struct device	*dev[2];
 	unsigned int	(*init_chipset)(struct pci_dev *);
@@ -1604,4 +1604,11 @@ static inline ide_drive_t *ide_get_pair_dev(ide_drive_t *drive)
 
 	return (peer->dev_flags & IDE_DFLAG_PRESENT) ? peer : NULL;
 }
+
+#define ide_port_for_each_dev(i, dev, port) \
+	for ((i) = 0; ((dev) = (port)->devices[i]) || (i) < MAX_DRIVES; (i)++)
+
+#define ide_host_for_each_port(i, port, host) \
+	for ((i) = 0; ((port) = (host)->ports[i]) || (i) < MAX_HOST_PORTS; (i)++)
+
 #endif /* _IDE_H */
-- 
cgit v1.2.3


From d6251d4488a361c93da2398818e1ec69cffb6073 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@gmail.com>
Date: Tue, 6 Jan 2009 17:20:58 +0100
Subject: ide-cd: convert to ide-atapi facilities

... and remove no longer needed cdrom_start_packet_command and
cdrom_transfer_packet_command.

Tested lightly with ide-cd and ide-floppy.

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-atapi.c |  5 ++-
 drivers/ide/ide-cd.c    | 96 ++-----------------------------------------------
 include/linux/ide.h     |  2 ++
 3 files changed, 8 insertions(+), 95 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index d6a50d67b7db..e96c01260598 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -549,7 +549,10 @@ static ide_startstop_t ide_transfer_pc(ide_drive_t *drive)
 	}
 
 	/* Set the interrupt routine */
-	ide_set_handler(drive, ide_pc_intr, timeout, expiry);
+	ide_set_handler(drive,
+			(dev_is_idecd(drive) ? drive->irq_handler
+					     : ide_pc_intr),
+			timeout, expiry);
 
 	/* Begin DMA, if necessary */
 	if (dev_is_idecd(drive)) {
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 2cb301dccd21..cae69372cf45 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -510,99 +510,6 @@ end_request:
 	return 1;
 }
 
-static ide_startstop_t cdrom_transfer_packet_command(ide_drive_t *);
-static ide_startstop_t cdrom_newpc_intr(ide_drive_t *);
-
-/*
- * Set up the device registers for transferring a packet command on DEV,
- * expecting to later transfer XFERLEN bytes.  HANDLER is the routine
- * which actually transfers the command to the drive.  If this is a
- * drq_interrupt device, this routine will arrange for HANDLER to be
- * called when the interrupt from the drive arrives.  Otherwise, HANDLER
- * will be called immediately after the drive is prepared for the transfer.
- */
-static ide_startstop_t cdrom_start_packet_command(ide_drive_t *drive)
-{
-	ide_hwif_t *hwif = drive->hwif;
-	struct request *rq = hwif->rq;
-	int xferlen;
-
-	xferlen = ide_cd_get_xferlen(rq);
-
-	ide_debug_log(IDE_DBG_PC, "Call %s, xferlen: %d\n", __func__, xferlen);
-
-	/* FIXME: for Virtual DMA we must check harder */
-	if (drive->dma)
-		drive->dma = !hwif->dma_ops->dma_setup(drive);
-
-	/* set up the controller registers */
-	ide_pktcmd_tf_load(drive, IDE_TFLAG_OUT_NSECT | IDE_TFLAG_OUT_LBAL,
-			   xferlen, drive->dma);
-
-	if (drive->atapi_flags & IDE_AFLAG_DRQ_INTERRUPT) {
-		/* waiting for CDB interrupt, not DMA yet. */
-		if (drive->dma)
-			drive->waiting_for_dma = 0;
-
-		/* packet command */
-		ide_execute_command(drive, ATA_CMD_PACKET,
-				    cdrom_transfer_packet_command,
-				    ATAPI_WAIT_PC, ide_cd_expiry);
-		return ide_started;
-	} else {
-		ide_execute_pkt_cmd(drive);
-
-		return cdrom_transfer_packet_command(drive);
-	}
-}
-
-/*
- * Send a packet command to DRIVE described by CMD_BUF and CMD_LEN. The device
- * registers must have already been prepared by cdrom_start_packet_command.
- * HANDLER is the interrupt handler to call when the command completes or
- * there's data ready.
- */
-#define ATAPI_MIN_CDB_BYTES 12
-static ide_startstop_t cdrom_transfer_packet_command(ide_drive_t *drive)
-{
-	ide_hwif_t *hwif = drive->hwif;
-	struct request *rq = hwif->rq;
-	int cmd_len;
-	ide_startstop_t startstop;
-
-	ide_debug_log(IDE_DBG_PC, "Call %s\n", __func__);
-
-	/* we must wait for DRQ to get set */
-	if (ide_wait_stat(&startstop, drive, ATA_DRQ, ATA_BUSY, WAIT_READY)) {
-		printk(KERN_ERR "%s: timeout while waiting for DRQ to assert\n",
-				drive->name);
-		return startstop;
-	}
-
-	if (drive->atapi_flags & IDE_AFLAG_DRQ_INTERRUPT) {
-		/* ok, next interrupt will be DMA interrupt */
-		if (drive->dma)
-			drive->waiting_for_dma = 1;
-	}
-
-	/* arm the interrupt handler */
-	ide_set_handler(drive, cdrom_newpc_intr, rq->timeout, ide_cd_expiry);
-
-	/* ATAPI commands get padded out to 12 bytes minimum */
-	cmd_len = COMMAND_SIZE(rq->cmd[0]);
-	if (cmd_len < ATAPI_MIN_CDB_BYTES)
-		cmd_len = ATAPI_MIN_CDB_BYTES;
-
-	/* start the DMA if need be */
-	if (drive->dma)
-		hwif->dma_ops->dma_start(drive);
-
-	/* send the command to the device */
-	hwif->tp_ops->output_data(drive, NULL, rq->cmd, cmd_len);
-
-	return ide_started;
-}
-
 /*
  * Check the contents of the interrupt reason register from the cdrom
  * and attempt to recover if there are problems.  Returns  0 if everything's
@@ -1174,7 +1081,7 @@ static ide_startstop_t ide_cd_do_request(ide_drive_t *drive, struct request *rq,
 		return ide_stopped;
 	}
 
-	return cdrom_start_packet_command(drive);
+	return ide_issue_pc(drive);
 }
 
 /*
@@ -2072,6 +1979,7 @@ static int ide_cd_probe(ide_drive_t *drive)
 	}
 
 	drive->debug_mask = debug_mask;
+	drive->irq_handler = cdrom_newpc_intr;
 
 	info = kzalloc(sizeof(struct cdrom_info), GFP_KERNEL);
 	if (info == NULL) {
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 4cecd923fc79..13deba5e0157 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -654,6 +654,8 @@ struct ide_drive_s {
 	int  (*pc_io_buffers)(struct ide_drive_s *, struct ide_atapi_pc *,
 			      unsigned int, int);
 
+	ide_startstop_t (*irq_handler)(struct ide_drive_s *);
+
 	unsigned long atapi_flags;
 
 	struct ide_atapi_pc request_sense_pc;
-- 
cgit v1.2.3


From 906ef986a71d541a726550fa40dcbc5c356f810e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:59 +0100
Subject: ide: struct ide_atapi_pc - remove unused fields and update
 documentation

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/linux/ide.h | 4 ----
 1 file changed, 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ide.h b/include/linux/ide.h
index 13deba5e0157..a7dbfd857115 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -426,13 +426,9 @@ struct ide_atapi_pc {
 	struct idetape_bh *bh;
 	char *b_data;
 
-	/* idescsi only for now */
 	struct scatterlist *sg;
 	unsigned int sg_cnt;
 
-	struct scsi_cmnd *scsi_cmd;
-	void (*done) (struct scsi_cmnd *);
-
 	unsigned long timeout;
 };
 
-- 
cgit v1.2.3


From 94c96445f32c16cfdc398b20b7e78945ab7e35f9 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 6 Jan 2009 17:20:59 +0100
Subject: ide: remove unused ide_hwif_t.sg_mapped field

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/linux/ide.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/ide.h b/include/linux/ide.h
index a7dbfd857115..ebc22a836520 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -817,7 +817,6 @@ typedef struct hwif_s {
 	unsigned	extra_ports;	/* number of extra dma ports */
 
 	unsigned	present    : 1;	/* this interface exists */
-	unsigned	sg_mapped  : 1;	/* sg_table and sg_nents are ready */
 	unsigned	busy	   : 1; /* serializes devices on a port */
 
 	struct device		gendev;
-- 
cgit v1.2.3


From 391ad1908a9c13d457ea12ce1508d6b8a7ba72ad Mon Sep 17 00:00:00 2001
From: Shane McDonald <mcdonald.shane@gmail.com>
Date: Tue, 6 Jan 2009 17:21:01 +0100
Subject: Resurrect IT8172 IDE controller driver

Support for the IT8172 IDE controller was removed from the kernel
sometime after 2.6.18.  Support for the only boards that used the IT8172
was removed from the kernel after 2.6.18, as they had never compiled
since 2.6.0.  However, there are a couple of platforms that use this
chip: the PMC-Sierra Xiao Hu thin-client computer, which is no longer
in production, and the Linksys NSS4000 Network Attached Storage box,
which is based on the Xiao Hu board.  I am attempting to add support
for the Xiao Hu to the kernel, and this IT8172 IDE controller is the
first bit of code in this effort.

This patch resurrects the IT8172 IDE controller code.  I began with
the 2.6.18 version of the it8172.c file, and have moved it forward so
that it works with the latest version of the kernel.  I have run this
driver on a PMC-Sierra Xiao Hu board with the 2.6.28 kernel, and
I have had no problems with it in my configuration.  The attached patch
applies cleanly against 2.6.28.

Signed-off-by: Shane McDonald <mcdonald.shane@gmail.com>
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: alan@lxorguk.ukuu.org.uk
[bart: s/HWIF(drive)/drive->hwif/]
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/Kconfig     |   7 ++
 drivers/ide/Makefile    |   1 +
 drivers/ide/it8172.c    | 166 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/pci_ids.h |   1 +
 4 files changed, 175 insertions(+)
 create mode 100644 drivers/ide/it8172.c

(limited to 'include/linux')

diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index 4ee85fcf9aaf..3f9503867e6b 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -511,6 +511,13 @@ config BLK_DEV_PIIX
 	  This allows the kernel to change PIO, DMA and UDMA speeds and to
 	  configure the chip to optimum performance.
 
+config BLK_DEV_IT8172
+	tristate "IT8172 IDE support"
+	select BLK_DEV_IDEDMA_PCI
+	help
+	  This driver adds support for the IDE controller on the
+	  IT8172 System Controller.
+
 config BLK_DEV_IT8213
 	tristate "IT8213 IDE support"
 	select BLK_DEV_IDEDMA_PCI
diff --git a/drivers/ide/Makefile b/drivers/ide/Makefile
index 410728992e6a..c2b9c93f0095 100644
--- a/drivers/ide/Makefile
+++ b/drivers/ide/Makefile
@@ -47,6 +47,7 @@ obj-$(CONFIG_BLK_DEV_SC1200)		+= sc1200.o
 obj-$(CONFIG_BLK_DEV_CY82C693)		+= cy82c693.o
 obj-$(CONFIG_BLK_DEV_DELKIN)		+= delkin_cb.o
 obj-$(CONFIG_BLK_DEV_HPT366)		+= hpt366.o
+obj-$(CONFIG_BLK_DEV_IT8172)		+= it8172.o
 obj-$(CONFIG_BLK_DEV_IT8213)		+= it8213.o
 obj-$(CONFIG_BLK_DEV_IT821X)		+= it821x.o
 obj-$(CONFIG_BLK_DEV_JMICRON)		+= jmicron.o
diff --git a/drivers/ide/it8172.c b/drivers/ide/it8172.c
new file mode 100644
index 000000000000..e021078cd06b
--- /dev/null
+++ b/drivers/ide/it8172.c
@@ -0,0 +1,166 @@
+/*
+ *
+ * BRIEF MODULE DESCRIPTION
+ *      IT8172 IDE controller support
+ *
+ * Copyright (C) 2000 MontaVista Software Inc.
+ * Copyright (C) 2008 Shane McDonald
+ *
+ *  This program is free software; you can redistribute  it and/or modify it
+ *  under  the terms of  the GNU General  Public License as published by the
+ *  Free Software Foundation;  either version 2 of the  License, or (at your
+ *  option) any later version.
+ *
+ *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
+ *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
+ *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
+ *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
+ *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
+ *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
+ *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
+ *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
+ *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *  You should have received a copy of the  GNU General Public License along
+ *  with this program; if not, write  to the Free Software Foundation, Inc.,
+ *  675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/ioport.h>
+#include <linux/pci.h>
+#include <linux/ide.h>
+#include <linux/init.h>
+
+#define DRV_NAME "IT8172"
+
+static void it8172_set_pio_mode(ide_drive_t *drive, const u8 pio)
+{
+	ide_hwif_t *hwif	= drive->hwif;
+	struct pci_dev *dev	= to_pci_dev(hwif->dev);
+	u16 drive_enables;
+	u32 drive_timing;
+
+	/*
+	 * The highest value of DIOR/DIOW pulse width and recovery time
+	 * that can be set in the IT8172 is 8 PCI clock cycles.  As a result,
+	 * it cannot be configured for PIO mode 0.  This table sets these
+	 * parameters to the maximum supported by the IT8172.
+	 */
+	static const u8 timings[] = { 0x3f, 0x3c, 0x1b, 0x12, 0x0a };
+
+	pci_read_config_word(dev, 0x40, &drive_enables);
+	pci_read_config_dword(dev, 0x44, &drive_timing);
+
+	/*
+	 * Enable port 0x44. The IT8172 spec is confused; it calls
+	 * this register the "Slave IDE Timing Register", but in fact,
+	 * it controls timing for both master and slave drives.
+	 */
+	drive_enables |= 0x4000;
+
+	drive_enables &= drive->dn ? 0xc006 : 0xc060;
+	if (drive->media == ide_disk)
+		/* enable prefetch */
+		drive_enables |= 0x0004 << (drive->dn * 4);
+	if (ata_id_has_iordy(drive->id))
+		/* enable IORDY sample-point */
+		drive_enables |= 0x0002 << (drive->dn * 4);
+
+	drive_timing &= drive->dn ? 0x00003f00 : 0x000fc000;
+	drive_timing |= timings[pio] << (drive->dn * 6 + 8);
+
+	pci_write_config_word(dev, 0x40, drive_enables);
+	pci_write_config_dword(dev, 0x44, drive_timing);
+}
+
+static void it8172_set_dma_mode(ide_drive_t *drive, const u8 speed)
+{
+	ide_hwif_t *hwif	= drive->hwif;
+	struct pci_dev *dev	= to_pci_dev(hwif->dev);
+	int a_speed		= 3 << (drive->dn * 4);
+	int u_flag		= 1 << drive->dn;
+	int u_speed		= 0;
+	u8 reg48, reg4a;
+
+	pci_read_config_byte(dev, 0x48, &reg48);
+	pci_read_config_byte(dev, 0x4a, &reg4a);
+
+	if (speed >= XFER_UDMA_0) {
+		u8 udma = speed - XFER_UDMA_0;
+		u_speed = udma << (drive->dn * 4);
+
+		pci_write_config_byte(dev, 0x48, reg48 | u_flag);
+		reg4a &= ~a_speed;
+		pci_write_config_byte(dev, 0x4a, reg4a | u_speed);
+	} else {
+		const u8 mwdma_to_pio[] = { 0, 3, 4 };
+		u8 pio;
+
+		pci_write_config_byte(dev, 0x48, reg48 & ~u_flag);
+		pci_write_config_byte(dev, 0x4a, reg4a & ~a_speed);
+
+		pio = mwdma_to_pio[speed - XFER_MW_DMA_0];
+
+		it8172_set_pio_mode(drive, pio);
+	}
+}
+
+
+static const struct ide_port_ops it8172_port_ops = {
+	.set_pio_mode	= it8172_set_pio_mode,
+	.set_dma_mode	= it8172_set_dma_mode,
+};
+
+static const struct ide_port_info it8172_port_info __devinitdata = {
+	.name		= DRV_NAME,
+	.port_ops	= &it8172_port_ops,
+	.enablebits	= { {0x41, 0x80, 0x80}, {0x00, 0x00, 0x00} },
+	.host_flags	= IDE_HFLAG_SINGLE,
+	.pio_mask	= ATA_PIO4 & ~ATA_PIO0,
+	.mwdma_mask	= ATA_MWDMA2,
+	.udma_mask	= ATA_UDMA2,
+};
+
+static int __devinit it8172_init_one(struct pci_dev *dev,
+					const struct pci_device_id *id)
+{
+	if ((dev->class >> 8) != PCI_CLASS_STORAGE_IDE)
+		return -ENODEV; /* IT8172 is more than an IDE controller */
+	return ide_pci_init_one(dev, &it8172_port_info, NULL);
+}
+
+static struct pci_device_id it8172_pci_tbl[] = {
+	{ PCI_VDEVICE(ITE, PCI_DEVICE_ID_ITE_8172), 0 },
+	{ 0, },
+};
+MODULE_DEVICE_TABLE(pci, it8172_pci_tbl);
+
+static struct pci_driver it8172_pci_driver = {
+	.name		= "IT8172_IDE",
+	.id_table	= it8172_pci_tbl,
+	.probe		= it8172_init_one,
+	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
+};
+
+static int __init it8172_ide_init(void)
+{
+	return ide_pci_register_driver(&it8172_pci_driver);
+}
+
+static void __exit it8172_ide_exit(void)
+{
+	pci_unregister_driver(&it8172_pci_driver);
+}
+
+module_init(it8172_ide_init);
+module_exit(it8172_ide_exit);
+
+MODULE_AUTHOR("Steve Longerbeam");
+MODULE_DESCRIPTION("PCI driver module for ITE 8172 IDE");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 218c73b1e6d4..d543365518ab 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1658,6 +1658,7 @@
 #define PCI_VENDOR_ID_ROCKWELL		0x127A
 
 #define PCI_VENDOR_ID_ITE		0x1283
+#define PCI_DEVICE_ID_ITE_8172		0x8172
 #define PCI_DEVICE_ID_ITE_8211		0x8211
 #define PCI_DEVICE_ID_ITE_8212		0x8212
 #define PCI_DEVICE_ID_ITE_8213		0x8213
-- 
cgit v1.2.3


From 592b5315219881c6c0af4785f96456ad2043193a Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Date: Tue, 6 Jan 2009 17:21:02 +0100
Subject: ide: move read_sff_dma_status() method to 'struct ide_dma_ops'

Move apparently misplaced read_sff_dma_status() method from 'struct ide_tp_ops'
to 'struct ide_dma_ops', renaming it to dma_sff_read_status() and making only
required for SFF-8038i compatible IDE controller drivers (greatly cutting down
the number of initializers) as its only user (outside ide-dma-sff.c and such
drivers) appears to be ide_pci_check_simplex() which is only called for such
controllers...

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/alim15x3.c     |  1 +
 drivers/ide/au1xxx-ide.c   |  1 -
 drivers/ide/cmd64x.c       |  3 +++
 drivers/ide/falconide.c    |  1 -
 drivers/ide/hpt366.c       |  3 +++
 drivers/ide/ide-dma-sff.c  | 22 +++++++++++++++++-----
 drivers/ide/ide-h8300.c    |  1 -
 drivers/ide/ide-iops.c     | 10 ----------
 drivers/ide/ide-probe.c    |  7 +++++--
 drivers/ide/it821x.c       |  1 +
 drivers/ide/ns87415.c      |  8 +++++---
 drivers/ide/pdc202xx_old.c |  2 ++
 drivers/ide/pmac.c         |  1 -
 drivers/ide/q40ide.c       |  1 -
 drivers/ide/sc1200.c       |  1 +
 drivers/ide/scc_pata.c     |  4 ++--
 drivers/ide/setup-pci.c    |  7 ++++---
 drivers/ide/sgiioc4.c      |  1 -
 drivers/ide/siimage.c      |  1 +
 drivers/ide/sl82c105.c     |  1 +
 drivers/ide/tc86c001.c     |  1 +
 drivers/ide/tx4939ide.c    | 21 ++++++++++++---------
 include/linux/ide.h        |  8 ++++++--
 23 files changed, 65 insertions(+), 42 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/alim15x3.c b/drivers/ide/alim15x3.c
index f1c817240501..66f43083408b 100644
--- a/drivers/ide/alim15x3.c
+++ b/drivers/ide/alim15x3.c
@@ -509,6 +509,7 @@ static const struct ide_dma_ops ali_dma_ops = {
 	.dma_test_irq		= ide_dma_test_irq,
 	.dma_lost_irq		= ide_dma_lost_irq,
 	.dma_timeout		= ide_dma_timeout,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 static const struct ide_port_info ali15x3_chipset __devinitdata = {
diff --git a/drivers/ide/au1xxx-ide.c b/drivers/ide/au1xxx-ide.c
index 11f2c8f3db48..79a2dfed8eb7 100644
--- a/drivers/ide/au1xxx-ide.c
+++ b/drivers/ide/au1xxx-ide.c
@@ -502,7 +502,6 @@ static const struct ide_tp_ops au1xxx_tp_ops = {
 	.exec_command		= ide_exec_command,
 	.read_status		= ide_read_status,
 	.read_altstatus		= ide_read_altstatus,
-	.read_sff_dma_status	= ide_read_sff_dma_status,
 
 	.set_irq		= ide_set_irq,
 
diff --git a/drivers/ide/cmd64x.c b/drivers/ide/cmd64x.c
index 265cf9268c63..2f9688d87ecd 100644
--- a/drivers/ide/cmd64x.c
+++ b/drivers/ide/cmd64x.c
@@ -385,6 +385,7 @@ static const struct ide_dma_ops cmd64x_dma_ops = {
 	.dma_test_irq		= cmd64x_dma_test_irq,
 	.dma_lost_irq		= ide_dma_lost_irq,
 	.dma_timeout		= ide_dma_timeout,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 static const struct ide_dma_ops cmd646_rev1_dma_ops = {
@@ -396,6 +397,7 @@ static const struct ide_dma_ops cmd646_rev1_dma_ops = {
 	.dma_test_irq		= ide_dma_test_irq,
 	.dma_lost_irq		= ide_dma_lost_irq,
 	.dma_timeout		= ide_dma_timeout,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 static const struct ide_dma_ops cmd648_dma_ops = {
@@ -407,6 +409,7 @@ static const struct ide_dma_ops cmd648_dma_ops = {
 	.dma_test_irq		= cmd648_dma_test_irq,
 	.dma_lost_irq		= ide_dma_lost_irq,
 	.dma_timeout		= ide_dma_timeout,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 static const struct ide_port_info cmd64x_chipsets[] __devinitdata = {
diff --git a/drivers/ide/falconide.c b/drivers/ide/falconide.c
index 39d500d84b07..a5ba820d69bb 100644
--- a/drivers/ide/falconide.c
+++ b/drivers/ide/falconide.c
@@ -70,7 +70,6 @@ static const struct ide_tp_ops falconide_tp_ops = {
 	.exec_command		= ide_exec_command,
 	.read_status		= ide_read_status,
 	.read_altstatus		= ide_read_altstatus,
-	.read_sff_dma_status	= ide_read_sff_dma_status,
 
 	.set_irq		= ide_set_irq,
 
diff --git a/drivers/ide/hpt366.c b/drivers/ide/hpt366.c
index 208614f4dc3c..3eb9b5c63a0f 100644
--- a/drivers/ide/hpt366.c
+++ b/drivers/ide/hpt366.c
@@ -1424,6 +1424,7 @@ static const struct ide_dma_ops hpt37x_dma_ops = {
 	.dma_test_irq		= hpt374_dma_test_irq,
 	.dma_lost_irq		= ide_dma_lost_irq,
 	.dma_timeout		= ide_dma_timeout,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 static const struct ide_dma_ops hpt370_dma_ops = {
@@ -1435,6 +1436,7 @@ static const struct ide_dma_ops hpt370_dma_ops = {
 	.dma_test_irq		= ide_dma_test_irq,
 	.dma_lost_irq		= ide_dma_lost_irq,
 	.dma_timeout		= hpt370_dma_timeout,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 static const struct ide_dma_ops hpt36x_dma_ops = {
@@ -1446,6 +1448,7 @@ static const struct ide_dma_ops hpt36x_dma_ops = {
 	.dma_test_irq		= ide_dma_test_irq,
 	.dma_lost_irq		= hpt366_dma_lost_irq,
 	.dma_timeout		= ide_dma_timeout,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 static const struct ide_port_info hpt366_chipsets[] __devinitdata = {
diff --git a/drivers/ide/ide-dma-sff.c b/drivers/ide/ide-dma-sff.c
index 623a82d1535d..bcdadc777564 100644
--- a/drivers/ide/ide-dma-sff.c
+++ b/drivers/ide/ide-dma-sff.c
@@ -50,6 +50,17 @@ int config_drive_for_dma(ide_drive_t *drive)
 	return 0;
 }
 
+u8 ide_dma_sff_read_status(ide_hwif_t *hwif)
+{
+	unsigned long addr = hwif->dma_base + ATA_DMA_STATUS;
+
+	if (hwif->host_flags & IDE_HFLAG_MMIO)
+		return readb((void __iomem *)addr);
+	else
+		return inb(addr);
+}
+EXPORT_SYMBOL_GPL(ide_dma_sff_read_status);
+
 /**
  *	ide_dma_host_set	-	Enable/disable DMA on a host
  *	@drive: drive to control
@@ -62,7 +73,7 @@ void ide_dma_host_set(ide_drive_t *drive, int on)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	u8 unit = drive->dn & 1;
-	u8 dma_stat = hwif->tp_ops->read_sff_dma_status(hwif);
+	u8 dma_stat = hwif->dma_ops->dma_sff_read_status(hwif);
 
 	if (on)
 		dma_stat |= (1 << (5 + unit));
@@ -200,7 +211,7 @@ int ide_dma_setup(ide_drive_t *drive)
 		outb(reading, hwif->dma_base + ATA_DMA_CMD);
 
 	/* read DMA status for INTR & ERROR flags */
-	dma_stat = hwif->tp_ops->read_sff_dma_status(hwif);
+	dma_stat = hwif->dma_ops->dma_sff_read_status(hwif);
 
 	/* clear INTR & ERROR flags */
 	if (mmio)
@@ -232,7 +243,7 @@ EXPORT_SYMBOL_GPL(ide_dma_setup);
 static int dma_timer_expiry(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	u8 dma_stat = hwif->tp_ops->read_sff_dma_status(hwif);
+	u8 dma_stat = hwif->dma_ops->dma_sff_read_status(hwif);
 
 	printk(KERN_WARNING "%s: %s: DMA status (0x%02x)\n",
 		drive->name, __func__, dma_stat);
@@ -305,7 +316,7 @@ int ide_dma_end(ide_drive_t *drive)
 	}
 
 	/* get DMA status */
-	dma_stat = hwif->tp_ops->read_sff_dma_status(hwif);
+	dma_stat = hwif->dma_ops->dma_sff_read_status(hwif);
 
 	if (mmio)
 		/* clear the INTR & ERROR bits */
@@ -331,7 +342,7 @@ EXPORT_SYMBOL_GPL(ide_dma_end);
 int ide_dma_test_irq(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	u8 dma_stat = hwif->tp_ops->read_sff_dma_status(hwif);
+	u8 dma_stat = hwif->dma_ops->dma_sff_read_status(hwif);
 
 	return (dma_stat & ATA_DMA_INTR) ? 1 : 0;
 }
@@ -346,5 +357,6 @@ const struct ide_dma_ops sff_dma_ops = {
 	.dma_test_irq		= ide_dma_test_irq,
 	.dma_timeout		= ide_dma_timeout,
 	.dma_lost_irq		= ide_dma_lost_irq,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 EXPORT_SYMBOL_GPL(sff_dma_ops);
diff --git a/drivers/ide/ide-h8300.c b/drivers/ide/ide-h8300.c
index e2cdd2e9cdec..9270d3255ee0 100644
--- a/drivers/ide/ide-h8300.c
+++ b/drivers/ide/ide-h8300.c
@@ -159,7 +159,6 @@ static const struct ide_tp_ops h8300_tp_ops = {
 	.exec_command		= ide_exec_command,
 	.read_status		= ide_read_status,
 	.read_altstatus		= ide_read_altstatus,
-	.read_sff_dma_status	= ide_read_sff_dma_status,
 
 	.set_irq		= ide_set_irq,
 
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 1a22c31ce7ac..e728cfe7273f 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -105,15 +105,6 @@ u8 ide_read_altstatus(ide_hwif_t *hwif)
 }
 EXPORT_SYMBOL_GPL(ide_read_altstatus);
 
-u8 ide_read_sff_dma_status(ide_hwif_t *hwif)
-{
-	if (hwif->host_flags & IDE_HFLAG_MMIO)
-		return readb((void __iomem *)(hwif->dma_base + ATA_DMA_STATUS));
-	else
-		return inb(hwif->dma_base + ATA_DMA_STATUS);
-}
-EXPORT_SYMBOL_GPL(ide_read_sff_dma_status);
-
 void ide_set_irq(ide_hwif_t *hwif, int on)
 {
 	u8 ctl = ATA_DEVCTL_OBS;
@@ -388,7 +379,6 @@ const struct ide_tp_ops default_tp_ops = {
 	.exec_command		= ide_exec_command,
 	.read_status		= ide_read_status,
 	.read_altstatus		= ide_read_altstatus,
-	.read_sff_dma_status	= ide_read_sff_dma_status,
 
 	.set_irq		= ide_set_irq,
 
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index ebb1b7f863f4..0ccbb4459fb9 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1229,6 +1229,8 @@ static void ide_init_port(ide_hwif_t *hwif, unsigned int port,
 	if ((d->host_flags & IDE_HFLAG_NO_DMA) == 0) {
 		int rc;
 
+		hwif->dma_ops = d->dma_ops;
+
 		if (d->init_dma)
 			rc = d->init_dma(hwif, d);
 		else
@@ -1236,12 +1238,13 @@ static void ide_init_port(ide_hwif_t *hwif, unsigned int port,
 
 		if (rc < 0) {
 			printk(KERN_INFO "%s: DMA disabled\n", hwif->name);
+
+			hwif->dma_ops = NULL;
 			hwif->dma_base = 0;
 			hwif->swdma_mask = 0;
 			hwif->mwdma_mask = 0;
 			hwif->ultra_mask = 0;
-		} else if (d->dma_ops)
-			hwif->dma_ops = d->dma_ops;
+		}
 	}
 
 	if ((d->host_flags & IDE_HFLAG_SERIALIZE) ||
diff --git a/drivers/ide/it821x.c b/drivers/ide/it821x.c
index 879056645352..0be27ac1f077 100644
--- a/drivers/ide/it821x.c
+++ b/drivers/ide/it821x.c
@@ -512,6 +512,7 @@ static struct ide_dma_ops it821x_pass_through_dma_ops = {
 	.dma_test_irq		= ide_dma_test_irq,
 	.dma_timeout		= ide_dma_timeout,
 	.dma_lost_irq		= ide_dma_lost_irq,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 /**
diff --git a/drivers/ide/ns87415.c b/drivers/ide/ns87415.c
index aceb2fcbe1d1..83643ed9a426 100644
--- a/drivers/ide/ns87415.c
+++ b/drivers/ide/ns87415.c
@@ -56,7 +56,7 @@ static u8 superio_read_status(ide_hwif_t *hwif)
 	return superio_ide_inb(hwif->io_ports.status_addr);
 }
 
-static u8 superio_read_sff_dma_status(ide_hwif_t *hwif)
+static u8 superio_dma_sff_read_status(ide_hwif_t *hwif)
 {
 	return superio_ide_inb(hwif->dma_base + ATA_DMA_STATUS);
 }
@@ -109,7 +109,6 @@ static const struct ide_tp_ops superio_tp_ops = {
 	.exec_command		= ide_exec_command,
 	.read_status		= superio_read_status,
 	.read_altstatus		= ide_read_altstatus,
-	.read_sff_dma_status	= superio_read_sff_dma_status,
 
 	.set_irq		= ide_set_irq,
 
@@ -132,6 +131,8 @@ static void __devinit superio_init_iops(struct hwif_s *hwif)
 	tmp = superio_ide_inb(dma_stat);
 	outb(tmp | 0x66, dma_stat);
 }
+#else
+#define superio_dma_sff_read_status ide_dma_sff_read_status
 #endif
 
 static unsigned int ns87415_count = 0, ns87415_control[MAX_HWIFS] = { 0 };
@@ -201,7 +202,7 @@ static int ns87415_dma_end(ide_drive_t *drive)
 	u8 dma_stat = 0, dma_cmd = 0;
 
 	drive->waiting_for_dma = 0;
-	dma_stat = hwif->tp_ops->read_sff_dma_status(hwif);
+	dma_stat = hwif->dma_ops->dma_sff_read_status(hwif);
 	/* get DMA command mode */
 	dma_cmd = inb(hwif->dma_base + ATA_DMA_CMD);
 	/* stop DMA */
@@ -308,6 +309,7 @@ static const struct ide_dma_ops ns87415_dma_ops = {
 	.dma_test_irq		= ide_dma_test_irq,
 	.dma_lost_irq		= ide_dma_lost_irq,
 	.dma_timeout		= ide_dma_timeout,
+	.dma_sff_read_status	= superio_dma_sff_read_status,
 };
 
 static const struct ide_port_info ns87415_chipset __devinitdata = {
diff --git a/drivers/ide/pdc202xx_old.c b/drivers/ide/pdc202xx_old.c
index e8e6b29d9e41..97193323aebf 100644
--- a/drivers/ide/pdc202xx_old.c
+++ b/drivers/ide/pdc202xx_old.c
@@ -337,6 +337,7 @@ static const struct ide_dma_ops pdc20246_dma_ops = {
 	.dma_test_irq		= pdc202xx_dma_test_irq,
 	.dma_lost_irq		= pdc202xx_dma_lost_irq,
 	.dma_timeout		= pdc202xx_dma_timeout,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 static const struct ide_dma_ops pdc2026x_dma_ops = {
@@ -348,6 +349,7 @@ static const struct ide_dma_ops pdc2026x_dma_ops = {
 	.dma_test_irq		= pdc202xx_dma_test_irq,
 	.dma_lost_irq		= pdc202xx_dma_lost_irq,
 	.dma_timeout		= pdc202xx_dma_timeout,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 #define DECLARE_PDC2026X_DEV(udma, sectors) \
diff --git a/drivers/ide/pmac.c b/drivers/ide/pmac.c
index ee52a21af1be..74625e821a43 100644
--- a/drivers/ide/pmac.c
+++ b/drivers/ide/pmac.c
@@ -955,7 +955,6 @@ static const struct ide_tp_ops pmac_tp_ops = {
 	.exec_command		= pmac_exec_command,
 	.read_status		= ide_read_status,
 	.read_altstatus		= ide_read_altstatus,
-	.read_sff_dma_status	= ide_read_sff_dma_status,
 
 	.set_irq		= pmac_set_irq,
 
diff --git a/drivers/ide/q40ide.c b/drivers/ide/q40ide.c
index 4af4a8ce4cdf..9f9c0b3cc3a3 100644
--- a/drivers/ide/q40ide.c
+++ b/drivers/ide/q40ide.c
@@ -99,7 +99,6 @@ static const struct ide_tp_ops q40ide_tp_ops = {
 	.exec_command		= ide_exec_command,
 	.read_status		= ide_read_status,
 	.read_altstatus		= ide_read_altstatus,
-	.read_sff_dma_status	= ide_read_sff_dma_status,
 
 	.set_irq		= ide_set_irq,
 
diff --git a/drivers/ide/sc1200.c b/drivers/ide/sc1200.c
index 1cf477aaae36..dbdd2985a0d8 100644
--- a/drivers/ide/sc1200.c
+++ b/drivers/ide/sc1200.c
@@ -292,6 +292,7 @@ static const struct ide_dma_ops sc1200_dma_ops = {
 	.dma_test_irq		= ide_dma_test_irq,
 	.dma_lost_irq		= ide_dma_lost_irq,
 	.dma_timeout		= ide_dma_timeout,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 static const struct ide_port_info sc1200_chipset __devinitdata = {
diff --git a/drivers/ide/scc_pata.c b/drivers/ide/scc_pata.c
index 5d53850c79d7..1cb43068455a 100644
--- a/drivers/ide/scc_pata.c
+++ b/drivers/ide/scc_pata.c
@@ -143,7 +143,7 @@ static u8 scc_read_altstatus(ide_hwif_t *hwif)
 	return (u8)in_be32((void *)hwif->io_ports.ctl_addr);
 }
 
-static u8 scc_read_sff_dma_status(ide_hwif_t *hwif)
+static u8 scc_dma_sff_read_status(ide_hwif_t *hwif)
 {
 	return (u8)in_be32((void *)(hwif->dma_base + 4));
 }
@@ -853,7 +853,6 @@ static const struct ide_tp_ops scc_tp_ops = {
 	.exec_command		= scc_exec_command,
 	.read_status		= scc_read_status,
 	.read_altstatus		= scc_read_altstatus,
-	.read_sff_dma_status	= scc_read_sff_dma_status,
 
 	.set_irq		= scc_set_irq,
 
@@ -880,6 +879,7 @@ static const struct ide_dma_ops scc_dma_ops = {
 	.dma_test_irq		= scc_dma_test_irq,
 	.dma_lost_irq		= ide_dma_lost_irq,
 	.dma_timeout		= ide_dma_timeout,
+	.dma_sff_read_status	= scc_dma_sff_read_status,
 };
 
 #define DECLARE_SCC_DEV(name_str)			\
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index bc37dff8c675..e85d1ed29c2a 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -130,7 +130,7 @@ int ide_pci_check_simplex(ide_hwif_t *hwif, const struct ide_port_info *d)
 	 * we tune the drive then try to grab DMA ownership if we want to be
 	 * the DMA end.  This has to be become dynamic to handle hot-plug.
 	 */
-	dma_stat = hwif->tp_ops->read_sff_dma_status(hwif);
+	dma_stat = hwif->dma_ops->dma_sff_read_status(hwif);
 	if ((dma_stat & 0x80) && hwif->mate && hwif->mate->dma_base) {
 		printk(KERN_INFO "%s %s: simplex device: DMA disabled\n",
 			d->name, pci_name(dev));
@@ -377,6 +377,9 @@ int ide_hwif_setup_dma(ide_hwif_t *hwif, const struct ide_port_info *d)
 
 		hwif->dma_base = base;
 
+		if (hwif->dma_ops == NULL)
+			hwif->dma_ops = &sff_dma_ops;
+
 		if (ide_pci_check_simplex(hwif, d) < 0)
 			return -1;
 
@@ -393,8 +396,6 @@ int ide_hwif_setup_dma(ide_hwif_t *hwif, const struct ide_port_info *d)
 
 		if (ide_allocate_dma_engine(hwif))
 			return -1;
-
-		hwif->dma_ops = &sff_dma_ops;
 	}
 
 	return 0;
diff --git a/drivers/ide/sgiioc4.c b/drivers/ide/sgiioc4.c
index 8e1ffd57a86d..fdb9d7037694 100644
--- a/drivers/ide/sgiioc4.c
+++ b/drivers/ide/sgiioc4.c
@@ -523,7 +523,6 @@ static const struct ide_tp_ops sgiioc4_tp_ops = {
 	.exec_command		= ide_exec_command,
 	.read_status		= sgiioc4_read_status,
 	.read_altstatus		= ide_read_altstatus,
-	.read_sff_dma_status	= ide_read_sff_dma_status,
 
 	.set_irq		= ide_set_irq,
 
diff --git a/drivers/ide/siimage.c b/drivers/ide/siimage.c
index 652b3a04620f..cb2b352b876b 100644
--- a/drivers/ide/siimage.c
+++ b/drivers/ide/siimage.c
@@ -717,6 +717,7 @@ static const struct ide_dma_ops sil_dma_ops = {
 	.dma_test_irq		= siimage_dma_test_irq,
 	.dma_timeout		= ide_dma_timeout,
 	.dma_lost_irq		= ide_dma_lost_irq,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 #define DECLARE_SII_DEV(p_ops)				\
diff --git a/drivers/ide/sl82c105.c b/drivers/ide/sl82c105.c
index 1ded01d81ab3..48cc748c5043 100644
--- a/drivers/ide/sl82c105.c
+++ b/drivers/ide/sl82c105.c
@@ -299,6 +299,7 @@ static const struct ide_dma_ops sl82c105_dma_ops = {
 	.dma_test_irq		= ide_dma_test_irq,
 	.dma_lost_irq		= sl82c105_dma_lost_irq,
 	.dma_timeout		= sl82c105_dma_timeout,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 static const struct ide_port_info sl82c105_chipset __devinitdata = {
diff --git a/drivers/ide/tc86c001.c b/drivers/ide/tc86c001.c
index d2c00fb928e4..84109f5a1632 100644
--- a/drivers/ide/tc86c001.c
+++ b/drivers/ide/tc86c001.c
@@ -188,6 +188,7 @@ static const struct ide_dma_ops tc86c001_dma_ops = {
 	.dma_test_irq		= ide_dma_test_irq,
 	.dma_lost_irq		= ide_dma_lost_irq,
 	.dma_timeout		= ide_dma_timeout,
+	.dma_sff_read_status	= ide_dma_sff_read_status,
 };
 
 static const struct ide_port_info tc86c001_chipset __devinitdata = {
diff --git a/drivers/ide/tx4939ide.c b/drivers/ide/tx4939ide.c
index 1ac27ac7283b..882f6f07c476 100644
--- a/drivers/ide/tx4939ide.c
+++ b/drivers/ide/tx4939ide.c
@@ -397,6 +397,17 @@ static int tx4939ide_dma_test_irq(ide_drive_t *drive)
 	return found;
 }
 
+#ifdef __BIG_ENDIAN
+static u8 tx4939ide_dma_sff_read_status(ide_hwif_t *hwif)
+{
+	void __iomem *base = TX4939IDE_BASE(hwif);
+
+	return tx4939ide_readb(base, TX4939IDE_DMA_Stat);
+}
+#else
+#define tx4939ide_dma_sff_read_status ide_dma_sff_read_status
+#endif
+
 static void tx4939ide_init_hwif(ide_hwif_t *hwif)
 {
 	void __iomem *base = TX4939IDE_BASE(hwif);
@@ -443,13 +454,6 @@ static void tx4939ide_tf_load_fixup(ide_drive_t *drive, ide_task_t *task)
 
 #ifdef __BIG_ENDIAN
 
-static u8 tx4939ide_read_sff_dma_status(ide_hwif_t *hwif)
-{
-	void __iomem *base = TX4939IDE_BASE(hwif);
-
-	return tx4939ide_readb(base, TX4939IDE_DMA_Stat);
-}
-
 /* custom iops (independent from SWAP_IO_SPACE) */
 static u8 tx4939ide_inb(unsigned long port)
 {
@@ -585,7 +589,6 @@ static const struct ide_tp_ops tx4939ide_tp_ops = {
 	.exec_command		= ide_exec_command,
 	.read_status		= ide_read_status,
 	.read_altstatus		= ide_read_altstatus,
-	.read_sff_dma_status	= tx4939ide_read_sff_dma_status,
 
 	.set_irq		= ide_set_irq,
 
@@ -609,7 +612,6 @@ static const struct ide_tp_ops tx4939ide_tp_ops = {
 	.exec_command		= ide_exec_command,
 	.read_status		= ide_read_status,
 	.read_altstatus		= ide_read_altstatus,
-	.read_sff_dma_status	= ide_read_sff_dma_status,
 
 	.set_irq		= ide_set_irq,
 
@@ -638,6 +640,7 @@ static const struct ide_dma_ops tx4939ide_dma_ops = {
 	.dma_test_irq		= tx4939ide_dma_test_irq,
 	.dma_lost_irq		= ide_dma_lost_irq,
 	.dma_timeout		= ide_dma_timeout,
+	.dma_sff_read_status	= tx4939ide_dma_sff_read_status,
 };
 
 static const struct ide_port_info tx4939ide_port_info __initdata = {
diff --git a/include/linux/ide.h b/include/linux/ide.h
index ebc22a836520..3644f6323384 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -674,7 +674,6 @@ struct ide_tp_ops {
 	void	(*exec_command)(struct hwif_s *, u8);
 	u8	(*read_status)(struct hwif_s *);
 	u8	(*read_altstatus)(struct hwif_s *);
-	u8	(*read_sff_dma_status)(struct hwif_s *);
 
 	void	(*set_irq)(struct hwif_s *, int);
 
@@ -735,6 +734,11 @@ struct ide_dma_ops {
 	int	(*dma_test_irq)(struct ide_drive_s *);
 	void	(*dma_lost_irq)(struct ide_drive_s *);
 	void	(*dma_timeout)(struct ide_drive_s *);
+	/*
+	 * The following method is optional and only required to be
+	 * implemented for the SFF-8038i compatible controllers.
+	 */
+	u8	(*dma_sff_read_status)(struct hwif_s *);
 };
 
 struct ide_host;
@@ -1177,7 +1181,6 @@ void ide_tf_dump(const char *, struct ide_taskfile *);
 void ide_exec_command(ide_hwif_t *, u8);
 u8 ide_read_status(ide_hwif_t *);
 u8 ide_read_altstatus(ide_hwif_t *);
-u8 ide_read_sff_dma_status(ide_hwif_t *);
 
 void ide_set_irq(ide_hwif_t *, int);
 
@@ -1458,6 +1461,7 @@ void ide_dma_exec_cmd(ide_drive_t *, u8);
 extern void ide_dma_start(ide_drive_t *);
 int ide_dma_end(ide_drive_t *);
 int ide_dma_test_irq(ide_drive_t *);
+u8 ide_dma_sff_read_status(ide_hwif_t *);
 extern const struct ide_dma_ops sff_dma_ops;
 #else
 static inline int config_drive_for_dma(ide_drive_t *drive) { return 0; }
-- 
cgit v1.2.3


From 548eaca46b3cf4419b6c2be839a106d8641ffb70 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Mon, 20 Oct 2008 17:48:43 -0400
Subject: nfsd: document new filehandle fsid types

Descriptions taken from mountd code (in nfs-utils/utils/mountd/cache.c).

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 include/linux/nfsd/nfsfh.h | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/nfsd/nfsfh.h b/include/linux/nfsd/nfsfh.h
index d1941cb965e9..b2e093870bc6 100644
--- a/include/linux/nfsd/nfsfh.h
+++ b/include/linux/nfsd/nfsfh.h
@@ -68,6 +68,10 @@ struct nfs_fhbase_old {
  *     1  - 4 byte user specified identifier
  *     2  - 4 byte major, 4 byte minor, 4 byte inode number - DEPRECATED
  *     3  - 4 byte device id, encoded for user-space, 4 byte inode number
+ *     4  - 4 byte inode number and 4 byte uuid
+ *     5  - 8 byte uuid
+ *     6  - 16 byte uuid
+ *     7  - 8 byte inode number and 16 byte uuid
  *
  * The fileid_type identified how the file within the filesystem is encoded.
  * This is (will be) passed to, and set by, the underlying filesystem if it supports
-- 
cgit v1.2.3


From c9233eb7b0b11ef176d4bf68da2ce85464b6ec39 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Mon, 20 Oct 2008 11:51:57 -0400
Subject: sunrpc: add sv_maxconn field to svc_serv (try #3)

svc_check_conn_limits() attempts to prevent denial of service attacks
by having the service close old connections once it reaches a
threshold. This threshold is based on the number of threads in the
service:

	(serv->sv_nrthreads + 3) * 20

Once we reach this, we drop the oldest connections and a printk pops
to warn the admin that they should increase the number of threads.

Increasing the number of threads isn't an option however for services
like lockd. We don't want to eliminate this check entirely for such
services but we need some way to increase this limit.

This patch adds a sv_maxconn field to the svc_serv struct. When it's
set to 0, we use the current method to calculate the max number of
connections. RPC services can then set this on an as-needed basis.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Acked-by: Neil Brown <neilb@suse.de>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 include/linux/sunrpc/svc.h |  5 ++++-
 net/sunrpc/svc_xprt.c      | 22 ++++++++++++++++------
 2 files changed, 20 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index 3afe7fb403b2..3435d24bfe55 100644
--- a/include/linux/sunrpc/svc.h
+++ b/include/linux/sunrpc/svc.h
@@ -58,10 +58,13 @@ struct svc_serv {
 	struct svc_stat *	sv_stats;	/* RPC statistics */
 	spinlock_t		sv_lock;
 	unsigned int		sv_nrthreads;	/* # of server threads */
+	unsigned int		sv_maxconn;	/* max connections allowed or
+						 * '0' causing max to be based
+						 * on number of threads. */
+
 	unsigned int		sv_max_payload;	/* datagram payload size */
 	unsigned int		sv_max_mesg;	/* max_payload + 1 page for overheads */
 	unsigned int		sv_xdrsize;	/* XDR buffer size */
-
 	struct list_head	sv_permsocks;	/* all permanent sockets */
 	struct list_head	sv_tempsocks;	/* all temporary sockets */
 	int			sv_tmpcnt;	/* count of temporary sockets */
diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c
index bf5b5cdafebf..3fe4f1004278 100644
--- a/net/sunrpc/svc_xprt.c
+++ b/net/sunrpc/svc_xprt.c
@@ -515,8 +515,10 @@ int svc_port_is_privileged(struct sockaddr *sin)
 }
 
 /*
- * Make sure that we don't have too many active connections.  If we
- * have, something must be dropped.
+ * Make sure that we don't have too many active connections. If we have,
+ * something must be dropped. It's not clear what will happen if we allow
+ * "too many" connections, but when dealing with network-facing software,
+ * we have to code defensively. Here we do that by imposing hard limits.
  *
  * There's no point in trying to do random drop here for DoS
  * prevention. The NFS clients does 1 reconnect in 15 seconds. An
@@ -525,19 +527,27 @@ int svc_port_is_privileged(struct sockaddr *sin)
  * The only somewhat efficient mechanism would be if drop old
  * connections from the same IP first. But right now we don't even
  * record the client IP in svc_sock.
+ *
+ * single-threaded services that expect a lot of clients will probably
+ * need to set sv_maxconn to override the default value which is based
+ * on the number of threads
  */
 static void svc_check_conn_limits(struct svc_serv *serv)
 {
-	if (serv->sv_tmpcnt > (serv->sv_nrthreads+3)*20) {
+	unsigned int limit = serv->sv_maxconn ? serv->sv_maxconn :
+				(serv->sv_nrthreads+3) * 20;
+
+	if (serv->sv_tmpcnt > limit) {
 		struct svc_xprt *xprt = NULL;
 		spin_lock_bh(&serv->sv_lock);
 		if (!list_empty(&serv->sv_tempsocks)) {
 			if (net_ratelimit()) {
 				/* Try to help the admin */
 				printk(KERN_NOTICE "%s: too many open  "
-				       "connections, consider increasing the "
-				       "number of nfsd threads\n",
-				       serv->sv_name);
+				       "connections, consider increasing %s\n",
+				       serv->sv_name, serv->sv_maxconn ?
+				       "the max number of connections." :
+				       "the number of threads.");
 			}
 			/*
 			 * Always select the oldest connection. It's not fair,
-- 
cgit v1.2.3


From 7538ce1eb656a1477bedd5b1c202226e7abf5e7b Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 4 Dec 2008 14:19:45 -0500
Subject: NLM: Use modern style for pointer fields in nlm_host

Clean up: I'm about to add another "char *" field to the nlm_host
structure.  The h_name field, for example, uses an older style of
declaring a "char *" field.  If I match that style for the new field,
checkpatch.pl will complain.

So, fix pointer fields to use the new style.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 include/linux/lockd/lockd.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 23da3fa69efa..3dbdd353156c 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -43,8 +43,8 @@ struct nlm_host {
 	struct sockaddr_storage	h_addr;		/* peer address */
 	size_t			h_addrlen;
 	struct sockaddr_storage	h_srcaddr;	/* our address (optional) */
-	struct rpc_clnt	*	h_rpcclnt;	/* RPC client to talk to peer */
-	char *			h_name;		/* remote hostname */
+	struct rpc_clnt		*h_rpcclnt;	/* RPC client to talk to peer */
+	char			*h_name;		/* remote hostname */
 	u32			h_version;	/* interface version */
 	unsigned short		h_proto;	/* transport proto */
 	unsigned short		h_reclaiming : 1,
@@ -64,7 +64,7 @@ struct nlm_host {
 	spinlock_t		h_lock;
 	struct list_head	h_granted;	/* Locks in GRANTED state */
 	struct list_head	h_reclaim;	/* Locks in RECLAIM state */
-	struct nsm_handle *	h_nsmhandle;	/* NSM status handle */
+	struct nsm_handle	*h_nsmhandle;	/* NSM status handle */
 
 	char			h_addrbuf[48],	/* address eyecatchers */
 				h_srcaddrbuf[48];
-- 
cgit v1.2.3


From 1df40b609ad5a622904eb652109c287fe9c93ec5 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 4 Dec 2008 14:19:53 -0500
Subject: NLM: Remove address eye-catcher buffers from nlm_host

The h_name field in struct nlm_host is a just copy of
h_nsmhandle->sm_name.  Likewise, the contents of the h_addrbuf field
should be identical to the sm_addrbuf field.

The h_srcaddrbuf field is used only in one place for debugging.  We can
live without this until we get %pI formatting for printk().

Currently these buffers are 48 bytes, but we need to support scope IDs
in IPv6 presentation addresses, which means making the buffers even
larger.  Instead, let's find ways to eliminate them to save space.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/host.c             | 10 +++-------
 include/linux/lockd/lockd.h |  4 +---
 2 files changed, 4 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index abdebf76b820..33bf67af7aba 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -206,6 +206,7 @@ static struct nlm_host *nlm_lookup_host(struct nlm_lookup_host_info *ni)
 		goto out;
 	}
 	host->h_name	   = nsm->sm_name;
+	host->h_addrbuf    = nsm->sm_addrbuf;
 	memcpy(nlm_addr(host), ni->sap, ni->salen);
 	host->h_addrlen = ni->salen;
 	nlm_clear_port(nlm_addr(host));
@@ -232,11 +233,6 @@ static struct nlm_host *nlm_lookup_host(struct nlm_lookup_host_info *ni)
 
 	nrhosts++;
 
-	nlm_display_address((struct sockaddr *)&host->h_addr,
-				host->h_addrbuf, sizeof(host->h_addrbuf));
-	nlm_display_address((struct sockaddr *)&host->h_srcaddr,
-				host->h_srcaddrbuf, sizeof(host->h_srcaddrbuf));
-
 	dprintk("lockd: nlm_lookup_host created host %s\n",
 			host->h_name);
 
@@ -378,8 +374,8 @@ nlm_bind_host(struct nlm_host *host)
 {
 	struct rpc_clnt	*clnt;
 
-	dprintk("lockd: nlm_bind_host %s (%s), my addr=%s\n",
-			host->h_name, host->h_addrbuf, host->h_srcaddrbuf);
+	dprintk("lockd: nlm_bind_host %s (%s)\n",
+			host->h_name, host->h_addrbuf);
 
 	/* Lock host handle */
 	mutex_lock(&host->h_mutex);
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 3dbdd353156c..dae22cb4c38d 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -65,9 +65,7 @@ struct nlm_host {
 	struct list_head	h_granted;	/* Locks in GRANTED state */
 	struct list_head	h_reclaim;	/* Locks in RECLAIM state */
 	struct nsm_handle	*h_nsmhandle;	/* NSM status handle */
-
-	char			h_addrbuf[48],	/* address eyecatchers */
-				h_srcaddrbuf[48];
+	char			*h_addrbuf;	/* address eyecatcher */
 };
 
 struct nsm_handle {
-- 
cgit v1.2.3


From bc995801a09d1fead0bec1356bfd836911c8eed7 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 4 Dec 2008 14:20:08 -0500
Subject: NLM: Support IPv6 scope IDs in nlm_display_address()

Scope ID support is needed since the kernel's NSM implementation is
about to use these displayed addresses as a mon_name in some cases.

When nsm_use_hostnames is zero, without scope ID support NSM will fail
to handle peers that contact us via a link-local address.  Link-local
addresses do not work without an interface ID, which is stored in the
sockaddr's sin6_scope_id field.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/host.c             | 21 +++++++++++++++------
 include/linux/lockd/lockd.h | 10 +++++++++-
 2 files changed, 24 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index beb5da810167..012e49aaecd1 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -105,22 +105,31 @@ static void nlm_clear_port(struct sockaddr *sap)
 	}
 }
 
+static void nlm_display_ipv6_address(const struct sockaddr *sap, char *buf,
+				     const size_t len)
+{
+	const struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap;
+
+	if (ipv6_addr_v4mapped(&sin6->sin6_addr))
+		snprintf(buf, len, "%pI4", &sin6->sin6_addr.s6_addr32[3]);
+	else if (sin6->sin6_scope_id != 0)
+		snprintf(buf, len, "%pI6%%%u", &sin6->sin6_addr,
+				sin6->sin6_scope_id);
+	else
+		snprintf(buf, len, "%pI6", &sin6->sin6_addr);
+}
+
 static void nlm_display_address(const struct sockaddr *sap,
 				char *buf, const size_t len)
 {
 	const struct sockaddr_in *sin = (struct sockaddr_in *)sap;
-	const struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap;
 
 	switch (sap->sa_family) {
 	case AF_INET:
 		snprintf(buf, len, "%pI4", &sin->sin_addr.s_addr);
 		break;
 	case AF_INET6:
-		if (ipv6_addr_v4mapped(&sin6->sin6_addr))
-			snprintf(buf, len, "%pI4",
-				 &sin6->sin6_addr.s6_addr32[3]);
-		else
-			snprintf(buf, len, "%pI6", &sin6->sin6_addr);
+		nlm_display_ipv6_address(sap, buf, len);
 		break;
 	default:
 		snprintf(buf, len, "unsupported address family");
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index dae22cb4c38d..80a0a2cff2b8 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -68,6 +68,14 @@ struct nlm_host {
 	char			*h_addrbuf;	/* address eyecatcher */
 };
 
+/*
+ * The largest string sm_addrbuf should hold is a full-size IPv6 address
+ * (no "::" anywhere) with a scope ID.  The buffer size is computed to
+ * hold eight groups of colon-separated four-hex-digit numbers, a
+ * percent sign, a scope id (at most 32 bits, in decimal), and NUL.
+ */
+#define NSM_ADDRBUF		((8 * 4 + 7) + (1 + 10) + 1)
+
 struct nsm_handle {
 	struct list_head	sm_link;
 	atomic_t		sm_count;
@@ -76,7 +84,7 @@ struct nsm_handle {
 	size_t			sm_addrlen;
 	unsigned int		sm_monitored : 1,
 				sm_sticky : 1;	/* don't unmonitor */
-	char			sm_addrbuf[48];	/* address eyecatcher */
+	char			sm_addrbuf[NSM_ADDRBUF];
 };
 
 /*
-- 
cgit v1.2.3


From f47534f7f0ac7727e05ec4274b764b181df2cf7f Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 4 Dec 2008 14:20:38 -0500
Subject: NSM: Use modern style for sm_name field in nsm_handle

Clean up: I'm about to add another "char *" field to the nsm_handle
structure.  The sm_name field uses an older style of declaring a
"char *" field.  If I match that style for the new field, checkpatch.pl
will complain.

So, fix the sm_name field to use the new style.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 include/linux/lockd/lockd.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 80a0a2cff2b8..54dbb458e73c 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -79,7 +79,7 @@ struct nlm_host {
 struct nsm_handle {
 	struct list_head	sm_link;
 	atomic_t		sm_count;
-	char *			sm_name;
+	char			*sm_name;
 	struct sockaddr_storage	sm_addr;
 	size_t			sm_addrlen;
 	unsigned int		sm_monitored : 1,
-- 
cgit v1.2.3


From 29ed1407ed81086b778ebf12145b048ac3f7e10e Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 4 Dec 2008 14:20:46 -0500
Subject: NSM: Support IPv6 version of mon_name

The "mon_name" argument of the NSMPROC_MON and NSMPROC_UNMON upcalls
is a string that contains the hostname or IP address of the remote peer
to be notified when this host has rebooted.  The sm-notify command uses
this identifier to contact the peer when we reboot, so it must be
either a well-qualified DNS hostname or a presentation format IP
address string.

When the "nsm_use_hostnames" sysctl is set to zero, the kernel's NSM
provides a presentation format IP address in the "mon_name" argument.
Otherwise, the "caller_name" argument from NLM requests is used,
which is usually just the DNS hostname of the peer.

To support IPv6 addresses for the mon_name argument, we use the
nsm_handle's address eye-catcher, which already contains an appropriate
presentation format address string.  Using the eye-catcher string
obviates the need to use a large buffer on the stack to form the
presentation address string for the upcall.

This patch also addresses a subtle bug.

An NSMPROC_MON request and the subsequent NSMPROC_UNMON request for the
same peer are required to use the same value for the "mon_name"
argument.  Otherwise, rpc.statd's NSMPROC_UNMON processing cannot
locate the database entry for that peer and remove it.

If the setting of nsm_use_hostnames is changed between the time the
kernel sends an NSMPROC_MON request and the time it sends the
NSMPROC_UNMON request for the same peer, the "mon_name" argument for
these two requests may not be the same.  This is because the value of
"mon_name" is currently chosen at the moment the call is made based on
the setting of nsm_use_hostnames

To ensure both requests pass identical contents in the "mon_name"
argument, we now select which string to use for the argument in the
nsm_monitor() function.  A pointer to this string is saved in the
nsm_handle so it can be used for a subsequent NSMPROC_UNMON upcall.

NB: There are other potential problems, such as how nlm_host_rebooted()
might behave if nsm_use_hostnames were changed while hosts are still
being monitored.  This patch does not attempt to address those
problems.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/mon.c              | 27 ++++++++-------------------
 include/linux/lockd/lockd.h |  1 +
 2 files changed, 9 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index 497dfea02e8a..a606fbbf804d 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -18,8 +18,6 @@
 
 #define NLMDBG_FACILITY		NLMDBG_MONITOR
 
-#define XDR_ADDRBUF_LEN		(20)
-
 static struct rpc_clnt *	nsm_create(void);
 
 static struct rpc_program	nsm_program;
@@ -42,7 +40,7 @@ nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res)
 		.prog		= NLM_PROGRAM,
 		.vers		= 3,
 		.proc		= NLMPROC_NSM_NOTIFY,
-		.mon_name	= nsm->sm_name,
+		.mon_name	= nsm->sm_mon_name,
 	};
 	struct rpc_message msg = {
 		.rpc_argp	= &args,
@@ -87,6 +85,12 @@ nsm_monitor(struct nlm_host *host)
 	if (nsm->sm_monitored)
 		return 0;
 
+	/*
+	 * Choose whether to record the caller_name or IP address of
+	 * this peer in the local rpc.statd's database.
+	 */
+	nsm->sm_mon_name = nsm_use_hostnames ? nsm->sm_name : nsm->sm_addrbuf;
+
 	status = nsm_mon_unmon(nsm, SM_MON, &res);
 
 	if (status < 0 || res.status != 0)
@@ -167,25 +171,10 @@ static __be32 *xdr_encode_nsm_string(__be32 *p, char *string)
 
 /*
  * "mon_name" specifies the host to be monitored.
- *
- * Linux uses a text version of the IP address of the remote
- * host as the host identifier (the "mon_name" argument).
- *
- * Linux statd always looks up the canonical hostname first for
- * whatever remote hostname it receives, so this works alright.
  */
 static __be32 *xdr_encode_mon_name(__be32 *p, struct nsm_args *argp)
 {
-	char	buffer[XDR_ADDRBUF_LEN + 1];
-	char	*name = argp->mon_name;
-
-	if (!nsm_use_hostnames) {
-		snprintf(buffer, XDR_ADDRBUF_LEN,
-			 "%pI4", &argp->addr);
-		name = buffer;
-	}
-
-	return xdr_encode_nsm_string(p, name);
+	return xdr_encode_nsm_string(p, argp->mon_name);
 }
 
 /*
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 54dbb458e73c..d3c7247d23e8 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -79,6 +79,7 @@ struct nlm_host {
 struct nsm_handle {
 	struct list_head	sm_link;
 	atomic_t		sm_count;
+	char			*sm_mon_name;
 	char			*sm_name;
 	struct sockaddr_storage	sm_addr;
 	size_t			sm_addrlen;
-- 
cgit v1.2.3


From 1e49323c4ab044d05bbc68cf13cadcbd4372468c Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 4 Dec 2008 14:21:24 -0500
Subject: NLM: Move the public declaration of nsm_monitor() to lockd.h

Clean up.

Make the nlm_host argument "const," and move the public declaration to
lockd.h with other NSM public function (nsm_release, eg) and global
variable declarations.

Add a documenting comment.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/mon.c                 | 15 +++++++++++----
 include/linux/lockd/lockd.h    |  4 ++++
 include/linux/lockd/sm_inter.h |  1 -
 3 files changed, 15 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index 07e16b81498d..aaaa08e7ae7a 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -69,11 +69,18 @@ nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res)
 	return status;
 }
 
-/*
- * Set up monitoring of a remote host
+/**
+ * nsm_monitor - Notify a peer in case we reboot
+ * @host: pointer to nlm_host of peer to notify
+ *
+ * If this peer is not already monitored, this function sends an
+ * upcall to the local rpc.statd to record the name/address of
+ * the peer to notify in case we reboot.
+ *
+ * Returns zero if the peer is monitored by the local rpc.statd;
+ * otherwise a negative errno value is returned.
  */
-int
-nsm_monitor(struct nlm_host *host)
+int nsm_monitor(const struct nlm_host *host)
 {
 	struct nsm_handle *nsm = host->h_nsmhandle;
 	struct nsm_res	res;
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index d3c7247d23e8..f15a4f5ccbfb 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -242,6 +242,10 @@ extern void	  nlm_host_rebooted(const struct sockaddr_in *, const char *,
 					unsigned int, u32);
 void		  nsm_release(struct nsm_handle *);
 
+/*
+ * Host monitoring
+ */
+int		  nsm_monitor(const struct nlm_host *host);
 
 /*
  * This is used in garbage collection and resource reclaim
diff --git a/include/linux/lockd/sm_inter.h b/include/linux/lockd/sm_inter.h
index 5a5448bdb17d..546b6102b0d7 100644
--- a/include/linux/lockd/sm_inter.h
+++ b/include/linux/lockd/sm_inter.h
@@ -41,7 +41,6 @@ struct nsm_res {
 	u32		state;
 };
 
-int		nsm_monitor(struct nlm_host *);
 int		nsm_unmonitor(struct nlm_host *);
 extern int	nsm_local_state;
 
-- 
cgit v1.2.3


From c8c23c423dec49cb439697d3dc714e1500ff1610 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 4 Dec 2008 14:21:31 -0500
Subject: NSM: Release nsmhandle in nlm_destroy_host

The nsm_handle's reference count is bumped in nlm_lookup_host().  It
should be decremented in nlm_destroy_host() to make it easier to see
the balance of these two operations.

Move the nsm_release() call to fs/lockd/host.c.

The h_nsmhandle pointer is set in nlm_lookup_host(), and never cleared.
The nlm_destroy_host() function is never called for the same nlm_host
twice, so h_nsmhandle won't ever be NULL when nsm_unmonitor() is
called.

All references to the nlm_host are gone before it is freed.  We can
skip making h_nsmhandle NULL just before the nlm_host is deallocated.

It's also likely we can remove the h_nsmhandle NULL check in
nlmsvc_is_client() as well, but we can do that later when rearchitect-
ing the nlm_host cache.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/host.c             | 8 +++-----
 fs/lockd/mon.c              | 5 -----
 include/linux/lockd/lockd.h | 1 -
 3 files changed, 3 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index 780918acd6f4..1d523c1a7b62 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -37,6 +37,7 @@ static struct nsm_handle	*nsm_find(const struct sockaddr *sap,
 						const char *hostname,
 						const size_t hostname_len,
 						const int create);
+static void			nsm_release(struct nsm_handle *nsm);
 
 struct nlm_lookup_host_info {
 	const int		server;		/* search for server|client */
@@ -263,10 +264,8 @@ nlm_destroy_host(struct nlm_host *host)
 	BUG_ON(!list_empty(&host->h_lockowners));
 	BUG_ON(atomic_read(&host->h_count));
 
-	/*
-	 * Release NSM handle and unmonitor host.
-	 */
 	nsm_unmonitor(host);
+	nsm_release(host->h_nsmhandle);
 
 	clnt = host->h_rpcclnt;
 	if (clnt != NULL)
@@ -711,8 +710,7 @@ found:
 /*
  * Release an NSM handle
  */
-void
-nsm_release(struct nsm_handle *nsm)
+static void nsm_release(struct nsm_handle *nsm)
 {
 	if (!nsm)
 		return;
diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index aaaa08e7ae7a..15fab22db028 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -117,10 +117,6 @@ nsm_unmonitor(struct nlm_host *host)
 	struct nsm_res	res;
 	int		status = 0;
 
-	if (nsm == NULL)
-		return 0;
-	host->h_nsmhandle = NULL;
-
 	if (atomic_read(&nsm->sm_count) == 1
 	 && nsm->sm_monitored && !nsm->sm_sticky) {
 		dprintk("lockd: nsm_unmonitor(%s)\n", nsm->sm_name);
@@ -132,7 +128,6 @@ nsm_unmonitor(struct nlm_host *host)
 		else
 			nsm->sm_monitored = 0;
 	}
-	nsm_release(nsm);
 	return status;
 }
 
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index f15a4f5ccbfb..30a6a9c1ce42 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -240,7 +240,6 @@ void		  nlm_release_host(struct nlm_host *);
 void		  nlm_shutdown_hosts(void);
 extern void	  nlm_host_rebooted(const struct sockaddr_in *, const char *,
 					unsigned int, u32);
-void		  nsm_release(struct nsm_handle *);
 
 /*
  * Host monitoring
-- 
cgit v1.2.3


From 356c3eb466fd1a12afd6448d90fba3922836e5f1 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 4 Dec 2008 14:21:38 -0500
Subject: NLM: Move the public declaration of nsm_unmonitor() to lockd.h

Clean up.

Make the nlm_host argument "const," and move the public declaration to
lockd.h.  Add a documenting comment.

Bruce observed that nsm_unmonitor()'s only caller doesn't care about
its return code, so make nsm_unmonitor() return void.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/mon.c                 | 15 +++++++++------
 include/linux/lockd/lockd.h    |  1 +
 include/linux/lockd/sm_inter.h |  1 -
 3 files changed, 10 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index 15fab22db028..d61cdc61cb50 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -107,15 +107,19 @@ int nsm_monitor(const struct nlm_host *host)
 	return status;
 }
 
-/*
- * Cease to monitor remote host
+/**
+ * nsm_unmonitor - Unregister peer notification
+ * @host: pointer to nlm_host of peer to stop monitoring
+ *
+ * If this peer is monitored, this function sends an upcall to
+ * tell the local rpc.statd not to send this peer a notification
+ * when we reboot.
  */
-int
-nsm_unmonitor(struct nlm_host *host)
+void nsm_unmonitor(const struct nlm_host *host)
 {
 	struct nsm_handle *nsm = host->h_nsmhandle;
 	struct nsm_res	res;
-	int		status = 0;
+	int status;
 
 	if (atomic_read(&nsm->sm_count) == 1
 	 && nsm->sm_monitored && !nsm->sm_sticky) {
@@ -128,7 +132,6 @@ nsm_unmonitor(struct nlm_host *host)
 		else
 			nsm->sm_monitored = 0;
 	}
-	return status;
 }
 
 /*
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 30a6a9c1ce42..38344bfb814a 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -245,6 +245,7 @@ extern void	  nlm_host_rebooted(const struct sockaddr_in *, const char *,
  * Host monitoring
  */
 int		  nsm_monitor(const struct nlm_host *host);
+void		  nsm_unmonitor(const struct nlm_host *host);
 
 /*
  * This is used in garbage collection and resource reclaim
diff --git a/include/linux/lockd/sm_inter.h b/include/linux/lockd/sm_inter.h
index 546b6102b0d7..896a5e303323 100644
--- a/include/linux/lockd/sm_inter.h
+++ b/include/linux/lockd/sm_inter.h
@@ -41,7 +41,6 @@ struct nsm_res {
 	u32		state;
 };
 
-int		nsm_unmonitor(struct nlm_host *);
 extern int	nsm_local_state;
 
 #endif /* LINUX_LOCKD_SM_INTER_H */
-- 
cgit v1.2.3


From 9c1bfd037f7ff8badaecb47418f109148d88bf45 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Fri, 5 Dec 2008 19:01:59 -0500
Subject: NSM: Move NSM-related XDR data structures to lockd's xdr.h

Clean up: NSM's XDR data structures are used only in fs/lockd/mon.c,
so move them there.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/mon.c                 | 14 ++++++++++++++
 include/linux/lockd/sm_inter.h | 20 --------------------
 2 files changed, 14 insertions(+), 20 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index 3bb71e1b1e1f..81308832e994 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -18,6 +18,20 @@
 
 #define NLMDBG_FACILITY		NLMDBG_MONITOR
 
+struct nsm_args {
+	__be32			addr;		/* remote address */
+	u32			prog;		/* RPC callback info */
+	u32			vers;
+	u32			proc;
+
+	char			*mon_name;
+};
+
+struct nsm_res {
+	u32			status;
+	u32			state;
+};
+
 static struct rpc_clnt *	nsm_create(void);
 
 static struct rpc_program	nsm_program;
diff --git a/include/linux/lockd/sm_inter.h b/include/linux/lockd/sm_inter.h
index 896a5e303323..dd9d8a5bb316 100644
--- a/include/linux/lockd/sm_inter.h
+++ b/include/linux/lockd/sm_inter.h
@@ -21,26 +21,6 @@
 #define SM_MAXSTRLEN	1024
 #define SM_PRIV_SIZE	16
 
-/*
- * Arguments for all calls to statd
- */
-struct nsm_args {
-	__be32		addr;		/* remote address */
-	u32		prog;		/* RPC callback info */
-	u32		vers;
-	u32		proc;
-
-	char *		mon_name;
-};
-
-/*
- * Result returned by statd
- */
-struct nsm_res {
-	u32		status;
-	u32		state;
-};
-
 extern int	nsm_local_state;
 
 #endif /* LINUX_LOCKD_SM_INTER_H */
-- 
cgit v1.2.3


From 36e8e668d3e6a61848a8921ddeb663b417299fa5 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Fri, 5 Dec 2008 19:02:07 -0500
Subject: NSM: Move NSM program and procedure numbers to fs/lockd/mon.c

Clean up: Move the RPC program and procedure numbers for NSM into the
one source file that needs them: fs/lockd/mon.c.

And, as with NLM, NFS, and rpcbind calls, use NSMPROC_FOO instead of
SM_FOO for NSM procedure numbers.

Finally, make a couple of comments more precise: what is referred to
here as SM_NOTIFY is really the NLM (lockd) NLMPROC_SM_NOTIFY downcall,
not NSMPROC_NOTIFY.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/mon.c                 | 42 +++++++++++++++++++++++++++---------------
 include/linux/lockd/sm_inter.h |  9 ---------
 2 files changed, 27 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index 81308832e994..0fc9836db4e7 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -17,6 +17,18 @@
 
 
 #define NLMDBG_FACILITY		NLMDBG_MONITOR
+#define NSM_PROGRAM		100024
+#define NSM_VERSION		1
+
+enum {
+	NSMPROC_NULL,
+	NSMPROC_STAT,
+	NSMPROC_MON,
+	NSMPROC_UNMON,
+	NSMPROC_UNMON_ALL,
+	NSMPROC_SIMU_CRASH,
+	NSMPROC_NOTIFY,
+};
 
 struct nsm_args {
 	__be32			addr;		/* remote address */
@@ -42,7 +54,7 @@ static struct rpc_program	nsm_program;
 int				nsm_local_state;
 
 /*
- * Common procedure for SM_MON/SM_UNMON calls
+ * Common procedure for NSMPROC_MON/NSMPROC_UNMON calls
  */
 static int
 nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res)
@@ -111,7 +123,7 @@ int nsm_monitor(const struct nlm_host *host)
 	 */
 	nsm->sm_mon_name = nsm_use_hostnames ? nsm->sm_name : nsm->sm_addrbuf;
 
-	status = nsm_mon_unmon(nsm, SM_MON, &res);
+	status = nsm_mon_unmon(nsm, NSMPROC_MON, &res);
 	if (res.status != 0)
 		status = -EIO;
 	if (status < 0)
@@ -139,7 +151,7 @@ void nsm_unmonitor(const struct nlm_host *host)
 	 && nsm->sm_monitored && !nsm->sm_sticky) {
 		dprintk("lockd: nsm_unmonitor(%s)\n", nsm->sm_name);
 
-		status = nsm_mon_unmon(nsm, SM_UNMON, &res);
+		status = nsm_mon_unmon(nsm, NSMPROC_UNMON, &res);
 		if (res.status != 0)
 			status = -EIO;
 		if (status < 0)
@@ -167,7 +179,7 @@ nsm_create(void)
 		.addrsize	= sizeof(sin),
 		.servername	= "localhost",
 		.program	= &nsm_program,
-		.version	= SM_VERSION,
+		.version	= NSM_VERSION,
 		.authflavor	= RPC_AUTH_NULL,
 	};
 
@@ -201,7 +213,7 @@ static __be32 *xdr_encode_mon_name(__be32 *p, struct nsm_args *argp)
 /*
  * The "my_id" argument specifies the hostname and RPC procedure
  * to be called when the status manager receives notification
- * (via the SM_NOTIFY call) that the state of host "mon_name"
+ * (via the NLMPROC_SM_NOTIFY call) that the state of host "mon_name"
  * has changed.
  */
 static __be32 *xdr_encode_my_id(__be32 *p, struct nsm_args *argp)
@@ -219,7 +231,7 @@ static __be32 *xdr_encode_my_id(__be32 *p, struct nsm_args *argp)
 
 /*
  * The "mon_id" argument specifies the non-private arguments
- * of an SM_MON or SM_UNMON call.
+ * of an NSMPROC_MON or NSMPROC_UNMON call.
  */
 static __be32 *xdr_encode_mon_id(__be32 *p, struct nsm_args *argp)
 {
@@ -232,8 +244,8 @@ static __be32 *xdr_encode_mon_id(__be32 *p, struct nsm_args *argp)
 
 /*
  * The "priv" argument may contain private information required
- * by the SM_MON call. This information will be supplied in the
- * SM_NOTIFY call.
+ * by the NSMPROC_MON call. This information will be supplied in the
+ * NLMPROC_SM_NOTIFY call.
  *
  * Linux provides the raw IP address of the monitored host,
  * left in network byte order.
@@ -300,22 +312,22 @@ xdr_decode_stat(struct rpc_rqst *rqstp, __be32 *p, struct nsm_res *resp)
 #define SM_unmonres_sz	1
 
 static struct rpc_procinfo	nsm_procedures[] = {
-[SM_MON] = {
-		.p_proc		= SM_MON,
+[NSMPROC_MON] = {
+		.p_proc		= NSMPROC_MON,
 		.p_encode	= (kxdrproc_t) xdr_encode_mon,
 		.p_decode	= (kxdrproc_t) xdr_decode_stat_res,
 		.p_arglen	= SM_mon_sz,
 		.p_replen	= SM_monres_sz,
-		.p_statidx	= SM_MON,
+		.p_statidx	= NSMPROC_MON,
 		.p_name		= "MONITOR",
 	},
-[SM_UNMON] = {
-		.p_proc		= SM_UNMON,
+[NSMPROC_UNMON] = {
+		.p_proc		= NSMPROC_UNMON,
 		.p_encode	= (kxdrproc_t) xdr_encode_unmon,
 		.p_decode	= (kxdrproc_t) xdr_decode_stat,
 		.p_arglen	= SM_mon_id_sz,
 		.p_replen	= SM_unmonres_sz,
-		.p_statidx	= SM_UNMON,
+		.p_statidx	= NSMPROC_UNMON,
 		.p_name		= "UNMONITOR",
 	},
 };
@@ -334,7 +346,7 @@ static struct rpc_stat		nsm_stats;
 
 static struct rpc_program	nsm_program = {
 		.name		= "statd",
-		.number		= SM_PROGRAM,
+		.number		= NSM_PROGRAM,
 		.nrvers		= ARRAY_SIZE(nsm_version),
 		.version	= nsm_version,
 		.stats		= &nsm_stats
diff --git a/include/linux/lockd/sm_inter.h b/include/linux/lockd/sm_inter.h
index dd9d8a5bb316..116bf38535a0 100644
--- a/include/linux/lockd/sm_inter.h
+++ b/include/linux/lockd/sm_inter.h
@@ -9,15 +9,6 @@
 #ifndef LINUX_LOCKD_SM_INTER_H
 #define LINUX_LOCKD_SM_INTER_H
 
-#define SM_PROGRAM	100024
-#define SM_VERSION	1
-#define SM_STAT		1
-#define SM_MON		2
-#define SM_UNMON	3
-#define SM_UNMON_ALL	4
-#define SM_SIMU_CRASH	5
-#define SM_NOTIFY	6
-
 #define SM_MAXSTRLEN	1024
 #define SM_PRIV_SIZE	16
 
-- 
cgit v1.2.3


From 67c6d107a689243979a2b5f15244b5261634a924 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Fri, 5 Dec 2008 19:02:45 -0500
Subject: NSM: Move nsm_find() to fs/lockd/mon.c

The nsm_find() function sets up fresh nsm_handle entries.  This is
where we will store the "priv" cookie used to lookup nsm_handles during
reboot recovery.  The cookie will be constructed when nsm_find()
creates a new nsm_handle.

As much as possible, I would like to keep everything that handles a
"priv" cookie in fs/lockd/mon.c so that all the smarts are in one
source file.  That organization should make it pretty simple to see how
all this works.

To me, it makes more sense than the current arrangement to keep
nsm_find() with nsm_monitor() and nsm_unmonitor().

So, start reorganizing by moving nsm_find() into fs/lockd/mon.c.  The
nsm_release() function comes along too, since it shares the nsm_lock
global variable.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/host.c             | 128 ------------------------------------------
 fs/lockd/mon.c              | 133 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/lockd/lockd.h |   6 ++
 3 files changed, 139 insertions(+), 128 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index 1d523c1a7b62..dbdeaa88d2fa 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -32,12 +32,6 @@ static int			nrhosts;
 static DEFINE_MUTEX(nlm_host_mutex);
 
 static void			nlm_gc_hosts(void);
-static struct nsm_handle	*nsm_find(const struct sockaddr *sap,
-						const size_t salen,
-						const char *hostname,
-						const size_t hostname_len,
-						const int create);
-static void			nsm_release(struct nsm_handle *nsm);
 
 struct nlm_lookup_host_info {
 	const int		server;		/* search for server|client */
@@ -106,43 +100,6 @@ static void nlm_clear_port(struct sockaddr *sap)
 	}
 }
 
-static void nlm_display_ipv4_address(const struct sockaddr *sap, char *buf,
-				     const size_t len)
-{
-	const struct sockaddr_in *sin = (struct sockaddr_in *)sap;
-	snprintf(buf, len, "%pI4", &sin->sin_addr.s_addr);
-}
-
-static void nlm_display_ipv6_address(const struct sockaddr *sap, char *buf,
-				     const size_t len)
-{
-	const struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap;
-
-	if (ipv6_addr_v4mapped(&sin6->sin6_addr))
-		snprintf(buf, len, "%pI4", &sin6->sin6_addr.s6_addr32[3]);
-	else if (sin6->sin6_scope_id != 0)
-		snprintf(buf, len, "%pI6%%%u", &sin6->sin6_addr,
-				sin6->sin6_scope_id);
-	else
-		snprintf(buf, len, "%pI6", &sin6->sin6_addr);
-}
-
-static void nlm_display_address(const struct sockaddr *sap,
-				char *buf, const size_t len)
-{
-	switch (sap->sa_family) {
-	case AF_INET:
-		nlm_display_ipv4_address(sap, buf, len);
-		break;
-	case AF_INET6:
-		nlm_display_ipv6_address(sap, buf, len);
-		break;
-	default:
-		snprintf(buf, len, "unsupported address family");
-		break;
-	}
-}
-
 /*
  * Common host lookup routine for server & client
  */
@@ -635,88 +592,3 @@ nlm_gc_hosts(void)
 
 	next_gc = jiffies + NLM_HOST_COLLECT;
 }
-
-
-/*
- * Manage NSM handles
- */
-static LIST_HEAD(nsm_handles);
-static DEFINE_SPINLOCK(nsm_lock);
-
-static struct nsm_handle *nsm_find(const struct sockaddr *sap,
-				   const size_t salen,
-				   const char *hostname,
-				   const size_t hostname_len,
-				   const int create)
-{
-	struct nsm_handle *nsm = NULL;
-	struct nsm_handle *pos;
-
-	if (!sap)
-		return NULL;
-
-	if (hostname && memchr(hostname, '/', hostname_len) != NULL) {
-		if (printk_ratelimit()) {
-			printk(KERN_WARNING "Invalid hostname \"%.*s\" "
-					    "in NFS lock request\n",
-				(int)hostname_len, hostname);
-		}
-		return NULL;
-	}
-
-retry:
-	spin_lock(&nsm_lock);
-	list_for_each_entry(pos, &nsm_handles, sm_link) {
-
-		if (hostname && nsm_use_hostnames) {
-			if (strlen(pos->sm_name) != hostname_len
-			 || memcmp(pos->sm_name, hostname, hostname_len))
-				continue;
-		} else if (!nlm_cmp_addr(nsm_addr(pos), sap))
-			continue;
-		atomic_inc(&pos->sm_count);
-		kfree(nsm);
-		nsm = pos;
-		goto found;
-	}
-	if (nsm) {
-		list_add(&nsm->sm_link, &nsm_handles);
-		goto found;
-	}
-	spin_unlock(&nsm_lock);
-
-	if (!create)
-		return NULL;
-
-	nsm = kzalloc(sizeof(*nsm) + hostname_len + 1, GFP_KERNEL);
-	if (nsm == NULL)
-		return NULL;
-
-	memcpy(nsm_addr(nsm), sap, salen);
-	nsm->sm_addrlen = salen;
-	nsm->sm_name = (char *) (nsm + 1);
-	memcpy(nsm->sm_name, hostname, hostname_len);
-	nsm->sm_name[hostname_len] = '\0';
-	nlm_display_address((struct sockaddr *)&nsm->sm_addr,
-				nsm->sm_addrbuf, sizeof(nsm->sm_addrbuf));
-	atomic_set(&nsm->sm_count, 1);
-	goto retry;
-
-found:
-	spin_unlock(&nsm_lock);
-	return nsm;
-}
-
-/*
- * Release an NSM handle
- */
-static void nsm_release(struct nsm_handle *nsm)
-{
-	if (!nsm)
-		return;
-	if (atomic_dec_and_lock(&nsm->sm_count, &nsm_lock)) {
-		list_del(&nsm->sm_link);
-		spin_unlock(&nsm_lock);
-		kfree(nsm);
-	}
-}
diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index 81e1cc14246f..8e68e799293c 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -47,12 +47,51 @@ struct nsm_res {
 static struct rpc_clnt *	nsm_create(void);
 
 static struct rpc_program	nsm_program;
+static				LIST_HEAD(nsm_handles);
+static				DEFINE_SPINLOCK(nsm_lock);
 
 /*
  * Local NSM state
  */
 int				nsm_local_state;
 
+static void nsm_display_ipv4_address(const struct sockaddr *sap, char *buf,
+				     const size_t len)
+{
+	const struct sockaddr_in *sin = (struct sockaddr_in *)sap;
+	snprintf(buf, len, "%pI4", &sin->sin_addr.s_addr);
+}
+
+static void nsm_display_ipv6_address(const struct sockaddr *sap, char *buf,
+				     const size_t len)
+{
+	const struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap;
+
+	if (ipv6_addr_v4mapped(&sin6->sin6_addr))
+		snprintf(buf, len, "%pI4", &sin6->sin6_addr.s6_addr32[3]);
+	else if (sin6->sin6_scope_id != 0)
+		snprintf(buf, len, "%pI6%%%u", &sin6->sin6_addr,
+				sin6->sin6_scope_id);
+	else
+		snprintf(buf, len, "%pI6", &sin6->sin6_addr);
+}
+
+static void nsm_display_address(const struct sockaddr *sap,
+				char *buf, const size_t len)
+{
+	switch (sap->sa_family) {
+	case AF_INET:
+		nsm_display_ipv4_address(sap, buf, len);
+		break;
+	case AF_INET6:
+		nsm_display_ipv6_address(sap, buf, len);
+		break;
+	default:
+		snprintf(buf, len, "unsupported address family");
+		break;
+	}
+}
+
 /*
  * Common procedure for NSMPROC_MON/NSMPROC_UNMON calls
  */
@@ -162,6 +201,100 @@ void nsm_unmonitor(const struct nlm_host *host)
 	}
 }
 
+/**
+ * nsm_find - Find or create a cached nsm_handle
+ * @sap: pointer to socket address of handle to find
+ * @salen: length of socket address
+ * @hostname: pointer to C string containing hostname to find
+ * @hostname_len: length of C string
+ * @create: one means create new handle if not found in cache
+ *
+ * Behavior is modulated by the global nsm_use_hostnames variable
+ * and by the @create argument.
+ *
+ * Returns a cached nsm_handle after bumping its ref count, or if
+ * @create is set, returns a fresh nsm_handle if a handle that
+ * matches @sap and/or @hostname cannot be found in the handle cache.
+ * Returns NULL if an error occurs.
+ */
+struct nsm_handle *nsm_find(const struct sockaddr *sap, const size_t salen,
+			    const char *hostname, const size_t hostname_len,
+			    const int create)
+{
+	struct nsm_handle *nsm = NULL;
+	struct nsm_handle *pos;
+
+	if (!sap)
+		return NULL;
+
+	if (hostname && memchr(hostname, '/', hostname_len) != NULL) {
+		if (printk_ratelimit()) {
+			printk(KERN_WARNING "Invalid hostname \"%.*s\" "
+					    "in NFS lock request\n",
+				(int)hostname_len, hostname);
+		}
+		return NULL;
+	}
+
+retry:
+	spin_lock(&nsm_lock);
+	list_for_each_entry(pos, &nsm_handles, sm_link) {
+
+		if (hostname && nsm_use_hostnames) {
+			if (strlen(pos->sm_name) != hostname_len
+			 || memcmp(pos->sm_name, hostname, hostname_len))
+				continue;
+		} else if (!nlm_cmp_addr(nsm_addr(pos), sap))
+			continue;
+		atomic_inc(&pos->sm_count);
+		kfree(nsm);
+		nsm = pos;
+		goto found;
+	}
+	if (nsm) {
+		list_add(&nsm->sm_link, &nsm_handles);
+		goto found;
+	}
+	spin_unlock(&nsm_lock);
+
+	if (!create)
+		return NULL;
+
+	nsm = kzalloc(sizeof(*nsm) + hostname_len + 1, GFP_KERNEL);
+	if (nsm == NULL)
+		return NULL;
+
+	memcpy(nsm_addr(nsm), sap, salen);
+	nsm->sm_addrlen = salen;
+	nsm->sm_name = (char *) (nsm + 1);
+	memcpy(nsm->sm_name, hostname, hostname_len);
+	nsm->sm_name[hostname_len] = '\0';
+	nsm_display_address((struct sockaddr *)&nsm->sm_addr,
+				nsm->sm_addrbuf, sizeof(nsm->sm_addrbuf));
+	atomic_set(&nsm->sm_count, 1);
+	goto retry;
+
+found:
+	spin_unlock(&nsm_lock);
+	return nsm;
+}
+
+/**
+ * nsm_release - Release an NSM handle
+ * @nsm: pointer to handle to be released
+ *
+ */
+void nsm_release(struct nsm_handle *nsm)
+{
+	if (!nsm)
+		return;
+	if (atomic_dec_and_lock(&nsm->sm_count, &nsm_lock)) {
+		list_del(&nsm->sm_link);
+		spin_unlock(&nsm_lock);
+		kfree(nsm);
+	}
+}
+
 /*
  * Create NSM client for the local host
  */
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 38344bfb814a..8d715363c6ac 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -247,6 +247,12 @@ extern void	  nlm_host_rebooted(const struct sockaddr_in *, const char *,
 int		  nsm_monitor(const struct nlm_host *host);
 void		  nsm_unmonitor(const struct nlm_host *host);
 
+struct nsm_handle *nsm_find(const struct sockaddr *sap, const size_t salen,
+					const char *hostname,
+					const size_t hostname_len,
+					const int create);
+void		  nsm_release(struct nsm_handle *nsm);
+
 /*
  * This is used in garbage collection and resource reclaim
  * A return value != 0 means destroy the lock/block/share
-- 
cgit v1.2.3


From 7e44d3bea21fbb9494930d1cd35ca92a9a4a3279 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Fri, 5 Dec 2008 19:03:16 -0500
Subject: NSM: Generate NSMPROC_MON's "priv" argument when nsm_handle is
 created

Introduce a new data type, used by both the in-kernel NLM and NSM
implementations, that is used to manage the opaque "priv" argument
for the NSMPROC_MON and NLMPROC_SM_NOTIFY calls.

Construct the "priv" cookie when the nsm_handle is created.

The nsm_init_private() function may look a little strange, but it is
roughly equivalent to how the XDR encoder formed the "priv" argument.
It's going to go away soon.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/mon.c                 | 16 ++++++++++++++++
 include/linux/lockd/lockd.h    |  1 +
 include/linux/lockd/sm_inter.h |  1 -
 include/linux/lockd/xdr.h      |  6 ++++++
 4 files changed, 23 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index 0792900b6281..c8d18cd22b8a 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -201,6 +201,21 @@ void nsm_unmonitor(const struct nlm_host *host)
 	}
 }
 
+/*
+ * Construct a unique cookie to match this nsm_handle to this monitored
+ * host.  It is passed to the local rpc.statd via NSMPROC_MON, and
+ * returned via NLMPROC_SM_NOTIFY, in the "priv" field of these
+ * requests.
+ *
+ * Linux provides the raw IP address of the monitored host,
+ * left in network byte order.
+ */
+static void nsm_init_private(struct nsm_handle *nsm)
+{
+	__be32 *p = (__be32 *)&nsm->sm_priv.data;
+	*p = nsm_addr_in(nsm)->sin_addr.s_addr;
+}
+
 /**
  * nsm_find - Find or create a cached nsm_handle
  * @sap: pointer to socket address of handle to find
@@ -271,6 +286,7 @@ retry:
 	nsm->sm_name = (char *) (nsm + 1);
 	memcpy(nsm->sm_name, hostname, hostname_len);
 	nsm->sm_name[hostname_len] = '\0';
+	nsm_init_private(nsm);
 	nsm_display_address((struct sockaddr *)&nsm->sm_addr,
 				nsm->sm_addrbuf, sizeof(nsm->sm_addrbuf));
 	atomic_set(&nsm->sm_count, 1);
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 8d715363c6ac..194fa8a66398 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -85,6 +85,7 @@ struct nsm_handle {
 	size_t			sm_addrlen;
 	unsigned int		sm_monitored : 1,
 				sm_sticky : 1;	/* don't unmonitor */
+	struct nsm_private	sm_priv;
 	char			sm_addrbuf[NSM_ADDRBUF];
 };
 
diff --git a/include/linux/lockd/sm_inter.h b/include/linux/lockd/sm_inter.h
index 116bf38535a0..5cef5a79dd94 100644
--- a/include/linux/lockd/sm_inter.h
+++ b/include/linux/lockd/sm_inter.h
@@ -10,7 +10,6 @@
 #define LINUX_LOCKD_SM_INTER_H
 
 #define SM_MAXSTRLEN	1024
-#define SM_PRIV_SIZE	16
 
 extern int	nsm_local_state;
 
diff --git a/include/linux/lockd/xdr.h b/include/linux/lockd/xdr.h
index d6b3a802c046..6b5199263858 100644
--- a/include/linux/lockd/xdr.h
+++ b/include/linux/lockd/xdr.h
@@ -13,6 +13,12 @@
 #include <linux/nfs.h>
 #include <linux/sunrpc/xdr.h>
 
+#define SM_PRIV_SIZE		16
+
+struct nsm_private {
+	unsigned char		data[SM_PRIV_SIZE];
+};
+
 struct svc_rqst;
 
 #define NLM_MAXCOOKIELEN    	32
-- 
cgit v1.2.3


From 7fefc9cb9d5f129c238d93166f705c96ca2e7e51 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Fri, 5 Dec 2008 19:03:31 -0500
Subject: NLM: Change nlm_host_rebooted() to take a single nlm_reboot argument

Pass the nlm_reboot data structure directly from the NLMPROC_SM_NOTIFY
XDR decoders to nlm_host_rebooted().  This eliminates some packing and
unpacking of the NLMPROC_SM_NOTIFY results, and prepares for passing
these results, including the "priv" cookie, directly to a lookup
routine in fs/lockd/mon.c.

This patch changes code organization but should not cause any
behavioral change.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/host.c             | 31 +++++++++++++++++--------------
 fs/lockd/svc4proc.c         | 11 +----------
 fs/lockd/svcproc.c          | 11 +----------
 include/linux/lockd/lockd.h |  3 +--
 4 files changed, 20 insertions(+), 36 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index dbdeaa88d2fa..ed103387964d 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -444,31 +444,34 @@ void nlm_release_host(struct nlm_host *host)
 	}
 }
 
-/*
- * We were notified that the host indicated by address &sin
- * has rebooted.
- * Release all resources held by that peer.
+/**
+ * nlm_host_rebooted - Release all resources held by rebooted host
+ * @info: pointer to decoded results of NLM_SM_NOTIFY call
+ *
+ * We were notified that the specified host has rebooted.  Release
+ * all resources held by that peer.
  */
-void nlm_host_rebooted(const struct sockaddr_in *sin,
-				const char *hostname,
-				unsigned int hostname_len,
-				u32 new_state)
+void nlm_host_rebooted(const struct nlm_reboot *info)
 {
+	const struct sockaddr_in sin = {
+		.sin_family		= AF_INET,
+		.sin_addr.s_addr	= info->addr,
+	};
 	struct hlist_head *chain;
 	struct hlist_node *pos;
 	struct nsm_handle *nsm;
 	struct nlm_host	*host;
 
-	nsm = nsm_find((struct sockaddr *)sin, sizeof(*sin),
-			hostname, hostname_len, 0);
+	nsm = nsm_find((struct sockaddr *)&sin, sizeof(sin),
+			info->mon, info->len, 0);
 	if (nsm == NULL) {
 		dprintk("lockd: never saw rebooted peer '%.*s' before\n",
-				hostname_len, hostname);
+				info->len, info->mon);
 		return;
 	}
 
 	dprintk("lockd: nlm_host_rebooted(%.*s, %s)\n",
-			hostname_len, hostname, nsm->sm_addrbuf);
+			info->len, info->mon, nsm->sm_addrbuf);
 
 	/* When reclaiming locks on this peer, make sure that
 	 * we set up a new notification */
@@ -483,8 +486,8 @@ again:	mutex_lock(&nlm_host_mutex);
 	for (chain = nlm_hosts; chain < nlm_hosts + NLM_HOST_NRHASH; ++chain) {
 		hlist_for_each_entry(host, pos, chain, h_hash) {
 			if (host->h_nsmhandle == nsm
-			 && host->h_nsmstate != new_state) {
-				host->h_nsmstate = new_state;
+			 && host->h_nsmstate != info->state) {
+				host->h_nsmstate = info->state;
 				host->h_state++;
 
 				nlm_get_host(host);
diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c
index 4dfdcbc6bf68..bb79a53e0608 100644
--- a/fs/lockd/svc4proc.c
+++ b/fs/lockd/svc4proc.c
@@ -419,8 +419,6 @@ static __be32
 nlm4svc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp,
 					      void	        *resp)
 {
-	struct sockaddr_in	saddr;
-
 	dprintk("lockd: SM_NOTIFY     called\n");
 
 	if (!nlm_privileged_requester(rqstp)) {
@@ -430,14 +428,7 @@ nlm4svc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp,
 		return rpc_system_err;
 	}
 
-	/* Obtain the host pointer for this NFS server and try to
-	 * reclaim all locks we hold on this server.
-	 */
-	memset(&saddr, 0, sizeof(saddr));
-	saddr.sin_family = AF_INET;
-	saddr.sin_addr.s_addr = argp->addr;
-	nlm_host_rebooted(&saddr, argp->mon, argp->len, argp->state);
-
+	nlm_host_rebooted(argp);
 	return rpc_success;
 }
 
diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c
index 3ca89e2a9381..e44310c0211c 100644
--- a/fs/lockd/svcproc.c
+++ b/fs/lockd/svcproc.c
@@ -451,8 +451,6 @@ static __be32
 nlmsvc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp,
 					      void	        *resp)
 {
-	struct sockaddr_in	saddr;
-
 	dprintk("lockd: SM_NOTIFY     called\n");
 
 	if (!nlm_privileged_requester(rqstp)) {
@@ -462,14 +460,7 @@ nlmsvc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp,
 		return rpc_system_err;
 	}
 
-	/* Obtain the host pointer for this NFS server and try to
-	 * reclaim all locks we hold on this server.
-	 */
-	memset(&saddr, 0, sizeof(saddr));
-	saddr.sin_family = AF_INET;
-	saddr.sin_addr.s_addr = argp->addr;
-	nlm_host_rebooted(&saddr, argp->mon, argp->len, argp->state);
-
+	nlm_host_rebooted(argp);
 	return rpc_success;
 }
 
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 194fa8a66398..2a3533ea38dd 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -239,8 +239,7 @@ void		  nlm_rebind_host(struct nlm_host *);
 struct nlm_host * nlm_get_host(struct nlm_host *);
 void		  nlm_release_host(struct nlm_host *);
 void		  nlm_shutdown_hosts(void);
-extern void	  nlm_host_rebooted(const struct sockaddr_in *, const char *,
-					unsigned int, u32);
+void		  nlm_host_rebooted(const struct nlm_reboot *);
 
 /*
  * Host monitoring
-- 
cgit v1.2.3


From 576df4634e37e46b441fefb91915184edb13bb94 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Fri, 5 Dec 2008 19:03:39 -0500
Subject: NLM: Decode "priv" argument of NLMPROC_SM_NOTIFY as an opaque

The NLM XDR decoders for the NLMPROC_SM_NOTIFY procedure should treat
their "priv" argument truly as an opaque, as defined by the protocol,
and let the upper layers figure out what is in it.

This will make it easier to modify the contents and interpretation of
the "priv" argument, and keep knowledge about what's in "priv" local
to fs/lockd/mon.c.

For now, the NLM and NSM implementations should behave exactly as they
did before.

The formation of the address of the rebooted host in
nlm_host_rebooted() may look a little strange, but it is the inverse
of how nsm_init_private() forms the private cookie.  Plus, it's
going away soon anyway.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/host.c           | 3 ++-
 fs/lockd/xdr.c            | 4 ++--
 fs/lockd/xdr4.c           | 4 ++--
 include/linux/lockd/xdr.h | 8 ++++----
 4 files changed, 10 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index ed103387964d..dc41e46ef74c 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -453,9 +453,10 @@ void nlm_release_host(struct nlm_host *host)
  */
 void nlm_host_rebooted(const struct nlm_reboot *info)
 {
+	__be32 *p = (__be32 *)&info->priv.data;
 	const struct sockaddr_in sin = {
 		.sin_family		= AF_INET,
-		.sin_addr.s_addr	= info->addr,
+		.sin_addr.s_addr	= *p,
 	};
 	struct hlist_head *chain;
 	struct hlist_node *pos;
diff --git a/fs/lockd/xdr.c b/fs/lockd/xdr.c
index 1f226290c67c..4cc7d01a1eb5 100644
--- a/fs/lockd/xdr.c
+++ b/fs/lockd/xdr.c
@@ -349,8 +349,8 @@ nlmsvc_decode_reboot(struct svc_rqst *rqstp, __be32 *p, struct nlm_reboot *argp)
 	if (!(p = xdr_decode_string_inplace(p, &argp->mon, &argp->len, SM_MAXSTRLEN)))
 		return 0;
 	argp->state = ntohl(*p++);
-	/* Preserve the address in network byte order */
-	argp->addr = *p++;
+	memcpy(&argp->priv.data, p, sizeof(argp->priv.data));
+	p += XDR_QUADLEN(SM_PRIV_SIZE);
 	return xdr_argsize_check(rqstp, p);
 }
 
diff --git a/fs/lockd/xdr4.c b/fs/lockd/xdr4.c
index 50c493a8ad8e..61d1714a470e 100644
--- a/fs/lockd/xdr4.c
+++ b/fs/lockd/xdr4.c
@@ -356,8 +356,8 @@ nlm4svc_decode_reboot(struct svc_rqst *rqstp, __be32 *p, struct nlm_reboot *argp
 	if (!(p = xdr_decode_string_inplace(p, &argp->mon, &argp->len, SM_MAXSTRLEN)))
 		return 0;
 	argp->state = ntohl(*p++);
-	/* Preserve the address in network byte order */
-	argp->addr  = *p++;
+	memcpy(&argp->priv.data, p, sizeof(argp->priv.data));
+	p += XDR_QUADLEN(SM_PRIV_SIZE);
 	return xdr_argsize_check(rqstp, p);
 }
 
diff --git a/include/linux/lockd/xdr.h b/include/linux/lockd/xdr.h
index 6b5199263858..6338866222a8 100644
--- a/include/linux/lockd/xdr.h
+++ b/include/linux/lockd/xdr.h
@@ -83,10 +83,10 @@ struct nlm_res {
  * statd callback when client has rebooted
  */
 struct nlm_reboot {
-	char *		mon;
-	unsigned int	len;
-	u32		state;
-	__be32		addr;
+	char			*mon;
+	unsigned int		len;
+	u32			state;
+	struct nsm_private	priv;
 };
 
 /*
-- 
cgit v1.2.3


From 3420a8c4359a189f7d854ed7075d151257415447 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Fri, 5 Dec 2008 19:03:46 -0500
Subject: NSM: Add nsm_lookup() function

Introduce a new API to fs/lockd/mon.c that allows nlm_host_rebooted()
to lookup up nsm_handles via the contents of an nlm_reboot struct.

The new function is equivalent to calling nsm_find() with @create set
to zero, but it takes a struct nlm_reboot instead of separate
arguments.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/mon.c              | 64 +++++++++++++++++++++++++++++++++++++++++++++
 include/linux/lockd/lockd.h |  1 +
 2 files changed, 65 insertions(+)

(limited to 'include/linux')

diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index 4424b0a5a51f..e46903995c99 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -201,6 +201,29 @@ void nsm_unmonitor(const struct nlm_host *host)
 	}
 }
 
+static struct nsm_handle *nsm_lookup_hostname(const char *hostname,
+					      const size_t len)
+{
+	struct nsm_handle *nsm;
+
+	list_for_each_entry(nsm, &nsm_handles, sm_link)
+		if (strlen(nsm->sm_name) == len &&
+		    memcmp(nsm->sm_name, hostname, len) == 0)
+			return nsm;
+	return NULL;
+}
+
+static struct nsm_handle *nsm_lookup_priv(const struct nsm_private *priv)
+{
+	struct nsm_handle *nsm;
+
+	list_for_each_entry(nsm, &nsm_handles, sm_link)
+		if (memcmp(nsm->sm_priv.data, priv->data,
+					sizeof(priv->data)) == 0)
+			return nsm;
+	return NULL;
+}
+
 /*
  * Construct a unique cookie to match this nsm_handle to this monitored
  * host.  It is passed to the local rpc.statd via NSMPROC_MON, and
@@ -297,6 +320,47 @@ found:
 	return nsm;
 }
 
+/**
+ * nsm_reboot_lookup - match NLMPROC_SM_NOTIFY arguments to an nsm_handle
+ * @info: pointer to NLMPROC_SM_NOTIFY arguments
+ *
+ * Returns a matching nsm_handle if found in the nsm cache; the returned
+ * nsm_handle's reference count is bumped and sm_monitored is cleared.
+ * Otherwise returns NULL if some error occurred.
+ */
+struct nsm_handle *nsm_reboot_lookup(const struct nlm_reboot *info)
+{
+	struct nsm_handle *cached;
+
+	spin_lock(&nsm_lock);
+
+	if (nsm_use_hostnames && info->mon != NULL)
+		cached = nsm_lookup_hostname(info->mon, info->len);
+	else
+		cached = nsm_lookup_priv(&info->priv);
+
+	if (unlikely(cached == NULL)) {
+		spin_unlock(&nsm_lock);
+		dprintk("lockd: never saw rebooted peer '%.*s' before\n",
+				info->len, info->mon);
+		return cached;
+	}
+
+	atomic_inc(&cached->sm_count);
+	spin_unlock(&nsm_lock);
+
+	/*
+	 * During subsequent lock activity, force a fresh
+	 * notification to be set up for this host.
+	 */
+	cached->sm_monitored = 0;
+
+	dprintk("lockd: host %s (%s) rebooted, cnt %d\n",
+			cached->sm_name, cached->sm_addrbuf,
+			atomic_read(&cached->sm_count));
+	return cached;
+}
+
 /**
  * nsm_release - Release an NSM handle
  * @nsm: pointer to handle to be released
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 2a3533ea38dd..5e3ad926de89 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -251,6 +251,7 @@ struct nsm_handle *nsm_find(const struct sockaddr *sap, const size_t salen,
 					const char *hostname,
 					const size_t hostname_len,
 					const int create);
+struct nsm_handle *nsm_reboot_lookup(const struct nlm_reboot *info);
 void		  nsm_release(struct nsm_handle *nsm);
 
 /*
-- 
cgit v1.2.3


From 92fd91b998a5216a6d6606704e71d541a180216c Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Fri, 5 Dec 2008 19:04:01 -0500
Subject: NLM: Remove "create" argument from nsm_find()

Clean up: nsm_find() now has only one caller, and that caller
unconditionally sets the @create argument. Thus the @create
argument is no longer needed.

Since nsm_find() now has a more specific purpose, pick a more
appropriate name for it.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/host.c             |  4 ++--
 fs/lockd/mon.c              | 23 +++++++++--------------
 include/linux/lockd/lockd.h |  6 +++---
 3 files changed, 14 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index 230de93fc048..e5a65df4c0cd 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -159,8 +159,8 @@ static struct nlm_host *nlm_lookup_host(struct nlm_lookup_host_info *ni)
 		atomic_inc(&nsm->sm_count);
 	else {
 		host = NULL;
-		nsm = nsm_find(ni->sap, ni->salen,
-				ni->hostname, ni->hostname_len, 1);
+		nsm = nsm_get_handle(ni->sap, ni->salen,
+					ni->hostname, ni->hostname_len);
 		if (!nsm) {
 			dprintk("lockd: nlm_lookup_host failed; "
 				"no nsm handle\n");
diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index e46903995c99..740702216042 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -240,24 +240,22 @@ static void nsm_init_private(struct nsm_handle *nsm)
 }
 
 /**
- * nsm_find - Find or create a cached nsm_handle
+ * nsm_get_handle - Find or create a cached nsm_handle
  * @sap: pointer to socket address of handle to find
  * @salen: length of socket address
  * @hostname: pointer to C string containing hostname to find
  * @hostname_len: length of C string
- * @create: one means create new handle if not found in cache
  *
- * Behavior is modulated by the global nsm_use_hostnames variable
- * and by the @create argument.
+ * Behavior is modulated by the global nsm_use_hostnames variable.
  *
- * Returns a cached nsm_handle after bumping its ref count, or if
- * @create is set, returns a fresh nsm_handle if a handle that
- * matches @sap and/or @hostname cannot be found in the handle cache.
- * Returns NULL if an error occurs.
+ * Returns a cached nsm_handle after bumping its ref count, or
+ * returns a fresh nsm_handle if a handle that matches @sap and/or
+ * @hostname cannot be found in the handle cache.  Returns NULL if
+ * an error occurs.
  */
-struct nsm_handle *nsm_find(const struct sockaddr *sap, const size_t salen,
-			    const char *hostname, const size_t hostname_len,
-			    const int create)
+struct nsm_handle *nsm_get_handle(const struct sockaddr *sap,
+				  const size_t salen, const char *hostname,
+				  const size_t hostname_len)
 {
 	struct nsm_handle *nsm = NULL;
 	struct nsm_handle *pos;
@@ -297,9 +295,6 @@ retry:
 	}
 	spin_unlock(&nsm_lock);
 
-	if (!create)
-		return NULL;
-
 	nsm = kzalloc(sizeof(*nsm) + hostname_len + 1, GFP_KERNEL);
 	if (nsm == NULL)
 		return NULL;
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 5e3ad926de89..1ccd49e97a7f 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -247,10 +247,10 @@ void		  nlm_host_rebooted(const struct nlm_reboot *);
 int		  nsm_monitor(const struct nlm_host *host);
 void		  nsm_unmonitor(const struct nlm_host *host);
 
-struct nsm_handle *nsm_find(const struct sockaddr *sap, const size_t salen,
+struct nsm_handle *nsm_get_handle(const struct sockaddr *sap,
+					const size_t salen,
 					const char *hostname,
-					const size_t hostname_len,
-					const int create);
+					const size_t hostname_len);
 struct nsm_handle *nsm_reboot_lookup(const struct nlm_reboot *info);
 void		  nsm_release(struct nsm_handle *nsm);
 
-- 
cgit v1.2.3


From e6765b83977f07983c7a10e6bbb19d6c7bbfc3a4 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 11 Dec 2008 17:56:14 -0500
Subject: NSM: Remove include/linux/lockd/sm_inter.h

Clean up: The include/linux/lockd/sm_inter.h header is nearly empty
now.  Remove it.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/clntproc.c            |  1 -
 fs/lockd/host.c                |  1 -
 fs/lockd/mon.c                 |  2 --
 fs/lockd/svc.c                 |  1 -
 fs/lockd/svc4proc.c            |  2 --
 fs/lockd/svcproc.c             |  2 --
 fs/lockd/svcsubs.c             |  1 -
 fs/lockd/xdr.c                 |  1 -
 fs/lockd/xdr4.c                |  1 -
 include/linux/lockd/lockd.h    |  1 +
 include/linux/lockd/sm_inter.h | 16 ----------------
 include/linux/lockd/xdr.h      |  1 +
 12 files changed, 2 insertions(+), 28 deletions(-)
 delete mode 100644 include/linux/lockd/sm_inter.h

(limited to 'include/linux')

diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c
index 5ce42e0ed4a0..dd7957064a8c 100644
--- a/fs/lockd/clntproc.c
+++ b/fs/lockd/clntproc.c
@@ -16,7 +16,6 @@
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/svc.h>
 #include <linux/lockd/lockd.h>
-#include <linux/lockd/sm_inter.h>
 
 #define NLMDBG_FACILITY		NLMDBG_CLIENT
 #define NLMCLNT_GRACE_WAIT	(5*HZ)
diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index e5a65df4c0cd..99d737bd4325 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -15,7 +15,6 @@
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/svc.h>
 #include <linux/lockd/lockd.h>
-#include <linux/lockd/sm_inter.h>
 #include <linux/mutex.h>
 
 #include <net/ipv6.h>
diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index 8ae4c02d7dfd..dfa9d80efcba 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -15,8 +15,6 @@
 #include <linux/sunrpc/xprtsock.h>
 #include <linux/sunrpc/svc.h>
 #include <linux/lockd/lockd.h>
-#include <linux/lockd/sm_inter.h>
-
 
 #define NLMDBG_FACILITY		NLMDBG_MONITOR
 #define NSM_PROGRAM		100024
diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c
index bc3c3cb62db5..0b13392931a6 100644
--- a/fs/lockd/svc.c
+++ b/fs/lockd/svc.c
@@ -35,7 +35,6 @@
 #include <linux/sunrpc/svcsock.h>
 #include <net/ip.h>
 #include <linux/lockd/lockd.h>
-#include <linux/lockd/sm_inter.h>
 #include <linux/nfs.h>
 
 #define NLMDBG_FACILITY		NLMDBG_SVC
diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c
index bb79a53e0608..1725037374c5 100644
--- a/fs/lockd/svc4proc.c
+++ b/fs/lockd/svc4proc.c
@@ -16,8 +16,6 @@
 #include <linux/nfsd/nfsd.h>
 #include <linux/lockd/lockd.h>
 #include <linux/lockd/share.h>
-#include <linux/lockd/sm_inter.h>
-
 
 #define NLMDBG_FACILITY		NLMDBG_CLIENT
 
diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c
index e44310c0211c..3688e55901fc 100644
--- a/fs/lockd/svcproc.c
+++ b/fs/lockd/svcproc.c
@@ -16,8 +16,6 @@
 #include <linux/nfsd/nfsd.h>
 #include <linux/lockd/lockd.h>
 #include <linux/lockd/share.h>
-#include <linux/lockd/sm_inter.h>
-
 
 #define NLMDBG_FACILITY		NLMDBG_CLIENT
 
diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c
index 34c2766e27c7..9e4d6aab611b 100644
--- a/fs/lockd/svcsubs.c
+++ b/fs/lockd/svcsubs.c
@@ -17,7 +17,6 @@
 #include <linux/nfsd/export.h>
 #include <linux/lockd/lockd.h>
 #include <linux/lockd/share.h>
-#include <linux/lockd/sm_inter.h>
 #include <linux/module.h>
 #include <linux/mount.h>
 
diff --git a/fs/lockd/xdr.c b/fs/lockd/xdr.c
index 4cc7d01a1eb5..0336f2beacde 100644
--- a/fs/lockd/xdr.c
+++ b/fs/lockd/xdr.c
@@ -16,7 +16,6 @@
 #include <linux/sunrpc/svc.h>
 #include <linux/sunrpc/stats.h>
 #include <linux/lockd/lockd.h>
-#include <linux/lockd/sm_inter.h>
 
 #define NLMDBG_FACILITY		NLMDBG_XDR
 
diff --git a/fs/lockd/xdr4.c b/fs/lockd/xdr4.c
index 61d1714a470e..e1d528653192 100644
--- a/fs/lockd/xdr4.c
+++ b/fs/lockd/xdr4.c
@@ -17,7 +17,6 @@
 #include <linux/sunrpc/svc.h>
 #include <linux/sunrpc/stats.h>
 #include <linux/lockd/lockd.h>
-#include <linux/lockd/sm_inter.h>
 
 #define NLMDBG_FACILITY		NLMDBG_XDR
 
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 1ccd49e97a7f..8b57467375cc 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -205,6 +205,7 @@ extern struct svc_procedure	nlmsvc_procedures4[];
 extern int			nlmsvc_grace_period;
 extern unsigned long		nlmsvc_timeout;
 extern int			nsm_use_hostnames;
+extern int			nsm_local_state;
 
 /*
  * Lockd client functions
diff --git a/include/linux/lockd/sm_inter.h b/include/linux/lockd/sm_inter.h
deleted file mode 100644
index 5cef5a79dd94..000000000000
--- a/include/linux/lockd/sm_inter.h
+++ /dev/null
@@ -1,16 +0,0 @@
-/*
- * linux/include/linux/lockd/sm_inter.h
- *
- * Declarations for the kernel statd client.
- *
- * Copyright (C) 1996, Olaf Kirch <okir@monad.swb.de>
- */
-
-#ifndef LINUX_LOCKD_SM_INTER_H
-#define LINUX_LOCKD_SM_INTER_H
-
-#define SM_MAXSTRLEN	1024
-
-extern int	nsm_local_state;
-
-#endif /* LINUX_LOCKD_SM_INTER_H */
diff --git a/include/linux/lockd/xdr.h b/include/linux/lockd/xdr.h
index 6338866222a8..7dc5b6cb44cd 100644
--- a/include/linux/lockd/xdr.h
+++ b/include/linux/lockd/xdr.h
@@ -13,6 +13,7 @@
 #include <linux/nfs.h>
 #include <linux/sunrpc/xdr.h>
 
+#define SM_MAXSTRLEN		1024
 #define SM_PRIV_SIZE		16
 
 struct nsm_private {
-- 
cgit v1.2.3


From 8529bc51d30b8f001734b29b21a51b579c260f5b Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 11 Dec 2008 17:56:22 -0500
Subject: NSM: Move nsm_addr() to fs/lockd/mon.c

Clean up: nsm_addr_in() is no longer used, and nsm_addr() is used only in
fs/lockd/mon.c, so move it there.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/mon.c              |  5 +++++
 include/linux/lockd/lockd.h | 10 ----------
 2 files changed, 5 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index dfa9d80efcba..43be31c4a2de 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -55,6 +55,11 @@ static				DEFINE_SPINLOCK(nsm_lock);
  */
 int				nsm_local_state;
 
+static inline struct sockaddr *nsm_addr(const struct nsm_handle *nsm)
+{
+	return (struct sockaddr *)&nsm->sm_addr;
+}
+
 static void nsm_display_ipv4_address(const struct sockaddr *sap, char *buf,
 				     const size_t len)
 {
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 8b57467375cc..6ab0449bc828 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -112,16 +112,6 @@ static inline struct sockaddr *nlm_srcaddr(const struct nlm_host *host)
 	return (struct sockaddr *)&host->h_srcaddr;
 }
 
-static inline struct sockaddr_in *nsm_addr_in(const struct nsm_handle *handle)
-{
-	return (struct sockaddr_in *)&handle->sm_addr;
-}
-
-static inline struct sockaddr *nsm_addr(const struct nsm_handle *handle)
-{
-	return (struct sockaddr *)&handle->sm_addr;
-}
-
 /*
  * Map an fl_owner_t into a unique 32-bit "pid"
  */
-- 
cgit v1.2.3


From d1208f70738c91f13b4eadb1b7a694082e439da2 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 11 Dec 2008 17:56:44 -0500
Subject: NLM: nlm_privileged_requester() doesn't recognize mapped loopback
 address

Commit b85e4676 added the nlm_privileged_requester() helper to check
whether an RPC request was sent from a local privileged caller.  It
recognizes IPv4 privileged callers (from "127.0.0.1"), and IPv6
privileged callers (from "::1").

However, IPV6_ADDR_LOOPBACK is not set for the mapped IPv4 loopback
address (::ffff:7f00:0001), so the test breaks when the kernel's RPC
service is IPv6-enabled but user space is calling via the IPv4
loopback address.  This is actually the most common case for IPv6-
enabled RPC services on Linux.

Rewrite the IPv6 check to handle the mapped IPv4 loopback address as
well as a normal IPv6 loopback address.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 include/linux/lockd/lockd.h | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 6ab0449bc828..80d7e8a8257d 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -299,8 +299,14 @@ static inline int __nlm_privileged_request4(const struct sockaddr *sap)
 static inline int __nlm_privileged_request6(const struct sockaddr *sap)
 {
 	const struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap;
-	return (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_LOOPBACK) &&
-			(ntohs(sin6->sin6_port) < 1024);
+
+	if (ntohs(sin6->sin6_port) > 1023)
+		return 0;
+
+	if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_MAPPED)
+		return ipv4_is_loopback(sin6->sin6_addr.s6_addr32[3]);
+
+	return ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_LOOPBACK;
 }
 #else	/* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */
 static inline int __nlm_privileged_request6(const struct sockaddr *sap)
-- 
cgit v1.2.3


From 57ef692588bc225853ca3267ca5b7cea2b07e058 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 11 Dec 2008 17:56:52 -0500
Subject: NLM: Rewrite IPv4 privileged requester's check

Clean up.

For consistency, rewrite the IPv4 check to match the same style as the
new IPv6 check.  Note that ipv4_is_loopback() is somewhat broader in
its interpretation of what is a loopback address than simply
"127.0.0.1".

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 include/linux/lockd/lockd.h | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 80d7e8a8257d..aa6fe7026de7 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -291,8 +291,11 @@ static inline struct inode *nlmsvc_file_inode(struct nlm_file *file)
 static inline int __nlm_privileged_request4(const struct sockaddr *sap)
 {
 	const struct sockaddr_in *sin = (struct sockaddr_in *)sap;
-	return (sin->sin_addr.s_addr == htonl(INADDR_LOOPBACK)) &&
-			(ntohs(sin->sin_port) < 1024);
+
+	if (ntohs(sin->sin_port) > 1023)
+		return 0;
+
+	return ipv4_is_loopback(sin->sin_addr.s_addr);
 }
 
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
-- 
cgit v1.2.3


From 6f49a57aa5a0c6d4e4e27c85f7af6c83325a12d1 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 6 Jan 2009 11:38:14 -0700
Subject: dmaengine: up-level reference counting to the module level

Simply, if a client wants any dmaengine channel then prevent all dmaengine
modules from being removed.  Once the clients are done re-enable module
removal.

Why?, beyond reducing complication:
1/ Tracking reference counts per-transaction in an efficient manner, as
   is currently done, requires a complicated scheme to avoid cache-line
   bouncing effects.
2/ Per-transaction ref-counting gives the false impression that a
   dma-driver can be gracefully removed ahead of its user (net, md, or
   dma-slave)
3/ None of the in-tree dma-drivers talk to hot pluggable hardware, but
   if such an engine were built one day we still would not need to notify
   clients of remove events.  The driver can simply return NULL to a
   ->prep() request, something that is much easier for a client to handle.

Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 crypto/async_tx/async_tx.c   |   4 -
 drivers/dma/dmaengine.c      | 205 +++++++++++++++++++++++++++----------------
 drivers/dma/dmatest.c        |   2 -
 drivers/dma/dw_dmac.c        |   2 +-
 drivers/mmc/host/atmel-mci.c |   4 +-
 include/linux/dmaengine.h    |  21 -----
 include/net/netdma.h         |   4 +-
 net/ipv4/tcp.c               |   1 -
 8 files changed, 132 insertions(+), 111 deletions(-)

(limited to 'include/linux')

diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c
index 8cfac182165d..43fe4cbe71e6 100644
--- a/crypto/async_tx/async_tx.c
+++ b/crypto/async_tx/async_tx.c
@@ -198,8 +198,6 @@ dma_channel_add_remove(struct dma_client *client,
 		/* add the channel to the generic management list */
 		master_ref = kmalloc(sizeof(*master_ref), GFP_KERNEL);
 		if (master_ref) {
-			/* keep a reference until async_tx is unloaded */
-			dma_chan_get(chan);
 			init_dma_chan_ref(master_ref, chan);
 			spin_lock_irqsave(&async_tx_lock, flags);
 			list_add_tail_rcu(&master_ref->node,
@@ -221,8 +219,6 @@ dma_channel_add_remove(struct dma_client *client,
 		spin_lock_irqsave(&async_tx_lock, flags);
 		list_for_each_entry(ref, &async_tx_master_list, node)
 			if (ref->chan == chan) {
-				/* permit backing devices to go away */
-				dma_chan_put(ref->chan);
 				list_del_rcu(&ref->node);
 				call_rcu(&ref->rcu, free_dma_chan_ref);
 				found = 1;
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index b9008932a8f3..d4d925912c47 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -74,6 +74,7 @@
 static DEFINE_MUTEX(dma_list_mutex);
 static LIST_HEAD(dma_device_list);
 static LIST_HEAD(dma_client_list);
+static long dmaengine_ref_count;
 
 /* --- sysfs implementation --- */
 
@@ -105,19 +106,8 @@ static ssize_t show_bytes_transferred(struct device *dev, struct device_attribut
 static ssize_t show_in_use(struct device *dev, struct device_attribute *attr, char *buf)
 {
 	struct dma_chan *chan = to_dma_chan(dev);
-	int in_use = 0;
-
-	if (unlikely(chan->slow_ref) &&
-		atomic_read(&chan->refcount.refcount) > 1)
-		in_use = 1;
-	else {
-		if (local_read(&(per_cpu_ptr(chan->local,
-			get_cpu())->refcount)) > 0)
-			in_use = 1;
-		put_cpu();
-	}
 
-	return sprintf(buf, "%d\n", in_use);
+	return sprintf(buf, "%d\n", chan->client_count);
 }
 
 static struct device_attribute dma_attrs[] = {
@@ -155,6 +145,78 @@ __dma_chan_satisfies_mask(struct dma_chan *chan, dma_cap_mask_t *want)
 	return bitmap_equal(want->bits, has.bits, DMA_TX_TYPE_END);
 }
 
+static struct module *dma_chan_to_owner(struct dma_chan *chan)
+{
+	return chan->device->dev->driver->owner;
+}
+
+/**
+ * balance_ref_count - catch up the channel reference count
+ * @chan - channel to balance ->client_count versus dmaengine_ref_count
+ *
+ * balance_ref_count must be called under dma_list_mutex
+ */
+static void balance_ref_count(struct dma_chan *chan)
+{
+	struct module *owner = dma_chan_to_owner(chan);
+
+	while (chan->client_count < dmaengine_ref_count) {
+		__module_get(owner);
+		chan->client_count++;
+	}
+}
+
+/**
+ * dma_chan_get - try to grab a dma channel's parent driver module
+ * @chan - channel to grab
+ *
+ * Must be called under dma_list_mutex
+ */
+static int dma_chan_get(struct dma_chan *chan)
+{
+	int err = -ENODEV;
+	struct module *owner = dma_chan_to_owner(chan);
+
+	if (chan->client_count) {
+		__module_get(owner);
+		err = 0;
+	} else if (try_module_get(owner))
+		err = 0;
+
+	if (err == 0)
+		chan->client_count++;
+
+	/* allocate upon first client reference */
+	if (chan->client_count == 1 && err == 0) {
+		int desc_cnt = chan->device->device_alloc_chan_resources(chan, NULL);
+
+		if (desc_cnt < 0) {
+			err = desc_cnt;
+			chan->client_count = 0;
+			module_put(owner);
+		} else
+			balance_ref_count(chan);
+	}
+
+	return err;
+}
+
+/**
+ * dma_chan_put - drop a reference to a dma channel's parent driver module
+ * @chan - channel to release
+ *
+ * Must be called under dma_list_mutex
+ */
+static void dma_chan_put(struct dma_chan *chan)
+{
+	if (!chan->client_count)
+		return; /* this channel failed alloc_chan_resources */
+	chan->client_count--;
+	module_put(dma_chan_to_owner(chan));
+	if (chan->client_count == 0)
+		chan->device->device_free_chan_resources(chan);
+}
+
 /**
  * dma_client_chan_alloc - try to allocate channels to a client
  * @client: &dma_client
@@ -165,7 +227,6 @@ static void dma_client_chan_alloc(struct dma_client *client)
 {
 	struct dma_device *device;
 	struct dma_chan *chan;
-	int desc;	/* allocated descriptor count */
 	enum dma_state_client ack;
 
 	/* Find a channel */
@@ -178,23 +239,16 @@ static void dma_client_chan_alloc(struct dma_client *client)
 		list_for_each_entry(chan, &device->channels, device_node) {
 			if (!dma_chan_satisfies_mask(chan, client->cap_mask))
 				continue;
+			if (!chan->client_count)
+				continue;
+			ack = client->event_callback(client, chan,
+						     DMA_RESOURCE_AVAILABLE);
 
-			desc = chan->device->device_alloc_chan_resources(
-					chan, client);
-			if (desc >= 0) {
-				ack = client->event_callback(client,
-						chan,
-						DMA_RESOURCE_AVAILABLE);
-
-				/* we are done once this client rejects
-				 * an available resource
-				 */
-				if (ack == DMA_ACK) {
-					dma_chan_get(chan);
-					chan->client_count++;
-				} else if (ack == DMA_NAK)
-					return;
-			}
+			/* we are done once this client rejects
+			 * an available resource
+			 */
+			if (ack == DMA_NAK)
+				return;
 		}
 	}
 }
@@ -224,7 +278,6 @@ EXPORT_SYMBOL(dma_sync_wait);
 void dma_chan_cleanup(struct kref *kref)
 {
 	struct dma_chan *chan = container_of(kref, struct dma_chan, refcount);
-	chan->device->device_free_chan_resources(chan);
 	kref_put(&chan->device->refcount, dma_async_device_cleanup);
 }
 EXPORT_SYMBOL(dma_chan_cleanup);
@@ -232,18 +285,12 @@ EXPORT_SYMBOL(dma_chan_cleanup);
 static void dma_chan_free_rcu(struct rcu_head *rcu)
 {
 	struct dma_chan *chan = container_of(rcu, struct dma_chan, rcu);
-	int bias = 0x7FFFFFFF;
-	int i;
-	for_each_possible_cpu(i)
-		bias -= local_read(&per_cpu_ptr(chan->local, i)->refcount);
-	atomic_sub(bias, &chan->refcount.refcount);
+
 	kref_put(&chan->refcount, dma_chan_cleanup);
 }
 
 static void dma_chan_release(struct dma_chan *chan)
 {
-	atomic_add(0x7FFFFFFF, &chan->refcount.refcount);
-	chan->slow_ref = 1;
 	call_rcu(&chan->rcu, dma_chan_free_rcu);
 }
 
@@ -262,44 +309,37 @@ static void dma_clients_notify_available(void)
 	mutex_unlock(&dma_list_mutex);
 }
 
-/**
- * dma_chans_notify_available - tell the clients that a channel is going away
- * @chan: channel on its way out
- */
-static void dma_clients_notify_removed(struct dma_chan *chan)
-{
-	struct dma_client *client;
-	enum dma_state_client ack;
-
-	mutex_lock(&dma_list_mutex);
-
-	list_for_each_entry(client, &dma_client_list, global_node) {
-		ack = client->event_callback(client, chan,
-				DMA_RESOURCE_REMOVED);
-
-		/* client was holding resources for this channel so
-		 * free it
-		 */
-		if (ack == DMA_ACK) {
-			dma_chan_put(chan);
-			chan->client_count--;
-		}
-	}
-
-	mutex_unlock(&dma_list_mutex);
-}
-
 /**
  * dma_async_client_register - register a &dma_client
  * @client: ptr to a client structure with valid 'event_callback' and 'cap_mask'
  */
 void dma_async_client_register(struct dma_client *client)
 {
+	struct dma_device *device, *_d;
+	struct dma_chan *chan;
+	int err;
+
 	/* validate client data */
 	BUG_ON(dma_has_cap(DMA_SLAVE, client->cap_mask) &&
 		!client->slave);
 
 	mutex_lock(&dma_list_mutex);
+	dmaengine_ref_count++;
+
+	/* try to grab channels */
+	list_for_each_entry_safe(device, _d, &dma_device_list, global_node)
+		list_for_each_entry(chan, &device->channels, device_node) {
+			err = dma_chan_get(chan);
+			if (err == -ENODEV) {
+				/* module removed before we could use it */
+				list_del_init(&device->global_node);
+				break;
+			} else if (err)
+				pr_err("dmaengine: failed to get %s: (%d)\n",
+				       dev_name(&chan->dev), err);
+		}
+
+
 	list_add_tail(&client->global_node, &dma_client_list);
 	mutex_unlock(&dma_list_mutex);
 }
@@ -315,23 +355,17 @@ void dma_async_client_unregister(struct dma_client *client)
 {
 	struct dma_device *device;
 	struct dma_chan *chan;
-	enum dma_state_client ack;
 
 	if (!client)
 		return;
 
 	mutex_lock(&dma_list_mutex);
-	/* free all channels the client is holding */
+	dmaengine_ref_count--;
+	BUG_ON(dmaengine_ref_count < 0);
+	/* drop channel references */
 	list_for_each_entry(device, &dma_device_list, global_node)
-		list_for_each_entry(chan, &device->channels, device_node) {
-			ack = client->event_callback(client, chan,
-				DMA_RESOURCE_REMOVED);
-
-			if (ack == DMA_ACK) {
-				dma_chan_put(chan);
-				chan->client_count--;
-			}
-		}
+		list_for_each_entry(chan, &device->channels, device_node)
+			dma_chan_put(chan);
 
 	list_del(&client->global_node);
 	mutex_unlock(&dma_list_mutex);
@@ -423,6 +457,21 @@ int dma_async_device_register(struct dma_device *device)
 	}
 
 	mutex_lock(&dma_list_mutex);
+	if (dmaengine_ref_count)
+		list_for_each_entry(chan, &device->channels, device_node) {
+			/* if clients are already waiting for channels we need
+			 * to take references on their behalf
+			 */
+			if (dma_chan_get(chan) == -ENODEV) {
+				/* note we can only get here for the first
+				 * channel as the remaining channels are
+				 * guaranteed to get a reference
+				 */
+				rc = -ENODEV;
+				mutex_unlock(&dma_list_mutex);
+				goto err_out;
+			}
+		}
 	list_add_tail(&device->global_node, &dma_device_list);
 	mutex_unlock(&dma_list_mutex);
 
@@ -456,7 +505,7 @@ static void dma_async_device_cleanup(struct kref *kref)
 }
 
 /**
- * dma_async_device_unregister - unregisters DMA devices
+ * dma_async_device_unregister - unregister a DMA device
  * @device: &dma_device
  */
 void dma_async_device_unregister(struct dma_device *device)
@@ -468,7 +517,9 @@ void dma_async_device_unregister(struct dma_device *device)
 	mutex_unlock(&dma_list_mutex);
 
 	list_for_each_entry(chan, &device->channels, device_node) {
-		dma_clients_notify_removed(chan);
+		WARN_ONCE(chan->client_count,
+			  "%s called while %d clients hold a reference\n",
+			  __func__, chan->client_count);
 		device_unregister(&chan->dev);
 		dma_chan_release(chan);
 	}
diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c
index ed9636bfb54a..db4050884713 100644
--- a/drivers/dma/dmatest.c
+++ b/drivers/dma/dmatest.c
@@ -215,7 +215,6 @@ static int dmatest_func(void *data)
 
 	smp_rmb();
 	chan = thread->chan;
-	dma_chan_get(chan);
 
 	while (!kthread_should_stop()) {
 		total_tests++;
@@ -293,7 +292,6 @@ static int dmatest_func(void *data)
 	}
 
 	ret = 0;
-	dma_chan_put(chan);
 	kfree(thread->dstbuf);
 err_dstbuf:
 	kfree(thread->srcbuf);
diff --git a/drivers/dma/dw_dmac.c b/drivers/dma/dw_dmac.c
index 0778d99aea7c..377dafa37a20 100644
--- a/drivers/dma/dw_dmac.c
+++ b/drivers/dma/dw_dmac.c
@@ -773,7 +773,7 @@ static int dwc_alloc_chan_resources(struct dma_chan *chan,
 	dev_vdbg(&chan->dev, "alloc_chan_resources\n");
 
 	/* Channels doing slave DMA can only handle one client. */
-	if (dwc->dws || client->slave) {
+	if (dwc->dws || (client && client->slave)) {
 		if (chan->client_count)
 			return -EBUSY;
 	}
diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
index 7a3f2436b011..6c11f4d4c4e9 100644
--- a/drivers/mmc/host/atmel-mci.c
+++ b/drivers/mmc/host/atmel-mci.c
@@ -593,10 +593,8 @@ atmci_submit_data_dma(struct atmel_mci *host, struct mmc_data *data)
 
 	/* If we don't have a channel, we can't do DMA */
 	chan = host->dma.chan;
-	if (chan) {
-		dma_chan_get(chan);
+	if (chan)
 		host->data_chan = chan;
-	}
 
 	if (!chan)
 		return -ENODEV;
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index e4ec7e7b8056..d18d37d1015d 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -165,7 +165,6 @@ struct dma_slave {
  */
 
 struct dma_chan_percpu {
-	local_t refcount;
 	/* stats */
 	unsigned long memcpy_count;
 	unsigned long bytes_transferred;
@@ -205,26 +204,6 @@ struct dma_chan {
 
 void dma_chan_cleanup(struct kref *kref);
 
-static inline void dma_chan_get(struct dma_chan *chan)
-{
-	if (unlikely(chan->slow_ref))
-		kref_get(&chan->refcount);
-	else {
-		local_inc(&(per_cpu_ptr(chan->local, get_cpu())->refcount));
-		put_cpu();
-	}
-}
-
-static inline void dma_chan_put(struct dma_chan *chan)
-{
-	if (unlikely(chan->slow_ref))
-		kref_put(&chan->refcount, dma_chan_cleanup);
-	else {
-		local_dec(&(per_cpu_ptr(chan->local, get_cpu())->refcount));
-		put_cpu();
-	}
-}
-
 /*
  * typedef dma_event_callback - function pointer to a DMA event callback
  * For each channel added to the system this routine is called for each client.
diff --git a/include/net/netdma.h b/include/net/netdma.h
index f28c6e064e8f..cbe2737f4a61 100644
--- a/include/net/netdma.h
+++ b/include/net/netdma.h
@@ -27,11 +27,11 @@
 static inline struct dma_chan *get_softnet_dma(void)
 {
 	struct dma_chan *chan;
+
 	rcu_read_lock();
 	chan = rcu_dereference(__get_cpu_var(softnet_data).net_dma);
-	if (chan)
-		dma_chan_get(chan);
 	rcu_read_unlock();
+
 	return chan;
 }
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index f28acf11fc67..75e0e0a2d8db 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1632,7 +1632,6 @@ skip_copy:
 
 		/* Safe to free early-copied skbs now */
 		__skb_queue_purge(&sk->sk_async_wait_queue);
-		dma_chan_put(tp->ucopy.dma_chan);
 		tp->ucopy.dma_chan = NULL;
 	}
 	if (tp->ucopy.pinned_list) {
-- 
cgit v1.2.3


From bec085134e446577a983f17f57d642a88d1af53b Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 6 Jan 2009 11:38:14 -0700
Subject: dmaengine: centralize channel allocation, introduce dma_find_channel

Allowing multiple clients to each define their own channel allocation
scheme quickly leads to a pathological situation.  For memory-to-memory
offload all clients can share a central allocator.

This simply moves the existing async_tx allocator to dmaengine with
minimal fixups:
* async_tx.c:get_chan_ref_by_cap --> dmaengine.c:nth_chan
* async_tx.c:async_tx_rebalance --> dmaengine.c:dma_channel_rebalance
* split out common code from async_tx.c:__async_tx_find_channel -->
  dma_find_channel

Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 crypto/async_tx/async_tx.c | 146 ++-------------------------------------
 drivers/dma/dmaengine.c    | 168 ++++++++++++++++++++++++++++++++++++++++++++-
 include/linux/dmaengine.h  |   3 +
 3 files changed, 174 insertions(+), 143 deletions(-)

(limited to 'include/linux')

diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c
index 43fe4cbe71e6..b88bb1f608fc 100644
--- a/crypto/async_tx/async_tx.c
+++ b/crypto/async_tx/async_tx.c
@@ -37,26 +37,11 @@ static struct dma_client async_tx_dma = {
 	/* .cap_mask == 0 defaults to all channels */
 };
 
-/**
- * dma_cap_mask_all - enable iteration over all operation types
- */
-static dma_cap_mask_t dma_cap_mask_all;
-
-/**
- * chan_ref_percpu - tracks channel allocations per core/opertion
- */
-struct chan_ref_percpu {
-	struct dma_chan_ref *ref;
-};
-
-static int channel_table_initialized;
-static struct chan_ref_percpu *channel_table[DMA_TX_TYPE_END];
-
 /**
  * async_tx_lock - protect modification of async_tx_master_list and serialize
  *	rebalance operations
  */
-static spinlock_t async_tx_lock;
+static DEFINE_SPINLOCK(async_tx_lock);
 
 static LIST_HEAD(async_tx_master_list);
 
@@ -89,85 +74,6 @@ init_dma_chan_ref(struct dma_chan_ref *ref, struct dma_chan *chan)
 	atomic_set(&ref->count, 0);
 }
 
-/**
- * get_chan_ref_by_cap - returns the nth channel of the given capability
- * 	defaults to returning the channel with the desired capability and the
- * 	lowest reference count if the index can not be satisfied
- * @cap: capability to match
- * @index: nth channel desired, passing -1 has the effect of forcing the
- *  default return value
- */
-static struct dma_chan_ref *
-get_chan_ref_by_cap(enum dma_transaction_type cap, int index)
-{
-	struct dma_chan_ref *ret_ref = NULL, *min_ref = NULL, *ref;
-
-	rcu_read_lock();
-	list_for_each_entry_rcu(ref, &async_tx_master_list, node)
-		if (dma_has_cap(cap, ref->chan->device->cap_mask)) {
-			if (!min_ref)
-				min_ref = ref;
-			else if (atomic_read(&ref->count) <
-				atomic_read(&min_ref->count))
-				min_ref = ref;
-
-			if (index-- == 0) {
-				ret_ref = ref;
-				break;
-			}
-		}
-	rcu_read_unlock();
-
-	if (!ret_ref)
-		ret_ref = min_ref;
-
-	if (ret_ref)
-		atomic_inc(&ret_ref->count);
-
-	return ret_ref;
-}
-
-/**
- * async_tx_rebalance - redistribute the available channels, optimize
- * for cpu isolation in the SMP case, and opertaion isolation in the
- * uniprocessor case
- */
-static void async_tx_rebalance(void)
-{
-	int cpu, cap, cpu_idx = 0;
-	unsigned long flags;
-
-	if (!channel_table_initialized)
-		return;
-
-	spin_lock_irqsave(&async_tx_lock, flags);
-
-	/* undo the last distribution */
-	for_each_dma_cap_mask(cap, dma_cap_mask_all)
-		for_each_possible_cpu(cpu) {
-			struct dma_chan_ref *ref =
-				per_cpu_ptr(channel_table[cap], cpu)->ref;
-			if (ref) {
-				atomic_set(&ref->count, 0);
-				per_cpu_ptr(channel_table[cap], cpu)->ref =
-									NULL;
-			}
-		}
-
-	for_each_dma_cap_mask(cap, dma_cap_mask_all)
-		for_each_online_cpu(cpu) {
-			struct dma_chan_ref *new;
-			if (NR_CPUS > 1)
-				new = get_chan_ref_by_cap(cap, cpu_idx++);
-			else
-				new = get_chan_ref_by_cap(cap, -1);
-
-			per_cpu_ptr(channel_table[cap], cpu)->ref = new;
-		}
-
-	spin_unlock_irqrestore(&async_tx_lock, flags);
-}
-
 static enum dma_state_client
 dma_channel_add_remove(struct dma_client *client,
 	struct dma_chan *chan, enum dma_state state)
@@ -211,8 +117,6 @@ dma_channel_add_remove(struct dma_client *client,
 				" (-ENOMEM)\n");
 			return 0;
 		}
-
-		async_tx_rebalance();
 		break;
 	case DMA_RESOURCE_REMOVED:
 		found = 0;
@@ -233,8 +137,6 @@ dma_channel_add_remove(struct dma_client *client,
 			ack = DMA_ACK;
 		else
 			break;
-
-		async_tx_rebalance();
 		break;
 	case DMA_RESOURCE_SUSPEND:
 	case DMA_RESOURCE_RESUME:
@@ -248,51 +150,18 @@ dma_channel_add_remove(struct dma_client *client,
 	return ack;
 }
 
-static int __init
-async_tx_init(void)
+static int __init async_tx_init(void)
 {
-	enum dma_transaction_type cap;
-
-	spin_lock_init(&async_tx_lock);
-	bitmap_fill(dma_cap_mask_all.bits, DMA_TX_TYPE_END);
-
-	/* an interrupt will never be an explicit operation type.
-	 * clearing this bit prevents allocation to a slot in 'channel_table'
-	 */
-	clear_bit(DMA_INTERRUPT, dma_cap_mask_all.bits);
-
-	for_each_dma_cap_mask(cap, dma_cap_mask_all) {
-		channel_table[cap] = alloc_percpu(struct chan_ref_percpu);
-		if (!channel_table[cap])
-			goto err;
-	}
-
-	channel_table_initialized = 1;
 	dma_async_client_register(&async_tx_dma);
 	dma_async_client_chan_request(&async_tx_dma);
 
 	printk(KERN_INFO "async_tx: api initialized (async)\n");
 
 	return 0;
-err:
-	printk(KERN_ERR "async_tx: initialization failure\n");
-
-	while (--cap >= 0)
-		free_percpu(channel_table[cap]);
-
-	return 1;
 }
 
 static void __exit async_tx_exit(void)
 {
-	enum dma_transaction_type cap;
-
-	channel_table_initialized = 0;
-
-	for_each_dma_cap_mask(cap, dma_cap_mask_all)
-		if (channel_table[cap])
-			free_percpu(channel_table[cap]);
-
 	dma_async_client_unregister(&async_tx_dma);
 }
 
@@ -308,16 +177,9 @@ __async_tx_find_channel(struct dma_async_tx_descriptor *depend_tx,
 {
 	/* see if we can keep the chain on one channel */
 	if (depend_tx &&
-		dma_has_cap(tx_type, depend_tx->chan->device->cap_mask))
+	    dma_has_cap(tx_type, depend_tx->chan->device->cap_mask))
 		return depend_tx->chan;
-	else if (likely(channel_table_initialized)) {
-		struct dma_chan_ref *ref;
-		int cpu = get_cpu();
-		ref = per_cpu_ptr(channel_table[tx_type], cpu)->ref;
-		put_cpu();
-		return ref ? ref->chan : NULL;
-	} else
-		return NULL;
+	return dma_find_channel(tx_type);
 }
 EXPORT_SYMBOL_GPL(__async_tx_find_channel);
 #else
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index d4d925912c47..87a8cd4791ed 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -294,6 +294,164 @@ static void dma_chan_release(struct dma_chan *chan)
 	call_rcu(&chan->rcu, dma_chan_free_rcu);
 }
 
+/**
+ * dma_cap_mask_all - enable iteration over all operation types
+ */
+static dma_cap_mask_t dma_cap_mask_all;
+
+/**
+ * dma_chan_tbl_ent - tracks channel allocations per core/operation
+ * @chan - associated channel for this entry
+ */
+struct dma_chan_tbl_ent {
+	struct dma_chan *chan;
+};
+
+/**
+ * channel_table - percpu lookup table for memory-to-memory offload providers
+ */
+static struct dma_chan_tbl_ent *channel_table[DMA_TX_TYPE_END];
+
+static int __init dma_channel_table_init(void)
+{
+	enum dma_transaction_type cap;
+	int err = 0;
+
+	bitmap_fill(dma_cap_mask_all.bits, DMA_TX_TYPE_END);
+
+	/* 'interrupt' and 'slave' are channel capabilities, but are not
+	 * associated with an operation so they do not need an entry in the
+	 * channel_table
+	 */
+	clear_bit(DMA_INTERRUPT, dma_cap_mask_all.bits);
+	clear_bit(DMA_SLAVE, dma_cap_mask_all.bits);
+
+	for_each_dma_cap_mask(cap, dma_cap_mask_all) {
+		channel_table[cap] = alloc_percpu(struct dma_chan_tbl_ent);
+		if (!channel_table[cap]) {
+			err = -ENOMEM;
+			break;
+		}
+	}
+
+	if (err) {
+		pr_err("dmaengine: initialization failure\n");
+		for_each_dma_cap_mask(cap, dma_cap_mask_all)
+			if (channel_table[cap])
+				free_percpu(channel_table[cap]);
+	}
+
+	return err;
+}
+subsys_initcall(dma_channel_table_init);
+
+/**
+ * dma_find_channel - find a channel to carry out the operation
+ * @tx_type: transaction type
+ */
+struct dma_chan *dma_find_channel(enum dma_transaction_type tx_type)
+{
+	struct dma_chan *chan;
+	int cpu;
+
+	WARN_ONCE(dmaengine_ref_count == 0,
+		  "client called %s without a reference", __func__);
+
+	cpu = get_cpu();
+	chan = per_cpu_ptr(channel_table[tx_type], cpu)->chan;
+	put_cpu();
+
+	return chan;
+}
+EXPORT_SYMBOL(dma_find_channel);
+
+/**
+ * nth_chan - returns the nth channel of the given capability
+ * @cap: capability to match
+ * @n: nth channel desired
+ *
+ * Defaults to returning the channel with the desired capability and the
+ * lowest reference count when 'n' cannot be satisfied.  Must be called
+ * under dma_list_mutex.
+ */
+static struct dma_chan *nth_chan(enum dma_transaction_type cap, int n)
+{
+	struct dma_device *device;
+	struct dma_chan *chan;
+	struct dma_chan *ret = NULL;
+	struct dma_chan *min = NULL;
+
+	list_for_each_entry(device, &dma_device_list, global_node) {
+		if (!dma_has_cap(cap, device->cap_mask))
+			continue;
+		list_for_each_entry(chan, &device->channels, device_node) {
+			if (!chan->client_count)
+				continue;
+			if (!min)
+				min = chan;
+			else if (chan->table_count < min->table_count)
+				min = chan;
+
+			if (n-- == 0) {
+				ret = chan;
+				break; /* done */
+			}
+		}
+		if (ret)
+			break; /* done */
+	}
+
+	if (!ret)
+		ret = min;
+
+	if (ret)
+		ret->table_count++;
+
+	return ret;
+}
+
+/**
+ * dma_channel_rebalance - redistribute the available channels
+ *
+ * Optimize for cpu isolation (each cpu gets a dedicated channel for an
+ * operation type) in the SMP case,  and operation isolation (avoid
+ * multi-tasking channels) in the non-SMP case.  Must be called under
+ * dma_list_mutex.
+ */
+static void dma_channel_rebalance(void)
+{
+	struct dma_chan *chan;
+	struct dma_device *device;
+	int cpu;
+	int cap;
+	int n;
+
+	/* undo the last distribution */
+	for_each_dma_cap_mask(cap, dma_cap_mask_all)
+		for_each_possible_cpu(cpu)
+			per_cpu_ptr(channel_table[cap], cpu)->chan = NULL;
+
+	list_for_each_entry(device, &dma_device_list, global_node)
+		list_for_each_entry(chan, &device->channels, device_node)
+			chan->table_count = 0;
+
+	/* don't populate the channel_table if no clients are available */
+	if (!dmaengine_ref_count)
+		return;
+
+	/* redistribute available channels */
+	n = 0;
+	for_each_dma_cap_mask(cap, dma_cap_mask_all)
+		for_each_online_cpu(cpu) {
+			if (num_possible_cpus() > 1)
+				chan = nth_chan(cap, n++);
+			else
+				chan = nth_chan(cap, -1);
+
+			per_cpu_ptr(channel_table[cap], cpu)->chan = chan;
+		}
+}
+
 /**
  * dma_chans_notify_available - broadcast available channels to the clients
  */
@@ -339,7 +497,12 @@ void dma_async_client_register(struct dma_client *client)
 				       dev_name(&chan->dev), err);
 		}
 
-
+	/* if this is the first reference and there were channels
+	 * waiting we need to rebalance to get those channels
+	 * incorporated into the channel table
+	 */
+	if (dmaengine_ref_count == 1)
+		dma_channel_rebalance();
 	list_add_tail(&client->global_node, &dma_client_list);
 	mutex_unlock(&dma_list_mutex);
 }
@@ -473,6 +636,7 @@ int dma_async_device_register(struct dma_device *device)
 			}
 		}
 	list_add_tail(&device->global_node, &dma_device_list);
+	dma_channel_rebalance();
 	mutex_unlock(&dma_list_mutex);
 
 	dma_clients_notify_available();
@@ -514,6 +678,7 @@ void dma_async_device_unregister(struct dma_device *device)
 
 	mutex_lock(&dma_list_mutex);
 	list_del(&device->global_node);
+	dma_channel_rebalance();
 	mutex_unlock(&dma_list_mutex);
 
 	list_for_each_entry(chan, &device->channels, device_node) {
@@ -768,3 +933,4 @@ static int __init dma_bus_init(void)
 }
 subsys_initcall(dma_bus_init);
 
+
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index d18d37d1015d..b466f02e2433 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -182,6 +182,7 @@ struct dma_chan_percpu {
  * @device_node: used to add this to the device chan list
  * @local: per-cpu pointer to a struct dma_chan_percpu
  * @client-count: how many clients are using this channel
+ * @table_count: number of appearances in the mem-to-mem allocation table
  */
 struct dma_chan {
 	struct dma_device *device;
@@ -198,6 +199,7 @@ struct dma_chan {
 	struct list_head device_node;
 	struct dma_chan_percpu *local;
 	int client_count;
+	int table_count;
 };
 
 #define to_dma_chan(p) container_of(p, struct dma_chan, dev)
@@ -468,6 +470,7 @@ static inline enum dma_status dma_wait_for_async_tx(struct dma_async_tx_descript
 int dma_async_device_register(struct dma_device *device);
 void dma_async_device_unregister(struct dma_device *device);
 void dma_run_dependencies(struct dma_async_tx_descriptor *tx);
+struct dma_chan *dma_find_channel(enum dma_transaction_type tx_type);
 
 /* --- Helper iov-locking functions --- */
 
-- 
cgit v1.2.3


From 2ba05622b8b143b0c95968ba59bddfbd6d2f2559 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 6 Jan 2009 11:38:14 -0700
Subject: dmaengine: provide a common 'issue_pending_all' implementation

async_tx and net_dma each have open-coded versions of issue_pending_all,
so provide a common routine in dmaengine.

The implementation needs to walk the global device list, so implement
rcu to allow dma_issue_pending_all to run lockless.  Clients protect
themselves from channel removal events by holding a dmaengine reference.

Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 crypto/async_tx/async_tx.c | 12 ------------
 drivers/dma/dmaengine.c    | 27 ++++++++++++++++++++++++---
 include/linux/async_tx.h   |  2 +-
 include/linux/dmaengine.h  |  1 +
 net/core/dev.c             |  9 +--------
 5 files changed, 27 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c
index b88bb1f608fc..2cdf7a0867b7 100644
--- a/crypto/async_tx/async_tx.c
+++ b/crypto/async_tx/async_tx.c
@@ -45,18 +45,6 @@ static DEFINE_SPINLOCK(async_tx_lock);
 
 static LIST_HEAD(async_tx_master_list);
 
-/* async_tx_issue_pending_all - start all transactions on all channels */
-void async_tx_issue_pending_all(void)
-{
-	struct dma_chan_ref *ref;
-
-	rcu_read_lock();
-	list_for_each_entry_rcu(ref, &async_tx_master_list, node)
-		ref->chan->device->device_issue_pending(ref->chan);
-	rcu_read_unlock();
-}
-EXPORT_SYMBOL_GPL(async_tx_issue_pending_all);
-
 static void
 free_dma_chan_ref(struct rcu_head *rcu)
 {
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 87a8cd4791ed..418eca28d472 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -70,6 +70,7 @@
 #include <linux/rcupdate.h>
 #include <linux/mutex.h>
 #include <linux/jiffies.h>
+#include <linux/rculist.h>
 
 static DEFINE_MUTEX(dma_list_mutex);
 static LIST_HEAD(dma_device_list);
@@ -365,6 +366,26 @@ struct dma_chan *dma_find_channel(enum dma_transaction_type tx_type)
 }
 EXPORT_SYMBOL(dma_find_channel);
 
+/**
+ * dma_issue_pending_all - flush all pending operations across all channels
+ */
+void dma_issue_pending_all(void)
+{
+	struct dma_device *device;
+	struct dma_chan *chan;
+
+	WARN_ONCE(dmaengine_ref_count == 0,
+		  "client called %s without a reference", __func__);
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(device, &dma_device_list, global_node)
+		list_for_each_entry(chan, &device->channels, device_node)
+			if (chan->client_count)
+				device->device_issue_pending(chan);
+	rcu_read_unlock();
+}
+EXPORT_SYMBOL(dma_issue_pending_all);
+
 /**
  * nth_chan - returns the nth channel of the given capability
  * @cap: capability to match
@@ -490,7 +511,7 @@ void dma_async_client_register(struct dma_client *client)
 			err = dma_chan_get(chan);
 			if (err == -ENODEV) {
 				/* module removed before we could use it */
-				list_del_init(&device->global_node);
+				list_del_rcu(&device->global_node);
 				break;
 			} else if (err)
 				pr_err("dmaengine: failed to get %s: (%d)\n",
@@ -635,7 +656,7 @@ int dma_async_device_register(struct dma_device *device)
 				goto err_out;
 			}
 		}
-	list_add_tail(&device->global_node, &dma_device_list);
+	list_add_tail_rcu(&device->global_node, &dma_device_list);
 	dma_channel_rebalance();
 	mutex_unlock(&dma_list_mutex);
 
@@ -677,7 +698,7 @@ void dma_async_device_unregister(struct dma_device *device)
 	struct dma_chan *chan;
 
 	mutex_lock(&dma_list_mutex);
-	list_del(&device->global_node);
+	list_del_rcu(&device->global_node);
 	dma_channel_rebalance();
 	mutex_unlock(&dma_list_mutex);
 
diff --git a/include/linux/async_tx.h b/include/linux/async_tx.h
index 1c816775f135..45f6297821bd 100644
--- a/include/linux/async_tx.h
+++ b/include/linux/async_tx.h
@@ -59,7 +59,7 @@ enum async_tx_flags {
 };
 
 #ifdef CONFIG_DMA_ENGINE
-void async_tx_issue_pending_all(void);
+#define async_tx_issue_pending_all dma_issue_pending_all
 #ifdef CONFIG_ARCH_HAS_ASYNC_TX_FIND_CHANNEL
 #include <asm/async_tx.h>
 #else
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index b466f02e2433..57a43adfc39e 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -471,6 +471,7 @@ int dma_async_device_register(struct dma_device *device);
 void dma_async_device_unregister(struct dma_device *device);
 void dma_run_dependencies(struct dma_async_tx_descriptor *tx);
 struct dma_chan *dma_find_channel(enum dma_transaction_type tx_type);
+void dma_issue_pending_all(void);
 
 /* --- Helper iov-locking functions --- */
 
diff --git a/net/core/dev.c b/net/core/dev.c
index 09c66a449da6..e40b0d57f8ff 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2635,14 +2635,7 @@ out:
 	 * There may not be any more sk_buffs coming right now, so push
 	 * any pending DMA copies to hardware
 	 */
-	if (!cpus_empty(net_dma.channel_mask)) {
-		int chan_idx;
-		for_each_cpu_mask_nr(chan_idx, net_dma.channel_mask) {
-			struct dma_chan *chan = net_dma.channels[chan_idx];
-			if (chan)
-				dma_async_memcpy_issue_pending(chan);
-		}
-	}
+	dma_issue_pending_all();
 #endif
 
 	return;
-- 
cgit v1.2.3


From f67b45999205164958de4ec0658d51fa4bee066d Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 6 Jan 2009 11:38:15 -0700
Subject: net_dma: convert to dma_find_channel

Use the general-purpose channel allocation provided by dmaengine.

Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 include/linux/netdevice.h |  3 ---
 include/net/netdma.h      | 11 -----------
 net/core/dev.c            | 40 ----------------------------------------
 net/ipv4/tcp.c            |  4 ++--
 net/ipv4/tcp_input.c      |  2 +-
 net/ipv4/tcp_ipv4.c       |  2 +-
 net/ipv6/tcp_ipv6.c       |  2 +-
 7 files changed, 5 insertions(+), 59 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 41e1224651cf..bac2c458d9b8 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1113,9 +1113,6 @@ struct softnet_data
 	struct sk_buff		*completion_queue;
 
 	struct napi_struct	backlog;
-#ifdef CONFIG_NET_DMA
-	struct dma_chan		*net_dma;
-#endif
 };
 
 DECLARE_PER_CPU(struct softnet_data,softnet_data);
diff --git a/include/net/netdma.h b/include/net/netdma.h
index cbe2737f4a61..8ba8ce284eeb 100644
--- a/include/net/netdma.h
+++ b/include/net/netdma.h
@@ -24,17 +24,6 @@
 #include <linux/dmaengine.h>
 #include <linux/skbuff.h>
 
-static inline struct dma_chan *get_softnet_dma(void)
-{
-	struct dma_chan *chan;
-
-	rcu_read_lock();
-	chan = rcu_dereference(__get_cpu_var(softnet_data).net_dma);
-	rcu_read_unlock();
-
-	return chan;
-}
-
 int dma_skb_copy_datagram_iovec(struct dma_chan* chan,
 		struct sk_buff *skb, int offset, struct iovec *to,
 		size_t len, struct dma_pinned_list *pinned_list);
diff --git a/net/core/dev.c b/net/core/dev.c
index e40b0d57f8ff..bbb07dbe1740 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4827,44 +4827,6 @@ static int dev_cpu_callback(struct notifier_block *nfb,
 }
 
 #ifdef CONFIG_NET_DMA
-/**
- * net_dma_rebalance - try to maintain one DMA channel per CPU
- * @net_dma: DMA client and associated data (lock, channels, channel_mask)
- *
- * This is called when the number of channels allocated to the net_dma client
- * changes.  The net_dma client tries to have one DMA channel per CPU.
- */
-
-static void net_dma_rebalance(struct net_dma *net_dma)
-{
-	unsigned int cpu, i, n, chan_idx;
-	struct dma_chan *chan;
-
-	if (cpus_empty(net_dma->channel_mask)) {
-		for_each_online_cpu(cpu)
-			rcu_assign_pointer(per_cpu(softnet_data, cpu).net_dma, NULL);
-		return;
-	}
-
-	i = 0;
-	cpu = first_cpu(cpu_online_map);
-
-	for_each_cpu_mask_nr(chan_idx, net_dma->channel_mask) {
-		chan = net_dma->channels[chan_idx];
-
-		n = ((num_online_cpus() / cpus_weight(net_dma->channel_mask))
-		   + (i < (num_online_cpus() %
-			cpus_weight(net_dma->channel_mask)) ? 1 : 0));
-
-		while(n) {
-			per_cpu(softnet_data, cpu).net_dma = chan;
-			cpu = next_cpu(cpu, cpu_online_map);
-			n--;
-		}
-		i++;
-	}
-}
-
 /**
  * netdev_dma_event - event callback for the net_dma_client
  * @client: should always be net_dma_client
@@ -4894,7 +4856,6 @@ netdev_dma_event(struct dma_client *client, struct dma_chan *chan,
 			ack = DMA_ACK;
 			net_dma->channels[pos] = chan;
 			cpu_set(pos, net_dma->channel_mask);
-			net_dma_rebalance(net_dma);
 		}
 		break;
 	case DMA_RESOURCE_REMOVED:
@@ -4909,7 +4870,6 @@ netdev_dma_event(struct dma_client *client, struct dma_chan *chan,
 			ack = DMA_ACK;
 			cpu_clear(pos, net_dma->channel_mask);
 			net_dma->channels[i] = NULL;
-			net_dma_rebalance(net_dma);
 		}
 		break;
 	default:
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 75e0e0a2d8db..9b275abc8eb9 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1317,7 +1317,7 @@ int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 		if ((available < target) &&
 		    (len > sysctl_tcp_dma_copybreak) && !(flags & MSG_PEEK) &&
 		    !sysctl_tcp_low_latency &&
-		    __get_cpu_var(softnet_data).net_dma) {
+		    dma_find_channel(DMA_MEMCPY)) {
 			preempt_enable_no_resched();
 			tp->ucopy.pinned_list =
 					dma_pin_iovec_pages(msg->msg_iov, len);
@@ -1527,7 +1527,7 @@ do_prequeue:
 		if (!(flags & MSG_TRUNC)) {
 #ifdef CONFIG_NET_DMA
 			if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list)
-				tp->ucopy.dma_chan = get_softnet_dma();
+				tp->ucopy.dma_chan = dma_find_channel(DMA_MEMCPY);
 
 			if (tp->ucopy.dma_chan) {
 				tp->ucopy.dma_cookie = dma_skb_copy_datagram_iovec(
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 99b7ecbe8893..a6961d75c7ea 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -5005,7 +5005,7 @@ static int tcp_dma_try_early_copy(struct sock *sk, struct sk_buff *skb,
 		return 0;
 
 	if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list)
-		tp->ucopy.dma_chan = get_softnet_dma();
+		tp->ucopy.dma_chan = dma_find_channel(DMA_MEMCPY);
 
 	if (tp->ucopy.dma_chan && skb_csum_unnecessary(skb)) {
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 9d839fa9331e..19d7b429a262 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1594,7 +1594,7 @@ process:
 #ifdef CONFIG_NET_DMA
 		struct tcp_sock *tp = tcp_sk(sk);
 		if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list)
-			tp->ucopy.dma_chan = get_softnet_dma();
+			tp->ucopy.dma_chan = dma_find_channel(DMA_MEMCPY);
 		if (tp->ucopy.dma_chan)
 			ret = tcp_v4_do_rcv(sk, skb);
 		else
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index e8b8337a8310..71cd70951d7d 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1640,7 +1640,7 @@ process:
 #ifdef CONFIG_NET_DMA
 		struct tcp_sock *tp = tcp_sk(sk);
 		if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list)
-			tp->ucopy.dma_chan = get_softnet_dma();
+			tp->ucopy.dma_chan = dma_find_channel(DMA_MEMCPY);
 		if (tp->ucopy.dma_chan)
 			ret = tcp_v6_do_rcv(sk, skb);
 		else
-- 
cgit v1.2.3


From 59b5ec21446b9239d706ab237fb261d525b75e81 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 6 Jan 2009 11:38:15 -0700
Subject: dmaengine: introduce dma_request_channel and private channels

This interface is primarily for device-to-memory clients which need to
search for dma channels with platform-specific characteristics.  The
prototype is:

struct dma_chan *dma_request_channel(dma_cap_mask_t mask,
                                     dma_filter_fn filter_fn,
                                     void *filter_param);

When the optional 'filter_fn' parameter is set to NULL
dma_request_channel simply returns the first channel that satisfies the
capability mask.  Otherwise, when the mask parameter is insufficient for
specifying the necessary channel, the filter_fn routine can be used to
disposition the available channels in the system. The filter_fn routine
is called once for each free channel in the system.  Upon seeing a
suitable channel filter_fn returns DMA_ACK which flags that channel to
be the return value from dma_request_channel.  A channel allocated via
this interface is exclusive to the caller, until dma_release_channel()
is called.

To ensure that all channels are not consumed by the general-purpose
allocator the DMA_PRIVATE capability is provided to exclude a dma_device
from general-purpose (memory-to-memory) consideration.

Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dma/dmaengine.c   | 155 +++++++++++++++++++++++++++++++++++++++++-----
 include/linux/dmaengine.h |  16 +++++
 2 files changed, 155 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 418eca28d472..7a0594f24a3f 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -134,14 +134,14 @@ static struct class dma_devclass = {
 
 /* --- client and device registration --- */
 
-#define dma_chan_satisfies_mask(chan, mask) \
-	__dma_chan_satisfies_mask((chan), &(mask))
+#define dma_device_satisfies_mask(device, mask) \
+	__dma_device_satisfies_mask((device), &(mask))
 static int
-__dma_chan_satisfies_mask(struct dma_chan *chan, dma_cap_mask_t *want)
+__dma_device_satisfies_mask(struct dma_device *device, dma_cap_mask_t *want)
 {
 	dma_cap_mask_t has;
 
-	bitmap_and(has.bits, want->bits, chan->device->cap_mask.bits,
+	bitmap_and(has.bits, want->bits, device->cap_mask.bits,
 		DMA_TX_TYPE_END);
 	return bitmap_equal(want->bits, has.bits, DMA_TX_TYPE_END);
 }
@@ -195,7 +195,7 @@ static int dma_chan_get(struct dma_chan *chan)
 			err = desc_cnt;
 			chan->client_count = 0;
 			module_put(owner);
-		} else
+		} else if (!dma_has_cap(DMA_PRIVATE, chan->device->cap_mask))
 			balance_ref_count(chan);
 	}
 
@@ -232,14 +232,16 @@ static void dma_client_chan_alloc(struct dma_client *client)
 
 	/* Find a channel */
 	list_for_each_entry(device, &dma_device_list, global_node) {
+		if (dma_has_cap(DMA_PRIVATE, device->cap_mask))
+			continue;
 		/* Does the client require a specific DMA controller? */
 		if (client->slave && client->slave->dma_dev
 				&& client->slave->dma_dev != device->dev)
 			continue;
+		if (!dma_device_satisfies_mask(device, client->cap_mask))
+			continue;
 
 		list_for_each_entry(chan, &device->channels, device_node) {
-			if (!dma_chan_satisfies_mask(chan, client->cap_mask))
-				continue;
 			if (!chan->client_count)
 				continue;
 			ack = client->event_callback(client, chan,
@@ -320,11 +322,12 @@ static int __init dma_channel_table_init(void)
 
 	bitmap_fill(dma_cap_mask_all.bits, DMA_TX_TYPE_END);
 
-	/* 'interrupt' and 'slave' are channel capabilities, but are not
-	 * associated with an operation so they do not need an entry in the
-	 * channel_table
+	/* 'interrupt', 'private', and 'slave' are channel capabilities,
+	 * but are not associated with an operation so they do not need
+	 * an entry in the channel_table
 	 */
 	clear_bit(DMA_INTERRUPT, dma_cap_mask_all.bits);
+	clear_bit(DMA_PRIVATE, dma_cap_mask_all.bits);
 	clear_bit(DMA_SLAVE, dma_cap_mask_all.bits);
 
 	for_each_dma_cap_mask(cap, dma_cap_mask_all) {
@@ -378,10 +381,13 @@ void dma_issue_pending_all(void)
 		  "client called %s without a reference", __func__);
 
 	rcu_read_lock();
-	list_for_each_entry_rcu(device, &dma_device_list, global_node)
+	list_for_each_entry_rcu(device, &dma_device_list, global_node) {
+		if (dma_has_cap(DMA_PRIVATE, device->cap_mask))
+			continue;
 		list_for_each_entry(chan, &device->channels, device_node)
 			if (chan->client_count)
 				device->device_issue_pending(chan);
+	}
 	rcu_read_unlock();
 }
 EXPORT_SYMBOL(dma_issue_pending_all);
@@ -403,7 +409,8 @@ static struct dma_chan *nth_chan(enum dma_transaction_type cap, int n)
 	struct dma_chan *min = NULL;
 
 	list_for_each_entry(device, &dma_device_list, global_node) {
-		if (!dma_has_cap(cap, device->cap_mask))
+		if (!dma_has_cap(cap, device->cap_mask) ||
+		    dma_has_cap(DMA_PRIVATE, device->cap_mask))
 			continue;
 		list_for_each_entry(chan, &device->channels, device_node) {
 			if (!chan->client_count)
@@ -452,9 +459,12 @@ static void dma_channel_rebalance(void)
 		for_each_possible_cpu(cpu)
 			per_cpu_ptr(channel_table[cap], cpu)->chan = NULL;
 
-	list_for_each_entry(device, &dma_device_list, global_node)
+	list_for_each_entry(device, &dma_device_list, global_node) {
+		if (dma_has_cap(DMA_PRIVATE, device->cap_mask))
+			continue;
 		list_for_each_entry(chan, &device->channels, device_node)
 			chan->table_count = 0;
+	}
 
 	/* don't populate the channel_table if no clients are available */
 	if (!dmaengine_ref_count)
@@ -473,6 +483,111 @@ static void dma_channel_rebalance(void)
 		}
 }
 
+static struct dma_chan *private_candidate(dma_cap_mask_t *mask, struct dma_device *dev)
+{
+	struct dma_chan *chan;
+	struct dma_chan *ret = NULL;
+
+	if (!__dma_device_satisfies_mask(dev, mask)) {
+		pr_debug("%s: wrong capabilities\n", __func__);
+		return NULL;
+	}
+	/* devices with multiple channels need special handling as we need to
+	 * ensure that all channels are either private or public.
+	 */
+	if (dev->chancnt > 1 && !dma_has_cap(DMA_PRIVATE, dev->cap_mask))
+		list_for_each_entry(chan, &dev->channels, device_node) {
+			/* some channels are already publicly allocated */
+			if (chan->client_count)
+				return NULL;
+		}
+
+	list_for_each_entry(chan, &dev->channels, device_node) {
+		if (chan->client_count) {
+			pr_debug("%s: %s busy\n",
+				 __func__, dev_name(&chan->dev));
+			continue;
+		}
+		ret = chan;
+		break;
+	}
+
+	return ret;
+}
+
+/**
+ * dma_request_channel - try to allocate an exclusive channel
+ * @mask: capabilities that the channel must satisfy
+ * @fn: optional callback to disposition available channels
+ * @fn_param: opaque parameter to pass to dma_filter_fn
+ */
+struct dma_chan *__dma_request_channel(dma_cap_mask_t *mask, dma_filter_fn fn, void *fn_param)
+{
+	struct dma_device *device, *_d;
+	struct dma_chan *chan = NULL;
+	enum dma_state_client ack;
+	int err;
+
+	/* Find a channel */
+	mutex_lock(&dma_list_mutex);
+	list_for_each_entry_safe(device, _d, &dma_device_list, global_node) {
+		chan = private_candidate(mask, device);
+		if (!chan)
+			continue;
+
+		if (fn)
+			ack = fn(chan, fn_param);
+		else
+			ack = DMA_ACK;
+
+		if (ack == DMA_ACK) {
+			/* Found a suitable channel, try to grab, prep, and
+			 * return it.  We first set DMA_PRIVATE to disable
+			 * balance_ref_count as this channel will not be
+			 * published in the general-purpose allocator
+			 */
+			dma_cap_set(DMA_PRIVATE, device->cap_mask);
+			err = dma_chan_get(chan);
+
+			if (err == -ENODEV) {
+				pr_debug("%s: %s module removed\n", __func__,
+					 dev_name(&chan->dev));
+				list_del_rcu(&device->global_node);
+			} else if (err)
+				pr_err("dmaengine: failed to get %s: (%d)\n",
+				       dev_name(&chan->dev), err);
+			else
+				break;
+		} else if (ack == DMA_DUP) {
+			pr_debug("%s: %s filter said DMA_DUP\n",
+				 __func__, dev_name(&chan->dev));
+		} else if (ack == DMA_NAK) {
+			pr_debug("%s: %s filter said DMA_NAK\n",
+				 __func__, dev_name(&chan->dev));
+			break;
+		} else
+			WARN_ONCE(1, "filter_fn: unknown response?\n");
+		chan = NULL;
+	}
+	mutex_unlock(&dma_list_mutex);
+
+	pr_debug("%s: %s (%s)\n", __func__, chan ? "success" : "fail",
+		 chan ? dev_name(&chan->dev) : NULL);
+
+	return chan;
+}
+EXPORT_SYMBOL_GPL(__dma_request_channel);
+
+void dma_release_channel(struct dma_chan *chan)
+{
+	mutex_lock(&dma_list_mutex);
+	WARN_ONCE(chan->client_count != 1,
+		  "chan reference count %d != 1\n", chan->client_count);
+	dma_chan_put(chan);
+	mutex_unlock(&dma_list_mutex);
+}
+EXPORT_SYMBOL_GPL(dma_release_channel);
+
 /**
  * dma_chans_notify_available - broadcast available channels to the clients
  */
@@ -506,7 +621,9 @@ void dma_async_client_register(struct dma_client *client)
 	dmaengine_ref_count++;
 
 	/* try to grab channels */
-	list_for_each_entry_safe(device, _d, &dma_device_list, global_node)
+	list_for_each_entry_safe(device, _d, &dma_device_list, global_node) {
+		if (dma_has_cap(DMA_PRIVATE, device->cap_mask))
+			continue;
 		list_for_each_entry(chan, &device->channels, device_node) {
 			err = dma_chan_get(chan);
 			if (err == -ENODEV) {
@@ -517,6 +634,7 @@ void dma_async_client_register(struct dma_client *client)
 				pr_err("dmaengine: failed to get %s: (%d)\n",
 				       dev_name(&chan->dev), err);
 		}
+	}
 
 	/* if this is the first reference and there were channels
 	 * waiting we need to rebalance to get those channels
@@ -547,9 +665,12 @@ void dma_async_client_unregister(struct dma_client *client)
 	dmaengine_ref_count--;
 	BUG_ON(dmaengine_ref_count < 0);
 	/* drop channel references */
-	list_for_each_entry(device, &dma_device_list, global_node)
+	list_for_each_entry(device, &dma_device_list, global_node) {
+		if (dma_has_cap(DMA_PRIVATE, device->cap_mask))
+			continue;
 		list_for_each_entry(chan, &device->channels, device_node)
 			dma_chan_put(chan);
+	}
 
 	list_del(&client->global_node);
 	mutex_unlock(&dma_list_mutex);
@@ -639,9 +760,11 @@ int dma_async_device_register(struct dma_device *device)
 		chan->slow_ref = 0;
 		INIT_RCU_HEAD(&chan->rcu);
 	}
+	device->chancnt = chancnt;
 
 	mutex_lock(&dma_list_mutex);
-	if (dmaengine_ref_count)
+	/* take references on public channels */
+	if (dmaengine_ref_count && !dma_has_cap(DMA_PRIVATE, device->cap_mask))
 		list_for_each_entry(chan, &device->channels, device_node) {
 			/* if clients are already waiting for channels we need
 			 * to take references on their behalf
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 57a43adfc39e..fe40bc020af6 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -89,6 +89,7 @@ enum dma_transaction_type {
 	DMA_MEMSET,
 	DMA_MEMCPY_CRC32C,
 	DMA_INTERRUPT,
+	DMA_PRIVATE,
 	DMA_SLAVE,
 };
 
@@ -223,6 +224,18 @@ struct dma_client;
 typedef enum dma_state_client (*dma_event_callback) (struct dma_client *client,
 		struct dma_chan *chan, enum dma_state state);
 
+/**
+ * typedef dma_filter_fn - callback filter for dma_request_channel
+ * @chan: channel to be reviewed
+ * @filter_param: opaque parameter passed through dma_request_channel
+ *
+ * When this optional parameter is specified in a call to dma_request_channel a
+ * suitable channel is passed to this routine for further dispositioning before
+ * being returned.  Where 'suitable' indicates a non-busy channel that
+ * satisfies the given capability mask.
+ */
+typedef enum dma_state_client (*dma_filter_fn)(struct dma_chan *chan, void *filter_param);
+
 /**
  * struct dma_client - info on the entity making use of DMA services
  * @event_callback: func ptr to call when something happens
@@ -472,6 +485,9 @@ void dma_async_device_unregister(struct dma_device *device);
 void dma_run_dependencies(struct dma_async_tx_descriptor *tx);
 struct dma_chan *dma_find_channel(enum dma_transaction_type tx_type);
 void dma_issue_pending_all(void);
+#define dma_request_channel(mask, x, y) __dma_request_channel(&(mask), x, y)
+struct dma_chan *__dma_request_channel(dma_cap_mask_t *mask, dma_filter_fn fn, void *fn_param);
+void dma_release_channel(struct dma_chan *chan);
 
 /* --- Helper iov-locking functions --- */
 
-- 
cgit v1.2.3


From 33df8ca068123457db56c316946a3c0e4ef787d6 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 6 Jan 2009 11:38:15 -0700
Subject: dmatest: convert to dma_request_channel

Replace the client registration infrastructure with a custom loop to
poll for channels.  Once dma_request_channel returns NULL stop asking
for channels.  A userspace side effect of this change if that loading
the dmatest module before loading a dma driver will result in no
channels being found, previously dmatest would get a callback.  To
facilitate testing in the built-in case dmatest_init is marked as a
late_initcall.  Another side effect is that channels under test can not
be used for any other purpose.

Cc: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dma/dmatest.c     | 115 +++++++++++++++++-----------------------------
 include/linux/dmaengine.h |   6 +++
 2 files changed, 49 insertions(+), 72 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c
index db4050884713..1d6e48f9cd02 100644
--- a/drivers/dma/dmatest.c
+++ b/drivers/dma/dmatest.c
@@ -35,7 +35,7 @@ MODULE_PARM_DESC(threads_per_chan,
 
 static unsigned int max_channels;
 module_param(max_channels, uint, S_IRUGO);
-MODULE_PARM_DESC(nr_channels,
+MODULE_PARM_DESC(max_channels,
 		"Maximum number of channels to use (default: all)");
 
 /*
@@ -71,7 +71,7 @@ struct dmatest_chan {
 
 /*
  * These are protected by dma_list_mutex since they're only used by
- * the DMA client event callback
+ * the DMA filter function callback
  */
 static LIST_HEAD(dmatest_channels);
 static unsigned int nr_channels;
@@ -317,21 +317,16 @@ static void dmatest_cleanup_channel(struct dmatest_chan *dtc)
 	kfree(dtc);
 }
 
-static enum dma_state_client dmatest_add_channel(struct dma_chan *chan)
+static int dmatest_add_channel(struct dma_chan *chan)
 {
 	struct dmatest_chan	*dtc;
 	struct dmatest_thread	*thread;
 	unsigned int		i;
 
-	/* Have we already been told about this channel? */
-	list_for_each_entry(dtc, &dmatest_channels, node)
-		if (dtc->chan == chan)
-			return DMA_DUP;
-
 	dtc = kmalloc(sizeof(struct dmatest_chan), GFP_KERNEL);
 	if (!dtc) {
 		pr_warning("dmatest: No memory for %s\n", dev_name(&chan->dev));
-		return DMA_NAK;
+		return -ENOMEM;
 	}
 
 	dtc->chan = chan;
@@ -365,81 +360,57 @@ static enum dma_state_client dmatest_add_channel(struct dma_chan *chan)
 	list_add_tail(&dtc->node, &dmatest_channels);
 	nr_channels++;
 
-	return DMA_ACK;
+	return 0;
 }
 
-static enum dma_state_client dmatest_remove_channel(struct dma_chan *chan)
+static enum dma_state_client filter(struct dma_chan *chan, void *param)
 {
-	struct dmatest_chan	*dtc, *_dtc;
-
-	list_for_each_entry_safe(dtc, _dtc, &dmatest_channels, node) {
-		if (dtc->chan == chan) {
-			list_del(&dtc->node);
-			dmatest_cleanup_channel(dtc);
-			pr_debug("dmatest: lost channel %s\n",
-					dev_name(&chan->dev));
-			return DMA_ACK;
-		}
-	}
-
-	return DMA_DUP;
-}
-
-/*
- * Start testing threads as new channels are assigned to us, and kill
- * them when the channels go away.
- *
- * When we unregister the client, all channels are removed so this
- * will also take care of cleaning things up when the module is
- * unloaded.
- */
-static enum dma_state_client
-dmatest_event(struct dma_client *client, struct dma_chan *chan,
-		enum dma_state state)
-{
-	enum dma_state_client	ack = DMA_NAK;
-
-	switch (state) {
-	case DMA_RESOURCE_AVAILABLE:
-		if (!dmatest_match_channel(chan)
-				|| !dmatest_match_device(chan->device))
-			ack = DMA_DUP;
-		else if (max_channels && nr_channels >= max_channels)
-			ack = DMA_NAK;
-		else
-			ack = dmatest_add_channel(chan);
-		break;
-
-	case DMA_RESOURCE_REMOVED:
-		ack = dmatest_remove_channel(chan);
-		break;
-
-	default:
-		pr_info("dmatest: Unhandled event %u (%s)\n",
-				state, dev_name(&chan->dev));
-		break;
-	}
-
-	return ack;
+	if (!dmatest_match_channel(chan) || !dmatest_match_device(chan->device))
+		return DMA_DUP;
+	else
+		return DMA_ACK;
 }
 
-static struct dma_client dmatest_client = {
-	.event_callback	= dmatest_event,
-};
-
 static int __init dmatest_init(void)
 {
-	dma_cap_set(DMA_MEMCPY, dmatest_client.cap_mask);
-	dma_async_client_register(&dmatest_client);
-	dma_async_client_chan_request(&dmatest_client);
+	dma_cap_mask_t mask;
+	struct dma_chan *chan;
+	int err = 0;
+
+	dma_cap_zero(mask);
+	dma_cap_set(DMA_MEMCPY, mask);
+	for (;;) {
+		chan = dma_request_channel(mask, filter, NULL);
+		if (chan) {
+			err = dmatest_add_channel(chan);
+			if (err == 0)
+				continue;
+			else {
+				dma_release_channel(chan);
+				break; /* add_channel failed, punt */
+			}
+		} else
+			break; /* no more channels available */
+		if (max_channels && nr_channels >= max_channels)
+			break; /* we have all we need */
+	}
 
-	return 0;
+	return err;
 }
-module_init(dmatest_init);
+/* when compiled-in wait for drivers to load first */
+late_initcall(dmatest_init);
 
 static void __exit dmatest_exit(void)
 {
-	dma_async_client_unregister(&dmatest_client);
+	struct dmatest_chan *dtc, *_dtc;
+
+	list_for_each_entry_safe(dtc, _dtc, &dmatest_channels, node) {
+		list_del(&dtc->node);
+		dmatest_cleanup_channel(dtc);
+		pr_debug("dmatest: dropped channel %s\n",
+			 dev_name(&dtc->chan->dev));
+		dma_release_channel(dtc->chan);
+	}
 }
 module_exit(dmatest_exit);
 
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index fe40bc020af6..6f2d070ac7f3 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -400,6 +400,12 @@ __dma_cap_set(enum dma_transaction_type tx_type, dma_cap_mask_t *dstp)
 	set_bit(tx_type, dstp->bits);
 }
 
+#define dma_cap_zero(mask) __dma_cap_zero(&(mask))
+static inline void __dma_cap_zero(dma_cap_mask_t *dstp)
+{
+	bitmap_zero(dstp->bits, DMA_TX_TYPE_END);
+}
+
 #define dma_has_cap(tx, mask) __dma_has_cap((tx), &(mask))
 static inline int
 __dma_has_cap(enum dma_transaction_type tx_type, dma_cap_mask_t *srcp)
-- 
cgit v1.2.3


From 74465b4ff9ac1da503025c0a0042e023bfa6505c Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 6 Jan 2009 11:38:16 -0700
Subject: atmel-mci: convert to dma_request_channel and down-level dma_slave

dma_request_channel provides an exclusive channel, so we no longer need to
pass slave data through dmaengine.

Cc: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 arch/avr32/include/asm/atmel-mci.h  |  6 +--
 arch/avr32/mach-at32ap/at32ap700x.c | 15 ++----
 drivers/dma/dmaengine.c             |  8 ---
 drivers/dma/dw_dmac.c               | 25 +++-------
 drivers/mmc/host/atmel-mci.c        | 98 ++++++++++---------------------------
 include/linux/dmaengine.h           | 38 --------------
 include/linux/dw_dmac.h             | 31 +++++++++---
 7 files changed, 62 insertions(+), 159 deletions(-)

(limited to 'include/linux')

diff --git a/arch/avr32/include/asm/atmel-mci.h b/arch/avr32/include/asm/atmel-mci.h
index 59f3fadd0b68..e5e54c6f35d9 100644
--- a/arch/avr32/include/asm/atmel-mci.h
+++ b/arch/avr32/include/asm/atmel-mci.h
@@ -3,7 +3,7 @@
 
 #define ATMEL_MCI_MAX_NR_SLOTS	2
 
-struct dma_slave;
+#include <linux/dw_dmac.h>
 
 /**
  * struct mci_slot_pdata - board-specific per-slot configuration
@@ -28,11 +28,11 @@ struct mci_slot_pdata {
 
 /**
  * struct mci_platform_data - board-specific MMC/SDcard configuration
- * @dma_slave: DMA slave interface to use in data transfers, or NULL.
+ * @dma_slave: DMA slave interface to use in data transfers.
  * @slot: Per-slot configuration data.
  */
 struct mci_platform_data {
-	struct dma_slave	*dma_slave;
+	struct dw_dma_slave	dma_slave;
 	struct mci_slot_pdata	slot[ATMEL_MCI_MAX_NR_SLOTS];
 };
 
diff --git a/arch/avr32/mach-at32ap/at32ap700x.c b/arch/avr32/mach-at32ap/at32ap700x.c
index 066252eebf61..414f174e38cd 100644
--- a/arch/avr32/mach-at32ap/at32ap700x.c
+++ b/arch/avr32/mach-at32ap/at32ap700x.c
@@ -1305,7 +1305,7 @@ struct platform_device *__init
 at32_add_device_mci(unsigned int id, struct mci_platform_data *data)
 {
 	struct platform_device		*pdev;
-	struct dw_dma_slave		*dws;
+	struct dw_dma_slave		*dws = &data->dma_slave;
 	u32				pioa_mask;
 	u32				piob_mask;
 
@@ -1324,22 +1324,13 @@ at32_add_device_mci(unsigned int id, struct mci_platform_data *data)
 				ARRAY_SIZE(atmel_mci0_resource)))
 		goto fail;
 
-	if (data->dma_slave)
-		dws = kmemdup(to_dw_dma_slave(data->dma_slave),
-				sizeof(struct dw_dma_slave), GFP_KERNEL);
-	else
-		dws = kzalloc(sizeof(struct dw_dma_slave), GFP_KERNEL);
-
-	dws->slave.dev = &pdev->dev;
-	dws->slave.dma_dev = &dw_dmac0_device.dev;
-	dws->slave.reg_width = DMA_SLAVE_WIDTH_32BIT;
+	dws->dma_dev = &dw_dmac0_device.dev;
+	dws->reg_width = DW_DMA_SLAVE_WIDTH_32BIT;
 	dws->cfg_hi = (DWC_CFGH_SRC_PER(0)
 				| DWC_CFGH_DST_PER(1));
 	dws->cfg_lo &= ~(DWC_CFGL_HS_DST_POL
 				| DWC_CFGL_HS_SRC_POL);
 
-	data->dma_slave = &dws->slave;
-
 	if (platform_device_add_data(pdev, data,
 				sizeof(struct mci_platform_data)))
 		goto fail;
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 7a0594f24a3f..90aca505a1df 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -234,10 +234,6 @@ static void dma_client_chan_alloc(struct dma_client *client)
 	list_for_each_entry(device, &dma_device_list, global_node) {
 		if (dma_has_cap(DMA_PRIVATE, device->cap_mask))
 			continue;
-		/* Does the client require a specific DMA controller? */
-		if (client->slave && client->slave->dma_dev
-				&& client->slave->dma_dev != device->dev)
-			continue;
 		if (!dma_device_satisfies_mask(device, client->cap_mask))
 			continue;
 
@@ -613,10 +609,6 @@ void dma_async_client_register(struct dma_client *client)
 	struct dma_chan *chan;
 	int err;
 
-	/* validate client data */
-	BUG_ON(dma_has_cap(DMA_SLAVE, client->cap_mask) &&
-		!client->slave);
-
 	mutex_lock(&dma_list_mutex);
 	dmaengine_ref_count++;
 
diff --git a/drivers/dma/dw_dmac.c b/drivers/dma/dw_dmac.c
index 377dafa37a20..dbd50804e5d2 100644
--- a/drivers/dma/dw_dmac.c
+++ b/drivers/dma/dw_dmac.c
@@ -567,7 +567,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 	if (unlikely(!dws || !sg_len))
 		return NULL;
 
-	reg_width = dws->slave.reg_width;
+	reg_width = dws->reg_width;
 	prev = first = NULL;
 
 	sg_len = dma_map_sg(chan->dev.parent, sgl, sg_len, direction);
@@ -579,7 +579,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 				| DWC_CTLL_DST_FIX
 				| DWC_CTLL_SRC_INC
 				| DWC_CTLL_FC_M2P);
-		reg = dws->slave.tx_reg;
+		reg = dws->tx_reg;
 		for_each_sg(sgl, sg, sg_len, i) {
 			struct dw_desc	*desc;
 			u32		len;
@@ -625,7 +625,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 				| DWC_CTLL_SRC_FIX
 				| DWC_CTLL_FC_P2M);
 
-		reg = dws->slave.rx_reg;
+		reg = dws->rx_reg;
 		for_each_sg(sgl, sg, sg_len, i) {
 			struct dw_desc	*desc;
 			u32		len;
@@ -764,7 +764,6 @@ static int dwc_alloc_chan_resources(struct dma_chan *chan,
 	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
 	struct dw_dma		*dw = to_dw_dma(chan->device);
 	struct dw_desc		*desc;
-	struct dma_slave	*slave;
 	struct dw_dma_slave	*dws;
 	int			i;
 	u32			cfghi;
@@ -772,12 +771,6 @@ static int dwc_alloc_chan_resources(struct dma_chan *chan,
 
 	dev_vdbg(&chan->dev, "alloc_chan_resources\n");
 
-	/* Channels doing slave DMA can only handle one client. */
-	if (dwc->dws || (client && client->slave)) {
-		if (chan->client_count)
-			return -EBUSY;
-	}
-
 	/* ASSERT:  channel is idle */
 	if (dma_readl(dw, CH_EN) & dwc->mask) {
 		dev_dbg(&chan->dev, "DMA channel not idle?\n");
@@ -789,23 +782,17 @@ static int dwc_alloc_chan_resources(struct dma_chan *chan,
 	cfghi = DWC_CFGH_FIFO_MODE;
 	cfglo = 0;
 
-	slave = client->slave;
-	if (slave) {
+	dws = dwc->dws;
+	if (dws) {
 		/*
 		 * We need controller-specific data to set up slave
 		 * transfers.
 		 */
-		BUG_ON(!slave->dma_dev || slave->dma_dev != dw->dma.dev);
-
-		dws = container_of(slave, struct dw_dma_slave, slave);
+		BUG_ON(!dws->dma_dev || dws->dma_dev != dw->dma.dev);
 
-		dwc->dws = dws;
 		cfghi = dws->cfg_hi;
 		cfglo = dws->cfg_lo;
-	} else {
-		dwc->dws = NULL;
 	}
-
 	channel_writel(dwc, CFG_LO, cfglo);
 	channel_writel(dwc, CFG_HI, cfghi);
 
diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
index 6c11f4d4c4e9..7a34118507db 100644
--- a/drivers/mmc/host/atmel-mci.c
+++ b/drivers/mmc/host/atmel-mci.c
@@ -1441,60 +1441,6 @@ static irqreturn_t atmci_detect_interrupt(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-#ifdef CONFIG_MMC_ATMELMCI_DMA
-
-static inline struct atmel_mci *
-dma_client_to_atmel_mci(struct dma_client *client)
-{
-	return container_of(client, struct atmel_mci, dma.client);
-}
-
-static enum dma_state_client atmci_dma_event(struct dma_client *client,
-		struct dma_chan *chan, enum dma_state state)
-{
-	struct atmel_mci	*host;
-	enum dma_state_client	ret = DMA_NAK;
-
-	host = dma_client_to_atmel_mci(client);
-
-	switch (state) {
-	case DMA_RESOURCE_AVAILABLE:
-		spin_lock_bh(&host->lock);
-		if (!host->dma.chan) {
-			host->dma.chan = chan;
-			ret = DMA_ACK;
-		}
-		spin_unlock_bh(&host->lock);
-
-		if (ret == DMA_ACK)
-			dev_info(&host->pdev->dev,
-					"Using %s for DMA transfers\n",
-					chan->dev.bus_id);
-		break;
-
-	case DMA_RESOURCE_REMOVED:
-		spin_lock_bh(&host->lock);
-		if (host->dma.chan == chan) {
-			host->dma.chan = NULL;
-			ret = DMA_ACK;
-		}
-		spin_unlock_bh(&host->lock);
-
-		if (ret == DMA_ACK)
-			dev_info(&host->pdev->dev,
-					"Lost %s, falling back to PIO\n",
-					chan->dev.bus_id);
-		break;
-
-	default:
-		break;
-	}
-
-
-	return ret;
-}
-#endif /* CONFIG_MMC_ATMELMCI_DMA */
-
 static int __init atmci_init_slot(struct atmel_mci *host,
 		struct mci_slot_pdata *slot_data, unsigned int id,
 		u32 sdc_reg)
@@ -1598,6 +1544,18 @@ static void __exit atmci_cleanup_slot(struct atmel_mci_slot *slot,
 	mmc_free_host(slot->mmc);
 }
 
+#ifdef CONFIG_MMC_ATMELMCI_DMA
+static enum dma_state_client filter(struct dma_chan *chan, void *slave)
+{
+	struct dw_dma_slave *dws = slave;
+
+	if (dws->dma_dev == chan->device->dev)
+		return DMA_ACK;
+	else
+		return DMA_DUP;
+}
+#endif
+
 static int __init atmci_probe(struct platform_device *pdev)
 {
 	struct mci_platform_data	*pdata;
@@ -1650,22 +1608,20 @@ static int __init atmci_probe(struct platform_device *pdev)
 		goto err_request_irq;
 
 #ifdef CONFIG_MMC_ATMELMCI_DMA
-	if (pdata->dma_slave) {
-		struct dma_slave *slave = pdata->dma_slave;
+	if (pdata->dma_slave.dma_dev) {
+		struct dw_dma_slave *dws = &pdata->dma_slave;
+		dma_cap_mask_t mask;
 
-		slave->tx_reg = regs->start + MCI_TDR;
-		slave->rx_reg = regs->start + MCI_RDR;
+		dws->tx_reg = regs->start + MCI_TDR;
+		dws->rx_reg = regs->start + MCI_RDR;
 
 		/* Try to grab a DMA channel */
-		host->dma.client.event_callback = atmci_dma_event;
-		dma_cap_set(DMA_SLAVE, host->dma.client.cap_mask);
-		host->dma.client.slave = slave;
-
-		dma_async_client_register(&host->dma.client);
-		dma_async_client_chan_request(&host->dma.client);
-	} else {
-		dev_notice(&pdev->dev, "DMA not available, using PIO\n");
+		dma_cap_zero(mask);
+		dma_cap_set(DMA_SLAVE, mask);
+		host->dma.chan = dma_request_channel(mask, filter, dws);
 	}
+	if (!host->dma.chan)
+		dev_notice(&pdev->dev, "DMA not available, using PIO\n");
 #endif /* CONFIG_MMC_ATMELMCI_DMA */
 
 	platform_set_drvdata(pdev, host);
@@ -1697,8 +1653,8 @@ static int __init atmci_probe(struct platform_device *pdev)
 
 err_init_slot:
 #ifdef CONFIG_MMC_ATMELMCI_DMA
-	if (pdata->dma_slave)
-		dma_async_client_unregister(&host->dma.client);
+	if (host->dma.chan)
+		dma_release_channel(host->dma.chan);
 #endif
 	free_irq(irq, host);
 err_request_irq:
@@ -1729,8 +1685,8 @@ static int __exit atmci_remove(struct platform_device *pdev)
 	clk_disable(host->mck);
 
 #ifdef CONFIG_MMC_ATMELMCI_DMA
-	if (host->dma.client.slave)
-		dma_async_client_unregister(&host->dma.client);
+	if (host->dma.chan)
+		dma_release_channel(host->dma.chan);
 #endif
 
 	free_irq(platform_get_irq(pdev, 0), host);
@@ -1759,7 +1715,7 @@ static void __exit atmci_exit(void)
 	platform_driver_unregister(&atmci_driver);
 }
 
-module_init(atmci_init);
+late_initcall(atmci_init); /* try to load after dma driver when built-in */
 module_exit(atmci_exit);
 
 MODULE_DESCRIPTION("Atmel Multimedia Card Interface driver");
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 6f2d070ac7f3..d63544cf8a1a 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -96,17 +96,6 @@ enum dma_transaction_type {
 /* last transaction type for creation of the capabilities mask */
 #define DMA_TX_TYPE_END (DMA_SLAVE + 1)
 
-/**
- * enum dma_slave_width - DMA slave register access width.
- * @DMA_SLAVE_WIDTH_8BIT: Do 8-bit slave register accesses
- * @DMA_SLAVE_WIDTH_16BIT: Do 16-bit slave register accesses
- * @DMA_SLAVE_WIDTH_32BIT: Do 32-bit slave register accesses
- */
-enum dma_slave_width {
-	DMA_SLAVE_WIDTH_8BIT,
-	DMA_SLAVE_WIDTH_16BIT,
-	DMA_SLAVE_WIDTH_32BIT,
-};
 
 /**
  * enum dma_ctrl_flags - DMA flags to augment operation preparation,
@@ -132,32 +121,6 @@ enum dma_ctrl_flags {
  */
 typedef struct { DECLARE_BITMAP(bits, DMA_TX_TYPE_END); } dma_cap_mask_t;
 
-/**
- * struct dma_slave - Information about a DMA slave
- * @dev: device acting as DMA slave
- * @dma_dev: required DMA master device. If non-NULL, the client can not be
- *	bound to other masters than this.
- * @tx_reg: physical address of data register used for
- *	memory-to-peripheral transfers
- * @rx_reg: physical address of data register used for
- *	peripheral-to-memory transfers
- * @reg_width: peripheral register width
- *
- * If dma_dev is non-NULL, the client can not be bound to other DMA
- * masters than the one corresponding to this device. The DMA master
- * driver may use this to determine if there is controller-specific
- * data wrapped around this struct. Drivers of platform code that sets
- * the dma_dev field must therefore make sure to use an appropriate
- * controller-specific dma slave structure wrapping this struct.
- */
-struct dma_slave {
-	struct device		*dev;
-	struct device		*dma_dev;
-	dma_addr_t		tx_reg;
-	dma_addr_t		rx_reg;
-	enum dma_slave_width	reg_width;
-};
-
 /**
  * struct dma_chan_percpu - the per-CPU part of struct dma_chan
  * @refcount: local_t used for open-coded "bigref" counting
@@ -248,7 +211,6 @@ typedef enum dma_state_client (*dma_filter_fn)(struct dma_chan *chan, void *filt
 struct dma_client {
 	dma_event_callback	event_callback;
 	dma_cap_mask_t		cap_mask;
-	struct dma_slave	*slave;
 	struct list_head	global_node;
 };
 
diff --git a/include/linux/dw_dmac.h b/include/linux/dw_dmac.h
index 04d217b442bf..d797dde247f7 100644
--- a/include/linux/dw_dmac.h
+++ b/include/linux/dw_dmac.h
@@ -21,15 +21,35 @@ struct dw_dma_platform_data {
 	unsigned int	nr_channels;
 };
 
+/**
+ * enum dw_dma_slave_width - DMA slave register access width.
+ * @DMA_SLAVE_WIDTH_8BIT: Do 8-bit slave register accesses
+ * @DMA_SLAVE_WIDTH_16BIT: Do 16-bit slave register accesses
+ * @DMA_SLAVE_WIDTH_32BIT: Do 32-bit slave register accesses
+ */
+enum dw_dma_slave_width {
+	DW_DMA_SLAVE_WIDTH_8BIT,
+	DW_DMA_SLAVE_WIDTH_16BIT,
+	DW_DMA_SLAVE_WIDTH_32BIT,
+};
+
 /**
  * struct dw_dma_slave - Controller-specific information about a slave
- * @slave: Generic information about the slave
- * @ctl_lo: Platform-specific initializer for the CTL_LO register
+ *
+ * @dma_dev: required DMA master device
+ * @tx_reg: physical address of data register used for
+ *	memory-to-peripheral transfers
+ * @rx_reg: physical address of data register used for
+ *	peripheral-to-memory transfers
+ * @reg_width: peripheral register width
  * @cfg_hi: Platform-specific initializer for the CFG_HI register
  * @cfg_lo: Platform-specific initializer for the CFG_LO register
  */
 struct dw_dma_slave {
-	struct dma_slave	slave;
+	struct device		*dma_dev;
+	dma_addr_t		tx_reg;
+	dma_addr_t		rx_reg;
+	enum dw_dma_slave_width	reg_width;
 	u32			cfg_hi;
 	u32			cfg_lo;
 };
@@ -54,9 +74,4 @@ struct dw_dma_slave {
 #define DWC_CFGL_HS_DST_POL	(1 << 18)	/* dst handshake active low */
 #define DWC_CFGL_HS_SRC_POL	(1 << 19)	/* src handshake active low */
 
-static inline struct dw_dma_slave *to_dw_dma_slave(struct dma_slave *slave)
-{
-	return container_of(slave, struct dw_dma_slave, slave);
-}
-
 #endif /* DW_DMAC_H */
-- 
cgit v1.2.3


From 209b84a88fe81341b4d8d465acc4a67cb7c3feb3 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 6 Jan 2009 11:38:17 -0700
Subject: dmaengine: replace dma_async_client_register with dmaengine_get

Now that clients no longer need to be notified of channel arrival
dma_async_client_register can simply increment the dmaengine_ref_count.

Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 crypto/async_tx/async_tx.c | 115 +--------------------------------------------
 drivers/dma/dmaengine.c    |  22 +++------
 include/linux/dmaengine.h  |   4 +-
 net/core/dev.c             |   3 +-
 4 files changed, 11 insertions(+), 133 deletions(-)

(limited to 'include/linux')

diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c
index 2cdf7a0867b7..f21147f3626a 100644
--- a/crypto/async_tx/async_tx.c
+++ b/crypto/async_tx/async_tx.c
@@ -28,120 +28,9 @@
 #include <linux/async_tx.h>
 
 #ifdef CONFIG_DMA_ENGINE
-static enum dma_state_client
-dma_channel_add_remove(struct dma_client *client,
-	struct dma_chan *chan, enum dma_state state);
-
-static struct dma_client async_tx_dma = {
-	.event_callback = dma_channel_add_remove,
-	/* .cap_mask == 0 defaults to all channels */
-};
-
-/**
- * async_tx_lock - protect modification of async_tx_master_list and serialize
- *	rebalance operations
- */
-static DEFINE_SPINLOCK(async_tx_lock);
-
-static LIST_HEAD(async_tx_master_list);
-
-static void
-free_dma_chan_ref(struct rcu_head *rcu)
-{
-	struct dma_chan_ref *ref;
-	ref = container_of(rcu, struct dma_chan_ref, rcu);
-	kfree(ref);
-}
-
-static void
-init_dma_chan_ref(struct dma_chan_ref *ref, struct dma_chan *chan)
-{
-	INIT_LIST_HEAD(&ref->node);
-	INIT_RCU_HEAD(&ref->rcu);
-	ref->chan = chan;
-	atomic_set(&ref->count, 0);
-}
-
-static enum dma_state_client
-dma_channel_add_remove(struct dma_client *client,
-	struct dma_chan *chan, enum dma_state state)
-{
-	unsigned long found, flags;
-	struct dma_chan_ref *master_ref, *ref;
-	enum dma_state_client ack = DMA_DUP; /* default: take no action */
-
-	switch (state) {
-	case DMA_RESOURCE_AVAILABLE:
-		found = 0;
-		rcu_read_lock();
-		list_for_each_entry_rcu(ref, &async_tx_master_list, node)
-			if (ref->chan == chan) {
-				found = 1;
-				break;
-			}
-		rcu_read_unlock();
-
-		pr_debug("async_tx: dma resource available [%s]\n",
-			found ? "old" : "new");
-
-		if (!found)
-			ack = DMA_ACK;
-		else
-			break;
-
-		/* add the channel to the generic management list */
-		master_ref = kmalloc(sizeof(*master_ref), GFP_KERNEL);
-		if (master_ref) {
-			init_dma_chan_ref(master_ref, chan);
-			spin_lock_irqsave(&async_tx_lock, flags);
-			list_add_tail_rcu(&master_ref->node,
-				&async_tx_master_list);
-			spin_unlock_irqrestore(&async_tx_lock,
-				flags);
-		} else {
-			printk(KERN_WARNING "async_tx: unable to create"
-				" new master entry in response to"
-				" a DMA_RESOURCE_ADDED event"
-				" (-ENOMEM)\n");
-			return 0;
-		}
-		break;
-	case DMA_RESOURCE_REMOVED:
-		found = 0;
-		spin_lock_irqsave(&async_tx_lock, flags);
-		list_for_each_entry(ref, &async_tx_master_list, node)
-			if (ref->chan == chan) {
-				list_del_rcu(&ref->node);
-				call_rcu(&ref->rcu, free_dma_chan_ref);
-				found = 1;
-				break;
-			}
-		spin_unlock_irqrestore(&async_tx_lock, flags);
-
-		pr_debug("async_tx: dma resource removed [%s]\n",
-			found ? "ours" : "not ours");
-
-		if (found)
-			ack = DMA_ACK;
-		else
-			break;
-		break;
-	case DMA_RESOURCE_SUSPEND:
-	case DMA_RESOURCE_RESUME:
-		printk(KERN_WARNING "async_tx: does not support dma channel"
-			" suspend/resume\n");
-		break;
-	default:
-		BUG();
-	}
-
-	return ack;
-}
-
 static int __init async_tx_init(void)
 {
-	dma_async_client_register(&async_tx_dma);
-	dma_async_client_chan_request(&async_tx_dma);
+	dmaengine_get();
 
 	printk(KERN_INFO "async_tx: api initialized (async)\n");
 
@@ -150,7 +39,7 @@ static int __init async_tx_init(void)
 
 static void __exit async_tx_exit(void)
 {
-	dma_async_client_unregister(&async_tx_dma);
+	dmaengine_put();
 }
 
 /**
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 90aca505a1df..3f1849b7f5ef 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -600,10 +600,9 @@ static void dma_clients_notify_available(void)
 }
 
 /**
- * dma_async_client_register - register a &dma_client
- * @client: ptr to a client structure with valid 'event_callback' and 'cap_mask'
+ * dmaengine_get - register interest in dma_channels
  */
-void dma_async_client_register(struct dma_client *client)
+void dmaengine_get(void)
 {
 	struct dma_device *device, *_d;
 	struct dma_chan *chan;
@@ -634,25 +633,18 @@ void dma_async_client_register(struct dma_client *client)
 	 */
 	if (dmaengine_ref_count == 1)
 		dma_channel_rebalance();
-	list_add_tail(&client->global_node, &dma_client_list);
 	mutex_unlock(&dma_list_mutex);
 }
-EXPORT_SYMBOL(dma_async_client_register);
+EXPORT_SYMBOL(dmaengine_get);
 
 /**
- * dma_async_client_unregister - unregister a client and free the &dma_client
- * @client: &dma_client to free
- *
- * Force frees any allocated DMA channels, frees the &dma_client memory
+ * dmaengine_put - let dma drivers be removed when ref_count == 0
  */
-void dma_async_client_unregister(struct dma_client *client)
+void dmaengine_put(void)
 {
 	struct dma_device *device;
 	struct dma_chan *chan;
 
-	if (!client)
-		return;
-
 	mutex_lock(&dma_list_mutex);
 	dmaengine_ref_count--;
 	BUG_ON(dmaengine_ref_count < 0);
@@ -663,11 +655,9 @@ void dma_async_client_unregister(struct dma_client *client)
 		list_for_each_entry(chan, &device->channels, device_node)
 			dma_chan_put(chan);
 	}
-
-	list_del(&client->global_node);
 	mutex_unlock(&dma_list_mutex);
 }
-EXPORT_SYMBOL(dma_async_client_unregister);
+EXPORT_SYMBOL(dmaengine_put);
 
 /**
  * dma_async_client_chan_request - send all available channels to the
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index d63544cf8a1a..37d95db156d3 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -318,8 +318,8 @@ struct dma_device {
 
 /* --- public DMA engine API --- */
 
-void dma_async_client_register(struct dma_client *client);
-void dma_async_client_unregister(struct dma_client *client);
+void dmaengine_get(void);
+void dmaengine_put(void);
 void dma_async_client_chan_request(struct dma_client *client);
 dma_cookie_t dma_async_memcpy_buf_to_buf(struct dma_chan *chan,
 	void *dest, void *src, size_t len);
diff --git a/net/core/dev.c b/net/core/dev.c
index bbb07dbe1740..7596fc9403c8 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4894,8 +4894,7 @@ static int __init netdev_dma_register(void)
 	}
 	spin_lock_init(&net_dma.lock);
 	dma_cap_set(DMA_MEMCPY, net_dma.client.cap_mask);
-	dma_async_client_register(&net_dma.client);
-	dma_async_client_chan_request(&net_dma.client);
+	dmaengine_get();
 	return 0;
 }
 
-- 
cgit v1.2.3


From aa1e6f1a385eb2b04171ec841f3b760091e4a8ee Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 6 Jan 2009 11:38:17 -0700
Subject: dmaengine: kill struct dma_client and supporting infrastructure

All users have been converted to either the general-purpose allocator,
dma_find_channel, or dma_request_channel.

Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dma/dmaengine.c      | 74 ++-------------------------------
 drivers/dma/dw_dmac.c        |  3 +-
 drivers/dma/fsldma.c         |  3 +-
 drivers/dma/ioat_dma.c       |  5 +--
 drivers/dma/iop-adma.c       |  7 ++--
 drivers/dma/mv_xor.c         |  7 ++--
 drivers/mmc/host/atmel-mci.c |  1 -
 include/linux/dmaengine.h    | 50 +---------------------
 net/core/dev.c               | 99 ++------------------------------------------
 9 files changed, 17 insertions(+), 232 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 3f1849b7f5ef..9fc91f973a9a 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -31,15 +31,12 @@
  *
  * LOCKING:
  *
- * The subsystem keeps two global lists, dma_device_list and dma_client_list.
- * Both of these are protected by a mutex, dma_list_mutex.
+ * The subsystem keeps a global list of dma_device structs it is protected by a
+ * mutex, dma_list_mutex.
  *
  * Each device has a channels list, which runs unlocked but is never modified
  * once the device is registered, it's just setup by the driver.
  *
- * Each client is responsible for keeping track of the channels it uses.  See
- * the definition of dma_event_callback in dmaengine.h.
- *
  * Each device has a kref, which is initialized to 1 when the device is
  * registered. A kref_get is done for each device registered.  When the
  * device is released, the corresponding kref_put is done in the release
@@ -74,7 +71,6 @@
 
 static DEFINE_MUTEX(dma_list_mutex);
 static LIST_HEAD(dma_device_list);
-static LIST_HEAD(dma_client_list);
 static long dmaengine_ref_count;
 
 /* --- sysfs implementation --- */
@@ -189,7 +185,7 @@ static int dma_chan_get(struct dma_chan *chan)
 
 	/* allocate upon first client reference */
 	if (chan->client_count == 1 && err == 0) {
-		int desc_cnt = chan->device->device_alloc_chan_resources(chan, NULL);
+		int desc_cnt = chan->device->device_alloc_chan_resources(chan);
 
 		if (desc_cnt < 0) {
 			err = desc_cnt;
@@ -218,40 +214,6 @@ static void dma_chan_put(struct dma_chan *chan)
 		chan->device->device_free_chan_resources(chan);
 }
 
-/**
- * dma_client_chan_alloc - try to allocate channels to a client
- * @client: &dma_client
- *
- * Called with dma_list_mutex held.
- */
-static void dma_client_chan_alloc(struct dma_client *client)
-{
-	struct dma_device *device;
-	struct dma_chan *chan;
-	enum dma_state_client ack;
-
-	/* Find a channel */
-	list_for_each_entry(device, &dma_device_list, global_node) {
-		if (dma_has_cap(DMA_PRIVATE, device->cap_mask))
-			continue;
-		if (!dma_device_satisfies_mask(device, client->cap_mask))
-			continue;
-
-		list_for_each_entry(chan, &device->channels, device_node) {
-			if (!chan->client_count)
-				continue;
-			ack = client->event_callback(client, chan,
-						     DMA_RESOURCE_AVAILABLE);
-
-			/* we are done once this client rejects
-			 * an available resource
-			 */
-			if (ack == DMA_NAK)
-				return;
-		}
-	}
-}
-
 enum dma_status dma_sync_wait(struct dma_chan *chan, dma_cookie_t cookie)
 {
 	enum dma_status status;
@@ -584,21 +546,6 @@ void dma_release_channel(struct dma_chan *chan)
 }
 EXPORT_SYMBOL_GPL(dma_release_channel);
 
-/**
- * dma_chans_notify_available - broadcast available channels to the clients
- */
-static void dma_clients_notify_available(void)
-{
-	struct dma_client *client;
-
-	mutex_lock(&dma_list_mutex);
-
-	list_for_each_entry(client, &dma_client_list, global_node)
-		dma_client_chan_alloc(client);
-
-	mutex_unlock(&dma_list_mutex);
-}
-
 /**
  * dmaengine_get - register interest in dma_channels
  */
@@ -659,19 +606,6 @@ void dmaengine_put(void)
 }
 EXPORT_SYMBOL(dmaengine_put);
 
-/**
- * dma_async_client_chan_request - send all available channels to the
- * client that satisfy the capability mask
- * @client - requester
- */
-void dma_async_client_chan_request(struct dma_client *client)
-{
-	mutex_lock(&dma_list_mutex);
-	dma_client_chan_alloc(client);
-	mutex_unlock(&dma_list_mutex);
-}
-EXPORT_SYMBOL(dma_async_client_chan_request);
-
 /**
  * dma_async_device_register - registers DMA devices found
  * @device: &dma_device
@@ -765,8 +699,6 @@ int dma_async_device_register(struct dma_device *device)
 	dma_channel_rebalance();
 	mutex_unlock(&dma_list_mutex);
 
-	dma_clients_notify_available();
-
 	return 0;
 
 err_out:
diff --git a/drivers/dma/dw_dmac.c b/drivers/dma/dw_dmac.c
index dbd50804e5d2..a29dda8f801b 100644
--- a/drivers/dma/dw_dmac.c
+++ b/drivers/dma/dw_dmac.c
@@ -758,8 +758,7 @@ static void dwc_issue_pending(struct dma_chan *chan)
 	spin_unlock_bh(&dwc->lock);
 }
 
-static int dwc_alloc_chan_resources(struct dma_chan *chan,
-		struct dma_client *client)
+static int dwc_alloc_chan_resources(struct dma_chan *chan)
 {
 	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
 	struct dw_dma		*dw = to_dw_dma(chan->device);
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index 0b95dcce447e..46e0128929a0 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -366,8 +366,7 @@ static struct fsl_desc_sw *fsl_dma_alloc_descriptor(
  *
  * Return - The number of descriptors allocated.
  */
-static int fsl_dma_alloc_chan_resources(struct dma_chan *chan,
-					struct dma_client *client)
+static int fsl_dma_alloc_chan_resources(struct dma_chan *chan)
 {
 	struct fsl_dma_chan *fsl_chan = to_fsl_chan(chan);
 
diff --git a/drivers/dma/ioat_dma.c b/drivers/dma/ioat_dma.c
index 6607fdd00b1c..e42e1aea0f18 100644
--- a/drivers/dma/ioat_dma.c
+++ b/drivers/dma/ioat_dma.c
@@ -734,8 +734,7 @@ static void ioat2_dma_massage_chan_desc(struct ioat_dma_chan *ioat_chan)
  * ioat_dma_alloc_chan_resources - returns the number of allocated descriptors
  * @chan: the channel to be filled out
  */
-static int ioat_dma_alloc_chan_resources(struct dma_chan *chan,
-					 struct dma_client *client)
+static int ioat_dma_alloc_chan_resources(struct dma_chan *chan)
 {
 	struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan);
 	struct ioat_desc_sw *desc;
@@ -1381,7 +1380,7 @@ static int ioat_dma_self_test(struct ioatdma_device *device)
 	dma_chan = container_of(device->common.channels.next,
 				struct dma_chan,
 				device_node);
-	if (device->common.device_alloc_chan_resources(dma_chan, NULL) < 1) {
+	if (device->common.device_alloc_chan_resources(dma_chan) < 1) {
 		dev_err(&device->pdev->dev,
 			"selftest cannot allocate chan resource\n");
 		err = -ENODEV;
diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c
index be9ea9f88805..c74ac9eb009a 100644
--- a/drivers/dma/iop-adma.c
+++ b/drivers/dma/iop-adma.c
@@ -470,8 +470,7 @@ static void iop_chan_start_null_xor(struct iop_adma_chan *iop_chan);
  * greater than 2x the number slots needed to satisfy a device->max_xor
  * request.
  * */
-static int iop_adma_alloc_chan_resources(struct dma_chan *chan,
-					 struct dma_client *client)
+static int iop_adma_alloc_chan_resources(struct dma_chan *chan)
 {
 	char *hw_desc;
 	int idx;
@@ -865,7 +864,7 @@ static int __devinit iop_adma_memcpy_self_test(struct iop_adma_device *device)
 	dma_chan = container_of(device->common.channels.next,
 				struct dma_chan,
 				device_node);
-	if (iop_adma_alloc_chan_resources(dma_chan, NULL) < 1) {
+	if (iop_adma_alloc_chan_resources(dma_chan) < 1) {
 		err = -ENODEV;
 		goto out;
 	}
@@ -963,7 +962,7 @@ iop_adma_xor_zero_sum_self_test(struct iop_adma_device *device)
 	dma_chan = container_of(device->common.channels.next,
 				struct dma_chan,
 				device_node);
-	if (iop_adma_alloc_chan_resources(dma_chan, NULL) < 1) {
+	if (iop_adma_alloc_chan_resources(dma_chan) < 1) {
 		err = -ENODEV;
 		goto out;
 	}
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c
index 3f46df3390c7..fbaa2f6225e2 100644
--- a/drivers/dma/mv_xor.c
+++ b/drivers/dma/mv_xor.c
@@ -606,8 +606,7 @@ submit_done:
 }
 
 /* returns the number of allocated descriptors */
-static int mv_xor_alloc_chan_resources(struct dma_chan *chan,
-				       struct dma_client *client)
+static int mv_xor_alloc_chan_resources(struct dma_chan *chan)
 {
 	char *hw_desc;
 	int idx;
@@ -957,7 +956,7 @@ static int __devinit mv_xor_memcpy_self_test(struct mv_xor_device *device)
 	dma_chan = container_of(device->common.channels.next,
 				struct dma_chan,
 				device_node);
-	if (mv_xor_alloc_chan_resources(dma_chan, NULL) < 1) {
+	if (mv_xor_alloc_chan_resources(dma_chan) < 1) {
 		err = -ENODEV;
 		goto out;
 	}
@@ -1052,7 +1051,7 @@ mv_xor_xor_self_test(struct mv_xor_device *device)
 	dma_chan = container_of(device->common.channels.next,
 				struct dma_chan,
 				device_node);
-	if (mv_xor_alloc_chan_resources(dma_chan, NULL) < 1) {
+	if (mv_xor_alloc_chan_resources(dma_chan) < 1) {
 		err = -ENODEV;
 		goto out;
 	}
diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
index 7a34118507db..4b567a0408e1 100644
--- a/drivers/mmc/host/atmel-mci.c
+++ b/drivers/mmc/host/atmel-mci.c
@@ -55,7 +55,6 @@ enum atmel_mci_state {
 
 struct atmel_mci_dma {
 #ifdef CONFIG_MMC_ATMELMCI_DMA
-	struct dma_client		client;
 	struct dma_chan			*chan;
 	struct dma_async_tx_descriptor	*data_desc;
 #endif
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 37d95db156d3..db050e97d2b4 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -28,20 +28,6 @@
 #include <linux/rcupdate.h>
 #include <linux/dma-mapping.h>
 
-/**
- * enum dma_state - resource PNP/power management state
- * @DMA_RESOURCE_SUSPEND: DMA device going into low power state
- * @DMA_RESOURCE_RESUME: DMA device returning to full power
- * @DMA_RESOURCE_AVAILABLE: DMA device available to the system
- * @DMA_RESOURCE_REMOVED: DMA device removed from the system
- */
-enum dma_state {
-	DMA_RESOURCE_SUSPEND,
-	DMA_RESOURCE_RESUME,
-	DMA_RESOURCE_AVAILABLE,
-	DMA_RESOURCE_REMOVED,
-};
-
 /**
  * enum dma_state_client - state of the channel in the client
  * @DMA_ACK: client would like to use, or was using this channel
@@ -170,23 +156,6 @@ struct dma_chan {
 
 void dma_chan_cleanup(struct kref *kref);
 
-/*
- * typedef dma_event_callback - function pointer to a DMA event callback
- * For each channel added to the system this routine is called for each client.
- * If the client would like to use the channel it returns '1' to signal (ack)
- * the dmaengine core to take out a reference on the channel and its
- * corresponding device.  A client must not 'ack' an available channel more
- * than once.  When a channel is removed all clients are notified.  If a client
- * is using the channel it must 'ack' the removal.  A client must not 'ack' a
- * removed channel more than once.
- * @client - 'this' pointer for the client context
- * @chan - channel to be acted upon
- * @state - available or removed
- */
-struct dma_client;
-typedef enum dma_state_client (*dma_event_callback) (struct dma_client *client,
-		struct dma_chan *chan, enum dma_state state);
-
 /**
  * typedef dma_filter_fn - callback filter for dma_request_channel
  * @chan: channel to be reviewed
@@ -199,21 +168,6 @@ typedef enum dma_state_client (*dma_event_callback) (struct dma_client *client,
  */
 typedef enum dma_state_client (*dma_filter_fn)(struct dma_chan *chan, void *filter_param);
 
-/**
- * struct dma_client - info on the entity making use of DMA services
- * @event_callback: func ptr to call when something happens
- * @cap_mask: only return channels that satisfy the requested capabilities
- *  a value of zero corresponds to any capability
- * @slave: data for preparing slave transfer. Must be non-NULL iff the
- *  DMA_SLAVE capability is requested.
- * @global_node: list_head for global dma_client_list
- */
-struct dma_client {
-	dma_event_callback	event_callback;
-	dma_cap_mask_t		cap_mask;
-	struct list_head	global_node;
-};
-
 typedef void (*dma_async_tx_callback)(void *dma_async_param);
 /**
  * struct dma_async_tx_descriptor - async transaction descriptor
@@ -285,8 +239,7 @@ struct dma_device {
 	int dev_id;
 	struct device *dev;
 
-	int (*device_alloc_chan_resources)(struct dma_chan *chan,
-			struct dma_client *client);
+	int (*device_alloc_chan_resources)(struct dma_chan *chan);
 	void (*device_free_chan_resources)(struct dma_chan *chan);
 
 	struct dma_async_tx_descriptor *(*device_prep_dma_memcpy)(
@@ -320,7 +273,6 @@ struct dma_device {
 
 void dmaengine_get(void);
 void dmaengine_put(void);
-void dma_async_client_chan_request(struct dma_client *client);
 dma_cookie_t dma_async_memcpy_buf_to_buf(struct dma_chan *chan,
 	void *dest, void *src, size_t len);
 dma_cookie_t dma_async_memcpy_buf_to_pg(struct dma_chan *chan,
diff --git a/net/core/dev.c b/net/core/dev.c
index 7596fc9403c8..ac55d84d6255 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -167,25 +167,6 @@ static DEFINE_SPINLOCK(ptype_lock);
 static struct list_head ptype_base[PTYPE_HASH_SIZE] __read_mostly;
 static struct list_head ptype_all __read_mostly;	/* Taps */
 
-#ifdef CONFIG_NET_DMA
-struct net_dma {
-	struct dma_client client;
-	spinlock_t lock;
-	cpumask_t channel_mask;
-	struct dma_chan **channels;
-};
-
-static enum dma_state_client
-netdev_dma_event(struct dma_client *client, struct dma_chan *chan,
-	enum dma_state state);
-
-static struct net_dma net_dma = {
-	.client = {
-		.event_callback = netdev_dma_event,
-	},
-};
-#endif
-
 /*
  * The @dev_base_head list is protected by @dev_base_lock and the rtnl
  * semaphore.
@@ -4826,81 +4807,6 @@ static int dev_cpu_callback(struct notifier_block *nfb,
 	return NOTIFY_OK;
 }
 
-#ifdef CONFIG_NET_DMA
-/**
- * netdev_dma_event - event callback for the net_dma_client
- * @client: should always be net_dma_client
- * @chan: DMA channel for the event
- * @state: DMA state to be handled
- */
-static enum dma_state_client
-netdev_dma_event(struct dma_client *client, struct dma_chan *chan,
-	enum dma_state state)
-{
-	int i, found = 0, pos = -1;
-	struct net_dma *net_dma =
-		container_of(client, struct net_dma, client);
-	enum dma_state_client ack = DMA_DUP; /* default: take no action */
-
-	spin_lock(&net_dma->lock);
-	switch (state) {
-	case DMA_RESOURCE_AVAILABLE:
-		for (i = 0; i < nr_cpu_ids; i++)
-			if (net_dma->channels[i] == chan) {
-				found = 1;
-				break;
-			} else if (net_dma->channels[i] == NULL && pos < 0)
-				pos = i;
-
-		if (!found && pos >= 0) {
-			ack = DMA_ACK;
-			net_dma->channels[pos] = chan;
-			cpu_set(pos, net_dma->channel_mask);
-		}
-		break;
-	case DMA_RESOURCE_REMOVED:
-		for (i = 0; i < nr_cpu_ids; i++)
-			if (net_dma->channels[i] == chan) {
-				found = 1;
-				pos = i;
-				break;
-			}
-
-		if (found) {
-			ack = DMA_ACK;
-			cpu_clear(pos, net_dma->channel_mask);
-			net_dma->channels[i] = NULL;
-		}
-		break;
-	default:
-		break;
-	}
-	spin_unlock(&net_dma->lock);
-
-	return ack;
-}
-
-/**
- * netdev_dma_register - register the networking subsystem as a DMA client
- */
-static int __init netdev_dma_register(void)
-{
-	net_dma.channels = kzalloc(nr_cpu_ids * sizeof(struct net_dma),
-								GFP_KERNEL);
-	if (unlikely(!net_dma.channels)) {
-		printk(KERN_NOTICE
-				"netdev_dma: no memory for net_dma.channels\n");
-		return -ENOMEM;
-	}
-	spin_lock_init(&net_dma.lock);
-	dma_cap_set(DMA_MEMCPY, net_dma.client.cap_mask);
-	dmaengine_get();
-	return 0;
-}
-
-#else
-static int __init netdev_dma_register(void) { return -ENODEV; }
-#endif /* CONFIG_NET_DMA */
 
 /**
  *	netdev_increment_features - increment feature set by one
@@ -5120,14 +5026,15 @@ static int __init net_dev_init(void)
 	if (register_pernet_device(&default_device_ops))
 		goto out;
 
-	netdev_dma_register();
-
 	open_softirq(NET_TX_SOFTIRQ, net_tx_action);
 	open_softirq(NET_RX_SOFTIRQ, net_rx_action);
 
 	hotcpu_notifier(dev_cpu_callback, 0);
 	dst_init();
 	dev_mcast_init();
+	#ifdef CONFIG_NET_DMA
+	dmaengine_get();
+	#endif
 	rc = 0;
 out:
 	return rc;
-- 
cgit v1.2.3


From f27c580c3628d79b17f38976d842a6d7f3616e2e Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 6 Jan 2009 11:38:18 -0700
Subject: dmaengine: remove 'bigref' infrastructure

Reference counting is done at the module level so clients need not worry
that a channel will leave while they are actively using dmaengine.

Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dma/dmaengine.c   | 87 +++++------------------------------------------
 drivers/dma/iop-adma.c    |  1 -
 drivers/dma/mv_xor.c      |  1 -
 include/linux/dmaengine.h |  7 ----
 4 files changed, 9 insertions(+), 87 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 9fc91f973a9a..b245c38dbec3 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -34,26 +34,15 @@
  * The subsystem keeps a global list of dma_device structs it is protected by a
  * mutex, dma_list_mutex.
  *
+ * A subsystem can get access to a channel by calling dmaengine_get() followed
+ * by dma_find_channel(), or if it has need for an exclusive channel it can call
+ * dma_request_channel().  Once a channel is allocated a reference is taken
+ * against its corresponding driver to disable removal.
+ *
  * Each device has a channels list, which runs unlocked but is never modified
  * once the device is registered, it's just setup by the driver.
  *
- * Each device has a kref, which is initialized to 1 when the device is
- * registered. A kref_get is done for each device registered.  When the
- * device is released, the corresponding kref_put is done in the release
- * method. Every time one of the device's channels is allocated to a client,
- * a kref_get occurs.  When the channel is freed, the corresponding kref_put
- * happens. The device's release function does a completion, so
- * unregister_device does a remove event, device_unregister, a kref_put
- * for the first reference, then waits on the completion for all other
- * references to finish.
- *
- * Each channel has an open-coded implementation of Rusty Russell's "bigref,"
- * with a kref and a per_cpu local_t.  A dma_chan_get is called when a client
- * signals that it wants to use a channel, and dma_chan_put is called when
- * a channel is removed or a client using it is unregistered.  A client can
- * take extra references per outstanding transaction, as is the case with
- * the NET DMA client.  The release function does a kref_put on the device.
- *	-ChrisL, DanW
+ * See Documentation/dmaengine.txt for more details
  */
 
 #include <linux/init.h>
@@ -114,18 +103,9 @@ static struct device_attribute dma_attrs[] = {
 	__ATTR_NULL
 };
 
-static void dma_async_device_cleanup(struct kref *kref);
-
-static void dma_dev_release(struct device *dev)
-{
-	struct dma_chan *chan = to_dma_chan(dev);
-	kref_put(&chan->device->refcount, dma_async_device_cleanup);
-}
-
 static struct class dma_devclass = {
 	.name		= "dma",
 	.dev_attrs	= dma_attrs,
-	.dev_release	= dma_dev_release,
 };
 
 /* --- client and device registration --- */
@@ -232,29 +212,6 @@ enum dma_status dma_sync_wait(struct dma_chan *chan, dma_cookie_t cookie)
 }
 EXPORT_SYMBOL(dma_sync_wait);
 
-/**
- * dma_chan_cleanup - release a DMA channel's resources
- * @kref: kernel reference structure that contains the DMA channel device
- */
-void dma_chan_cleanup(struct kref *kref)
-{
-	struct dma_chan *chan = container_of(kref, struct dma_chan, refcount);
-	kref_put(&chan->device->refcount, dma_async_device_cleanup);
-}
-EXPORT_SYMBOL(dma_chan_cleanup);
-
-static void dma_chan_free_rcu(struct rcu_head *rcu)
-{
-	struct dma_chan *chan = container_of(rcu, struct dma_chan, rcu);
-
-	kref_put(&chan->refcount, dma_chan_cleanup);
-}
-
-static void dma_chan_release(struct dma_chan *chan)
-{
-	call_rcu(&chan->rcu, dma_chan_free_rcu);
-}
-
 /**
  * dma_cap_mask_all - enable iteration over all operation types
  */
@@ -641,9 +598,6 @@ int dma_async_device_register(struct dma_device *device)
 	BUG_ON(!device->device_issue_pending);
 	BUG_ON(!device->dev);
 
-	init_completion(&device->done);
-	kref_init(&device->refcount);
-
 	mutex_lock(&dma_list_mutex);
 	device->dev_id = id++;
 	mutex_unlock(&dma_list_mutex);
@@ -662,19 +616,11 @@ int dma_async_device_register(struct dma_device *device)
 
 		rc = device_register(&chan->dev);
 		if (rc) {
-			chancnt--;
 			free_percpu(chan->local);
 			chan->local = NULL;
 			goto err_out;
 		}
-
-		/* One for the channel, one of the class device */
-		kref_get(&device->refcount);
-		kref_get(&device->refcount);
-		kref_init(&chan->refcount);
 		chan->client_count = 0;
-		chan->slow_ref = 0;
-		INIT_RCU_HEAD(&chan->rcu);
 	}
 	device->chancnt = chancnt;
 
@@ -705,30 +651,19 @@ err_out:
 	list_for_each_entry(chan, &device->channels, device_node) {
 		if (chan->local == NULL)
 			continue;
-		kref_put(&device->refcount, dma_async_device_cleanup);
 		device_unregister(&chan->dev);
-		chancnt--;
 		free_percpu(chan->local);
 	}
 	return rc;
 }
 EXPORT_SYMBOL(dma_async_device_register);
 
-/**
- * dma_async_device_cleanup - function called when all references are released
- * @kref: kernel reference object
- */
-static void dma_async_device_cleanup(struct kref *kref)
-{
-	struct dma_device *device;
-
-	device = container_of(kref, struct dma_device, refcount);
-	complete(&device->done);
-}
-
 /**
  * dma_async_device_unregister - unregister a DMA device
  * @device: &dma_device
+ *
+ * This routine is called by dma driver exit routines, dmaengine holds module
+ * references to prevent it being called while channels are in use.
  */
 void dma_async_device_unregister(struct dma_device *device)
 {
@@ -744,11 +679,7 @@ void dma_async_device_unregister(struct dma_device *device)
 			  "%s called while %d clients hold a reference\n",
 			  __func__, chan->client_count);
 		device_unregister(&chan->dev);
-		dma_chan_release(chan);
 	}
-
-	kref_put(&device->refcount, dma_async_device_cleanup);
-	wait_for_completion(&device->done);
 }
 EXPORT_SYMBOL(dma_async_device_unregister);
 
diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c
index c74ac9eb009a..df0e37d70b31 100644
--- a/drivers/dma/iop-adma.c
+++ b/drivers/dma/iop-adma.c
@@ -1253,7 +1253,6 @@ static int __devinit iop_adma_probe(struct platform_device *pdev)
 	spin_lock_init(&iop_chan->lock);
 	INIT_LIST_HEAD(&iop_chan->chain);
 	INIT_LIST_HEAD(&iop_chan->all_slots);
-	INIT_RCU_HEAD(&iop_chan->common.rcu);
 	iop_chan->common.device = dma_dev;
 	list_add_tail(&iop_chan->common.device_node, &dma_dev->channels);
 
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c
index fbaa2f6225e2..d35cbd1ff0b3 100644
--- a/drivers/dma/mv_xor.c
+++ b/drivers/dma/mv_xor.c
@@ -1219,7 +1219,6 @@ static int __devinit mv_xor_probe(struct platform_device *pdev)
 	INIT_LIST_HEAD(&mv_chan->chain);
 	INIT_LIST_HEAD(&mv_chan->completed_slots);
 	INIT_LIST_HEAD(&mv_chan->all_slots);
-	INIT_RCU_HEAD(&mv_chan->common.rcu);
 	mv_chan->common.device = dma_dev;
 
 	list_add_tail(&mv_chan->common.device_node, &dma_dev->channels);
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index db050e97d2b4..bca2fc758894 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -142,10 +142,6 @@ struct dma_chan {
 	int chan_id;
 	struct device dev;
 
-	struct kref refcount;
-	int slow_ref;
-	struct rcu_head rcu;
-
 	struct list_head device_node;
 	struct dma_chan_percpu *local;
 	int client_count;
@@ -233,9 +229,6 @@ struct dma_device {
 	dma_cap_mask_t  cap_mask;
 	int max_xor;
 
-	struct kref refcount;
-	struct completion done;
-
 	int dev_id;
 	struct device *dev;
 
-- 
cgit v1.2.3


From 7dd602510128d7a64b11ff3b7d4f30ac8e3946ce Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 6 Jan 2009 11:38:19 -0700
Subject: dmaengine: kill enum dma_state_client

DMA_NAK is now useless.  We can just use a bool instead.

Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dma/dmaengine.c      | 16 +++++-----------
 drivers/dma/dmatest.c        |  6 +++---
 drivers/mmc/host/atmel-mci.c |  6 +++---
 include/linux/dmaengine.h    | 17 +++--------------
 4 files changed, 14 insertions(+), 31 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index b245c38dbec3..cdc8ecfc2c2c 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -440,7 +440,7 @@ struct dma_chan *__dma_request_channel(dma_cap_mask_t *mask, dma_filter_fn fn, v
 {
 	struct dma_device *device, *_d;
 	struct dma_chan *chan = NULL;
-	enum dma_state_client ack;
+	bool ack;
 	int err;
 
 	/* Find a channel */
@@ -453,9 +453,9 @@ struct dma_chan *__dma_request_channel(dma_cap_mask_t *mask, dma_filter_fn fn, v
 		if (fn)
 			ack = fn(chan, fn_param);
 		else
-			ack = DMA_ACK;
+			ack = true;
 
-		if (ack == DMA_ACK) {
+		if (ack) {
 			/* Found a suitable channel, try to grab, prep, and
 			 * return it.  We first set DMA_PRIVATE to disable
 			 * balance_ref_count as this channel will not be
@@ -473,15 +473,9 @@ struct dma_chan *__dma_request_channel(dma_cap_mask_t *mask, dma_filter_fn fn, v
 				       dev_name(&chan->dev), err);
 			else
 				break;
-		} else if (ack == DMA_DUP) {
-			pr_debug("%s: %s filter said DMA_DUP\n",
-				 __func__, dev_name(&chan->dev));
-		} else if (ack == DMA_NAK) {
-			pr_debug("%s: %s filter said DMA_NAK\n",
-				 __func__, dev_name(&chan->dev));
-			break;
 		} else
-			WARN_ONCE(1, "filter_fn: unknown response?\n");
+			pr_debug("%s: %s filter said false\n",
+				 __func__, dev_name(&chan->dev));
 		chan = NULL;
 	}
 	mutex_unlock(&dma_list_mutex);
diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c
index 1d6e48f9cd02..c77d47c4ec5b 100644
--- a/drivers/dma/dmatest.c
+++ b/drivers/dma/dmatest.c
@@ -363,12 +363,12 @@ static int dmatest_add_channel(struct dma_chan *chan)
 	return 0;
 }
 
-static enum dma_state_client filter(struct dma_chan *chan, void *param)
+static bool filter(struct dma_chan *chan, void *param)
 {
 	if (!dmatest_match_channel(chan) || !dmatest_match_device(chan->device))
-		return DMA_DUP;
+		return false;
 	else
-		return DMA_ACK;
+		return true;
 }
 
 static int __init dmatest_init(void)
diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
index 4b567a0408e1..b0042d06eaf7 100644
--- a/drivers/mmc/host/atmel-mci.c
+++ b/drivers/mmc/host/atmel-mci.c
@@ -1544,14 +1544,14 @@ static void __exit atmci_cleanup_slot(struct atmel_mci_slot *slot,
 }
 
 #ifdef CONFIG_MMC_ATMELMCI_DMA
-static enum dma_state_client filter(struct dma_chan *chan, void *slave)
+static bool filter(struct dma_chan *chan, void *slave)
 {
 	struct dw_dma_slave *dws = slave;
 
 	if (dws->dma_dev == chan->device->dev)
-		return DMA_ACK;
+		return true;
 	else
-		return DMA_DUP;
+		return false;
 }
 #endif
 
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index bca2fc758894..1419a5094478 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -28,18 +28,6 @@
 #include <linux/rcupdate.h>
 #include <linux/dma-mapping.h>
 
-/**
- * enum dma_state_client - state of the channel in the client
- * @DMA_ACK: client would like to use, or was using this channel
- * @DMA_DUP: client has already seen this channel, or is not using this channel
- * @DMA_NAK: client does not want to see any more channels
- */
-enum dma_state_client {
-	DMA_ACK,
-	DMA_DUP,
-	DMA_NAK,
-};
-
 /**
  * typedef dma_cookie_t - an opaque DMA cookie
  *
@@ -160,9 +148,10 @@ void dma_chan_cleanup(struct kref *kref);
  * When this optional parameter is specified in a call to dma_request_channel a
  * suitable channel is passed to this routine for further dispositioning before
  * being returned.  Where 'suitable' indicates a non-busy channel that
- * satisfies the given capability mask.
+ * satisfies the given capability mask.  It returns 'true' to indicate that the
+ * channel is suitable.
  */
-typedef enum dma_state_client (*dma_filter_fn)(struct dma_chan *chan, void *filter_param);
+typedef bool (*dma_filter_fn)(struct dma_chan *chan, void *filter_param);
 
 typedef void (*dma_async_tx_callback)(void *dma_async_param);
 /**
-- 
cgit v1.2.3


From 41d5e59c1299f27983977bcfe3b360600996051c Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 6 Jan 2009 11:38:21 -0700
Subject: dmaengine: add a release for dma class devices and dependent
 infrastructure

Resolves:
WARNING: at drivers/base/core.c:122 device_release+0x4d/0x52()
Device 'dma0chan0' does not have a release() function, it is broken and must be fixed.

The dma_chan_dev object is introduced to gear-match sysfs kobject and
dmaengine channel lifetimes.  When a channel is removed access to the
sysfs entries return -ENODEV until the kobject can be released.

The bulk of the change is updates to existing code to handle the extra
layer of indirection between a dma_chan and its struct device.

Reported-by: Alexander Beregalov <a.beregalov@gmail.com>
Acked-by: Stephen Hemminger <shemminger@vyatta.com>
Cc: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dma/dmaengine.c   | 106 ++++++++++++++++++++++++++++++++++++----------
 drivers/dma/dmatest.c     |  14 +++---
 drivers/dma/dw_dmac.c     |  91 +++++++++++++++++++++------------------
 drivers/dma/fsldma.c      |   2 +-
 include/linux/dmaengine.h |  19 +++++++--
 5 files changed, 157 insertions(+), 75 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index cdc8ecfc2c2c..93c4c9ac8997 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -64,36 +64,75 @@ static long dmaengine_ref_count;
 
 /* --- sysfs implementation --- */
 
+/**
+ * dev_to_dma_chan - convert a device pointer to the its sysfs container object
+ * @dev - device node
+ *
+ * Must be called under dma_list_mutex
+ */
+static struct dma_chan *dev_to_dma_chan(struct device *dev)
+{
+	struct dma_chan_dev *chan_dev;
+
+	chan_dev = container_of(dev, typeof(*chan_dev), device);
+	return chan_dev->chan;
+}
+
 static ssize_t show_memcpy_count(struct device *dev, struct device_attribute *attr, char *buf)
 {
-	struct dma_chan *chan = to_dma_chan(dev);
+	struct dma_chan *chan;
 	unsigned long count = 0;
 	int i;
+	int err;
 
-	for_each_possible_cpu(i)
-		count += per_cpu_ptr(chan->local, i)->memcpy_count;
+	mutex_lock(&dma_list_mutex);
+	chan = dev_to_dma_chan(dev);
+	if (chan) {
+		for_each_possible_cpu(i)
+			count += per_cpu_ptr(chan->local, i)->memcpy_count;
+		err = sprintf(buf, "%lu\n", count);
+	} else
+		err = -ENODEV;
+	mutex_unlock(&dma_list_mutex);
 
-	return sprintf(buf, "%lu\n", count);
+	return err;
 }
 
 static ssize_t show_bytes_transferred(struct device *dev, struct device_attribute *attr,
 				      char *buf)
 {
-	struct dma_chan *chan = to_dma_chan(dev);
+	struct dma_chan *chan;
 	unsigned long count = 0;
 	int i;
+	int err;
 
-	for_each_possible_cpu(i)
-		count += per_cpu_ptr(chan->local, i)->bytes_transferred;
+	mutex_lock(&dma_list_mutex);
+	chan = dev_to_dma_chan(dev);
+	if (chan) {
+		for_each_possible_cpu(i)
+			count += per_cpu_ptr(chan->local, i)->bytes_transferred;
+		err = sprintf(buf, "%lu\n", count);
+	} else
+		err = -ENODEV;
+	mutex_unlock(&dma_list_mutex);
 
-	return sprintf(buf, "%lu\n", count);
+	return err;
 }
 
 static ssize_t show_in_use(struct device *dev, struct device_attribute *attr, char *buf)
 {
-	struct dma_chan *chan = to_dma_chan(dev);
+	struct dma_chan *chan;
+	int err;
 
-	return sprintf(buf, "%d\n", chan->client_count);
+	mutex_lock(&dma_list_mutex);
+	chan = dev_to_dma_chan(dev);
+	if (chan)
+		err = sprintf(buf, "%d\n", chan->client_count);
+	else
+		err = -ENODEV;
+	mutex_unlock(&dma_list_mutex);
+
+	return err;
 }
 
 static struct device_attribute dma_attrs[] = {
@@ -103,9 +142,18 @@ static struct device_attribute dma_attrs[] = {
 	__ATTR_NULL
 };
 
+static void chan_dev_release(struct device *dev)
+{
+	struct dma_chan_dev *chan_dev;
+
+	chan_dev = container_of(dev, typeof(*chan_dev), device);
+	kfree(chan_dev);
+}
+
 static struct class dma_devclass = {
 	.name		= "dma",
 	.dev_attrs	= dma_attrs,
+	.dev_release	= chan_dev_release,
 };
 
 /* --- client and device registration --- */
@@ -420,7 +468,7 @@ static struct dma_chan *private_candidate(dma_cap_mask_t *mask, struct dma_devic
 	list_for_each_entry(chan, &dev->channels, device_node) {
 		if (chan->client_count) {
 			pr_debug("%s: %s busy\n",
-				 __func__, dev_name(&chan->dev));
+				 __func__, dma_chan_name(chan));
 			continue;
 		}
 		ret = chan;
@@ -466,22 +514,22 @@ struct dma_chan *__dma_request_channel(dma_cap_mask_t *mask, dma_filter_fn fn, v
 
 			if (err == -ENODEV) {
 				pr_debug("%s: %s module removed\n", __func__,
-					 dev_name(&chan->dev));
+					 dma_chan_name(chan));
 				list_del_rcu(&device->global_node);
 			} else if (err)
 				pr_err("dmaengine: failed to get %s: (%d)\n",
-				       dev_name(&chan->dev), err);
+				       dma_chan_name(chan), err);
 			else
 				break;
 		} else
 			pr_debug("%s: %s filter said false\n",
-				 __func__, dev_name(&chan->dev));
+				 __func__, dma_chan_name(chan));
 		chan = NULL;
 	}
 	mutex_unlock(&dma_list_mutex);
 
 	pr_debug("%s: %s (%s)\n", __func__, chan ? "success" : "fail",
-		 chan ? dev_name(&chan->dev) : NULL);
+		 chan ? dma_chan_name(chan) : NULL);
 
 	return chan;
 }
@@ -521,7 +569,7 @@ void dmaengine_get(void)
 				break;
 			} else if (err)
 				pr_err("dmaengine: failed to get %s: (%d)\n",
-				       dev_name(&chan->dev), err);
+				       dma_chan_name(chan), err);
 		}
 	}
 
@@ -601,14 +649,20 @@ int dma_async_device_register(struct dma_device *device)
 		chan->local = alloc_percpu(typeof(*chan->local));
 		if (chan->local == NULL)
 			continue;
+		chan->dev = kzalloc(sizeof(*chan->dev), GFP_KERNEL);
+		if (chan->dev == NULL) {
+			free_percpu(chan->local);
+			continue;
+		}
 
 		chan->chan_id = chancnt++;
-		chan->dev.class = &dma_devclass;
-		chan->dev.parent = device->dev;
-		dev_set_name(&chan->dev, "dma%dchan%d",
+		chan->dev->device.class = &dma_devclass;
+		chan->dev->device.parent = device->dev;
+		chan->dev->chan = chan;
+		dev_set_name(&chan->dev->device, "dma%dchan%d",
 			     device->dev_id, chan->chan_id);
 
-		rc = device_register(&chan->dev);
+		rc = device_register(&chan->dev->device);
 		if (rc) {
 			free_percpu(chan->local);
 			chan->local = NULL;
@@ -645,7 +699,10 @@ err_out:
 	list_for_each_entry(chan, &device->channels, device_node) {
 		if (chan->local == NULL)
 			continue;
-		device_unregister(&chan->dev);
+		mutex_lock(&dma_list_mutex);
+		chan->dev->chan = NULL;
+		mutex_unlock(&dma_list_mutex);
+		device_unregister(&chan->dev->device);
 		free_percpu(chan->local);
 	}
 	return rc;
@@ -672,7 +729,10 @@ void dma_async_device_unregister(struct dma_device *device)
 		WARN_ONCE(chan->client_count,
 			  "%s called while %d clients hold a reference\n",
 			  __func__, chan->client_count);
-		device_unregister(&chan->dev);
+		mutex_lock(&dma_list_mutex);
+		chan->dev->chan = NULL;
+		mutex_unlock(&dma_list_mutex);
+		device_unregister(&chan->dev->device);
 	}
 }
 EXPORT_SYMBOL(dma_async_device_unregister);
@@ -845,7 +905,7 @@ dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx)
 		return DMA_SUCCESS;
 
 	WARN_ONCE(tx->parent, "%s: speculatively walking dependency chain for"
-		  " %s\n", __func__, dev_name(&tx->chan->dev));
+		  " %s\n", __func__, dma_chan_name(tx->chan));
 
 	/* poll through the dependency chain, return when tx is complete */
 	do {
diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c
index c77d47c4ec5b..3603f1ea5b28 100644
--- a/drivers/dma/dmatest.c
+++ b/drivers/dma/dmatest.c
@@ -80,7 +80,7 @@ static bool dmatest_match_channel(struct dma_chan *chan)
 {
 	if (test_channel[0] == '\0')
 		return true;
-	return strcmp(dev_name(&chan->dev), test_channel) == 0;
+	return strcmp(dma_chan_name(chan), test_channel) == 0;
 }
 
 static bool dmatest_match_device(struct dma_device *device)
@@ -325,7 +325,7 @@ static int dmatest_add_channel(struct dma_chan *chan)
 
 	dtc = kmalloc(sizeof(struct dmatest_chan), GFP_KERNEL);
 	if (!dtc) {
-		pr_warning("dmatest: No memory for %s\n", dev_name(&chan->dev));
+		pr_warning("dmatest: No memory for %s\n", dma_chan_name(chan));
 		return -ENOMEM;
 	}
 
@@ -336,16 +336,16 @@ static int dmatest_add_channel(struct dma_chan *chan)
 		thread = kzalloc(sizeof(struct dmatest_thread), GFP_KERNEL);
 		if (!thread) {
 			pr_warning("dmatest: No memory for %s-test%u\n",
-				   dev_name(&chan->dev), i);
+				   dma_chan_name(chan), i);
 			break;
 		}
 		thread->chan = dtc->chan;
 		smp_wmb();
 		thread->task = kthread_run(dmatest_func, thread, "%s-test%u",
-				dev_name(&chan->dev), i);
+				dma_chan_name(chan), i);
 		if (IS_ERR(thread->task)) {
 			pr_warning("dmatest: Failed to run thread %s-test%u\n",
-					dev_name(&chan->dev), i);
+					dma_chan_name(chan), i);
 			kfree(thread);
 			break;
 		}
@@ -355,7 +355,7 @@ static int dmatest_add_channel(struct dma_chan *chan)
 		list_add_tail(&thread->node, &dtc->threads);
 	}
 
-	pr_info("dmatest: Started %u threads using %s\n", i, dev_name(&chan->dev));
+	pr_info("dmatest: Started %u threads using %s\n", i, dma_chan_name(chan));
 
 	list_add_tail(&dtc->node, &dmatest_channels);
 	nr_channels++;
@@ -408,7 +408,7 @@ static void __exit dmatest_exit(void)
 		list_del(&dtc->node);
 		dmatest_cleanup_channel(dtc);
 		pr_debug("dmatest: dropped channel %s\n",
-			 dev_name(&dtc->chan->dev));
+			 dma_chan_name(dtc->chan));
 		dma_release_channel(dtc->chan);
 	}
 }
diff --git a/drivers/dma/dw_dmac.c b/drivers/dma/dw_dmac.c
index a29dda8f801b..6b702cc46b3d 100644
--- a/drivers/dma/dw_dmac.c
+++ b/drivers/dma/dw_dmac.c
@@ -70,6 +70,15 @@
  * the controller, though.
  */
 
+static struct device *chan2dev(struct dma_chan *chan)
+{
+	return &chan->dev->device;
+}
+static struct device *chan2parent(struct dma_chan *chan)
+{
+	return chan->dev->device.parent;
+}
+
 static struct dw_desc *dwc_first_active(struct dw_dma_chan *dwc)
 {
 	return list_entry(dwc->active_list.next, struct dw_desc, desc_node);
@@ -93,12 +102,12 @@ static struct dw_desc *dwc_desc_get(struct dw_dma_chan *dwc)
 			ret = desc;
 			break;
 		}
-		dev_dbg(&dwc->chan.dev, "desc %p not ACKed\n", desc);
+		dev_dbg(chan2dev(&dwc->chan), "desc %p not ACKed\n", desc);
 		i++;
 	}
 	spin_unlock_bh(&dwc->lock);
 
-	dev_vdbg(&dwc->chan.dev, "scanned %u descriptors on freelist\n", i);
+	dev_vdbg(chan2dev(&dwc->chan), "scanned %u descriptors on freelist\n", i);
 
 	return ret;
 }
@@ -108,10 +117,10 @@ static void dwc_sync_desc_for_cpu(struct dw_dma_chan *dwc, struct dw_desc *desc)
 	struct dw_desc	*child;
 
 	list_for_each_entry(child, &desc->txd.tx_list, desc_node)
-		dma_sync_single_for_cpu(dwc->chan.dev.parent,
+		dma_sync_single_for_cpu(chan2parent(&dwc->chan),
 				child->txd.phys, sizeof(child->lli),
 				DMA_TO_DEVICE);
-	dma_sync_single_for_cpu(dwc->chan.dev.parent,
+	dma_sync_single_for_cpu(chan2parent(&dwc->chan),
 			desc->txd.phys, sizeof(desc->lli),
 			DMA_TO_DEVICE);
 }
@@ -129,11 +138,11 @@ static void dwc_desc_put(struct dw_dma_chan *dwc, struct dw_desc *desc)
 
 		spin_lock_bh(&dwc->lock);
 		list_for_each_entry(child, &desc->txd.tx_list, desc_node)
-			dev_vdbg(&dwc->chan.dev,
+			dev_vdbg(chan2dev(&dwc->chan),
 					"moving child desc %p to freelist\n",
 					child);
 		list_splice_init(&desc->txd.tx_list, &dwc->free_list);
-		dev_vdbg(&dwc->chan.dev, "moving desc %p to freelist\n", desc);
+		dev_vdbg(chan2dev(&dwc->chan), "moving desc %p to freelist\n", desc);
 		list_add(&desc->desc_node, &dwc->free_list);
 		spin_unlock_bh(&dwc->lock);
 	}
@@ -163,9 +172,9 @@ static void dwc_dostart(struct dw_dma_chan *dwc, struct dw_desc *first)
 
 	/* ASSERT:  channel is idle */
 	if (dma_readl(dw, CH_EN) & dwc->mask) {
-		dev_err(&dwc->chan.dev,
+		dev_err(chan2dev(&dwc->chan),
 			"BUG: Attempted to start non-idle channel\n");
-		dev_err(&dwc->chan.dev,
+		dev_err(chan2dev(&dwc->chan),
 			"  SAR: 0x%x DAR: 0x%x LLP: 0x%x CTL: 0x%x:%08x\n",
 			channel_readl(dwc, SAR),
 			channel_readl(dwc, DAR),
@@ -193,7 +202,7 @@ dwc_descriptor_complete(struct dw_dma_chan *dwc, struct dw_desc *desc)
 	void				*param;
 	struct dma_async_tx_descriptor	*txd = &desc->txd;
 
-	dev_vdbg(&dwc->chan.dev, "descriptor %u complete\n", txd->cookie);
+	dev_vdbg(chan2dev(&dwc->chan), "descriptor %u complete\n", txd->cookie);
 
 	dwc->completed = txd->cookie;
 	callback = txd->callback;
@@ -208,11 +217,11 @@ dwc_descriptor_complete(struct dw_dma_chan *dwc, struct dw_desc *desc)
 	 * mapped before they were submitted...
 	 */
 	if (!(txd->flags & DMA_COMPL_SKIP_DEST_UNMAP))
-		dma_unmap_page(dwc->chan.dev.parent, desc->lli.dar, desc->len,
-				DMA_FROM_DEVICE);
+		dma_unmap_page(chan2parent(&dwc->chan), desc->lli.dar,
+			       desc->len, DMA_FROM_DEVICE);
 	if (!(txd->flags & DMA_COMPL_SKIP_SRC_UNMAP))
-		dma_unmap_page(dwc->chan.dev.parent, desc->lli.sar, desc->len,
-				DMA_TO_DEVICE);
+		dma_unmap_page(chan2parent(&dwc->chan), desc->lli.sar,
+			       desc->len, DMA_TO_DEVICE);
 
 	/*
 	 * The API requires that no submissions are done from a
@@ -228,7 +237,7 @@ static void dwc_complete_all(struct dw_dma *dw, struct dw_dma_chan *dwc)
 	LIST_HEAD(list);
 
 	if (dma_readl(dw, CH_EN) & dwc->mask) {
-		dev_err(&dwc->chan.dev,
+		dev_err(chan2dev(&dwc->chan),
 			"BUG: XFER bit set, but channel not idle!\n");
 
 		/* Try to continue after resetting the channel... */
@@ -273,7 +282,7 @@ static void dwc_scan_descriptors(struct dw_dma *dw, struct dw_dma_chan *dwc)
 		return;
 	}
 
-	dev_vdbg(&dwc->chan.dev, "scan_descriptors: llp=0x%x\n", llp);
+	dev_vdbg(chan2dev(&dwc->chan), "scan_descriptors: llp=0x%x\n", llp);
 
 	list_for_each_entry_safe(desc, _desc, &dwc->active_list, desc_node) {
 		if (desc->lli.llp == llp)
@@ -292,7 +301,7 @@ static void dwc_scan_descriptors(struct dw_dma *dw, struct dw_dma_chan *dwc)
 		dwc_descriptor_complete(dwc, desc);
 	}
 
-	dev_err(&dwc->chan.dev,
+	dev_err(chan2dev(&dwc->chan),
 		"BUG: All descriptors done, but channel not idle!\n");
 
 	/* Try to continue after resetting the channel... */
@@ -308,7 +317,7 @@ static void dwc_scan_descriptors(struct dw_dma *dw, struct dw_dma_chan *dwc)
 
 static void dwc_dump_lli(struct dw_dma_chan *dwc, struct dw_lli *lli)
 {
-	dev_printk(KERN_CRIT, &dwc->chan.dev,
+	dev_printk(KERN_CRIT, chan2dev(&dwc->chan),
 			"  desc: s0x%x d0x%x l0x%x c0x%x:%x\n",
 			lli->sar, lli->dar, lli->llp,
 			lli->ctlhi, lli->ctllo);
@@ -342,9 +351,9 @@ static void dwc_handle_error(struct dw_dma *dw, struct dw_dma_chan *dwc)
 	 * controller flagged an error instead of scribbling over
 	 * random memory locations.
 	 */
-	dev_printk(KERN_CRIT, &dwc->chan.dev,
+	dev_printk(KERN_CRIT, chan2dev(&dwc->chan),
 			"Bad descriptor submitted for DMA!\n");
-	dev_printk(KERN_CRIT, &dwc->chan.dev,
+	dev_printk(KERN_CRIT, chan2dev(&dwc->chan),
 			"  cookie: %d\n", bad_desc->txd.cookie);
 	dwc_dump_lli(dwc, &bad_desc->lli);
 	list_for_each_entry(child, &bad_desc->txd.tx_list, desc_node)
@@ -442,12 +451,12 @@ static dma_cookie_t dwc_tx_submit(struct dma_async_tx_descriptor *tx)
 	 * for DMA. But this is hard to do in a race-free manner.
 	 */
 	if (list_empty(&dwc->active_list)) {
-		dev_vdbg(&tx->chan->dev, "tx_submit: started %u\n",
+		dev_vdbg(chan2dev(tx->chan), "tx_submit: started %u\n",
 				desc->txd.cookie);
 		dwc_dostart(dwc, desc);
 		list_add_tail(&desc->desc_node, &dwc->active_list);
 	} else {
-		dev_vdbg(&tx->chan->dev, "tx_submit: queued %u\n",
+		dev_vdbg(chan2dev(tx->chan), "tx_submit: queued %u\n",
 				desc->txd.cookie);
 
 		list_add_tail(&desc->desc_node, &dwc->queue);
@@ -472,11 +481,11 @@ dwc_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
 	unsigned int		dst_width;
 	u32			ctllo;
 
-	dev_vdbg(&chan->dev, "prep_dma_memcpy d0x%x s0x%x l0x%zx f0x%lx\n",
+	dev_vdbg(chan2dev(chan), "prep_dma_memcpy d0x%x s0x%x l0x%zx f0x%lx\n",
 			dest, src, len, flags);
 
 	if (unlikely(!len)) {
-		dev_dbg(&chan->dev, "prep_dma_memcpy: length is zero!\n");
+		dev_dbg(chan2dev(chan), "prep_dma_memcpy: length is zero!\n");
 		return NULL;
 	}
 
@@ -516,7 +525,7 @@ dwc_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
 			first = desc;
 		} else {
 			prev->lli.llp = desc->txd.phys;
-			dma_sync_single_for_device(chan->dev.parent,
+			dma_sync_single_for_device(chan2parent(chan),
 					prev->txd.phys, sizeof(prev->lli),
 					DMA_TO_DEVICE);
 			list_add_tail(&desc->desc_node,
@@ -531,7 +540,7 @@ dwc_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
 		prev->lli.ctllo |= DWC_CTLL_INT_EN;
 
 	prev->lli.llp = 0;
-	dma_sync_single_for_device(chan->dev.parent,
+	dma_sync_single_for_device(chan2parent(chan),
 			prev->txd.phys, sizeof(prev->lli),
 			DMA_TO_DEVICE);
 
@@ -562,7 +571,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 	struct scatterlist	*sg;
 	size_t			total_len = 0;
 
-	dev_vdbg(&chan->dev, "prep_dma_slave\n");
+	dev_vdbg(chan2dev(chan), "prep_dma_slave\n");
 
 	if (unlikely(!dws || !sg_len))
 		return NULL;
@@ -570,7 +579,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 	reg_width = dws->reg_width;
 	prev = first = NULL;
 
-	sg_len = dma_map_sg(chan->dev.parent, sgl, sg_len, direction);
+	sg_len = dma_map_sg(chan2parent(chan), sgl, sg_len, direction);
 
 	switch (direction) {
 	case DMA_TO_DEVICE:
@@ -587,7 +596,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 
 			desc = dwc_desc_get(dwc);
 			if (!desc) {
-				dev_err(&chan->dev,
+				dev_err(chan2dev(chan),
 					"not enough descriptors available\n");
 				goto err_desc_get;
 			}
@@ -607,7 +616,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 				first = desc;
 			} else {
 				prev->lli.llp = desc->txd.phys;
-				dma_sync_single_for_device(chan->dev.parent,
+				dma_sync_single_for_device(chan2parent(chan),
 						prev->txd.phys,
 						sizeof(prev->lli),
 						DMA_TO_DEVICE);
@@ -633,7 +642,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 
 			desc = dwc_desc_get(dwc);
 			if (!desc) {
-				dev_err(&chan->dev,
+				dev_err(chan2dev(chan),
 					"not enough descriptors available\n");
 				goto err_desc_get;
 			}
@@ -653,7 +662,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 				first = desc;
 			} else {
 				prev->lli.llp = desc->txd.phys;
-				dma_sync_single_for_device(chan->dev.parent,
+				dma_sync_single_for_device(chan2parent(chan),
 						prev->txd.phys,
 						sizeof(prev->lli),
 						DMA_TO_DEVICE);
@@ -673,7 +682,7 @@ dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 		prev->lli.ctllo |= DWC_CTLL_INT_EN;
 
 	prev->lli.llp = 0;
-	dma_sync_single_for_device(chan->dev.parent,
+	dma_sync_single_for_device(chan2parent(chan),
 			prev->txd.phys, sizeof(prev->lli),
 			DMA_TO_DEVICE);
 
@@ -768,11 +777,11 @@ static int dwc_alloc_chan_resources(struct dma_chan *chan)
 	u32			cfghi;
 	u32			cfglo;
 
-	dev_vdbg(&chan->dev, "alloc_chan_resources\n");
+	dev_vdbg(chan2dev(chan), "alloc_chan_resources\n");
 
 	/* ASSERT:  channel is idle */
 	if (dma_readl(dw, CH_EN) & dwc->mask) {
-		dev_dbg(&chan->dev, "DMA channel not idle?\n");
+		dev_dbg(chan2dev(chan), "DMA channel not idle?\n");
 		return -EIO;
 	}
 
@@ -808,7 +817,7 @@ static int dwc_alloc_chan_resources(struct dma_chan *chan)
 
 		desc = kzalloc(sizeof(struct dw_desc), GFP_KERNEL);
 		if (!desc) {
-			dev_info(&chan->dev,
+			dev_info(chan2dev(chan),
 				"only allocated %d descriptors\n", i);
 			spin_lock_bh(&dwc->lock);
 			break;
@@ -818,7 +827,7 @@ static int dwc_alloc_chan_resources(struct dma_chan *chan)
 		desc->txd.tx_submit = dwc_tx_submit;
 		desc->txd.flags = DMA_CTRL_ACK;
 		INIT_LIST_HEAD(&desc->txd.tx_list);
-		desc->txd.phys = dma_map_single(chan->dev.parent, &desc->lli,
+		desc->txd.phys = dma_map_single(chan2parent(chan), &desc->lli,
 				sizeof(desc->lli), DMA_TO_DEVICE);
 		dwc_desc_put(dwc, desc);
 
@@ -833,7 +842,7 @@ static int dwc_alloc_chan_resources(struct dma_chan *chan)
 
 	spin_unlock_bh(&dwc->lock);
 
-	dev_dbg(&chan->dev,
+	dev_dbg(chan2dev(chan),
 		"alloc_chan_resources allocated %d descriptors\n", i);
 
 	return i;
@@ -846,7 +855,7 @@ static void dwc_free_chan_resources(struct dma_chan *chan)
 	struct dw_desc		*desc, *_desc;
 	LIST_HEAD(list);
 
-	dev_dbg(&chan->dev, "free_chan_resources (descs allocated=%u)\n",
+	dev_dbg(chan2dev(chan), "free_chan_resources (descs allocated=%u)\n",
 			dwc->descs_allocated);
 
 	/* ASSERT:  channel is idle */
@@ -867,13 +876,13 @@ static void dwc_free_chan_resources(struct dma_chan *chan)
 	spin_unlock_bh(&dwc->lock);
 
 	list_for_each_entry_safe(desc, _desc, &list, desc_node) {
-		dev_vdbg(&chan->dev, "  freeing descriptor %p\n", desc);
-		dma_unmap_single(chan->dev.parent, desc->txd.phys,
+		dev_vdbg(chan2dev(chan), "  freeing descriptor %p\n", desc);
+		dma_unmap_single(chan2parent(chan), desc->txd.phys,
 				sizeof(desc->lli), DMA_TO_DEVICE);
 		kfree(desc);
 	}
 
-	dev_vdbg(&chan->dev, "free_chan_resources done\n");
+	dev_vdbg(chan2dev(chan), "free_chan_resources done\n");
 }
 
 /*----------------------------------------------------------------------*/
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index 46e0128929a0..ca70a21afc68 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -822,7 +822,7 @@ static int __devinit fsl_dma_chan_probe(struct fsl_dma_device *fdev,
 	 */
 	WARN_ON(fdev->feature != new_fsl_chan->feature);
 
-	new_fsl_chan->dev = &new_fsl_chan->common.dev;
+	new_fsl_chan->dev = &new_fsl_chan->common.dev->device;
 	new_fsl_chan->reg_base = ioremap(new_fsl_chan->reg.start,
 			new_fsl_chan->reg.end - new_fsl_chan->reg.start + 1);
 
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 1419a5094478..d6b6bff355f4 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -113,7 +113,7 @@ struct dma_chan_percpu {
  * @device: ptr to the dma device who supplies this channel, always !%NULL
  * @cookie: last cookie value returned to client
  * @chan_id: channel ID for sysfs
- * @class_dev: class device for sysfs
+ * @dev: class device for sysfs
  * @refcount: kref, used in "bigref" slow-mode
  * @slow_ref: indicates that the DMA channel is free
  * @rcu: the DMA channel's RCU head
@@ -128,7 +128,7 @@ struct dma_chan {
 
 	/* sysfs */
 	int chan_id;
-	struct device dev;
+	struct dma_chan_dev *dev;
 
 	struct list_head device_node;
 	struct dma_chan_percpu *local;
@@ -136,7 +136,20 @@ struct dma_chan {
 	int table_count;
 };
 
-#define to_dma_chan(p) container_of(p, struct dma_chan, dev)
+/**
+ * struct dma_chan_dev - relate sysfs device node to backing channel device
+ * @chan - driver channel device
+ * @device - sysfs device
+ */
+struct dma_chan_dev {
+	struct dma_chan *chan;
+	struct device device;
+};
+
+static inline const char *dma_chan_name(struct dma_chan *chan)
+{
+	return dev_name(&chan->dev->device);
+}
 
 void dma_chan_cleanup(struct kref *kref);
 
-- 
cgit v1.2.3


From 864498aaa9fef69ee166da023d12413a7776342d Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Tue, 6 Jan 2009 11:38:21 -0700
Subject: dmaengine: use idr for registering dma device numbers

This brings some predictability to dma device numbers, i.e. an rmmod/insmod
cycle may now result in /sys/class/dma/dma0chan0 being restored rather than
/sys/class/dma/dma1chan0 appearing.

Cc: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/dma/dmaengine.c   | 27 +++++++++++++++++++++++++--
 include/linux/dmaengine.h |  4 ++++
 2 files changed, 29 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 93c4c9ac8997..dd43410c1019 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -57,10 +57,12 @@
 #include <linux/mutex.h>
 #include <linux/jiffies.h>
 #include <linux/rculist.h>
+#include <linux/idr.h>
 
 static DEFINE_MUTEX(dma_list_mutex);
 static LIST_HEAD(dma_device_list);
 static long dmaengine_ref_count;
+static struct idr dma_idr;
 
 /* --- sysfs implementation --- */
 
@@ -147,6 +149,12 @@ static void chan_dev_release(struct device *dev)
 	struct dma_chan_dev *chan_dev;
 
 	chan_dev = container_of(dev, typeof(*chan_dev), device);
+	if (atomic_dec_and_test(chan_dev->idr_ref)) {
+		mutex_lock(&dma_list_mutex);
+		idr_remove(&dma_idr, chan_dev->dev_id);
+		mutex_unlock(&dma_list_mutex);
+		kfree(chan_dev->idr_ref);
+	}
 	kfree(chan_dev);
 }
 
@@ -611,9 +619,9 @@ EXPORT_SYMBOL(dmaengine_put);
  */
 int dma_async_device_register(struct dma_device *device)
 {
-	static int id;
 	int chancnt = 0, rc;
 	struct dma_chan* chan;
+	atomic_t *idr_ref;
 
 	if (!device)
 		return -ENODEV;
@@ -640,9 +648,20 @@ int dma_async_device_register(struct dma_device *device)
 	BUG_ON(!device->device_issue_pending);
 	BUG_ON(!device->dev);
 
+	idr_ref = kmalloc(sizeof(*idr_ref), GFP_KERNEL);
+	if (!idr_ref)
+		return -ENOMEM;
+	atomic_set(idr_ref, 0);
+ idr_retry:
+	if (!idr_pre_get(&dma_idr, GFP_KERNEL))
+		return -ENOMEM;
 	mutex_lock(&dma_list_mutex);
-	device->dev_id = id++;
+	rc = idr_get_new(&dma_idr, NULL, &device->dev_id);
 	mutex_unlock(&dma_list_mutex);
+	if (rc == -EAGAIN)
+		goto idr_retry;
+	else if (rc != 0)
+		return rc;
 
 	/* represent channels in sysfs. Probably want devs too */
 	list_for_each_entry(chan, &device->channels, device_node) {
@@ -659,6 +678,9 @@ int dma_async_device_register(struct dma_device *device)
 		chan->dev->device.class = &dma_devclass;
 		chan->dev->device.parent = device->dev;
 		chan->dev->chan = chan;
+		chan->dev->idr_ref = idr_ref;
+		chan->dev->dev_id = device->dev_id;
+		atomic_inc(idr_ref);
 		dev_set_name(&chan->dev->device, "dma%dchan%d",
 			     device->dev_id, chan->chan_id);
 
@@ -971,6 +993,7 @@ EXPORT_SYMBOL_GPL(dma_run_dependencies);
 
 static int __init dma_bus_init(void)
 {
+	idr_init(&dma_idr);
 	mutex_init(&dma_list_mutex);
 	return class_register(&dma_devclass);
 }
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index d6b6bff355f4..64dea2ab326c 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -140,10 +140,14 @@ struct dma_chan {
  * struct dma_chan_dev - relate sysfs device node to backing channel device
  * @chan - driver channel device
  * @device - sysfs device
+ * @dev_id - parent dma_device dev_id
+ * @idr_ref - reference count to gate release of dma_device dev_id
  */
 struct dma_chan_dev {
 	struct dma_chan *chan;
 	struct device device;
+	int dev_id;
+	atomic_t *idr_ref;
 };
 
 static inline const char *dma_chan_name(struct dma_chan *chan)
-- 
cgit v1.2.3


From adf094931ffb25ef4b381559918f1a34181a5273 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 6 Oct 2008 22:46:05 +0200
Subject: PM: Simplify the new suspend/hibernation framework for devices

PM: Simplify the new suspend/hibernation framework for devices

Following the discussion at the Kernel Summit, simplify the new
device PM framework by merging 'struct pm_ops' and
'struct pm_ext_ops' and removing pointers to 'struct pm_ext_ops'
from 'struct platform_driver' and 'struct pci_driver'.

After this change, the suspend/hibernation callbacks will only
reside in 'struct device_driver' as well as at the bus type/
device class/device type level.  Accordingly, PCI and platform
device drivers are now expected to put their suspend/hibernation
callbacks into the 'struct device_driver' embedded in
'struct pci_driver' or 'struct platform_driver', respectively.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@suse.cz>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/platform.c         | 115 +++++++++++++++++++++-------------------
 drivers/base/power/main.c       |  19 +++----
 drivers/pci/pci-driver.c        |  46 +++++++---------
 drivers/usb/core/usb.c          |   4 +-
 include/linux/device.h          |   8 +--
 include/linux/pci.h             |   1 -
 include/linux/platform_device.h |   1 -
 include/linux/pm.h              |  76 +++++++++-----------------
 8 files changed, 119 insertions(+), 151 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/platform.c b/drivers/base/platform.c
index dfcbfe504867..6c743b6008d9 100644
--- a/drivers/base/platform.c
+++ b/drivers/base/platform.c
@@ -503,8 +503,6 @@ int platform_driver_register(struct platform_driver *drv)
 		drv->driver.suspend = platform_drv_suspend;
 	if (drv->resume)
 		drv->driver.resume = platform_drv_resume;
-	if (drv->pm)
-		drv->driver.pm = &drv->pm->base;
 	return driver_register(&drv->driver);
 }
 EXPORT_SYMBOL_GPL(platform_driver_register);
@@ -686,7 +684,10 @@ static int platform_pm_suspend(struct device *dev)
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->suspend)
 			ret = drv->pm->suspend(dev);
 	} else {
@@ -698,16 +699,15 @@ static int platform_pm_suspend(struct device *dev)
 
 static int platform_pm_suspend_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->suspend_noirq)
-			ret = pdrv->pm->suspend_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->suspend_noirq)
+			ret = drv->pm->suspend_noirq(dev);
 	} else {
 		ret = platform_legacy_suspend_late(dev, PMSG_SUSPEND);
 	}
@@ -720,7 +720,10 @@ static int platform_pm_resume(struct device *dev)
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->resume)
 			ret = drv->pm->resume(dev);
 	} else {
@@ -732,16 +735,15 @@ static int platform_pm_resume(struct device *dev)
 
 static int platform_pm_resume_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->resume_noirq)
-			ret = pdrv->pm->resume_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->resume_noirq)
+			ret = drv->pm->resume_noirq(dev);
 	} else {
 		ret = platform_legacy_resume_early(dev);
 	}
@@ -780,16 +782,15 @@ static int platform_pm_freeze(struct device *dev)
 
 static int platform_pm_freeze_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->freeze_noirq)
-			ret = pdrv->pm->freeze_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->freeze_noirq)
+			ret = drv->pm->freeze_noirq(dev);
 	} else {
 		ret = platform_legacy_suspend_late(dev, PMSG_FREEZE);
 	}
@@ -802,7 +803,10 @@ static int platform_pm_thaw(struct device *dev)
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->thaw)
 			ret = drv->pm->thaw(dev);
 	} else {
@@ -814,16 +818,15 @@ static int platform_pm_thaw(struct device *dev)
 
 static int platform_pm_thaw_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->thaw_noirq)
-			ret = pdrv->pm->thaw_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->thaw_noirq)
+			ret = drv->pm->thaw_noirq(dev);
 	} else {
 		ret = platform_legacy_resume_early(dev);
 	}
@@ -836,7 +839,10 @@ static int platform_pm_poweroff(struct device *dev)
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->poweroff)
 			ret = drv->pm->poweroff(dev);
 	} else {
@@ -848,16 +854,15 @@ static int platform_pm_poweroff(struct device *dev)
 
 static int platform_pm_poweroff_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->poweroff_noirq)
-			ret = pdrv->pm->poweroff_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->poweroff_noirq)
+			ret = drv->pm->poweroff_noirq(dev);
 	} else {
 		ret = platform_legacy_suspend_late(dev, PMSG_HIBERNATE);
 	}
@@ -870,7 +875,10 @@ static int platform_pm_restore(struct device *dev)
 	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (drv && drv->pm) {
+	if (!drv)
+		return 0;
+
+	if (drv->pm) {
 		if (drv->pm->restore)
 			ret = drv->pm->restore(dev);
 	} else {
@@ -882,16 +890,15 @@ static int platform_pm_restore(struct device *dev)
 
 static int platform_pm_restore_noirq(struct device *dev)
 {
-	struct platform_driver *pdrv;
+	struct device_driver *drv = dev->driver;
 	int ret = 0;
 
-	if (!dev->driver)
+	if (!drv)
 		return 0;
 
-	pdrv = to_platform_driver(dev->driver);
-	if (pdrv->pm) {
-		if (pdrv->pm->restore_noirq)
-			ret = pdrv->pm->restore_noirq(dev);
+	if (drv->pm) {
+		if (drv->pm->restore_noirq)
+			ret = drv->pm->restore_noirq(dev);
 	} else {
 		ret = platform_legacy_resume_early(dev);
 	}
@@ -912,17 +919,15 @@ static int platform_pm_restore_noirq(struct device *dev)
 
 #endif /* !CONFIG_HIBERNATION */
 
-static struct pm_ext_ops platform_pm_ops = {
-	.base = {
-		.prepare = platform_pm_prepare,
-		.complete = platform_pm_complete,
-		.suspend = platform_pm_suspend,
-		.resume = platform_pm_resume,
-		.freeze = platform_pm_freeze,
-		.thaw = platform_pm_thaw,
-		.poweroff = platform_pm_poweroff,
-		.restore = platform_pm_restore,
-	},
+static struct dev_pm_ops platform_dev_pm_ops = {
+	.prepare = platform_pm_prepare,
+	.complete = platform_pm_complete,
+	.suspend = platform_pm_suspend,
+	.resume = platform_pm_resume,
+	.freeze = platform_pm_freeze,
+	.thaw = platform_pm_thaw,
+	.poweroff = platform_pm_poweroff,
+	.restore = platform_pm_restore,
 	.suspend_noirq = platform_pm_suspend_noirq,
 	.resume_noirq = platform_pm_resume_noirq,
 	.freeze_noirq = platform_pm_freeze_noirq,
@@ -931,7 +936,7 @@ static struct pm_ext_ops platform_pm_ops = {
 	.restore_noirq = platform_pm_restore_noirq,
 };
 
-#define PLATFORM_PM_OPS_PTR	&platform_pm_ops
+#define PLATFORM_PM_OPS_PTR	(&platform_dev_pm_ops)
 
 #else /* !CONFIG_PM_SLEEP */
 
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index 692c20ba5144..a8e4dcbcaf7a 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -112,7 +112,8 @@ void device_pm_remove(struct device *dev)
  *	@ops:	PM operations to choose from.
  *	@state:	PM transition of the system being carried out.
  */
-static int pm_op(struct device *dev, struct pm_ops *ops, pm_message_t state)
+static int pm_op(struct device *dev, struct dev_pm_ops *ops,
+			pm_message_t state)
 {
 	int error = 0;
 
@@ -174,7 +175,7 @@ static int pm_op(struct device *dev, struct pm_ops *ops, pm_message_t state)
  *	The operation is executed with interrupts disabled by the only remaining
  *	functional CPU in the system.
  */
-static int pm_noirq_op(struct device *dev, struct pm_ext_ops *ops,
+static int pm_noirq_op(struct device *dev, struct dev_pm_ops *ops,
 			pm_message_t state)
 {
 	int error = 0;
@@ -354,7 +355,7 @@ static int resume_device(struct device *dev, pm_message_t state)
 	if (dev->bus) {
 		if (dev->bus->pm) {
 			pm_dev_dbg(dev, state, "");
-			error = pm_op(dev, &dev->bus->pm->base, state);
+			error = pm_op(dev, dev->bus->pm, state);
 		} else if (dev->bus->resume) {
 			pm_dev_dbg(dev, state, "legacy ");
 			error = dev->bus->resume(dev);
@@ -451,9 +452,9 @@ static void complete_device(struct device *dev, pm_message_t state)
 		dev->type->pm->complete(dev);
 	}
 
-	if (dev->bus && dev->bus->pm && dev->bus->pm->base.complete) {
+	if (dev->bus && dev->bus->pm && dev->bus->pm->complete) {
 		pm_dev_dbg(dev, state, "completing ");
-		dev->bus->pm->base.complete(dev);
+		dev->bus->pm->complete(dev);
 	}
 
 	up(&dev->sem);
@@ -624,7 +625,7 @@ static int suspend_device(struct device *dev, pm_message_t state)
 	if (dev->bus) {
 		if (dev->bus->pm) {
 			pm_dev_dbg(dev, state, "");
-			error = pm_op(dev, &dev->bus->pm->base, state);
+			error = pm_op(dev, dev->bus->pm, state);
 		} else if (dev->bus->suspend) {
 			pm_dev_dbg(dev, state, "legacy ");
 			error = dev->bus->suspend(dev, state);
@@ -685,10 +686,10 @@ static int prepare_device(struct device *dev, pm_message_t state)
 
 	down(&dev->sem);
 
-	if (dev->bus && dev->bus->pm && dev->bus->pm->base.prepare) {
+	if (dev->bus && dev->bus->pm && dev->bus->pm->prepare) {
 		pm_dev_dbg(dev, state, "preparing ");
-		error = dev->bus->pm->base.prepare(dev);
-		suspend_report_result(dev->bus->pm->base.prepare, error);
+		error = dev->bus->pm->prepare(dev);
+		suspend_report_result(dev->bus->pm->prepare, error);
 		if (error)
 			goto End;
 	}
diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index b4cdd690ae71..4042d211c3e5 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -433,8 +433,7 @@ static int pci_pm_suspend(struct device *dev)
 
 static int pci_pm_suspend_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	if (drv && drv->pm) {
@@ -469,11 +468,10 @@ static int pci_pm_resume(struct device *dev)
 
 static int pci_pm_resume_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
-	pci_fixup_device(pci_fixup_resume_early, pci_dev);
+	pci_fixup_device(pci_fixup_resume_early, to_pci_dev(dev));
 
 	if (drv && drv->pm) {
 		if (drv->pm->resume_noirq)
@@ -519,8 +517,7 @@ static int pci_pm_freeze(struct device *dev)
 
 static int pci_pm_freeze_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	if (drv && drv->pm) {
@@ -553,15 +550,14 @@ static int pci_pm_thaw(struct device *dev)
 
 static int pci_pm_thaw_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	if (drv && drv->pm) {
 		if (drv->pm->thaw_noirq)
 			error = drv->pm->thaw_noirq(dev);
 	} else {
-		pci_fixup_device(pci_fixup_resume_early, pci_dev);
+		pci_fixup_device(pci_fixup_resume_early, to_pci_dev(dev));
 		error = pci_legacy_resume_early(dev);
 	}
 
@@ -589,8 +585,7 @@ static int pci_pm_poweroff(struct device *dev)
 
 static int pci_pm_poweroff_noirq(struct device *dev)
 {
-	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	if (drv && drv->pm) {
@@ -625,7 +620,7 @@ static int pci_pm_restore(struct device *dev)
 static int pci_pm_restore_noirq(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
-	struct pci_driver *drv = pci_dev->driver;
+	struct device_driver *drv = dev->driver;
 	int error = 0;
 
 	pci_fixup_device(pci_fixup_resume, pci_dev);
@@ -654,17 +649,15 @@ static int pci_pm_restore_noirq(struct device *dev)
 
 #endif /* !CONFIG_HIBERNATION */
 
-struct pm_ext_ops pci_pm_ops = {
-	.base = {
-		.prepare = pci_pm_prepare,
-		.complete = pci_pm_complete,
-		.suspend = pci_pm_suspend,
-		.resume = pci_pm_resume,
-		.freeze = pci_pm_freeze,
-		.thaw = pci_pm_thaw,
-		.poweroff = pci_pm_poweroff,
-		.restore = pci_pm_restore,
-	},
+struct dev_pm_ops pci_dev_pm_ops = {
+	.prepare = pci_pm_prepare,
+	.complete = pci_pm_complete,
+	.suspend = pci_pm_suspend,
+	.resume = pci_pm_resume,
+	.freeze = pci_pm_freeze,
+	.thaw = pci_pm_thaw,
+	.poweroff = pci_pm_poweroff,
+	.restore = pci_pm_restore,
 	.suspend_noirq = pci_pm_suspend_noirq,
 	.resume_noirq = pci_pm_resume_noirq,
 	.freeze_noirq = pci_pm_freeze_noirq,
@@ -673,7 +666,7 @@ struct pm_ext_ops pci_pm_ops = {
 	.restore_noirq = pci_pm_restore_noirq,
 };
 
-#define PCI_PM_OPS_PTR	&pci_pm_ops
+#define PCI_PM_OPS_PTR	(&pci_dev_pm_ops)
 
 #else /* !CONFIG_PM_SLEEP */
 
@@ -703,9 +696,6 @@ int __pci_register_driver(struct pci_driver *drv, struct module *owner,
 	drv->driver.owner = owner;
 	drv->driver.mod_name = mod_name;
 
-	if (drv->pm)
-		drv->driver.pm = &drv->pm->base;
-
 	spin_lock_init(&drv->dynids.lock);
 	INIT_LIST_HEAD(&drv->dynids.list);
 
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index be1fa0723f2c..399e15fc5052 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -286,7 +286,7 @@ static int usb_dev_restore(struct device *dev)
 	return usb_resume(dev);
 }
 
-static struct pm_ops usb_device_pm_ops = {
+static struct dev_pm_ops usb_device_pm_ops = {
 	.prepare =	usb_dev_prepare,
 	.complete =	usb_dev_complete,
 	.suspend =	usb_dev_suspend,
@@ -301,7 +301,7 @@ static struct pm_ops usb_device_pm_ops = {
 
 #define ksuspend_usb_init()	0
 #define ksuspend_usb_cleanup()	do {} while (0)
-#define usb_device_pm_ops	(*(struct pm_ops *)0)
+#define usb_device_pm_ops	(*(struct dev_pm_ops *)0)
 
 #endif	/* CONFIG_PM */
 
diff --git a/include/linux/device.h b/include/linux/device.h
index 1a3686d15f98..4a520051c315 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -65,7 +65,7 @@ struct bus_type {
 	int (*resume_early)(struct device *dev);
 	int (*resume)(struct device *dev);
 
-	struct pm_ext_ops *pm;
+	struct dev_pm_ops *pm;
 
 	struct bus_type_private *p;
 };
@@ -133,7 +133,7 @@ struct device_driver {
 	int (*resume) (struct device *dev);
 	struct attribute_group **groups;
 
-	struct pm_ops *pm;
+	struct dev_pm_ops *pm;
 
 	struct driver_private *p;
 };
@@ -198,7 +198,7 @@ struct class {
 	int (*suspend)(struct device *dev, pm_message_t state);
 	int (*resume)(struct device *dev);
 
-	struct pm_ops *pm;
+	struct dev_pm_ops *pm;
 	struct class_private *p;
 };
 
@@ -291,7 +291,7 @@ struct device_type {
 	int (*suspend)(struct device *dev, pm_message_t state);
 	int (*resume)(struct device *dev);
 
-	struct pm_ops *pm;
+	struct dev_pm_ops *pm;
 };
 
 /* interface for exporting device attributes */
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 03b0b8c3c81b..4bb156ba854a 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -421,7 +421,6 @@ struct pci_driver {
 	int  (*resume_early) (struct pci_dev *dev);
 	int  (*resume) (struct pci_dev *dev);	                /* Device woken up */
 	void (*shutdown) (struct pci_dev *dev);
-	struct pm_ext_ops *pm;
 	struct pci_error_handlers *err_handler;
 	struct device_driver	driver;
 	struct pci_dynids dynids;
diff --git a/include/linux/platform_device.h b/include/linux/platform_device.h
index 4b8cc6a32479..9a342699c607 100644
--- a/include/linux/platform_device.h
+++ b/include/linux/platform_device.h
@@ -55,7 +55,6 @@ struct platform_driver {
 	int (*suspend_late)(struct platform_device *, pm_message_t state);
 	int (*resume_early)(struct platform_device *);
 	int (*resume)(struct platform_device *);
-	struct pm_ext_ops *pm;
 	struct device_driver driver;
 };
 
diff --git a/include/linux/pm.h b/include/linux/pm.h
index 42de4003c4ee..5785666d0cc4 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -41,7 +41,7 @@ typedef struct pm_message {
 } pm_message_t;
 
 /**
- * struct pm_ops - device PM callbacks
+ * struct dev_pm_ops - device PM callbacks
  *
  * Several driver power state transitions are externally visible, affecting
  * the state of pending I/O queues and (for drivers that touch hardware)
@@ -126,46 +126,6 @@ typedef struct pm_message {
  *	On most platforms, there are no restrictions on availability of
  *	resources like clocks during @restore().
  *
- * All of the above callbacks, except for @complete(), return error codes.
- * However, the error codes returned by the resume operations, @resume(),
- * @thaw(), and @restore(), do not cause the PM core to abort the resume
- * transition during which they are returned.  The error codes returned in
- * that cases are only printed by the PM core to the system logs for debugging
- * purposes.  Still, it is recommended that drivers only return error codes
- * from their resume methods in case of an unrecoverable failure (i.e. when the
- * device being handled refuses to resume and becomes unusable) to allow us to
- * modify the PM core in the future, so that it can avoid attempting to handle
- * devices that failed to resume and their children.
- *
- * It is allowed to unregister devices while the above callbacks are being
- * executed.  However, it is not allowed to unregister a device from within any
- * of its own callbacks.
- */
-
-struct pm_ops {
-	int (*prepare)(struct device *dev);
-	void (*complete)(struct device *dev);
-	int (*suspend)(struct device *dev);
-	int (*resume)(struct device *dev);
-	int (*freeze)(struct device *dev);
-	int (*thaw)(struct device *dev);
-	int (*poweroff)(struct device *dev);
-	int (*restore)(struct device *dev);
-};
-
-/**
- * struct pm_ext_ops - extended device PM callbacks
- *
- * Some devices require certain operations related to suspend and hibernation
- * to be carried out with interrupts disabled.  Thus, 'struct pm_ext_ops' below
- * is defined, adding callbacks to be executed with interrupts disabled to
- * 'struct pm_ops'.
- *
- * The following callbacks included in 'struct pm_ext_ops' are executed with
- * the nonboot CPUs switched off and with interrupts disabled on the only
- * functional CPU.  They also are executed with the PM core list of devices
- * locked, so they must NOT unregister any devices.
- *
  * @suspend_noirq: Complete the operations of ->suspend() by carrying out any
  *	actions required for suspending the device that need interrupts to be
  *	disabled
@@ -190,18 +150,32 @@ struct pm_ops {
  *	actions required for restoring the operations of the device that need
  *	interrupts to be disabled
  *
- * All of the above callbacks return error codes, but the error codes returned
- * by the resume operations, @resume_noirq(), @thaw_noirq(), and
- * @restore_noirq(), do not cause the PM core to abort the resume transition
- * during which they are returned.  The error codes returned in that cases are
- * only printed by the PM core to the system logs for debugging purposes.
- * Still, as stated above, it is recommended that drivers only return error
- * codes from their resume methods if the device being handled fails to resume
- * and is not usable any more.
+ * All of the above callbacks, except for @complete(), return error codes.
+ * However, the error codes returned by the resume operations, @resume(),
+ * @thaw(), @restore(), @resume_noirq(), @thaw_noirq(), and @restore_noirq() do
+ * not cause the PM core to abort the resume transition during which they are
+ * returned.  The error codes returned in that cases are only printed by the PM
+ * core to the system logs for debugging purposes.  Still, it is recommended
+ * that drivers only return error codes from their resume methods in case of an
+ * unrecoverable failure (i.e. when the device being handled refuses to resume
+ * and becomes unusable) to allow us to modify the PM core in the future, so
+ * that it can avoid attempting to handle devices that failed to resume and
+ * their children.
+ *
+ * It is allowed to unregister devices while the above callbacks are being
+ * executed.  However, it is not allowed to unregister a device from within any
+ * of its own callbacks.
  */
 
-struct pm_ext_ops {
-	struct pm_ops base;
+struct dev_pm_ops {
+	int (*prepare)(struct device *dev);
+	void (*complete)(struct device *dev);
+	int (*suspend)(struct device *dev);
+	int (*resume)(struct device *dev);
+	int (*freeze)(struct device *dev);
+	int (*thaw)(struct device *dev);
+	int (*poweroff)(struct device *dev);
+	int (*restore)(struct device *dev);
 	int (*suspend_noirq)(struct device *dev);
 	int (*resume_noirq)(struct device *dev);
 	int (*freeze_noirq)(struct device *dev);
-- 
cgit v1.2.3


From 7f4f5d4516b441d712fa0ffe5380618fb7fc545e Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 17 Nov 2008 11:14:19 -0500
Subject: Fix misspellings in pm.h macros

This patch (as1167) fixes some misspellings in various recently-added
macros in pm.h.  Fortunately these macros are not yet used anywhere.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Acked-by: Rafael J. Wysocki <rjw@sisk.pl>
---
 include/linux/pm.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/pm.h b/include/linux/pm.h
index 5785666d0cc4..de2e0a8f6728 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -252,7 +252,7 @@ struct dev_pm_ops {
 #define PM_EVENT_SLEEP		(PM_EVENT_SUSPEND | PM_EVENT_HIBERNATE)
 #define PM_EVENT_USER_SUSPEND	(PM_EVENT_USER | PM_EVENT_SUSPEND)
 #define PM_EVENT_USER_RESUME	(PM_EVENT_USER | PM_EVENT_RESUME)
-#define PM_EVENT_REMOTE_WAKEUP	(PM_EVENT_REMOTE | PM_EVENT_RESUME)
+#define PM_EVENT_REMOTE_RESUME	(PM_EVENT_REMOTE | PM_EVENT_RESUME)
 #define PM_EVENT_AUTO_SUSPEND	(PM_EVENT_AUTO | PM_EVENT_SUSPEND)
 #define PM_EVENT_AUTO_RESUME	(PM_EVENT_AUTO | PM_EVENT_RESUME)
 
@@ -265,15 +265,15 @@ struct dev_pm_ops {
 #define PMSG_THAW	((struct pm_message){ .event = PM_EVENT_THAW, })
 #define PMSG_RESTORE	((struct pm_message){ .event = PM_EVENT_RESTORE, })
 #define PMSG_RECOVER	((struct pm_message){ .event = PM_EVENT_RECOVER, })
-#define PMSG_USER_SUSPEND	((struct pm_messge) \
+#define PMSG_USER_SUSPEND	((struct pm_message) \
 					{ .event = PM_EVENT_USER_SUSPEND, })
-#define PMSG_USER_RESUME	((struct pm_messge) \
+#define PMSG_USER_RESUME	((struct pm_message) \
 					{ .event = PM_EVENT_USER_RESUME, })
-#define PMSG_REMOTE_RESUME	((struct pm_messge) \
+#define PMSG_REMOTE_RESUME	((struct pm_message) \
 					{ .event = PM_EVENT_REMOTE_RESUME, })
-#define PMSG_AUTO_SUSPEND	((struct pm_messge) \
+#define PMSG_AUTO_SUSPEND	((struct pm_message) \
 					{ .event = PM_EVENT_AUTO_SUSPEND, })
-#define PMSG_AUTO_RESUME		((struct pm_messge) \
+#define PMSG_AUTO_RESUME	((struct pm_message) \
 					{ .event = PM_EVENT_AUTO_RESUME, })
 
 /**
-- 
cgit v1.2.3


From 929d2fa5955ab27aa21fac615b23e0e92e8dc3a0 Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Thu, 16 Oct 2008 15:51:35 -0600
Subject: driver core: Rearrange struct device for better packing

This minor rearrangement saves 16 bytes from sizeof(struct device)
according to pahole.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/device.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/device.h b/include/linux/device.h
index 4a520051c315..4e14fad41430 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -373,9 +373,9 @@ struct device {
 
 	struct kobject kobj;
 	char	bus_id[BUS_ID_SIZE];	/* position on parent bus */
+	unsigned		uevent_suppress:1;
 	const char		*init_name; /* initial name of the device */
 	struct device_type	*type;
-	unsigned		uevent_suppress:1;
 
 	struct semaphore	sem;	/* semaphore to synchronize calls to
 					 * its driver.
@@ -408,12 +408,13 @@ struct device {
 	/* arch specific additions */
 	struct dev_archdata	archdata;
 
+	dev_t			devt;	/* dev_t, creates the sysfs "dev" */
+
 	spinlock_t		devres_lock;
 	struct list_head	devres_head;
 
 	struct klist_node	knode_class;
 	struct class		*class;
-	dev_t			devt;	/* dev_t, creates the sysfs "dev" */
 	struct attribute_group	**groups;	/* optional groups */
 
 	void	(*release)(struct device *dev);
-- 
cgit v1.2.3


From 210272a28465a7a31bcd580d2f9529f924965aa5 Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Thu, 16 Oct 2008 14:57:54 -0600
Subject: driver core: Remove completion from struct klist_node

Removing the completion from klist_node reduces its size from 64 bytes
to 28 on x86-64.  To maintain the semantics of klist_remove(), we add
a single list of klist nodes which are pending deletion and scan them.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/klist.h |  2 --
 lib/klist.c           | 43 ++++++++++++++++++++++++++++++++++++++++---
 2 files changed, 40 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/klist.h b/include/linux/klist.h
index 8ea98db223e5..d5a27af9dba5 100644
--- a/include/linux/klist.h
+++ b/include/linux/klist.h
@@ -13,7 +13,6 @@
 #define _LINUX_KLIST_H
 
 #include <linux/spinlock.h>
-#include <linux/completion.h>
 #include <linux/kref.h>
 #include <linux/list.h>
 
@@ -41,7 +40,6 @@ struct klist_node {
 	void			*n_klist;	/* never access directly */
 	struct list_head	n_node;
 	struct kref		n_ref;
-	struct completion	n_removed;
 };
 
 extern void klist_add_tail(struct klist_node *n, struct klist *k);
diff --git a/lib/klist.c b/lib/klist.c
index bbdd3015c2c7..573d6068a42e 100644
--- a/lib/klist.c
+++ b/lib/klist.c
@@ -36,6 +36,7 @@
 
 #include <linux/klist.h>
 #include <linux/module.h>
+#include <linux/sched.h>
 
 /*
  * Use the lowest bit of n_klist to mark deleted nodes and exclude
@@ -108,7 +109,6 @@ static void add_tail(struct klist *k, struct klist_node *n)
 static void klist_node_init(struct klist *k, struct klist_node *n)
 {
 	INIT_LIST_HEAD(&n->n_node);
-	init_completion(&n->n_removed);
 	kref_init(&n->n_ref);
 	knode_set_klist(n, k);
 	if (k->get)
@@ -171,13 +171,34 @@ void klist_add_before(struct klist_node *n, struct klist_node *pos)
 }
 EXPORT_SYMBOL_GPL(klist_add_before);
 
+struct klist_waiter {
+	struct list_head list;
+	struct klist_node *node;
+	struct task_struct *process;
+	int woken;
+};
+
+static DEFINE_SPINLOCK(klist_remove_lock);
+static LIST_HEAD(klist_remove_waiters);
+
 static void klist_release(struct kref *kref)
 {
+	struct klist_waiter *waiter, *tmp;
 	struct klist_node *n = container_of(kref, struct klist_node, n_ref);
 
 	WARN_ON(!knode_dead(n));
 	list_del(&n->n_node);
-	complete(&n->n_removed);
+	spin_lock(&klist_remove_lock);
+	list_for_each_entry_safe(waiter, tmp, &klist_remove_waiters, list) {
+		if (waiter->node != n)
+			continue;
+
+		waiter->woken = 1;
+		mb();
+		wake_up_process(waiter->process);
+		list_del(&waiter->list);
+	}
+	spin_unlock(&klist_remove_lock);
 	knode_set_klist(n, NULL);
 }
 
@@ -217,8 +238,24 @@ EXPORT_SYMBOL_GPL(klist_del);
  */
 void klist_remove(struct klist_node *n)
 {
+	struct klist_waiter waiter;
+
+	waiter.node = n;
+	waiter.process = current;
+	waiter.woken = 0;
+	spin_lock(&klist_remove_lock);
+	list_add(&waiter.list, &klist_remove_waiters);
+	spin_unlock(&klist_remove_lock);
+
 	klist_del(n);
-	wait_for_completion(&n->n_removed);
+
+	for (;;) {
+		set_current_state(TASK_UNINTERRUPTIBLE);
+		if (waiter.woken)
+			break;
+		schedule();
+	}
+	__set_current_state(TASK_RUNNING);
 }
 EXPORT_SYMBOL_GPL(klist_remove);
 
-- 
cgit v1.2.3


From 2831fe6f9cc4e16c103504ee09a47a084297c0f3 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 16 Dec 2008 12:23:36 -0800
Subject: driver core: create a private portion of struct device

This is to be used to move things out of struct device that no code
outside of the driver core should ever touch.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/base.h    | 12 ++++++++++++
 drivers/base/core.c    |  8 ++++++++
 include/linux/device.h |  3 +++
 3 files changed, 23 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/base/base.h b/drivers/base/base.h
index 0a5f055dffba..6b20809b5fd4 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -63,6 +63,18 @@ struct class_private {
 #define to_class(obj)	\
 	container_of(obj, struct class_private, class_subsys.kobj)
 
+/**
+ * struct device_private - structure to hold the private to the driver core portions of the device structure.
+ *
+ * @device - pointer back to the struct class that this structure is
+ * associated with.
+ *
+ * Nothing outside of the driver core should ever touch these fields.
+ */
+struct device_private {
+	struct device *device;
+};
+
 /* initialisation functions */
 extern int devices_init(void);
 extern int buses_init(void);
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 14aa2b6953c9..6657787e64f8 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -109,6 +109,7 @@ static struct sysfs_ops dev_sysfs_ops = {
 static void device_release(struct kobject *kobj)
 {
 	struct device *dev = to_dev(kobj);
+	struct device_private *p = dev->p;
 
 	if (dev->release)
 		dev->release(dev);
@@ -120,6 +121,7 @@ static void device_release(struct kobject *kobj)
 		WARN(1, KERN_ERR "Device '%s' does not have a release() "
 			"function, it is broken and must be fixed.\n",
 			dev_name(dev));
+	kfree(p);
 }
 
 static struct kobj_type device_ktype = {
@@ -536,6 +538,12 @@ static void klist_children_put(struct klist_node *n)
  */
 void device_initialize(struct device *dev)
 {
+	dev->p = kzalloc(sizeof(*dev->p), GFP_KERNEL);
+	if (!dev->p) {
+		WARN_ON(1);
+		return;
+	}
+	dev->p->device = dev;
 	dev->kobj.kset = devices_kset;
 	kobject_init(&dev->kobj, &device_ktype);
 	klist_init(&dev->klist_children, klist_children_get,
diff --git a/include/linux/device.h b/include/linux/device.h
index 4e14fad41430..d6d34084fd37 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -28,6 +28,7 @@
 #define BUS_ID_SIZE		20
 
 struct device;
+struct device_private;
 struct device_driver;
 struct driver_private;
 struct class;
@@ -371,6 +372,8 @@ struct device {
 	struct klist_node	knode_bus;
 	struct device		*parent;
 
+	struct device_private	*p;
+
 	struct kobject kobj;
 	char	bus_id[BUS_ID_SIZE];	/* position on parent bus */
 	unsigned		uevent_suppress:1;
-- 
cgit v1.2.3


From 11c3b5c3e08f4d855cbef52883c266b9ab9df879 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 16 Dec 2008 12:24:56 -0800
Subject: driver core: move klist_children into private structure

Nothing outside of the driver core should ever touch klist_children, or
knode_parent, so move them out of the public eye.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/base.h    |  6 ++++++
 drivers/base/core.c    | 37 ++++++++++++++++++++++++-------------
 include/linux/device.h |  2 --
 3 files changed, 30 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/base.h b/drivers/base/base.h
index 6b20809b5fd4..f5cf31c664d7 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -66,14 +66,20 @@ struct class_private {
 /**
  * struct device_private - structure to hold the private to the driver core portions of the device structure.
  *
+ * @klist_children - klist containing all children of this device
+ * @knode_parent - node in sibling list
  * @device - pointer back to the struct class that this structure is
  * associated with.
  *
  * Nothing outside of the driver core should ever touch these fields.
  */
 struct device_private {
+	struct klist klist_children;
+	struct klist_node knode_parent;
 	struct device *device;
 };
+#define to_device_private_parent(obj)	\
+	container_of(obj, struct device_private, knode_parent)
 
 /* initialisation functions */
 extern int devices_init(void);
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 6657787e64f8..180ff84ea26c 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -509,14 +509,16 @@ EXPORT_SYMBOL_GPL(device_schedule_callback_owner);
 
 static void klist_children_get(struct klist_node *n)
 {
-	struct device *dev = container_of(n, struct device, knode_parent);
+	struct device_private *p = to_device_private_parent(n);
+	struct device *dev = p->device;
 
 	get_device(dev);
 }
 
 static void klist_children_put(struct klist_node *n)
 {
-	struct device *dev = container_of(n, struct device, knode_parent);
+	struct device_private *p = to_device_private_parent(n);
+	struct device *dev = p->device;
 
 	put_device(dev);
 }
@@ -546,7 +548,7 @@ void device_initialize(struct device *dev)
 	dev->p->device = dev;
 	dev->kobj.kset = devices_kset;
 	kobject_init(&dev->kobj, &device_ktype);
-	klist_init(&dev->klist_children, klist_children_get,
+	klist_init(&dev->p->klist_children, klist_children_get,
 		   klist_children_put);
 	INIT_LIST_HEAD(&dev->dma_pools);
 	init_MUTEX(&dev->sem);
@@ -927,7 +929,8 @@ int device_add(struct device *dev)
 	kobject_uevent(&dev->kobj, KOBJ_ADD);
 	bus_attach_device(dev);
 	if (parent)
-		klist_add_tail(&dev->knode_parent, &parent->klist_children);
+		klist_add_tail(&dev->p->knode_parent,
+			       &parent->p->klist_children);
 
 	if (dev->class) {
 		mutex_lock(&dev->class->p->class_mutex);
@@ -1038,7 +1041,7 @@ void device_del(struct device *dev)
 	device_pm_remove(dev);
 	dpm_sysfs_remove(dev);
 	if (parent)
-		klist_del(&dev->knode_parent);
+		klist_del(&dev->p->knode_parent);
 	if (MAJOR(dev->devt)) {
 		device_remove_sys_dev_entry(dev);
 		device_remove_file(dev, &devt_attr);
@@ -1102,7 +1105,14 @@ void device_unregister(struct device *dev)
 static struct device *next_device(struct klist_iter *i)
 {
 	struct klist_node *n = klist_next(i);
-	return n ? container_of(n, struct device, knode_parent) : NULL;
+	struct device *dev = NULL;
+	struct device_private *p;
+
+	if (n) {
+		p = to_device_private_parent(n);
+		dev = p->device;
+	}
+	return dev;
 }
 
 /**
@@ -1124,7 +1134,7 @@ int device_for_each_child(struct device *parent, void *data,
 	struct device *child;
 	int error = 0;
 
-	klist_iter_init(&parent->klist_children, &i);
+	klist_iter_init(&parent->p->klist_children, &i);
 	while ((child = next_device(&i)) && !error)
 		error = fn(child, data);
 	klist_iter_exit(&i);
@@ -1155,7 +1165,7 @@ struct device *device_find_child(struct device *parent, void *data,
 	if (!parent)
 		return NULL;
 
-	klist_iter_init(&parent->klist_children, &i);
+	klist_iter_init(&parent->p->klist_children, &i);
 	while ((child = next_device(&i)))
 		if (match(child, data) && get_device(child))
 			break;
@@ -1478,9 +1488,10 @@ int device_move(struct device *dev, struct device *new_parent)
 	old_parent = dev->parent;
 	dev->parent = new_parent;
 	if (old_parent)
-		klist_remove(&dev->knode_parent);
+		klist_remove(&dev->p->knode_parent);
 	if (new_parent) {
-		klist_add_tail(&dev->knode_parent, &new_parent->klist_children);
+		klist_add_tail(&dev->p->knode_parent,
+			       &new_parent->p->klist_children);
 		set_dev_node(dev, dev_to_node(new_parent));
 	}
 
@@ -1492,11 +1503,11 @@ int device_move(struct device *dev, struct device *new_parent)
 		device_move_class_links(dev, new_parent, old_parent);
 		if (!kobject_move(&dev->kobj, &old_parent->kobj)) {
 			if (new_parent)
-				klist_remove(&dev->knode_parent);
+				klist_remove(&dev->p->knode_parent);
 			dev->parent = old_parent;
 			if (old_parent) {
-				klist_add_tail(&dev->knode_parent,
-					       &old_parent->klist_children);
+				klist_add_tail(&dev->p->knode_parent,
+					       &old_parent->p->klist_children);
 				set_dev_node(dev, dev_to_node(old_parent));
 			}
 		}
diff --git a/include/linux/device.h b/include/linux/device.h
index d6d34084fd37..60423e687205 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -366,8 +366,6 @@ struct device_dma_parameters {
 };
 
 struct device {
-	struct klist		klist_children;
-	struct klist_node	knode_parent;	/* node in sibling list */
 	struct klist_node	knode_driver;
 	struct klist_node	knode_bus;
 	struct device		*parent;
-- 
cgit v1.2.3


From 93e746db183b3bdbbda67900f79b5835f9cb388f Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 16 Dec 2008 12:25:49 -0800
Subject: driver core: move knode_driver into private structure

Nothing outside of the driver core should ever touch knode_driver, so
move it out of the public eye.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/base.h    |  4 ++++
 drivers/base/dd.c      | 13 ++++++++-----
 drivers/base/driver.c  | 13 ++++++++++---
 include/linux/device.h |  1 -
 4 files changed, 22 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/base.h b/drivers/base/base.h
index f5cf31c664d7..8af0bb2c0aa8 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -68,6 +68,7 @@ struct class_private {
  *
  * @klist_children - klist containing all children of this device
  * @knode_parent - node in sibling list
+ * @knode_driver - node in driver list
  * @device - pointer back to the struct class that this structure is
  * associated with.
  *
@@ -76,10 +77,13 @@ struct class_private {
 struct device_private {
 	struct klist klist_children;
 	struct klist_node knode_parent;
+	struct klist_node knode_driver;
 	struct device *device;
 };
 #define to_device_private_parent(obj)	\
 	container_of(obj, struct device_private, knode_parent)
+#define to_device_private_driver(obj)	\
+	container_of(obj, struct device_private, knode_driver)
 
 /* initialisation functions */
 extern int devices_init(void);
diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 17a8e45cf9c6..bb5f1eb83c03 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -28,7 +28,7 @@
 
 static void driver_bound(struct device *dev)
 {
-	if (klist_node_attached(&dev->knode_driver)) {
+	if (klist_node_attached(&dev->p->knode_driver)) {
 		printk(KERN_WARNING "%s: device %s already bound\n",
 			__func__, kobject_name(&dev->kobj));
 		return;
@@ -41,7 +41,7 @@ static void driver_bound(struct device *dev)
 		blocking_notifier_call_chain(&dev->bus->p->bus_notifier,
 					     BUS_NOTIFY_BOUND_DRIVER, dev);
 
-	klist_add_tail(&dev->knode_driver, &dev->driver->p->klist_devices);
+	klist_add_tail(&dev->p->knode_driver, &dev->driver->p->klist_devices);
 }
 
 static int driver_sysfs_add(struct device *dev)
@@ -311,7 +311,7 @@ static void __device_release_driver(struct device *dev)
 			drv->remove(dev);
 		devres_release_all(dev);
 		dev->driver = NULL;
-		klist_remove(&dev->knode_driver);
+		klist_remove(&dev->p->knode_driver);
 	}
 }
 
@@ -341,6 +341,7 @@ EXPORT_SYMBOL_GPL(device_release_driver);
  */
 void driver_detach(struct device_driver *drv)
 {
+	struct device_private *dev_prv;
 	struct device *dev;
 
 	for (;;) {
@@ -349,8 +350,10 @@ void driver_detach(struct device_driver *drv)
 			spin_unlock(&drv->p->klist_devices.k_lock);
 			break;
 		}
-		dev = list_entry(drv->p->klist_devices.k_list.prev,
-				struct device, knode_driver.n_node);
+		dev_prv = list_entry(drv->p->klist_devices.k_list.prev,
+				     struct device_private,
+				     knode_driver.n_node);
+		dev = dev_prv->device;
 		get_device(dev);
 		spin_unlock(&drv->p->klist_devices.k_lock);
 
diff --git a/drivers/base/driver.c b/drivers/base/driver.c
index 1e2bda780e48..b76cc69f1106 100644
--- a/drivers/base/driver.c
+++ b/drivers/base/driver.c
@@ -19,7 +19,14 @@
 static struct device *next_device(struct klist_iter *i)
 {
 	struct klist_node *n = klist_next(i);
-	return n ? container_of(n, struct device, knode_driver) : NULL;
+	struct device *dev = NULL;
+	struct device_private *dev_prv;
+
+	if (n) {
+		dev_prv = to_device_private_driver(n);
+		dev = dev_prv->device;
+	}
+	return dev;
 }
 
 /**
@@ -42,7 +49,7 @@ int driver_for_each_device(struct device_driver *drv, struct device *start,
 		return -EINVAL;
 
 	klist_iter_init_node(&drv->p->klist_devices, &i,
-			     start ? &start->knode_driver : NULL);
+			     start ? &start->p->knode_driver : NULL);
 	while ((dev = next_device(&i)) && !error)
 		error = fn(dev, data);
 	klist_iter_exit(&i);
@@ -76,7 +83,7 @@ struct device *driver_find_device(struct device_driver *drv,
 		return NULL;
 
 	klist_iter_init_node(&drv->p->klist_devices, &i,
-			     (start ? &start->knode_driver : NULL));
+			     (start ? &start->p->knode_driver : NULL));
 	while ((dev = next_device(&i)))
 		if (match(dev, data) && get_device(dev))
 			break;
diff --git a/include/linux/device.h b/include/linux/device.h
index 60423e687205..e3630222c3c1 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -366,7 +366,6 @@ struct device_dma_parameters {
 };
 
 struct device {
-	struct klist_node	knode_driver;
 	struct klist_node	knode_bus;
 	struct device		*parent;
 
-- 
cgit v1.2.3


From b9daa99ee533578e3f88231e7a16784dcb44ec42 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 16 Dec 2008 12:26:21 -0800
Subject: driver core: move knode_bus into private structure

Nothing outside of the driver core should ever touch knode_bus, so
move it out of the public eye.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/base.h    |  4 ++++
 drivers/base/bus.c     | 40 +++++++++++++++++++++++++++-------------
 include/linux/device.h |  1 -
 3 files changed, 31 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/base.h b/drivers/base/base.h
index 8af0bb2c0aa8..b676f8f801f8 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -69,6 +69,7 @@ struct class_private {
  * @klist_children - klist containing all children of this device
  * @knode_parent - node in sibling list
  * @knode_driver - node in driver list
+ * @knode_bus - node in bus list
  * @device - pointer back to the struct class that this structure is
  * associated with.
  *
@@ -78,12 +79,15 @@ struct device_private {
 	struct klist klist_children;
 	struct klist_node knode_parent;
 	struct klist_node knode_driver;
+	struct klist_node knode_bus;
 	struct device *device;
 };
 #define to_device_private_parent(obj)	\
 	container_of(obj, struct device_private, knode_parent)
 #define to_device_private_driver(obj)	\
 	container_of(obj, struct device_private, knode_driver)
+#define to_device_private_bus(obj)	\
+	container_of(obj, struct device_private, knode_bus)
 
 /* initialisation functions */
 extern int devices_init(void);
diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 83f32b891fa9..0f0a50444672 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -253,7 +253,14 @@ static ssize_t store_drivers_probe(struct bus_type *bus,
 static struct device *next_device(struct klist_iter *i)
 {
 	struct klist_node *n = klist_next(i);
-	return n ? container_of(n, struct device, knode_bus) : NULL;
+	struct device *dev = NULL;
+	struct device_private *dev_prv;
+
+	if (n) {
+		dev_prv = to_device_private_bus(n);
+		dev = dev_prv->device;
+	}
+	return dev;
 }
 
 /**
@@ -286,7 +293,7 @@ int bus_for_each_dev(struct bus_type *bus, struct device *start,
 		return -EINVAL;
 
 	klist_iter_init_node(&bus->p->klist_devices, &i,
-			     (start ? &start->knode_bus : NULL));
+			     (start ? &start->p->knode_bus : NULL));
 	while ((dev = next_device(&i)) && !error)
 		error = fn(dev, data);
 	klist_iter_exit(&i);
@@ -320,7 +327,7 @@ struct device *bus_find_device(struct bus_type *bus,
 		return NULL;
 
 	klist_iter_init_node(&bus->p->klist_devices, &i,
-			     (start ? &start->knode_bus : NULL));
+			     (start ? &start->p->knode_bus : NULL));
 	while ((dev = next_device(&i)))
 		if (match(dev, data) && get_device(dev))
 			break;
@@ -507,7 +514,8 @@ void bus_attach_device(struct device *dev)
 			ret = device_attach(dev);
 		WARN_ON(ret < 0);
 		if (ret >= 0)
-			klist_add_tail(&dev->knode_bus, &bus->p->klist_devices);
+			klist_add_tail(&dev->p->knode_bus,
+				       &bus->p->klist_devices);
 	}
 }
 
@@ -528,8 +536,8 @@ void bus_remove_device(struct device *dev)
 		sysfs_remove_link(&dev->bus->p->devices_kset->kobj,
 				  dev_name(dev));
 		device_remove_attrs(dev->bus, dev);
-		if (klist_node_attached(&dev->knode_bus))
-			klist_del(&dev->knode_bus);
+		if (klist_node_attached(&dev->p->knode_bus))
+			klist_del(&dev->p->knode_bus);
 
 		pr_debug("bus: '%s': remove device %s\n",
 			 dev->bus->name, dev_name(dev));
@@ -831,14 +839,16 @@ static void bus_remove_attrs(struct bus_type *bus)
 
 static void klist_devices_get(struct klist_node *n)
 {
-	struct device *dev = container_of(n, struct device, knode_bus);
+	struct device_private *dev_prv = to_device_private_bus(n);
+	struct device *dev = dev_prv->device;
 
 	get_device(dev);
 }
 
 static void klist_devices_put(struct klist_node *n)
 {
-	struct device *dev = container_of(n, struct device, knode_bus);
+	struct device_private *dev_prv = to_device_private_bus(n);
+	struct device *dev = dev_prv->device;
 
 	put_device(dev);
 }
@@ -993,18 +1003,20 @@ static void device_insertion_sort_klist(struct device *a, struct list_head *list
 {
 	struct list_head *pos;
 	struct klist_node *n;
+	struct device_private *dev_prv;
 	struct device *b;
 
 	list_for_each(pos, list) {
 		n = container_of(pos, struct klist_node, n_node);
-		b = container_of(n, struct device, knode_bus);
+		dev_prv = to_device_private_bus(n);
+		b = dev_prv->device;
 		if (compare(a, b) <= 0) {
-			list_move_tail(&a->knode_bus.n_node,
-				       &b->knode_bus.n_node);
+			list_move_tail(&a->p->knode_bus.n_node,
+				       &b->p->knode_bus.n_node);
 			return;
 		}
 	}
-	list_move_tail(&a->knode_bus.n_node, list);
+	list_move_tail(&a->p->knode_bus.n_node, list);
 }
 
 void bus_sort_breadthfirst(struct bus_type *bus,
@@ -1014,6 +1026,7 @@ void bus_sort_breadthfirst(struct bus_type *bus,
 	LIST_HEAD(sorted_devices);
 	struct list_head *pos, *tmp;
 	struct klist_node *n;
+	struct device_private *dev_prv;
 	struct device *dev;
 	struct klist *device_klist;
 
@@ -1022,7 +1035,8 @@ void bus_sort_breadthfirst(struct bus_type *bus,
 	spin_lock(&device_klist->k_lock);
 	list_for_each_safe(pos, tmp, &device_klist->k_list) {
 		n = container_of(pos, struct klist_node, n_node);
-		dev = container_of(n, struct device, knode_bus);
+		dev_prv = to_device_private_bus(n);
+		dev = dev_prv->device;
 		device_insertion_sort_klist(dev, &sorted_devices, compare);
 	}
 	list_splice(&sorted_devices, &device_klist->k_list);
diff --git a/include/linux/device.h b/include/linux/device.h
index e3630222c3c1..e21b5d69d67c 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -366,7 +366,6 @@ struct device_dma_parameters {
 };
 
 struct device {
-	struct klist_node	knode_bus;
 	struct device		*parent;
 
 	struct device_private	*p;
-- 
cgit v1.2.3


From d0d85ff989222f08dd1fa66321fef5567bbc4a7b Mon Sep 17 00:00:00 2001
From: Cornelia Huck <cornelia.huck@de.ibm.com>
Date: Thu, 4 Dec 2008 16:55:47 +0100
Subject: Make DEBUG take precedence over DYNAMIC_PRINTK_DEBUG

Statically defined DEBUG should take precedence over
dynamically enabled debugging; otherwise adding DEBUG
(like, for example, via CONFIG_DEBUG_KOBJECT) does not
have the expected result of printing pr_debug() and dev_dbg()
messages unconditionally.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Acked-by: Jason Baron <jbaron@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/device.h | 8 ++++----
 include/linux/kernel.h | 8 ++++----
 2 files changed, 8 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/device.h b/include/linux/device.h
index e21b5d69d67c..b97a0cf1eb05 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -553,13 +553,13 @@ extern const char *dev_driver_string(const struct device *dev);
 #define dev_info(dev, format, arg...)		\
 	dev_printk(KERN_INFO , dev , format , ## arg)
 
-#if defined(CONFIG_DYNAMIC_PRINTK_DEBUG)
+#if defined(DEBUG)
+#define dev_dbg(dev, format, arg...)		\
+	dev_printk(KERN_DEBUG , dev , format , ## arg)
+#elif defined(CONFIG_DYNAMIC_PRINTK_DEBUG)
 #define dev_dbg(dev, format, ...) do { \
 	dynamic_dev_dbg(dev, format, ##__VA_ARGS__); \
 	} while (0)
-#elif defined(DEBUG)
-#define dev_dbg(dev, format, arg...)		\
-	dev_printk(KERN_DEBUG , dev , format , ## arg)
 #else
 #define dev_dbg(dev, format, arg...)		\
 	({ if (0) dev_printk(KERN_DEBUG, dev, format, ##arg); 0; })
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index ca9ff6411dfa..d242fe1381fd 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -349,13 +349,13 @@ static inline char *pack_hex_byte(char *buf, u8 byte)
         printk(KERN_INFO pr_fmt(fmt), ##__VA_ARGS__)
 
 /* If you are writing a driver, please use dev_dbg instead */
-#if defined(CONFIG_DYNAMIC_PRINTK_DEBUG)
+#if defined(DEBUG)
+#define pr_debug(fmt, ...) \
+	printk(KERN_DEBUG pr_fmt(fmt), ##__VA_ARGS__)
+#elif defined(CONFIG_DYNAMIC_PRINTK_DEBUG)
 #define pr_debug(fmt, ...) do { \
 	dynamic_pr_debug(pr_fmt(fmt), ##__VA_ARGS__); \
 	} while (0)
-#elif defined(DEBUG)
-#define pr_debug(fmt, ...) \
-	printk(KERN_DEBUG pr_fmt(fmt), ##__VA_ARGS__)
 #else
 #define pr_debug(fmt, ...) \
 	({ if (0) printk(KERN_DEBUG pr_fmt(fmt), ##__VA_ARGS__); 0; })
-- 
cgit v1.2.3


From 0aa0dc41bfd993491c2344870eee7a3b218551fb Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Mon, 15 Dec 2008 12:58:26 +0000
Subject: driver core: add root_device_register()

Add support for allocating root device objects which group
device objects under /sys/devices directories.

Also add a sysfs 'module' symlink which points to the owner
of the root device object. This symlink will be used in virtio
to allow userspace to determine which virtio bus implementation
a given device is associated with.

[Includes suggestions from Cornelia Huck]

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 91 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/device.h | 11 ++++++
 2 files changed, 102 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/base/core.c b/drivers/base/core.c
index ee555d7d5c3f..61df508fa62b 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -1217,6 +1217,97 @@ EXPORT_SYMBOL_GPL(put_device);
 EXPORT_SYMBOL_GPL(device_create_file);
 EXPORT_SYMBOL_GPL(device_remove_file);
 
+struct root_device
+{
+	struct device dev;
+	struct module *owner;
+};
+
+#define to_root_device(dev) container_of(dev, struct root_device, dev)
+
+static void root_device_release(struct device *dev)
+{
+	kfree(to_root_device(dev));
+}
+
+/**
+ * __root_device_register - allocate and register a root device
+ * @name: root device name
+ * @owner: owner module of the root device, usually THIS_MODULE
+ *
+ * This function allocates a root device and registers it
+ * using device_register(). In order to free the returned
+ * device, use root_device_unregister().
+ *
+ * Root devices are dummy devices which allow other devices
+ * to be grouped under /sys/devices. Use this function to
+ * allocate a root device and then use it as the parent of
+ * any device which should appear under /sys/devices/{name}
+ *
+ * The /sys/devices/{name} directory will also contain a
+ * 'module' symlink which points to the @owner directory
+ * in sysfs.
+ *
+ * Note: You probably want to use root_device_register().
+ */
+struct device *__root_device_register(const char *name, struct module *owner)
+{
+	struct root_device *root;
+	int err = -ENOMEM;
+
+	root = kzalloc(sizeof(struct root_device), GFP_KERNEL);
+	if (!root)
+		return ERR_PTR(err);
+
+	err = dev_set_name(&root->dev, name);
+	if (err) {
+		kfree(root);
+		return ERR_PTR(err);
+	}
+
+	root->dev.release = root_device_release;
+
+	err = device_register(&root->dev);
+	if (err) {
+		put_device(&root->dev);
+		return ERR_PTR(err);
+	}
+
+#ifdef CONFIG_MODULE	/* gotta find a "cleaner" way to do this */
+	if (owner) {
+		struct module_kobject *mk = &owner->mkobj;
+
+		err = sysfs_create_link(&root->dev.kobj, &mk->kobj, "module");
+		if (err) {
+			device_unregister(&root->dev);
+			return ERR_PTR(err);
+		}
+		root->owner = owner;
+	}
+#endif
+
+	return &root->dev;
+}
+EXPORT_SYMBOL_GPL(__root_device_register);
+
+/**
+ * root_device_unregister - unregister and free a root device
+ * @root: device going away.
+ *
+ * This function unregisters and cleans up a device that was created by
+ * root_device_register().
+ */
+void root_device_unregister(struct device *dev)
+{
+	struct root_device *root = to_root_device(dev);
+
+	if (root->owner)
+		sysfs_remove_link(&root->dev.kobj, "module");
+
+	device_unregister(dev);
+}
+EXPORT_SYMBOL_GPL(root_device_unregister);
+
 
 static void device_create_release(struct device *dev)
 {
diff --git a/include/linux/device.h b/include/linux/device.h
index b97a0cf1eb05..7d9da4b4993f 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -482,6 +482,17 @@ extern struct device *device_find_child(struct device *dev, void *data,
 extern int device_rename(struct device *dev, char *new_name);
 extern int device_move(struct device *dev, struct device *new_parent);
 
+/*
+ * Root device objects for grouping under /sys/devices
+ */
+extern struct device *__root_device_register(const char *name,
+					     struct module *owner);
+static inline struct device *root_device_register(const char *name)
+{
+	return __root_device_register(name, THIS_MODULE);
+}
+extern void root_device_unregister(struct device *root);
+
 /*
  * Manual binding of a device to driver. See drivers/base/bus.c
  * for information on use.
-- 
cgit v1.2.3


From 475b44c19913b877537c8bc19799f75b0b142641 Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Tue, 6 Jan 2009 10:44:38 -0800
Subject: mtd: struct device - replace bus_id with dev_name(), dev_set_name()

CC: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/mtd/devices/m25p80.c        | 16 ++++++++--------
 drivers/mtd/devices/mtd_dataflash.c | 30 +++++++++++++++---------------
 drivers/mtd/maps/integrator-flash.c |  2 +-
 drivers/mtd/maps/ixp2000.c          |  4 ++--
 drivers/mtd/maps/ixp4xx.c           |  2 +-
 drivers/mtd/maps/omap_nor.c         |  2 +-
 drivers/mtd/maps/physmap.c          |  6 +++---
 drivers/mtd/maps/physmap_of.c       |  4 ++--
 drivers/mtd/mtdconcat.c             |  2 +-
 drivers/mtd/nand/fsl_upm.c          |  2 +-
 drivers/mtd/nand/plat_nand.c        |  2 +-
 drivers/mtd/nand/tmio_nand.c        |  2 +-
 drivers/mtd/onenand/generic.c       |  2 +-
 drivers/mtd/onenand/omap2.c         |  2 +-
 drivers/mtd/ubi/build.c             |  2 +-
 drivers/mtd/ubi/vmt.c               |  4 ++--
 include/linux/mtd/concat.h          |  2 +-
 17 files changed, 43 insertions(+), 43 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/mtd/devices/m25p80.c b/drivers/mtd/devices/m25p80.c
index 6659b2275c0c..5733f0643843 100644
--- a/drivers/mtd/devices/m25p80.c
+++ b/drivers/mtd/devices/m25p80.c
@@ -170,7 +170,7 @@ static int wait_till_ready(struct m25p *flash)
 static int erase_chip(struct m25p *flash)
 {
 	DEBUG(MTD_DEBUG_LEVEL3, "%s: %s %dKiB\n",
-			flash->spi->dev.bus_id, __func__,
+			dev_name(&flash->spi->dev), __func__,
 			flash->mtd.size / 1024);
 
 	/* Wait until finished previous write command. */
@@ -197,7 +197,7 @@ static int erase_chip(struct m25p *flash)
 static int erase_sector(struct m25p *flash, u32 offset)
 {
 	DEBUG(MTD_DEBUG_LEVEL3, "%s: %s %dKiB at 0x%08x\n",
-			flash->spi->dev.bus_id, __func__,
+			dev_name(&flash->spi->dev), __func__,
 			flash->mtd.erasesize / 1024, offset);
 
 	/* Wait until finished previous write command. */
@@ -234,7 +234,7 @@ static int m25p80_erase(struct mtd_info *mtd, struct erase_info *instr)
 	u32 addr,len;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: %s %s 0x%08x, len %d\n",
-			flash->spi->dev.bus_id, __func__, "at",
+			dev_name(&flash->spi->dev), __func__, "at",
 			(u32)instr->addr, instr->len);
 
 	/* sanity checks */
@@ -295,7 +295,7 @@ static int m25p80_read(struct mtd_info *mtd, loff_t from, size_t len,
 	struct spi_message m;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: %s %s 0x%08x, len %zd\n",
-			flash->spi->dev.bus_id, __func__, "from",
+			dev_name(&flash->spi->dev), __func__, "from",
 			(u32)from, len);
 
 	/* sanity checks */
@@ -367,7 +367,7 @@ static int m25p80_write(struct mtd_info *mtd, loff_t to, size_t len,
 	struct spi_message m;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: %s %s 0x%08x, len %zd\n",
-			flash->spi->dev.bus_id, __func__, "to",
+			dev_name(&flash->spi->dev), __func__, "to",
 			(u32)to, len);
 
 	if (retlen)
@@ -563,7 +563,7 @@ static struct flash_info *__devinit jedec_probe(struct spi_device *spi)
 	tmp = spi_write_then_read(spi, &code, 1, id, 5);
 	if (tmp < 0) {
 		DEBUG(MTD_DEBUG_LEVEL0, "%s: error %d reading JEDEC ID\n",
-			spi->dev.bus_id, tmp);
+			dev_name(&spi->dev), tmp);
 		return NULL;
 	}
 	jedec = id[0];
@@ -617,7 +617,7 @@ static int __devinit m25p_probe(struct spi_device *spi)
 		/* unrecognized chip? */
 		if (i == ARRAY_SIZE(m25p_data)) {
 			DEBUG(MTD_DEBUG_LEVEL0, "%s: unrecognized id %s\n",
-					spi->dev.bus_id, data->type);
+					dev_name(&spi->dev), data->type);
 			info = NULL;
 
 		/* recognized; is that chip really what's there? */
@@ -658,7 +658,7 @@ static int __devinit m25p_probe(struct spi_device *spi)
 	if (data && data->name)
 		flash->mtd.name = data->name;
 	else
-		flash->mtd.name = spi->dev.bus_id;
+		flash->mtd.name = dev_name(&spi->dev);
 
 	flash->mtd.type = MTD_NORFLASH;
 	flash->mtd.writesize = 1;
diff --git a/drivers/mtd/devices/mtd_dataflash.c b/drivers/mtd/devices/mtd_dataflash.c
index 6dd9aff8bb2d..65126cd668ff 100644
--- a/drivers/mtd/devices/mtd_dataflash.c
+++ b/drivers/mtd/devices/mtd_dataflash.c
@@ -128,7 +128,7 @@ static int dataflash_waitready(struct spi_device *spi)
 		status = dataflash_status(spi);
 		if (status < 0) {
 			DEBUG(MTD_DEBUG_LEVEL1, "%s: status %d?\n",
-					spi->dev.bus_id, status);
+					dev_name(&spi->dev), status);
 			status = 0;
 		}
 
@@ -154,7 +154,7 @@ static int dataflash_erase(struct mtd_info *mtd, struct erase_info *instr)
 	uint8_t			*command;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: erase addr=0x%x len 0x%x\n",
-			spi->dev.bus_id,
+			dev_name(&spi->dev),
 			instr->addr, instr->len);
 
 	/* Sanity checks */
@@ -197,7 +197,7 @@ static int dataflash_erase(struct mtd_info *mtd, struct erase_info *instr)
 
 		if (status < 0) {
 			printk(KERN_ERR "%s: erase %x, err %d\n",
-				spi->dev.bus_id, pageaddr, status);
+				dev_name(&spi->dev), pageaddr, status);
 			/* REVISIT:  can retry instr->retries times; or
 			 * giveup and instr->fail_addr = instr->addr;
 			 */
@@ -239,7 +239,7 @@ static int dataflash_read(struct mtd_info *mtd, loff_t from, size_t len,
 	int			status;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: read 0x%x..0x%x\n",
-		priv->spi->dev.bus_id, (unsigned)from, (unsigned)(from + len));
+		dev_name(&priv->spi->dev), (unsigned)from, (unsigned)(from + len));
 
 	*retlen = 0;
 
@@ -288,7 +288,7 @@ static int dataflash_read(struct mtd_info *mtd, loff_t from, size_t len,
 		status = 0;
 	} else
 		DEBUG(MTD_DEBUG_LEVEL1, "%s: read %x..%x --> %d\n",
-			priv->spi->dev.bus_id,
+			dev_name(&priv->spi->dev),
 			(unsigned)from, (unsigned)(from + len),
 			status);
 	return status;
@@ -315,7 +315,7 @@ static int dataflash_write(struct mtd_info *mtd, loff_t to, size_t len,
 	uint8_t			*command;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: write 0x%x..0x%x\n",
-		spi->dev.bus_id, (unsigned)to, (unsigned)(to + len));
+		dev_name(&spi->dev), (unsigned)to, (unsigned)(to + len));
 
 	*retlen = 0;
 
@@ -374,7 +374,7 @@ static int dataflash_write(struct mtd_info *mtd, loff_t to, size_t len,
 			status = spi_sync(spi, &msg);
 			if (status < 0)
 				DEBUG(MTD_DEBUG_LEVEL1, "%s: xfer %u -> %d \n",
-					spi->dev.bus_id, addr, status);
+					dev_name(&spi->dev), addr, status);
 
 			(void) dataflash_waitready(priv->spi);
 		}
@@ -396,7 +396,7 @@ static int dataflash_write(struct mtd_info *mtd, loff_t to, size_t len,
 		spi_transfer_del(x + 1);
 		if (status < 0)
 			DEBUG(MTD_DEBUG_LEVEL1, "%s: pgm %u/%u -> %d \n",
-				spi->dev.bus_id, addr, writelen, status);
+				dev_name(&spi->dev), addr, writelen, status);
 
 		(void) dataflash_waitready(priv->spi);
 
@@ -416,14 +416,14 @@ static int dataflash_write(struct mtd_info *mtd, loff_t to, size_t len,
 		status = spi_sync(spi, &msg);
 		if (status < 0)
 			DEBUG(MTD_DEBUG_LEVEL1, "%s: compare %u -> %d \n",
-				spi->dev.bus_id, addr, status);
+				dev_name(&spi->dev), addr, status);
 
 		status = dataflash_waitready(priv->spi);
 
 		/* Check result of the compare operation */
 		if (status & (1 << 6)) {
 			printk(KERN_ERR "%s: compare page %u, err %d\n",
-				spi->dev.bus_id, pageaddr, status);
+				dev_name(&spi->dev), pageaddr, status);
 			remaining = 0;
 			status = -EIO;
 			break;
@@ -779,7 +779,7 @@ static struct flash_info *__devinit jedec_probe(struct spi_device *spi)
 	tmp = spi_write_then_read(spi, &code, 1, id, 3);
 	if (tmp < 0) {
 		DEBUG(MTD_DEBUG_LEVEL0, "%s: error %d reading JEDEC ID\n",
-			spi->dev.bus_id, tmp);
+			dev_name(&spi->dev), tmp);
 		return ERR_PTR(tmp);
 	}
 	if (id[0] != 0x1f)
@@ -869,7 +869,7 @@ static int __devinit dataflash_probe(struct spi_device *spi)
 	status = dataflash_status(spi);
 	if (status <= 0 || status == 0xff) {
 		DEBUG(MTD_DEBUG_LEVEL1, "%s: status error %d\n",
-				spi->dev.bus_id, status);
+				dev_name(&spi->dev), status);
 		if (status == 0 || status == 0xff)
 			status = -ENODEV;
 		return status;
@@ -905,13 +905,13 @@ static int __devinit dataflash_probe(struct spi_device *spi)
 	/* obsolete AT45DB1282 not (yet?) supported */
 	default:
 		DEBUG(MTD_DEBUG_LEVEL1, "%s: unsupported device (%x)\n",
-				spi->dev.bus_id, status & 0x3c);
+				dev_name(&spi->dev), status & 0x3c);
 		status = -ENODEV;
 	}
 
 	if (status < 0)
 		DEBUG(MTD_DEBUG_LEVEL1, "%s: add_dataflash --> %d\n",
-				spi->dev.bus_id, status);
+				dev_name(&spi->dev), status);
 
 	return status;
 }
@@ -921,7 +921,7 @@ static int __devexit dataflash_remove(struct spi_device *spi)
 	struct dataflash	*flash = dev_get_drvdata(&spi->dev);
 	int			status;
 
-	DEBUG(MTD_DEBUG_LEVEL1, "%s: remove\n", spi->dev.bus_id);
+	DEBUG(MTD_DEBUG_LEVEL1, "%s: remove\n", dev_name(&spi->dev));
 
 	if (mtd_has_partitions() && flash->partitioned)
 		status = del_mtd_partitions(&flash->mtd);
diff --git a/drivers/mtd/maps/integrator-flash.c b/drivers/mtd/maps/integrator-flash.c
index 7100ee3c7b01..d2ec262666c7 100644
--- a/drivers/mtd/maps/integrator-flash.c
+++ b/drivers/mtd/maps/integrator-flash.c
@@ -105,7 +105,7 @@ static int armflash_probe(struct platform_device *dev)
 	info->map.bankwidth	= plat->width;
 	info->map.phys		= res->start;
 	info->map.virt		= base;
-	info->map.name		= dev->dev.bus_id;
+	info->map.name		= dev_name(&dev->dev);
 	info->map.set_vpp	= armflash_set_vpp;
 
 	simple_map_init(&info->map);
diff --git a/drivers/mtd/maps/ixp2000.c b/drivers/mtd/maps/ixp2000.c
index 3ea1de9be720..d4fb9a3ab4df 100644
--- a/drivers/mtd/maps/ixp2000.c
+++ b/drivers/mtd/maps/ixp2000.c
@@ -188,7 +188,7 @@ static int ixp2000_flash_probe(struct platform_device *dev)
  	 */
 	info->map.map_priv_2 = (unsigned long) ixp_data->bank_setup;
 
-	info->map.name = dev->dev.bus_id;
+	info->map.name = dev_name(&dev->dev);
 	info->map.read = ixp2000_flash_read8;
 	info->map.write = ixp2000_flash_write8;
 	info->map.copy_from = ixp2000_flash_copy_from;
@@ -196,7 +196,7 @@ static int ixp2000_flash_probe(struct platform_device *dev)
 
 	info->res = request_mem_region(dev->resource->start,
 			dev->resource->end - dev->resource->start + 1,
-			dev->dev.bus_id);
+			dev_name(&dev->dev));
 	if (!info->res) {
 		dev_err(&dev->dev, "Could not reserve memory region\n");
 		err = -ENOMEM;
diff --git a/drivers/mtd/maps/ixp4xx.c b/drivers/mtd/maps/ixp4xx.c
index 16555cbeaea4..7214b876feba 100644
--- a/drivers/mtd/maps/ixp4xx.c
+++ b/drivers/mtd/maps/ixp4xx.c
@@ -218,7 +218,7 @@ static int ixp4xx_flash_probe(struct platform_device *dev)
 	 * handle that.
 	 */
 	info->map.bankwidth = 2;
-	info->map.name = dev->dev.bus_id;
+	info->map.name = dev_name(&dev->dev);
 	info->map.read = ixp4xx_read16,
 	info->map.write = ixp4xx_probe_write16,
 	info->map.copy_from = ixp4xx_copy_from,
diff --git a/drivers/mtd/maps/omap_nor.c b/drivers/mtd/maps/omap_nor.c
index 05f276af15da..7e50e9b1b781 100644
--- a/drivers/mtd/maps/omap_nor.c
+++ b/drivers/mtd/maps/omap_nor.c
@@ -101,7 +101,7 @@ static int __init omapflash_probe(struct platform_device *pdev)
 		err = -ENOMEM;
 		goto out_release_mem_region;
 	}
-	info->map.name		= pdev->dev.bus_id;
+	info->map.name		= dev_name(&pdev->dev);
 	info->map.phys		= res->start;
 	info->map.size		= size;
 	info->map.bankwidth	= pdata->width;
diff --git a/drivers/mtd/maps/physmap.c b/drivers/mtd/maps/physmap.c
index dfbf3f270cea..1db16e549e38 100644
--- a/drivers/mtd/maps/physmap.c
+++ b/drivers/mtd/maps/physmap.c
@@ -108,13 +108,13 @@ static int physmap_flash_probe(struct platform_device *dev)
 		if (!devm_request_mem_region(&dev->dev,
 			dev->resource[i].start,
 			dev->resource[i].end - dev->resource[i].start + 1,
-			dev->dev.bus_id)) {
+			dev_name(&dev->dev))) {
 			dev_err(&dev->dev, "Could not reserve memory region\n");
 			err = -ENOMEM;
 			goto err_out;
 		}
 
-		info->map[i].name = dev->dev.bus_id;
+		info->map[i].name = dev_name(&dev->dev);
 		info->map[i].phys = dev->resource[i].start;
 		info->map[i].size = dev->resource[i].end - dev->resource[i].start + 1;
 		info->map[i].bankwidth = physmap_data->width;
@@ -150,7 +150,7 @@ static int physmap_flash_probe(struct platform_device *dev)
 		 * We detected multiple devices. Concatenate them together.
 		 */
 #ifdef CONFIG_MTD_CONCAT
-		info->cmtd = mtd_concat_create(info->mtd, devices_found, dev->dev.bus_id);
+		info->cmtd = mtd_concat_create(info->mtd, devices_found, dev_name(&dev->dev));
 		if (info->cmtd == NULL)
 			err = -ENXIO;
 #else
diff --git a/drivers/mtd/maps/physmap_of.c b/drivers/mtd/maps/physmap_of.c
index 5fcfec034a94..fbf0ca939d72 100644
--- a/drivers/mtd/maps/physmap_of.c
+++ b/drivers/mtd/maps/physmap_of.c
@@ -183,7 +183,7 @@ static int __devinit of_flash_probe(struct of_device *dev,
 
 	err = -EBUSY;
 	info->res = request_mem_region(res.start, res.end - res.start + 1,
-				       dev->dev.bus_id);
+				       dev_name(&dev->dev));
 	if (!info->res)
 		goto err_out;
 
@@ -194,7 +194,7 @@ static int __devinit of_flash_probe(struct of_device *dev,
 		goto err_out;
 	}
 
-	info->map.name = dev->dev.bus_id;
+	info->map.name = dev_name(&dev->dev);
 	info->map.phys = res.start;
 	info->map.size = res.end - res.start + 1;
 	info->map.bankwidth = *width;
diff --git a/drivers/mtd/mtdconcat.c b/drivers/mtd/mtdconcat.c
index 789842d0e6f2..1a05cf37851e 100644
--- a/drivers/mtd/mtdconcat.c
+++ b/drivers/mtd/mtdconcat.c
@@ -691,7 +691,7 @@ static int concat_block_markbad(struct mtd_info *mtd, loff_t ofs)
  */
 struct mtd_info *mtd_concat_create(struct mtd_info *subdev[],	/* subdevices to concatenate */
 				   int num_devs,	/* number of subdevices      */
-				   char *name)
+				   const char *name)
 {				/* name for the new device   */
 	int i;
 	size_t size;
diff --git a/drivers/mtd/nand/fsl_upm.c b/drivers/mtd/nand/fsl_upm.c
index a83192f80eba..7815a404a632 100644
--- a/drivers/mtd/nand/fsl_upm.c
+++ b/drivers/mtd/nand/fsl_upm.c
@@ -222,7 +222,7 @@ static int __devinit fun_probe(struct of_device *ofdev,
 
 	fun->rnb_gpio = of_get_gpio(ofdev->node, 0);
 	if (fun->rnb_gpio >= 0) {
-		ret = gpio_request(fun->rnb_gpio, ofdev->dev.bus_id);
+		ret = gpio_request(fun->rnb_gpio, dev_name(&ofdev->dev));
 		if (ret) {
 			dev_err(&ofdev->dev, "can't request RNB gpio\n");
 			goto err2;
diff --git a/drivers/mtd/nand/plat_nand.c b/drivers/mtd/nand/plat_nand.c
index f674c5427b17..75f9f4874ecf 100644
--- a/drivers/mtd/nand/plat_nand.c
+++ b/drivers/mtd/nand/plat_nand.c
@@ -54,7 +54,7 @@ static int __init plat_nand_probe(struct platform_device *pdev)
 	data->chip.priv = &data;
 	data->mtd.priv = &data->chip;
 	data->mtd.owner = THIS_MODULE;
-	data->mtd.name = pdev->dev.bus_id;
+	data->mtd.name = dev_name(&pdev->dev);
 
 	data->chip.IO_ADDR_R = data->io_base;
 	data->chip.IO_ADDR_W = data->io_base;
diff --git a/drivers/mtd/nand/tmio_nand.c b/drivers/mtd/nand/tmio_nand.c
index edb1e322113d..daa6a4c3b8ce 100644
--- a/drivers/mtd/nand/tmio_nand.c
+++ b/drivers/mtd/nand/tmio_nand.c
@@ -433,7 +433,7 @@ static int tmio_probe(struct platform_device *dev)
 	nand_chip->chip_delay = 15;
 
 	retval = request_irq(irq, &tmio_irq,
-				IRQF_DISABLED, dev->dev.bus_id, tmio);
+				IRQF_DISABLED, dev_name(&dev->dev), tmio);
 	if (retval) {
 		dev_err(&dev->dev, "request_irq error %d\n", retval);
 		goto err_irq;
diff --git a/drivers/mtd/onenand/generic.c b/drivers/mtd/onenand/generic.c
index ad81ab8e95e2..5b69e7773c6c 100644
--- a/drivers/mtd/onenand/generic.c
+++ b/drivers/mtd/onenand/generic.c
@@ -63,7 +63,7 @@ static int __devinit generic_onenand_probe(struct device *dev)
 	info->onenand.mmcontrol = pdata->mmcontrol;
 	info->onenand.irq = platform_get_irq(pdev, 0);
 
-	info->mtd.name = pdev->dev.bus_id;
+	info->mtd.name = dev_name(&pdev->dev);
 	info->mtd.priv = &info->onenand;
 	info->mtd.owner = THIS_MODULE;
 
diff --git a/drivers/mtd/onenand/omap2.c b/drivers/mtd/onenand/omap2.c
index d1e0b8e7224b..96ecc1766fa8 100644
--- a/drivers/mtd/onenand/omap2.c
+++ b/drivers/mtd/onenand/omap2.c
@@ -668,7 +668,7 @@ static int __devinit omap2_onenand_probe(struct platform_device *pdev)
 		 c->onenand.base);
 
 	c->pdev = pdev;
-	c->mtd.name = pdev->dev.bus_id;
+	c->mtd.name = dev_name(&pdev->dev);
 	c->mtd.priv = &c->onenand;
 	c->mtd.owner = THIS_MODULE;
 
diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index ba0bd3d5775b..7caf22cd5ad0 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -280,7 +280,7 @@ static int ubi_sysfs_init(struct ubi_device *ubi)
 	ubi->dev.release = dev_release;
 	ubi->dev.devt = ubi->cdev.dev;
 	ubi->dev.class = ubi_class;
-	sprintf(&ubi->dev.bus_id[0], UBI_NAME_STR"%d", ubi->ubi_num);
+	dev_set_name(&ubi->dev, UBI_NAME_STR"%d", ubi->ubi_num);
 	err = device_register(&ubi->dev);
 	if (err)
 		return err;
diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c
index 3531ca9a1e24..22e1d7398fce 100644
--- a/drivers/mtd/ubi/vmt.c
+++ b/drivers/mtd/ubi/vmt.c
@@ -329,7 +329,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
 	vol->dev.devt = dev;
 	vol->dev.class = ubi_class;
 
-	sprintf(&vol->dev.bus_id[0], "%s_%d", ubi->ubi_name, vol->vol_id);
+	dev_set_name(&vol->dev, "%s_%d", ubi->ubi_name, vol->vol_id);
 	err = device_register(&vol->dev);
 	if (err) {
 		ubi_err("cannot register device");
@@ -678,7 +678,7 @@ int ubi_add_volume(struct ubi_device *ubi, struct ubi_volume *vol)
 	vol->dev.parent = &ubi->dev;
 	vol->dev.devt = dev;
 	vol->dev.class = ubi_class;
-	sprintf(&vol->dev.bus_id[0], "%s_%d", ubi->ubi_name, vol->vol_id);
+	dev_set_name(&vol->dev, "%s_%d", ubi->ubi_name, vol->vol_id);
 	err = device_register(&vol->dev);
 	if (err)
 		goto out_gluebi;
diff --git a/include/linux/mtd/concat.h b/include/linux/mtd/concat.h
index c02f3d264ecf..e80c674daeb3 100644
--- a/include/linux/mtd/concat.h
+++ b/include/linux/mtd/concat.h
@@ -13,7 +13,7 @@
 struct mtd_info *mtd_concat_create(
     struct mtd_info *subdev[],  /* subdevices to concatenate */
     int num_devs,               /* number of subdevices      */
-    char *name);                /* name for the new device   */
+    const char *name);          /* name for the new device   */
 
 void mtd_concat_destroy(struct mtd_info *mtd);
 
-- 
cgit v1.2.3


From e70c412ee45332db2636a8f5a35a0685efb0e4aa Mon Sep 17 00:00:00 2001
From: "Hans J. Koch" <hjk@linutronix.de>
Date: Sat, 6 Dec 2008 02:23:13 +0100
Subject: UIO: Pass information about ioports to userspace (V2)

Devices sometimes have memory where all or parts of it can not be mapped to
userspace. But it might still be possible to access this memory from
userspace by other means. An example are PCI cards that advertise not only
mappable memory but also ioport ranges. On x86 architectures, these can be
accessed with ioperm, iopl, inb, outb, and friends. Mike Frysinger (CCed)
reported a similar problem on Blackfin arch where it doesn't seem to be easy
to mmap non-cached memory but it can still be accessed from userspace.

This patch allows kernel drivers to pass information about such ports to
userspace. Similar to the existing mem[] array, it adds a port[] array to
struct uio_info. Each port range is described by start, size, and porttype.

If a driver fills in at least one such port range, the UIO core will simply
pass this information to userspace by creating a new directory "portio"
underneath /sys/class/uio/uioN/. Similar to the "mem" directory, it will
contain a subdirectory (portX) for each port range given.

Note that UIO simply passes this information to userspace, it performs no
action whatsoever with this data. It's userspace's responsibility to obtain
access to these ports and to solve arch dependent issues. The "porttype"
attribute tells userspace what kind of port it is dealing with.

This mechanism could also be used to give userspace information about GPIOs
related to a device. You frequently find such hardware in embedded devices,
so I added a UIO_PORT_GPIO definition. I'm not really sure if this is a good
idea since there are other solutions to this problem, but it won't hurt much
anyway.

Signed-off-by: Hans J. Koch <hjk@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/uio/uio.c          | 159 ++++++++++++++++++++++++++++++++++++++++-----
 include/linux/uio_driver.h |  26 ++++++++
 2 files changed, 168 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/uio/uio.c b/drivers/uio/uio.c
index 2d2440cd57a9..4ca85a113aa2 100644
--- a/drivers/uio/uio.c
+++ b/drivers/uio/uio.c
@@ -35,6 +35,7 @@ struct uio_device {
 	int			vma_count;
 	struct uio_info		*info;
 	struct kobject		*map_dir;
+	struct kobject		*portio_dir;
 };
 
 static int uio_major;
@@ -75,17 +76,17 @@ static ssize_t map_offset_show(struct uio_mem *mem, char *buf)
 	return sprintf(buf, "0x%lx\n", mem->addr & ~PAGE_MASK);
 }
 
-struct uio_sysfs_entry {
+struct map_sysfs_entry {
 	struct attribute attr;
 	ssize_t (*show)(struct uio_mem *, char *);
 	ssize_t (*store)(struct uio_mem *, const char *, size_t);
 };
 
-static struct uio_sysfs_entry addr_attribute =
+static struct map_sysfs_entry addr_attribute =
 	__ATTR(addr, S_IRUGO, map_addr_show, NULL);
-static struct uio_sysfs_entry size_attribute =
+static struct map_sysfs_entry size_attribute =
 	__ATTR(size, S_IRUGO, map_size_show, NULL);
-static struct uio_sysfs_entry offset_attribute =
+static struct map_sysfs_entry offset_attribute =
 	__ATTR(offset, S_IRUGO, map_offset_show, NULL);
 
 static struct attribute *attrs[] = {
@@ -106,9 +107,9 @@ static ssize_t map_type_show(struct kobject *kobj, struct attribute *attr,
 {
 	struct uio_map *map = to_map(kobj);
 	struct uio_mem *mem = map->mem;
-	struct uio_sysfs_entry *entry;
+	struct map_sysfs_entry *entry;
 
-	entry = container_of(attr, struct uio_sysfs_entry, attr);
+	entry = container_of(attr, struct map_sysfs_entry, attr);
 
 	if (!entry->show)
 		return -EIO;
@@ -116,16 +117,93 @@ static ssize_t map_type_show(struct kobject *kobj, struct attribute *attr,
 	return entry->show(mem, buf);
 }
 
-static struct sysfs_ops uio_sysfs_ops = {
+static struct sysfs_ops map_sysfs_ops = {
 	.show = map_type_show,
 };
 
 static struct kobj_type map_attr_type = {
 	.release	= map_release,
-	.sysfs_ops	= &uio_sysfs_ops,
+	.sysfs_ops	= &map_sysfs_ops,
 	.default_attrs	= attrs,
 };
 
+struct uio_portio {
+	struct kobject kobj;
+	struct uio_port *port;
+};
+#define to_portio(portio) container_of(portio, struct uio_portio, kobj)
+
+static ssize_t portio_start_show(struct uio_port *port, char *buf)
+{
+	return sprintf(buf, "0x%lx\n", port->start);
+}
+
+static ssize_t portio_size_show(struct uio_port *port, char *buf)
+{
+	return sprintf(buf, "0x%lx\n", port->size);
+}
+
+static ssize_t portio_porttype_show(struct uio_port *port, char *buf)
+{
+	const char *porttypes[] = {"none", "x86", "gpio", "other"};
+
+	if ((port->porttype < 0) || (port->porttype > UIO_PORT_OTHER))
+		return -EINVAL;
+
+	return sprintf(buf, "port_%s\n", porttypes[port->porttype]);
+}
+
+struct portio_sysfs_entry {
+	struct attribute attr;
+	ssize_t (*show)(struct uio_port *, char *);
+	ssize_t (*store)(struct uio_port *, const char *, size_t);
+};
+
+static struct portio_sysfs_entry portio_start_attribute =
+	__ATTR(start, S_IRUGO, portio_start_show, NULL);
+static struct portio_sysfs_entry portio_size_attribute =
+	__ATTR(size, S_IRUGO, portio_size_show, NULL);
+static struct portio_sysfs_entry portio_porttype_attribute =
+	__ATTR(porttype, S_IRUGO, portio_porttype_show, NULL);
+
+static struct attribute *portio_attrs[] = {
+	&portio_start_attribute.attr,
+	&portio_size_attribute.attr,
+	&portio_porttype_attribute.attr,
+	NULL,
+};
+
+static void portio_release(struct kobject *kobj)
+{
+	struct uio_portio *portio = to_portio(kobj);
+	kfree(portio);
+}
+
+static ssize_t portio_type_show(struct kobject *kobj, struct attribute *attr,
+			     char *buf)
+{
+	struct uio_portio *portio = to_portio(kobj);
+	struct uio_port *port = portio->port;
+	struct portio_sysfs_entry *entry;
+
+	entry = container_of(attr, struct portio_sysfs_entry, attr);
+
+	if (!entry->show)
+		return -EIO;
+
+	return entry->show(port, buf);
+}
+
+static struct sysfs_ops portio_sysfs_ops = {
+	.show = portio_type_show,
+};
+
+static struct kobj_type portio_attr_type = {
+	.release	= portio_release,
+	.sysfs_ops	= &portio_sysfs_ops,
+	.default_attrs	= portio_attrs,
+};
+
 static ssize_t show_name(struct device *dev,
 			 struct device_attribute *attr, char *buf)
 {
@@ -177,10 +255,13 @@ static struct attribute_group uio_attr_grp = {
 static int uio_dev_add_attributes(struct uio_device *idev)
 {
 	int ret;
-	int mi;
+	int mi, pi;
 	int map_found = 0;
+	int portio_found = 0;
 	struct uio_mem *mem;
 	struct uio_map *map;
+	struct uio_port *port;
+	struct uio_portio *portio;
 
 	ret = sysfs_create_group(&idev->dev->kobj, &uio_attr_grp);
 	if (ret)
@@ -195,25 +276,58 @@ static int uio_dev_add_attributes(struct uio_device *idev)
 			idev->map_dir = kobject_create_and_add("maps",
 							&idev->dev->kobj);
 			if (!idev->map_dir)
-				goto err;
+				goto err_map;
 		}
 		map = kzalloc(sizeof(*map), GFP_KERNEL);
 		if (!map)
-			goto err;
+			goto err_map;
 		kobject_init(&map->kobj, &map_attr_type);
 		map->mem = mem;
 		mem->map = map;
 		ret = kobject_add(&map->kobj, idev->map_dir, "map%d", mi);
 		if (ret)
-			goto err;
+			goto err_map;
 		ret = kobject_uevent(&map->kobj, KOBJ_ADD);
 		if (ret)
-			goto err;
+			goto err_map;
+	}
+
+	for (pi = 0; pi < MAX_UIO_PORT_REGIONS; pi++) {
+		port = &idev->info->port[pi];
+		if (port->size == 0)
+			break;
+		if (!portio_found) {
+			portio_found = 1;
+			idev->portio_dir = kobject_create_and_add("portio",
+							&idev->dev->kobj);
+			if (!idev->portio_dir)
+				goto err_portio;
+		}
+		portio = kzalloc(sizeof(*portio), GFP_KERNEL);
+		if (!portio)
+			goto err_portio;
+		kobject_init(&portio->kobj, &portio_attr_type);
+		portio->port = port;
+		port->portio = portio;
+		ret = kobject_add(&portio->kobj, idev->portio_dir,
+							"port%d", pi);
+		if (ret)
+			goto err_portio;
+		ret = kobject_uevent(&portio->kobj, KOBJ_ADD);
+		if (ret)
+			goto err_portio;
 	}
 
 	return 0;
 
-err:
+err_portio:
+	for (pi--; pi >= 0; pi--) {
+		port = &idev->info->port[pi];
+		portio = port->portio;
+		kobject_put(&portio->kobj);
+	}
+	kobject_put(idev->portio_dir);
+err_map:
 	for (mi--; mi>=0; mi--) {
 		mem = &idev->info->mem[mi];
 		map = mem->map;
@@ -228,15 +342,26 @@ err_group:
 
 static void uio_dev_del_attributes(struct uio_device *idev)
 {
-	int mi;
+	int i;
 	struct uio_mem *mem;
-	for (mi = 0; mi < MAX_UIO_MAPS; mi++) {
-		mem = &idev->info->mem[mi];
+	struct uio_port *port;
+
+	for (i = 0; i < MAX_UIO_MAPS; i++) {
+		mem = &idev->info->mem[i];
 		if (mem->size == 0)
 			break;
 		kobject_put(&mem->map->kobj);
 	}
 	kobject_put(idev->map_dir);
+
+	for (i = 0; i < MAX_UIO_PORT_REGIONS; i++) {
+		port = &idev->info->port[i];
+		if (port->size == 0)
+			break;
+		kobject_put(&port->portio->kobj);
+	}
+	kobject_put(idev->portio_dir);
+
 	sysfs_remove_group(&idev->dev->kobj, &uio_attr_grp);
 }
 
diff --git a/include/linux/uio_driver.h b/include/linux/uio_driver.h
index cdf338d94b7f..20be327bfbb4 100644
--- a/include/linux/uio_driver.h
+++ b/include/linux/uio_driver.h
@@ -38,6 +38,24 @@ struct uio_mem {
 
 #define MAX_UIO_MAPS	5
 
+struct uio_portio;
+
+/**
+ * struct uio_port - description of a UIO port region
+ * @start:		start of port region
+ * @size:		size of port region
+ * @porttype:		type of port (see UIO_PORT_* below)
+ * @portio:		for use by the UIO core only.
+ */
+struct uio_port {
+	unsigned long		start;
+	unsigned long		size;
+	int			porttype;
+	struct uio_portio	*portio;
+};
+
+#define MAX_UIO_PORT_REGIONS	5
+
 struct uio_device;
 
 /**
@@ -46,6 +64,7 @@ struct uio_device;
  * @name:		device name
  * @version:		device driver version
  * @mem:		list of mappable memory regions, size==0 for end of list
+ * @port:		list of port regions, size==0 for end of list
  * @irq:		interrupt number or UIO_IRQ_CUSTOM
  * @irq_flags:		flags for request_irq()
  * @priv:		optional private data
@@ -60,6 +79,7 @@ struct uio_info {
 	char			*name;
 	char			*version;
 	struct uio_mem		mem[MAX_UIO_MAPS];
+	struct uio_port		port[MAX_UIO_PORT_REGIONS];
 	long			irq;
 	unsigned long		irq_flags;
 	void			*priv;
@@ -92,4 +112,10 @@ extern void uio_event_notify(struct uio_info *info);
 #define UIO_MEM_LOGICAL	2
 #define UIO_MEM_VIRTUAL 3
 
+/* defines for uio_port->porttype */
+#define UIO_PORT_NONE	0
+#define UIO_PORT_X86	1
+#define UIO_PORT_GPIO	2
+#define UIO_PORT_OTHER	3
+
 #endif /* _LINUX_UIO_DRIVER_H_ */
-- 
cgit v1.2.3


From b8ac9fc0e8cda9f9776019c5b0464b0c6d2d4c90 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Fri, 12 Dec 2008 11:44:21 +0100
Subject: uio: make uio_info's name and version const

These are only ever assigned constant strings and never modified.

This was noticed because Wolfram Sang needed to cast the result of
of_get_property() in order to assign it to the name field of a struct
uio_info.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Hans J. Koch <hjk@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/DocBook/uio-howto.tmpl | 4 ++--
 include/linux/uio_driver.h           | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/DocBook/uio-howto.tmpl b/Documentation/DocBook/uio-howto.tmpl
index 6116b93608df..b787e4721c90 100644
--- a/Documentation/DocBook/uio-howto.tmpl
+++ b/Documentation/DocBook/uio-howto.tmpl
@@ -393,12 +393,12 @@ offset = N * getpagesize();
 
 <itemizedlist>
 <listitem><para>
-<varname>char *name</varname>: Required. The name of your driver as
+<varname>const char *name</varname>: Required. The name of your driver as
 it will appear in sysfs. I recommend using the name of your module for this.
 </para></listitem>
 
 <listitem><para>
-<varname>char *version</varname>: Required. This string appears in
+<varname>const char *version</varname>: Required. This string appears in
 <filename>/sys/class/uio/uioX/version</filename>.
 </para></listitem>
 
diff --git a/include/linux/uio_driver.h b/include/linux/uio_driver.h
index 20be327bfbb4..a0bb6bd2e5c1 100644
--- a/include/linux/uio_driver.h
+++ b/include/linux/uio_driver.h
@@ -76,8 +76,8 @@ struct uio_device;
  */
 struct uio_info {
 	struct uio_device	*uio_dev;
-	char			*name;
-	char			*version;
+	const char		*name;
+	const char		*version;
 	struct uio_mem		mem[MAX_UIO_MAPS];
 	struct uio_port		port[MAX_UIO_PORT_REGIONS];
 	long			irq;
-- 
cgit v1.2.3


From 96e93eab20337d063c70d537bd7bc70d90e04fa3 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 6 Jan 2009 10:49:34 -0800
Subject: gro: Add internal interfaces for VLAN

Previously GRO's only entry point from the outside is through
napi_gro_receive and napi_gro_frags.  These interfaces are for
device drivers.

This patch rearranges things to provide a new set of interfaces
for VLANs.  These interfaces are for internal use only.  The
VLAN code itself can then provide a set of entry points for
device drivers.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  6 ++++
 net/core/dev.c            | 82 ++++++++++++++++++++++++++++++++++-------------
 2 files changed, 65 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c28bbba3c23d..114091be8872 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1373,8 +1373,14 @@ extern int		netif_rx_ni(struct sk_buff *skb);
 #define HAVE_NETIF_RECEIVE_SKB 1
 extern int		netif_receive_skb(struct sk_buff *skb);
 extern void		napi_gro_flush(struct napi_struct *napi);
+extern int		dev_gro_receive(struct napi_struct *napi,
+					struct sk_buff *skb);
 extern int		napi_gro_receive(struct napi_struct *napi,
 					 struct sk_buff *skb);
+extern void		napi_reuse_skb(struct napi_struct *napi,
+				       struct sk_buff *skb);
+extern struct sk_buff *	napi_fraginfo_skb(struct napi_struct *napi,
+					  struct napi_gro_fraginfo *info);
 extern int		napi_gro_frags(struct napi_struct *napi,
 				       struct napi_gro_fraginfo *info);
 extern void		netif_nit_deliver(struct sk_buff *skb);
diff --git a/net/core/dev.c b/net/core/dev.c
index 382df6c09eec..bab8bcedd62e 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2387,7 +2387,7 @@ void napi_gro_flush(struct napi_struct *napi)
 }
 EXPORT_SYMBOL(napi_gro_flush);
 
-static int __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
+int dev_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 {
 	struct sk_buff **pp = NULL;
 	struct packet_type *ptype;
@@ -2417,11 +2417,14 @@ static int __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 
 		for (p = napi->gro_list; p; p = p->next) {
 			count++;
-			NAPI_GRO_CB(p)->same_flow =
-				p->mac_len == mac_len &&
-				!memcmp(skb_mac_header(p), skb_mac_header(skb),
-					mac_len);
-			NAPI_GRO_CB(p)->flush = 0;
+
+			if (!NAPI_GRO_CB(p)->same_flow)
+				continue;
+
+			if (p->mac_len != mac_len ||
+			    memcmp(skb_mac_header(p), skb_mac_header(skb),
+				   mac_len))
+				NAPI_GRO_CB(p)->same_flow = 0;
 		}
 
 		pp = ptype->gro_receive(&napi->gro_list, skb);
@@ -2463,6 +2466,19 @@ ok:
 normal:
 	return -1;
 }
+EXPORT_SYMBOL(dev_gro_receive);
+
+static int __napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
+{
+	struct sk_buff *p;
+
+	for (p = napi->gro_list; p; p = p->next) {
+		NAPI_GRO_CB(p)->same_flow = 1;
+		NAPI_GRO_CB(p)->flush = 0;
+	}
+
+	return dev_gro_receive(napi, skb);
+}
 
 int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 {
@@ -2479,11 +2495,26 @@ int napi_gro_receive(struct napi_struct *napi, struct sk_buff *skb)
 }
 EXPORT_SYMBOL(napi_gro_receive);
 
-int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info)
+void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb)
+{
+	skb_shinfo(skb)->nr_frags = 0;
+
+	skb->len -= skb->data_len;
+	skb->truesize -= skb->data_len;
+	skb->data_len = 0;
+
+	__skb_pull(skb, skb_headlen(skb));
+	skb_reserve(skb, NET_IP_ALIGN - skb_headroom(skb));
+
+	napi->skb = skb;
+}
+EXPORT_SYMBOL(napi_reuse_skb);
+
+struct sk_buff *napi_fraginfo_skb(struct napi_struct *napi,
+				  struct napi_gro_fraginfo *info)
 {
 	struct net_device *dev = napi->dev;
 	struct sk_buff *skb = napi->skb;
-	int err = NET_RX_DROP;
 
 	napi->skb = NULL;
 
@@ -2503,16 +2534,31 @@ int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info)
 	skb->len += info->len;
 	skb->truesize += info->len;
 
-	if (!pskb_may_pull(skb, ETH_HLEN))
-		goto reuse;
-
-	err = NET_RX_SUCCESS;
+	if (!pskb_may_pull(skb, ETH_HLEN)) {
+		napi_reuse_skb(napi, skb);
+		goto out;
+	}
 
 	skb->protocol = eth_type_trans(skb, dev);
 
 	skb->ip_summed = info->ip_summed;
 	skb->csum = info->csum;
 
+out:
+	return skb;
+}
+EXPORT_SYMBOL(napi_fraginfo_skb);
+
+int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info)
+{
+	struct sk_buff *skb = napi_fraginfo_skb(napi, info);
+	int err = NET_RX_DROP;
+
+	if (!skb)
+		goto out;
+
+	err = NET_RX_SUCCESS;
+
 	switch (__napi_gro_receive(napi, skb)) {
 	case -1:
 		return netif_receive_skb(skb);
@@ -2521,17 +2567,7 @@ int napi_gro_frags(struct napi_struct *napi, struct napi_gro_fraginfo *info)
 		goto out;
 	}
 
-reuse:
-	skb_shinfo(skb)->nr_frags = 0;
-
-	skb->len -= skb->data_len;
-	skb->truesize -= skb->data_len;
-	skb->data_len = 0;
-
-	__skb_pull(skb, skb_headlen(skb));
-	skb_reserve(skb, NET_IP_ALIGN - skb_headroom(skb));
-
-	napi->skb = skb;
+	napi_reuse_skb(napi, skb);
 
 out:
 	return err;
-- 
cgit v1.2.3


From e1c096e251e52773afeffbbcb74d0a072be47ea3 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 6 Jan 2009 10:50:09 -0800
Subject: vlan: Add GRO interfaces

This patch adds GRO interfaces for hardware-assisted VLAN reception.
With this in place we're now at parity with LRO as far as the
interface is concerned.  That is, you can now take any LRO driver
and convert it over to GRO.

As the CB memory clashes with GRO's use of CB, I've removed it
entirely by storing dev in skb->dev.  This is OK because VLAN
gets called first thing in netif_receive_skb and skb->dev is
not used in between us calling netif_rx and netif_receive_skb
getting called.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_vlan.h |  19 +++++++++
 net/8021q/vlan_core.c   | 111 ++++++++++++++++++++++++++++++++++++++----------
 2 files changed, 107 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index a5cb0c3f6dcf..f8ff918c208f 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -115,6 +115,11 @@ extern u16 vlan_dev_vlan_id(const struct net_device *dev);
 extern int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 			     u16 vlan_tci, int polling);
 extern int vlan_hwaccel_do_receive(struct sk_buff *skb);
+extern int vlan_gro_receive(struct napi_struct *napi, struct vlan_group *grp,
+			    unsigned int vlan_tci, struct sk_buff *skb);
+extern int vlan_gro_frags(struct napi_struct *napi, struct vlan_group *grp,
+			  unsigned int vlan_tci,
+			  struct napi_gro_fraginfo *info);
 
 #else
 static inline struct net_device *vlan_dev_real_dev(const struct net_device *dev)
@@ -140,6 +145,20 @@ static inline int vlan_hwaccel_do_receive(struct sk_buff *skb)
 {
 	return 0;
 }
+
+static inline int vlan_gro_receive(struct napi_struct *napi,
+				   struct vlan_group *grp,
+				   unsigned int vlan_tci, struct sk_buff *skb)
+{
+	return NET_RX_DROP;
+}
+
+static inline int vlan_gro_frags(struct napi_struct *napi,
+				 struct vlan_group *grp, unsigned int vlan_tci,
+				 struct napi_gro_fraginfo *info)
+{
+	return NET_RX_DROP;
+}
 #endif
 
 /**
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index dd86a1dc4cd0..6c1323940263 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -3,46 +3,35 @@
 #include <linux/if_vlan.h>
 #include "vlan.h"
 
-struct vlan_hwaccel_cb {
-	struct net_device	*dev;
-};
-
-static inline struct vlan_hwaccel_cb *vlan_hwaccel_cb(struct sk_buff *skb)
-{
-	return (struct vlan_hwaccel_cb *)skb->cb;
-}
-
 /* VLAN rx hw acceleration helper.  This acts like netif_{rx,receive_skb}(). */
 int __vlan_hwaccel_rx(struct sk_buff *skb, struct vlan_group *grp,
 		      u16 vlan_tci, int polling)
 {
-	struct vlan_hwaccel_cb *cb = vlan_hwaccel_cb(skb);
-
-	if (skb_bond_should_drop(skb)) {
-		dev_kfree_skb_any(skb);
-		return NET_RX_DROP;
-	}
+	if (skb_bond_should_drop(skb))
+		goto drop;
 
 	skb->vlan_tci = vlan_tci;
-	cb->dev = vlan_group_get_device(grp, vlan_tci & VLAN_VID_MASK);
+	skb->dev = vlan_group_get_device(grp, vlan_tci & VLAN_VID_MASK);
+
+	if (!skb->dev)
+		goto drop;
 
 	return (polling ? netif_receive_skb(skb) : netif_rx(skb));
+
+drop:
+	dev_kfree_skb_any(skb);
+	return NET_RX_DROP;
 }
 EXPORT_SYMBOL(__vlan_hwaccel_rx);
 
 int vlan_hwaccel_do_receive(struct sk_buff *skb)
 {
-	struct vlan_hwaccel_cb *cb = vlan_hwaccel_cb(skb);
-	struct net_device *dev = cb->dev;
+	struct net_device *dev = skb->dev;
 	struct net_device_stats *stats;
 
+	skb->dev = vlan_dev_info(dev)->real_dev;
 	netif_nit_deliver(skb);
 
-	if (dev == NULL) {
-		kfree_skb(skb);
-		return -1;
-	}
-
 	skb->dev = dev;
 	skb->priority = vlan_get_ingress_priority(dev, skb->vlan_tci);
 	skb->vlan_tci = 0;
@@ -80,3 +69,79 @@ u16 vlan_dev_vlan_id(const struct net_device *dev)
 	return vlan_dev_info(dev)->vlan_id;
 }
 EXPORT_SYMBOL_GPL(vlan_dev_vlan_id);
+
+static int vlan_gro_common(struct napi_struct *napi, struct vlan_group *grp,
+			   unsigned int vlan_tci, struct sk_buff *skb)
+{
+	struct sk_buff *p;
+
+	if (skb_bond_should_drop(skb))
+		goto drop;
+
+	skb->vlan_tci = vlan_tci;
+	skb->dev = vlan_group_get_device(grp, vlan_tci & VLAN_VID_MASK);
+
+	if (!skb->dev)
+		goto drop;
+
+	for (p = napi->gro_list; p; p = p->next) {
+		NAPI_GRO_CB(p)->same_flow = p->dev == skb->dev;
+		NAPI_GRO_CB(p)->flush = 0;
+	}
+
+	return dev_gro_receive(napi, skb);
+
+drop:
+	return 2;
+}
+
+int vlan_gro_receive(struct napi_struct *napi, struct vlan_group *grp,
+		     unsigned int vlan_tci, struct sk_buff *skb)
+{
+	int err = NET_RX_SUCCESS;
+
+	switch (vlan_gro_common(napi, grp, vlan_tci, skb)) {
+	case -1:
+		return netif_receive_skb(skb);
+
+	case 2:
+		err = NET_RX_DROP;
+		/* fall through */
+
+	case 1:
+		kfree_skb(skb);
+		break;
+	}
+
+	return err;
+}
+EXPORT_SYMBOL(vlan_gro_receive);
+
+int vlan_gro_frags(struct napi_struct *napi, struct vlan_group *grp,
+		   unsigned int vlan_tci, struct napi_gro_fraginfo *info)
+{
+	struct sk_buff *skb = napi_fraginfo_skb(napi, info);
+	int err = NET_RX_DROP;
+
+	if (!skb)
+		goto out;
+
+	err = NET_RX_SUCCESS;
+
+	switch (vlan_gro_common(napi, grp, vlan_tci, skb)) {
+	case -1:
+		return netif_receive_skb(skb);
+
+	case 2:
+		err = NET_RX_DROP;
+		/* fall through */
+
+	case 1:
+		napi_reuse_skb(napi, skb);
+		break;
+	}
+
+out:
+	return err;
+}
+EXPORT_SYMBOL(vlan_gro_frags);
-- 
cgit v1.2.3


From 1fa17d4ba43d7e5aab5e90777b07da06524f6748 Mon Sep 17 00:00:00 2001
From: Oliver Hartkopp <oliver@hartkopp.net>
Date: Tue, 6 Jan 2009 11:07:54 -0800
Subject: can: omit unneeded skb_clone() calls

The AF_CAN core delivered always cloned sk_buffs to the AF_CAN
protocols, although this was _only_ needed by the can-raw protocol.
With this (additionally documented) change, the AF_CAN core calls the
callback functions of the registered AF_CAN protocols with the original
(uncloned) sk_buff pointer and let's the can-raw protocol do the
skb_clone() itself which omits all unneeded skb_clone() calls for other
AF_CAN protocols.

Signed-off-by: Oliver Hartkopp <oliver@hartkopp.net>
Signed-off-by: Urs Thuermann <urs.thuermann@volkswagen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/can/core.h |  2 +-
 net/can/af_can.c         | 15 ++++++++-------
 net/can/bcm.c            | 12 +++++-------
 net/can/raw.c            | 15 ++++++++-------
 4 files changed, 22 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/can/core.h b/include/linux/can/core.h
index f50785ad4781..25085cbadcfc 100644
--- a/include/linux/can/core.h
+++ b/include/linux/can/core.h
@@ -19,7 +19,7 @@
 #include <linux/skbuff.h>
 #include <linux/netdevice.h>
 
-#define CAN_VERSION "20081130"
+#define CAN_VERSION "20090105"
 
 /* increment this number each time you change some user-space interface */
 #define CAN_ABI_VERSION "8"
diff --git a/net/can/af_can.c b/net/can/af_can.c
index 3dadb338addd..fa417ca6cbe6 100644
--- a/net/can/af_can.c
+++ b/net/can/af_can.c
@@ -414,6 +414,12 @@ static struct hlist_head *find_rcv_list(canid_t *can_id, canid_t *mask,
  *  The filter can be inverted (CAN_INV_FILTER bit set in can_id) or it can
  *  filter for error frames (CAN_ERR_FLAG bit set in mask).
  *
+ *  The provided pointer to the sk_buff is guaranteed to be valid as long as
+ *  the callback function is running. The callback function must *not* free
+ *  the given sk_buff while processing it's task. When the given sk_buff is
+ *  needed after the end of the callback function it must be cloned inside
+ *  the callback function with skb_clone().
+ *
  * Return:
  *  0 on success
  *  -ENOMEM on missing cache mem to create subscription entry
@@ -569,13 +575,8 @@ EXPORT_SYMBOL(can_rx_unregister);
 
 static inline void deliver(struct sk_buff *skb, struct receiver *r)
 {
-	struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC);
-
-	if (clone) {
-		clone->sk = skb->sk;
-		r->func(clone, r->data);
-		r->matches++;
-	}
+	r->func(skb, r->data);
+	r->matches++;
 }
 
 static int can_rcv_filter(struct dev_rcv_lists *d, struct sk_buff *skb)
diff --git a/net/can/bcm.c b/net/can/bcm.c
index 6248ae2502c7..1649c8ab2c2f 100644
--- a/net/can/bcm.c
+++ b/net/can/bcm.c
@@ -633,7 +633,7 @@ static void bcm_rx_handler(struct sk_buff *skb, void *data)
 	hrtimer_cancel(&op->timer);
 
 	if (op->can_id != rxframe->can_id)
-		goto rx_freeskb;
+		return;
 
 	/* save rx timestamp */
 	op->rx_stamp = skb->tstamp;
@@ -645,19 +645,19 @@ static void bcm_rx_handler(struct sk_buff *skb, void *data)
 	if (op->flags & RX_RTR_FRAME) {
 		/* send reply for RTR-request (placed in op->frames[0]) */
 		bcm_can_tx(op);
-		goto rx_freeskb;
+		return;
 	}
 
 	if (op->flags & RX_FILTER_ID) {
 		/* the easiest case */
 		bcm_rx_update_and_send(op, &op->last_frames[0], rxframe);
-		goto rx_freeskb_starttimer;
+		goto rx_starttimer;
 	}
 
 	if (op->nframes == 1) {
 		/* simple compare with index 0 */
 		bcm_rx_cmp_to_index(op, 0, rxframe);
-		goto rx_freeskb_starttimer;
+		goto rx_starttimer;
 	}
 
 	if (op->nframes > 1) {
@@ -678,10 +678,8 @@ static void bcm_rx_handler(struct sk_buff *skb, void *data)
 		}
 	}
 
-rx_freeskb_starttimer:
+rx_starttimer:
 	bcm_rx_starttimer(op);
-rx_freeskb:
-	kfree_skb(skb);
 }
 
 /*
diff --git a/net/can/raw.c b/net/can/raw.c
index 27aab63df467..0703cba4bf9f 100644
--- a/net/can/raw.c
+++ b/net/can/raw.c
@@ -99,13 +99,14 @@ static void raw_rcv(struct sk_buff *skb, void *data)
 	struct raw_sock *ro = raw_sk(sk);
 	struct sockaddr_can *addr;
 
-	if (!ro->recv_own_msgs) {
-		/* check the received tx sock reference */
-		if (skb->sk == sk) {
-			kfree_skb(skb);
-			return;
-		}
-	}
+	/* check the received tx sock reference */
+	if (!ro->recv_own_msgs && skb->sk == sk)
+		return;
+
+	/* clone the given skb to be able to enqueue it into the rcv queue */
+	skb = skb_clone(skb, GFP_ATOMIC);
+	if (!skb)
+		return;
 
 	/*
 	 *  Put the datagram to the queue so that raw_recvmsg() can
-- 
cgit v1.2.3


From 08fba69986e20c1c9e5fe2e6064d146cc4f42480 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Tue, 6 Jan 2009 14:38:53 -0800
Subject: mm: report the pagesize backing a VMA in /proc/pid/smaps

It is useful to verify a hugepage-aware application is using the expected
pagesizes for its memory regions. This patch creates an entry called
KernelPageSize in /proc/pid/smaps that is the size of page used by the
kernel to back a VMA. The entry is not called PageSize as it is possible
the MMU uses a different size. This extension should not break any sensible
parser that skips lines containing unrecognised information.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: "KOSAKI Motohiro" <kosaki.motohiro@jp.fujitsu.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/proc/task_mmu.c      |  6 ++++--
 include/linux/hugetlb.h |  3 +++
 mm/hugetlb.c            | 16 ++++++++++++++++
 3 files changed, 23 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 3a8bdd7f5756..41ef5f23e779 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -396,7 +396,8 @@ static int show_smap(struct seq_file *m, void *v)
 		   "Private_Clean:  %8lu kB\n"
 		   "Private_Dirty:  %8lu kB\n"
 		   "Referenced:     %8lu kB\n"
-		   "Swap:           %8lu kB\n",
+		   "Swap:           %8lu kB\n"
+		   "KernelPageSize: %8lu kB\n",
 		   (vma->vm_end - vma->vm_start) >> 10,
 		   mss.resident >> 10,
 		   (unsigned long)(mss.pss >> (10 + PSS_SHIFT)),
@@ -405,7 +406,8 @@ static int show_smap(struct seq_file *m, void *v)
 		   mss.private_clean >> 10,
 		   mss.private_dirty >> 10,
 		   mss.referenced >> 10,
-		   mss.swap >> 10);
+		   mss.swap >> 10,
+		   vma_kernel_pagesize(vma) >> 10);
 
 	if (m->count < m->size)  /* vma is copied successfully */
 		m->version = (vma != get_gate_vma(task)) ? vma->vm_start : 0;
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index e1c8afc002c0..648e1e25979e 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -233,6 +233,8 @@ static inline unsigned long huge_page_size(struct hstate *h)
 	return (unsigned long)PAGE_SIZE << h->order;
 }
 
+extern unsigned long vma_kernel_pagesize(struct vm_area_struct *vma);
+
 static inline unsigned long huge_page_mask(struct hstate *h)
 {
 	return h->mask;
@@ -273,6 +275,7 @@ struct hstate {};
 #define hstate_inode(i) NULL
 #define huge_page_size(h) PAGE_SIZE
 #define huge_page_mask(h) PAGE_MASK
+#define vma_kernel_pagesize(v) PAGE_SIZE
 #define huge_page_order(h) 0
 #define huge_page_shift(h) PAGE_SHIFT
 static inline unsigned int pages_per_huge_page(struct hstate *h)
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 6058b53dcb89..5cb8bc7c80f7 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -219,6 +219,22 @@ static pgoff_t vma_hugecache_offset(struct hstate *h,
 			(vma->vm_pgoff >> huge_page_order(h));
 }
 
+/*
+ * Return the size of the pages allocated when backing a VMA. In the majority
+ * cases this will be same size as used by the page table entries.
+ */
+unsigned long vma_kernel_pagesize(struct vm_area_struct *vma)
+{
+	struct hstate *hstate;
+
+	if (!is_vm_hugetlb_page(vma))
+		return PAGE_SIZE;
+
+	hstate = hstate_vma(vma);
+
+	return 1UL << (hstate->order + PAGE_SHIFT);
+}
+
 /*
  * Flags for MAP_PRIVATE reservations.  These are stored in the bottom
  * bits of the reservation map pointer, which are always clear due to
-- 
cgit v1.2.3


From 3340289ddf29ca75c3acfb3a6b72f234b2f74d5c Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Tue, 6 Jan 2009 14:38:54 -0800
Subject: mm: report the MMU pagesize in /proc/pid/smaps

The KernelPageSize entry in /proc/pid/smaps is the pagesize used by the
kernel to back a VMA.  This matches the size used by the MMU in the
majority of cases.  However, one counter-example occurs on PPC64 kernels
whereby a kernel using 64K as a base pagesize may still use 4K pages for
the MMU on older processor.  To distinguish, this patch reports
MMUPageSize as the pagesize used by the MMU in /proc/pid/smaps.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: "KOSAKI Motohiro" <kosaki.motohiro@jp.fujitsu.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/powerpc/include/asm/hugetlb.h |  6 ++++++
 arch/powerpc/mm/hugetlbpage.c      |  7 +++++++
 fs/proc/task_mmu.c                 |  6 ++++--
 include/linux/hugetlb.h            |  3 +++
 mm/hugetlb.c                       | 13 +++++++++++++
 5 files changed, 33 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/arch/powerpc/include/asm/hugetlb.h b/arch/powerpc/include/asm/hugetlb.h
index 26f0d0ab27a5..b1dafb6a9743 100644
--- a/arch/powerpc/include/asm/hugetlb.h
+++ b/arch/powerpc/include/asm/hugetlb.h
@@ -17,6 +17,12 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
 pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
 			      pte_t *ptep);
 
+/*
+ * The version of vma_mmu_pagesize() in arch/powerpc/mm/hugetlbpage.c needs
+ * to override the version in mm/hugetlb.c
+ */
+#define vma_mmu_pagesize vma_mmu_pagesize
+
 /*
  * If the arch doesn't supply something else, assume that hugepage
  * size aligned regions are ok without further preparation.
diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c
index 201c7a5486cb..9920d6a7cf29 100644
--- a/arch/powerpc/mm/hugetlbpage.c
+++ b/arch/powerpc/mm/hugetlbpage.c
@@ -512,6 +512,13 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 	return slice_get_unmapped_area(addr, len, flags, mmu_psize, 1, 0);
 }
 
+unsigned long vma_mmu_pagesize(struct vm_area_struct *vma)
+{
+	unsigned int psize = get_slice_psize(vma->vm_mm, vma->vm_start);
+
+	return 1UL << mmu_psize_to_shift(psize);
+}
+
 /*
  * Called by asm hashtable.S for doing lazy icache flush
  */
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 41ef5f23e779..94063840832a 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -397,7 +397,8 @@ static int show_smap(struct seq_file *m, void *v)
 		   "Private_Dirty:  %8lu kB\n"
 		   "Referenced:     %8lu kB\n"
 		   "Swap:           %8lu kB\n"
-		   "KernelPageSize: %8lu kB\n",
+		   "KernelPageSize: %8lu kB\n"
+		   "MMUPageSize:    %8lu kB\n",
 		   (vma->vm_end - vma->vm_start) >> 10,
 		   mss.resident >> 10,
 		   (unsigned long)(mss.pss >> (10 + PSS_SHIFT)),
@@ -407,7 +408,8 @@ static int show_smap(struct seq_file *m, void *v)
 		   mss.private_dirty >> 10,
 		   mss.referenced >> 10,
 		   mss.swap >> 10,
-		   vma_kernel_pagesize(vma) >> 10);
+		   vma_kernel_pagesize(vma) >> 10,
+		   vma_mmu_pagesize(vma) >> 10);
 
 	if (m->count < m->size)  /* vma is copied successfully */
 		m->version = (vma != get_gate_vma(task)) ? vma->vm_start : 0;
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 648e1e25979e..f1d2fba19ea0 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -235,6 +235,8 @@ static inline unsigned long huge_page_size(struct hstate *h)
 
 extern unsigned long vma_kernel_pagesize(struct vm_area_struct *vma);
 
+extern unsigned long vma_mmu_pagesize(struct vm_area_struct *vma);
+
 static inline unsigned long huge_page_mask(struct hstate *h)
 {
 	return h->mask;
@@ -276,6 +278,7 @@ struct hstate {};
 #define huge_page_size(h) PAGE_SIZE
 #define huge_page_mask(h) PAGE_MASK
 #define vma_kernel_pagesize(v) PAGE_SIZE
+#define vma_mmu_pagesize(v) PAGE_SIZE
 #define huge_page_order(h) 0
 #define huge_page_shift(h) PAGE_SHIFT
 static inline unsigned int pages_per_huge_page(struct hstate *h)
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 5cb8bc7c80f7..9595278b5ab4 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -235,6 +235,19 @@ unsigned long vma_kernel_pagesize(struct vm_area_struct *vma)
 	return 1UL << (hstate->order + PAGE_SHIFT);
 }
 
+/*
+ * Return the page size being used by the MMU to back a VMA. In the majority
+ * of cases, the page size used by the kernel matches the MMU size. On
+ * architectures where it differs, an architecture-specific version of this
+ * function is required.
+ */
+#ifndef vma_mmu_pagesize
+unsigned long vma_mmu_pagesize(struct vm_area_struct *vma)
+{
+	return vma_kernel_pagesize(vma);
+}
+#endif
+
 /*
  * Flags for MAP_PRIVATE reservations.  These are stored in the bottom
  * bits of the reservation map pointer, which are always clear due to
-- 
cgit v1.2.3


From 1c0fe6e3bda0464728c23c8d84aa47567e8b716c Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Tue, 6 Jan 2009 14:38:59 -0800
Subject: mm: invoke oom-killer from page fault

Rather than have the pagefault handler kill a process directly if it gets
a VM_FAULT_OOM, have it call into the OOM killer.

With increasingly sophisticated oom behaviour (cpusets, memory cgroups,
oom killing throttling, oom priority adjustment or selective disabling,
panic on oom, etc), it's silly to unconditionally kill the faulting
process at page fault time.  Create a hook for pagefault oom path to call
into instead.

Only converted x86 and uml so far.

[akpm@linux-foundation.org: make __out_of_memory() static]
[akpm@linux-foundation.org: fix comment]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Jeff Dike <jdike@addtoit.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/um/kernel/trap.c | 24 +++++--------
 arch/x86/mm/fault.c   | 24 ++++---------
 include/linux/mm.h    |  5 +++
 mm/oom_kill.c         | 94 +++++++++++++++++++++++++++++++++++----------------
 4 files changed, 84 insertions(+), 63 deletions(-)

(limited to 'include/linux')

diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c
index 44e490419495..7384d8accfe7 100644
--- a/arch/um/kernel/trap.c
+++ b/arch/um/kernel/trap.c
@@ -64,11 +64,10 @@ good_area:
 
 	do {
 		int fault;
-survive:
+
 		fault = handle_mm_fault(mm, vma, address, is_write);
 		if (unlikely(fault & VM_FAULT_ERROR)) {
 			if (fault & VM_FAULT_OOM) {
-				err = -ENOMEM;
 				goto out_of_memory;
 			} else if (fault & VM_FAULT_SIGBUS) {
 				err = -EACCES;
@@ -104,18 +103,14 @@ out:
 out_nosemaphore:
 	return err;
 
-/*
- * We ran out of memory, or some other thing happened to us that made
- * us unable to handle the page fault gracefully.
- */
 out_of_memory:
-	if (is_global_init(current)) {
-		up_read(&mm->mmap_sem);
-		yield();
-		down_read(&mm->mmap_sem);
-		goto survive;
-	}
-	goto out;
+	/*
+	 * We ran out of memory, call the OOM killer, and return the userspace
+	 * (which will retry the fault, or kill us if we got oom-killed).
+	 */
+	up_read(&mm->mmap_sem);
+	pagefault_out_of_memory();
+	return 0;
 }
 
 static void bad_segv(struct faultinfo fi, unsigned long ip)
@@ -214,9 +209,6 @@ unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user,
 		si.si_addr = (void __user *)address;
 		current->thread.arch.faultinfo = fi;
 		force_sig_info(SIGBUS, &si, current);
-	} else if (err == -ENOMEM) {
-		printk(KERN_INFO "VM: killing process %s\n", current->comm);
-		do_exit(SIGKILL);
 	} else {
 		BUG_ON(err != -EFAULT);
 		si.si_signo = SIGSEGV;
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 57ec8c86a877..9e268b6b204e 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -667,7 +667,6 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code)
 	if (unlikely(in_atomic() || !mm))
 		goto bad_area_nosemaphore;
 
-again:
 	/*
 	 * When running in the kernel we expect faults to occur only to
 	 * addresses in user space.  All other faults represent errors in the
@@ -859,25 +858,14 @@ no_context:
 	oops_end(flags, regs, sig);
 #endif
 
-/*
- * We ran out of memory, or some other thing happened to us that made
- * us unable to handle the page fault gracefully.
- */
 out_of_memory:
+	/*
+	 * We ran out of memory, call the OOM killer, and return the userspace
+	 * (which will retry the fault, or kill us if we got oom-killed).
+	 */
 	up_read(&mm->mmap_sem);
-	if (is_global_init(tsk)) {
-		yield();
-		/*
-		 * Re-lookup the vma - in theory the vma tree might
-		 * have changed:
-		 */
-		goto again;
-	}
-
-	printk("VM: killing process %s\n", tsk->comm);
-	if (error_code & PF_USER)
-		do_group_exit(SIGKILL);
-	goto no_context;
+	pagefault_out_of_memory();
+	return;
 
 do_sigbus:
 	up_read(&mm->mmap_sem);
diff --git a/include/linux/mm.h b/include/linux/mm.h
index aaa8b843be28..4a3d28c86443 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -717,6 +717,11 @@ static inline int page_mapped(struct page *page)
 
 #define VM_FAULT_ERROR	(VM_FAULT_OOM | VM_FAULT_SIGBUS)
 
+/*
+ * Can be called by the pagefault handler when it gets a VM_FAULT_OOM.
+ */
+extern void pagefault_out_of_memory(void);
+
 #define offset_in_page(p)	((unsigned long)(p) & ~PAGE_MASK)
 
 extern void show_free_areas(void);
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 558f9afe6e4e..c592965dab2f 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -509,6 +509,69 @@ void clear_zonelist_oom(struct zonelist *zonelist, gfp_t gfp_mask)
 	spin_unlock(&zone_scan_mutex);
 }
 
+/*
+ * Must be called with tasklist_lock held for read.
+ */
+static void __out_of_memory(gfp_t gfp_mask, int order)
+{
+	if (sysctl_oom_kill_allocating_task) {
+		oom_kill_process(current, gfp_mask, order, 0, NULL,
+				"Out of memory (oom_kill_allocating_task)");
+
+	} else {
+		unsigned long points;
+		struct task_struct *p;
+
+retry:
+		/*
+		 * Rambo mode: Shoot down a process and hope it solves whatever
+		 * issues we may have.
+		 */
+		p = select_bad_process(&points, NULL);
+
+		if (PTR_ERR(p) == -1UL)
+			return;
+
+		/* Found nothing?!?! Either we hang forever, or we panic. */
+		if (!p) {
+			read_unlock(&tasklist_lock);
+			panic("Out of memory and no killable processes...\n");
+		}
+
+		if (oom_kill_process(p, gfp_mask, order, points, NULL,
+				     "Out of memory"))
+			goto retry;
+	}
+}
+
+/*
+ * pagefault handler calls into here because it is out of memory but
+ * doesn't know exactly how or why.
+ */
+void pagefault_out_of_memory(void)
+{
+	unsigned long freed = 0;
+
+	blocking_notifier_call_chain(&oom_notify_list, 0, &freed);
+	if (freed > 0)
+		/* Got some memory back in the last second. */
+		return;
+
+	if (sysctl_panic_on_oom)
+		panic("out of memory from page fault. panic_on_oom is selected.\n");
+
+	read_lock(&tasklist_lock);
+	__out_of_memory(0, 0); /* unknown gfp_mask and order */
+	read_unlock(&tasklist_lock);
+
+	/*
+	 * Give "p" a good chance of killing itself before we
+	 * retry to allocate memory.
+	 */
+	if (!test_thread_flag(TIF_MEMDIE))
+		schedule_timeout_uninterruptible(1);
+}
+
 /**
  * out_of_memory - kill the "best" process when we run out of memory
  * @zonelist: zonelist pointer
@@ -522,8 +585,6 @@ void clear_zonelist_oom(struct zonelist *zonelist, gfp_t gfp_mask)
  */
 void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order)
 {
-	struct task_struct *p;
-	unsigned long points = 0;
 	unsigned long freed = 0;
 	enum oom_constraint constraint;
 
@@ -544,7 +605,7 @@ void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order)
 
 	switch (constraint) {
 	case CONSTRAINT_MEMORY_POLICY:
-		oom_kill_process(current, gfp_mask, order, points, NULL,
+		oom_kill_process(current, gfp_mask, order, 0, NULL,
 				"No available memory (MPOL_BIND)");
 		break;
 
@@ -553,35 +614,10 @@ void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order)
 			panic("out of memory. panic_on_oom is selected\n");
 		/* Fall-through */
 	case CONSTRAINT_CPUSET:
-		if (sysctl_oom_kill_allocating_task) {
-			oom_kill_process(current, gfp_mask, order, points, NULL,
-					"Out of memory (oom_kill_allocating_task)");
-			break;
-		}
-retry:
-		/*
-		 * Rambo mode: Shoot down a process and hope it solves whatever
-		 * issues we may have.
-		 */
-		p = select_bad_process(&points, NULL);
-
-		if (PTR_ERR(p) == -1UL)
-			goto out;
-
-		/* Found nothing?!?! Either we hang forever, or we panic. */
-		if (!p) {
-			read_unlock(&tasklist_lock);
-			panic("Out of memory and no killable processes...\n");
-		}
-
-		if (oom_kill_process(p, gfp_mask, order, points, NULL,
-				     "Out of memory"))
-			goto retry;
-
+		__out_of_memory(gfp_mask, order);
 		break;
 	}
 
-out:
 	read_unlock(&tasklist_lock);
 
 	/*
-- 
cgit v1.2.3


From 75aa199410359dc5fbcf9025ff7af98a9d20f0d5 Mon Sep 17 00:00:00 2001
From: David Rientjes <rientjes@google.com>
Date: Tue, 6 Jan 2009 14:39:01 -0800
Subject: oom: print triggering task's cpuset and mems allowed

When cpusets are enabled, it's necessary to print the triggering task's
set of allowable nodes so the subsequently printed meminfo can be
interpreted correctly.

We also print the task's cpuset name for informational purposes.

[rientjes@google.com: task lock current before dereferencing cpuset]
Cc: Paul Menage <menage@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cpuset.h |  6 ++++++
 kernel/cpuset.c        | 34 ++++++++++++++++++++++++++++++++++
 mm/oom_kill.c          |  3 +++
 3 files changed, 43 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index 8e540d32c9fe..51ea2bdea0f9 100644
--- a/include/linux/cpuset.h
+++ b/include/linux/cpuset.h
@@ -78,6 +78,8 @@ extern int current_cpuset_is_being_rebound(void);
 
 extern void rebuild_sched_domains(void);
 
+extern void cpuset_print_task_mems_allowed(struct task_struct *p);
+
 #else /* !CONFIG_CPUSETS */
 
 static inline int cpuset_init_early(void) { return 0; }
@@ -159,6 +161,10 @@ static inline void rebuild_sched_domains(void)
 	partition_sched_domains(1, NULL, NULL);
 }
 
+static inline void cpuset_print_task_mems_allowed(struct task_struct *p)
+{
+}
+
 #endif /* !CONFIG_CPUSETS */
 
 #endif /* _LINUX_CPUSET_H */
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 39c1a4c1c5a9..345ace5117de 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -239,6 +239,17 @@ static struct cpuset top_cpuset = {
 
 static DEFINE_MUTEX(callback_mutex);
 
+/*
+ * cpuset_buffer_lock protects both the cpuset_name and cpuset_nodelist
+ * buffers.  They are statically allocated to prevent using excess stack
+ * when calling cpuset_print_task_mems_allowed().
+ */
+#define CPUSET_NAME_LEN		(128)
+#define	CPUSET_NODELIST_LEN	(256)
+static char cpuset_name[CPUSET_NAME_LEN];
+static char cpuset_nodelist[CPUSET_NODELIST_LEN];
+static DEFINE_SPINLOCK(cpuset_buffer_lock);
+
 /*
  * This is ugly, but preserves the userspace API for existing cpuset
  * users. If someone tries to mount the "cpuset" filesystem, we
@@ -2356,6 +2367,29 @@ int cpuset_mems_allowed_intersects(const struct task_struct *tsk1,
 	return nodes_intersects(tsk1->mems_allowed, tsk2->mems_allowed);
 }
 
+/**
+ * cpuset_print_task_mems_allowed - prints task's cpuset and mems_allowed
+ * @task: pointer to task_struct of some task.
+ *
+ * Description: Prints @task's name, cpuset name, and cached copy of its
+ * mems_allowed to the kernel log.  Must hold task_lock(task) to allow
+ * dereferencing task_cs(task).
+ */
+void cpuset_print_task_mems_allowed(struct task_struct *tsk)
+{
+	struct dentry *dentry;
+
+	dentry = task_cs(tsk)->css.cgroup->dentry;
+	spin_lock(&cpuset_buffer_lock);
+	snprintf(cpuset_name, CPUSET_NAME_LEN,
+		 dentry ? (const char *)dentry->d_name.name : "/");
+	nodelist_scnprintf(cpuset_nodelist, CPUSET_NODELIST_LEN,
+			   tsk->mems_allowed);
+	printk(KERN_INFO "%s cpuset=%s mems_allowed=%s\n",
+	       tsk->comm, cpuset_name, cpuset_nodelist);
+	spin_unlock(&cpuset_buffer_lock);
+}
+
 /*
  * Collection of memory_pressure is suppressed unless
  * this flag is enabled by writing "1" to the special
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index e5f50cfdca4d..6b9e758c98a5 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -392,6 +392,9 @@ static int oom_kill_process(struct task_struct *p, gfp_t gfp_mask, int order,
 		printk(KERN_WARNING "%s invoked oom-killer: "
 			"gfp_mask=0x%x, order=%d, oomkilladj=%d\n",
 			current->comm, gfp_mask, order, current->oomkilladj);
+		task_lock(current);
+		cpuset_print_task_mems_allowed(current);
+		task_unlock(current);
 		dump_stack();
 		show_mem();
 		if (sysctl_oom_dump_tasks)
-- 
cgit v1.2.3


From c04fc586c1a480ba198f03ae7b6cbd7b57380b91 Mon Sep 17 00:00:00 2001
From: Gary Hade <garyhade@us.ibm.com>
Date: Tue, 6 Jan 2009 14:39:14 -0800
Subject: mm: show node to memory section relationship with symlinks in sysfs

Show node to memory section relationship with symlinks in sysfs

Add /sys/devices/system/node/nodeX/memoryY symlinks for all
the memory sections located on nodeX.  For example:
/sys/devices/system/node/node1/memory135 -> ../../memory/memory135
indicates that memory section 135 resides on node1.

Also revises documentation to cover this change as well as updating
Documentation/ABI/testing/sysfs-devices-memory to include descriptions
of memory hotremove files 'phys_device', 'phys_index', and 'state'
that were previously not described there.

In addition to it always being a good policy to provide users with
the maximum possible amount of physical location information for
resources that can be hot-added and/or hot-removed, the following
are some (but likely not all) of the user benefits provided by
this change.
Immediate:
  - Provides information needed to determine the specific node
    on which a defective DIMM is located.  This will reduce system
    downtime when the node or defective DIMM is swapped out.
  - Prevents unintended onlining of a memory section that was
    previously offlined due to a defective DIMM.  This could happen
    during node hot-add when the user or node hot-add assist script
    onlines _all_ offlined sections due to user or script inability
    to identify the specific memory sections located on the hot-added
    node.  The consequences of reintroducing the defective memory
    could be ugly.
  - Provides information needed to vary the amount and distribution
    of memory on specific nodes for testing or debugging purposes.
Future:
  - Will provide information needed to identify the memory
    sections that need to be offlined prior to physical removal
    of a specific node.

Symlink creation during boot was tested on 2-node x86_64, 2-node
ppc64, and 2-node ia64 systems.  Symlink creation during physical
memory hot-add tested on a 2-node x86_64 system.

Signed-off-by: Gary Hade <garyhade@us.ibm.com>
Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/ABI/testing/sysfs-devices-memory |  51 +++++++++++-
 Documentation/memory-hotplug.txt               |  16 +++-
 arch/ia64/mm/init.c                            |   2 +-
 arch/powerpc/mm/mem.c                          |   2 +-
 arch/s390/mm/init.c                            |   2 +-
 arch/sh/mm/init.c                              |   3 +-
 arch/x86/mm/init_32.c                          |   2 +-
 arch/x86/mm/init_64.c                          |   2 +-
 drivers/base/memory.c                          |  19 +++--
 drivers/base/node.c                            | 103 +++++++++++++++++++++++++
 include/linux/memory.h                         |   6 +-
 include/linux/memory_hotplug.h                 |   2 +-
 include/linux/node.h                           |  13 ++++
 mm/memory_hotplug.c                            |  11 +--
 14 files changed, 209 insertions(+), 25 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/ABI/testing/sysfs-devices-memory b/Documentation/ABI/testing/sysfs-devices-memory
index 7a16fe1e2270..9fe91c02ee40 100644
--- a/Documentation/ABI/testing/sysfs-devices-memory
+++ b/Documentation/ABI/testing/sysfs-devices-memory
@@ -6,7 +6,6 @@ Description:
 		internal state of the kernel memory blocks. Files could be
 		added or removed dynamically to represent hot-add/remove
 		operations.
-
 Users:		hotplug memory add/remove tools
 		https://w3.opensource.ibm.com/projects/powerpc-utils/
 
@@ -19,6 +18,56 @@ Description:
 		This is useful for a user-level agent to determine
 		identify removable sections of the memory before attempting
 		potentially expensive hot-remove memory operation
+Users:		hotplug memory remove tools
+		https://w3.opensource.ibm.com/projects/powerpc-utils/
+
+What:		/sys/devices/system/memory/memoryX/phys_device
+Date:		September 2008
+Contact:	Badari Pulavarty <pbadari@us.ibm.com>
+Description:
+		The file /sys/devices/system/memory/memoryX/phys_device
+		is read-only and is designed to show the name of physical
+		memory device.  Implementation is currently incomplete.
 
+What:		/sys/devices/system/memory/memoryX/phys_index
+Date:		September 2008
+Contact:	Badari Pulavarty <pbadari@us.ibm.com>
+Description:
+		The file /sys/devices/system/memory/memoryX/phys_index
+		is read-only and contains the section ID in hexadecimal
+		which is equivalent to decimal X contained in the
+		memory section directory name.
+
+What:		/sys/devices/system/memory/memoryX/state
+Date:		September 2008
+Contact:	Badari Pulavarty <pbadari@us.ibm.com>
+Description:
+		The file /sys/devices/system/memory/memoryX/state
+		is read-write.  When read, it's contents show the
+		online/offline state of the memory section.  When written,
+		root can toggle the the online/offline state of a removable
+		memory section (see removable file description above)
+		using the following commands.
+		# echo online > /sys/devices/system/memory/memoryX/state
+		# echo offline > /sys/devices/system/memory/memoryX/state
+
+		For example, if /sys/devices/system/memory/memory22/removable
+		contains a value of 1 and
+		/sys/devices/system/memory/memory22/state contains the
+		string "online" the following command can be executed by
+		by root to offline that section.
+		# echo offline > /sys/devices/system/memory/memory22/state
 Users:		hotplug memory remove tools
 		https://w3.opensource.ibm.com/projects/powerpc-utils/
+
+What:		/sys/devices/system/node/nodeX/memoryY
+Date:		September 2008
+Contact:	Gary Hade <garyhade@us.ibm.com>
+Description:
+		When CONFIG_NUMA is enabled
+		/sys/devices/system/node/nodeX/memoryY is a symbolic link that
+		points to the corresponding /sys/devices/system/memory/memoryY
+		memory section directory.  For example, the following symbolic
+		link is created for memory section 9 on node0.
+		/sys/devices/system/node/node0/memory9 -> ../../memory/memory9
+
diff --git a/Documentation/memory-hotplug.txt b/Documentation/memory-hotplug.txt
index 168117bd6ee8..4c2ecf537a4a 100644
--- a/Documentation/memory-hotplug.txt
+++ b/Documentation/memory-hotplug.txt
@@ -124,7 +124,7 @@ config options.
     This option can be kernel module too.
 
 --------------------------------
-3 sysfs files for memory hotplug
+4 sysfs files for memory hotplug
 --------------------------------
 All sections have their device information under /sys/devices/system/memory as
 
@@ -138,11 +138,12 @@ For example, assume 1GiB section size. A device for a memory starting at
 (0x100000000 / 1Gib = 4)
 This device covers address range [0x100000000 ... 0x140000000)
 
-Under each section, you can see 3 files.
+Under each section, you can see 4 files.
 
 /sys/devices/system/memory/memoryXXX/phys_index
 /sys/devices/system/memory/memoryXXX/phys_device
 /sys/devices/system/memory/memoryXXX/state
+/sys/devices/system/memory/memoryXXX/removable
 
 'phys_index' : read-only and contains section id, same as XXX.
 'state'      : read-write
@@ -150,10 +151,20 @@ Under each section, you can see 3 files.
                at write: user can specify "online", "offline" command
 'phys_device': read-only: designed to show the name of physical memory device.
                This is not well implemented now.
+'removable'  : read-only: contains an integer value indicating
+               whether the memory section is removable or not
+               removable.  A value of 1 indicates that the memory
+               section is removable and a value of 0 indicates that
+               it is not removable.
 
 NOTE:
   These directories/files appear after physical memory hotplug phase.
 
+If CONFIG_NUMA is enabled the
+/sys/devices/system/memory/memoryXXX memory section
+directories can also be accessed via symbolic links located in
+the /sys/devices/system/node/node* directories.  For example:
+/sys/devices/system/node/node0/memory9 -> ../../memory/memory9
 
 --------------------------------
 4. Physical memory hot-add phase
@@ -365,7 +376,6 @@ node if necessary.
   - allowing memory hot-add to ZONE_MOVABLE. maybe we need some switch like
     sysctl or new control file.
   - showing memory section and physical device relationship.
-  - showing memory section and node relationship (maybe good for NUMA)
   - showing memory section is under ZONE_MOVABLE or not
   - test and make it better memory offlining.
   - support HugeTLB page migration and offlining.
diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c
index 054bcd9439aa..56e12903973c 100644
--- a/arch/ia64/mm/init.c
+++ b/arch/ia64/mm/init.c
@@ -692,7 +692,7 @@ int arch_add_memory(int nid, u64 start, u64 size)
 	pgdat = NODE_DATA(nid);
 
 	zone = pgdat->node_zones + ZONE_NORMAL;
-	ret = __add_pages(zone, start_pfn, nr_pages);
+	ret = __add_pages(nid, zone, start_pfn, nr_pages);
 
 	if (ret)
 		printk("%s: Problem encountered in __add_pages() as ret=%d\n",
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index 53b06ebb3f2f..f00f09a77f12 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -132,7 +132,7 @@ int arch_add_memory(int nid, u64 start, u64 size)
 	/* this should work for most non-highmem platforms */
 	zone = pgdata->node_zones;
 
-	return __add_pages(zone, start_pfn, nr_pages);
+	return __add_pages(nid, zone, start_pfn, nr_pages);
 }
 #endif /* CONFIG_MEMORY_HOTPLUG */
 
diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c
index 158b0d6d7046..f0258ca3b17e 100644
--- a/arch/s390/mm/init.c
+++ b/arch/s390/mm/init.c
@@ -183,7 +183,7 @@ int arch_add_memory(int nid, u64 start, u64 size)
 	rc = vmem_add_mapping(start, size);
 	if (rc)
 		return rc;
-	rc = __add_pages(zone, PFN_DOWN(start), PFN_DOWN(size));
+	rc = __add_pages(nid, zone, PFN_DOWN(start), PFN_DOWN(size));
 	if (rc)
 		vmem_remove_mapping(start, size);
 	return rc;
diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index 6cbef8caeb56..3edf297c829b 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -311,7 +311,8 @@ int arch_add_memory(int nid, u64 start, u64 size)
 	pgdat = NODE_DATA(nid);
 
 	/* We only have ZONE_NORMAL, so this is easy.. */
-	ret = __add_pages(pgdat->node_zones + ZONE_NORMAL, start_pfn, nr_pages);
+	ret = __add_pages(nid, pgdat->node_zones + ZONE_NORMAL,
+				start_pfn, nr_pages);
 	if (unlikely(ret))
 		printk("%s: Failed, __add_pages() == %d\n", __func__, ret);
 
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index f99a6c6c432e..544d724caeee 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -1079,7 +1079,7 @@ int arch_add_memory(int nid, u64 start, u64 size)
 	unsigned long start_pfn = start >> PAGE_SHIFT;
 	unsigned long nr_pages = size >> PAGE_SHIFT;
 
-	return __add_pages(zone, start_pfn, nr_pages);
+	return __add_pages(nid, zone, start_pfn, nr_pages);
 }
 #endif
 
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 9f7a0d24d42a..54c437e96541 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -857,7 +857,7 @@ int arch_add_memory(int nid, u64 start, u64 size)
 	if (last_mapped_pfn > max_pfn_mapped)
 		max_pfn_mapped = last_mapped_pfn;
 
-	ret = __add_pages(zone, start_pfn, nr_pages);
+	ret = __add_pages(nid, zone, start_pfn, nr_pages);
 	WARN_ON_ONCE(ret);
 
 	return ret;
diff --git a/drivers/base/memory.c b/drivers/base/memory.c
index 5260e9e0df48..989429cfed88 100644
--- a/drivers/base/memory.c
+++ b/drivers/base/memory.c
@@ -347,8 +347,9 @@ static inline int memory_probe_init(void)
  * section belongs to...
  */
 
-static int add_memory_block(unsigned long node_id, struct mem_section *section,
-		     unsigned long state, int phys_device)
+static int add_memory_block(int nid, struct mem_section *section,
+			unsigned long state, int phys_device,
+			enum mem_add_context context)
 {
 	struct memory_block *mem = kzalloc(sizeof(*mem), GFP_KERNEL);
 	int ret = 0;
@@ -370,6 +371,10 @@ static int add_memory_block(unsigned long node_id, struct mem_section *section,
 		ret = mem_create_simple_file(mem, phys_device);
 	if (!ret)
 		ret = mem_create_simple_file(mem, removable);
+	if (!ret) {
+		if (context == HOTPLUG)
+			ret = register_mem_sect_under_node(mem, nid);
+	}
 
 	return ret;
 }
@@ -382,7 +387,7 @@ static int add_memory_block(unsigned long node_id, struct mem_section *section,
  *
  * This could be made generic for all sysdev classes.
  */
-static struct memory_block *find_memory_block(struct mem_section *section)
+struct memory_block *find_memory_block(struct mem_section *section)
 {
 	struct kobject *kobj;
 	struct sys_device *sysdev;
@@ -411,6 +416,7 @@ int remove_memory_block(unsigned long node_id, struct mem_section *section,
 	struct memory_block *mem;
 
 	mem = find_memory_block(section);
+	unregister_mem_sect_under_nodes(mem);
 	mem_remove_simple_file(mem, phys_index);
 	mem_remove_simple_file(mem, state);
 	mem_remove_simple_file(mem, phys_device);
@@ -424,9 +430,9 @@ int remove_memory_block(unsigned long node_id, struct mem_section *section,
  * need an interface for the VM to add new memory regions,
  * but without onlining it.
  */
-int register_new_memory(struct mem_section *section)
+int register_new_memory(int nid, struct mem_section *section)
 {
-	return add_memory_block(0, section, MEM_OFFLINE, 0);
+	return add_memory_block(nid, section, MEM_OFFLINE, 0, HOTPLUG);
 }
 
 int unregister_memory_section(struct mem_section *section)
@@ -458,7 +464,8 @@ int __init memory_dev_init(void)
 	for (i = 0; i < NR_MEM_SECTIONS; i++) {
 		if (!present_section_nr(i))
 			continue;
-		err = add_memory_block(0, __nr_to_section(i), MEM_ONLINE, 0);
+		err = add_memory_block(0, __nr_to_section(i), MEM_ONLINE,
+					0, BOOT);
 		if (!ret)
 			ret = err;
 	}
diff --git a/drivers/base/node.c b/drivers/base/node.c
index 91636cd8b6c9..43fa90b837ee 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -6,6 +6,7 @@
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/mm.h>
+#include <linux/memory.h>
 #include <linux/node.h>
 #include <linux/hugetlb.h>
 #include <linux/cpumask.h>
@@ -248,6 +249,105 @@ int unregister_cpu_under_node(unsigned int cpu, unsigned int nid)
 	return 0;
 }
 
+#ifdef CONFIG_MEMORY_HOTPLUG_SPARSE
+#define page_initialized(page)  (page->lru.next)
+
+static int get_nid_for_pfn(unsigned long pfn)
+{
+	struct page *page;
+
+	if (!pfn_valid_within(pfn))
+		return -1;
+	page = pfn_to_page(pfn);
+	if (!page_initialized(page))
+		return -1;
+	return pfn_to_nid(pfn);
+}
+
+/* register memory section under specified node if it spans that node */
+int register_mem_sect_under_node(struct memory_block *mem_blk, int nid)
+{
+	unsigned long pfn, sect_start_pfn, sect_end_pfn;
+
+	if (!mem_blk)
+		return -EFAULT;
+	if (!node_online(nid))
+		return 0;
+	sect_start_pfn = section_nr_to_pfn(mem_blk->phys_index);
+	sect_end_pfn = sect_start_pfn + PAGES_PER_SECTION - 1;
+	for (pfn = sect_start_pfn; pfn <= sect_end_pfn; pfn++) {
+		int page_nid;
+
+		page_nid = get_nid_for_pfn(pfn);
+		if (page_nid < 0)
+			continue;
+		if (page_nid != nid)
+			continue;
+		return sysfs_create_link_nowarn(&node_devices[nid].sysdev.kobj,
+					&mem_blk->sysdev.kobj,
+					kobject_name(&mem_blk->sysdev.kobj));
+	}
+	/* mem section does not span the specified node */
+	return 0;
+}
+
+/* unregister memory section under all nodes that it spans */
+int unregister_mem_sect_under_nodes(struct memory_block *mem_blk)
+{
+	nodemask_t unlinked_nodes;
+	unsigned long pfn, sect_start_pfn, sect_end_pfn;
+
+	if (!mem_blk)
+		return -EFAULT;
+	nodes_clear(unlinked_nodes);
+	sect_start_pfn = section_nr_to_pfn(mem_blk->phys_index);
+	sect_end_pfn = sect_start_pfn + PAGES_PER_SECTION - 1;
+	for (pfn = sect_start_pfn; pfn <= sect_end_pfn; pfn++) {
+		unsigned int nid;
+
+		nid = get_nid_for_pfn(pfn);
+		if (nid < 0)
+			continue;
+		if (!node_online(nid))
+			continue;
+		if (node_test_and_set(nid, unlinked_nodes))
+			continue;
+		sysfs_remove_link(&node_devices[nid].sysdev.kobj,
+			 kobject_name(&mem_blk->sysdev.kobj));
+	}
+	return 0;
+}
+
+static int link_mem_sections(int nid)
+{
+	unsigned long start_pfn = NODE_DATA(nid)->node_start_pfn;
+	unsigned long end_pfn = start_pfn + NODE_DATA(nid)->node_spanned_pages;
+	unsigned long pfn;
+	int err = 0;
+
+	for (pfn = start_pfn; pfn < end_pfn; pfn += PAGES_PER_SECTION) {
+		unsigned long section_nr = pfn_to_section_nr(pfn);
+		struct mem_section *mem_sect;
+		struct memory_block *mem_blk;
+		int ret;
+
+		if (!present_section_nr(section_nr))
+			continue;
+		mem_sect = __nr_to_section(section_nr);
+		mem_blk = find_memory_block(mem_sect);
+		ret = register_mem_sect_under_node(mem_blk, nid);
+		if (!err)
+			err = ret;
+
+		/* discard ref obtained in find_memory_block() */
+		kobject_put(&mem_blk->sysdev.kobj);
+	}
+	return err;
+}
+#else
+static int link_mem_sections(int nid) { return 0; }
+#endif /* CONFIG_MEMORY_HOTPLUG_SPARSE */
+
 int register_one_node(int nid)
 {
 	int error = 0;
@@ -267,6 +367,9 @@ int register_one_node(int nid)
 			if (cpu_to_node(cpu) == nid)
 				register_cpu_under_node(cpu, nid);
 		}
+
+		/* link memory sections under this node */
+		error = link_mem_sections(nid);
 	}
 
 	return error;
diff --git a/include/linux/memory.h b/include/linux/memory.h
index 36c82c9e6ea7..3fdc10806d31 100644
--- a/include/linux/memory.h
+++ b/include/linux/memory.h
@@ -79,14 +79,14 @@ static inline int memory_notify(unsigned long val, void *v)
 #else
 extern int register_memory_notifier(struct notifier_block *nb);
 extern void unregister_memory_notifier(struct notifier_block *nb);
-extern int register_new_memory(struct mem_section *);
+extern int register_new_memory(int, struct mem_section *);
 extern int unregister_memory_section(struct mem_section *);
 extern int memory_dev_init(void);
 extern int remove_memory_block(unsigned long, struct mem_section *, int);
 extern int memory_notify(unsigned long val, void *v);
+extern struct memory_block *find_memory_block(struct mem_section *);
 #define CONFIG_MEM_BLOCK_SIZE	(PAGES_PER_SECTION<<PAGE_SHIFT)
-
-
+enum mem_add_context { BOOT, HOTPLUG };
 #endif /* CONFIG_MEMORY_HOTPLUG_SPARSE */
 
 #ifdef CONFIG_MEMORY_HOTPLUG
diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h
index 763ba81fc0f0..d95f72e79b82 100644
--- a/include/linux/memory_hotplug.h
+++ b/include/linux/memory_hotplug.h
@@ -72,7 +72,7 @@ extern void __offline_isolated_pages(unsigned long, unsigned long);
 extern int offline_pages(unsigned long, unsigned long, unsigned long);
 
 /* reasonably generic interface to expand the physical pages in a zone  */
-extern int __add_pages(struct zone *zone, unsigned long start_pfn,
+extern int __add_pages(int nid, struct zone *zone, unsigned long start_pfn,
 	unsigned long nr_pages);
 extern int __remove_pages(struct zone *zone, unsigned long start_pfn,
 	unsigned long nr_pages);
diff --git a/include/linux/node.h b/include/linux/node.h
index bc001bc225c3..681a697b9a86 100644
--- a/include/linux/node.h
+++ b/include/linux/node.h
@@ -26,6 +26,7 @@ struct node {
 	struct sys_device	sysdev;
 };
 
+struct memory_block;
 extern struct node node_devices[];
 
 extern int register_node(struct node *, int, struct node *);
@@ -35,6 +36,9 @@ extern int register_one_node(int nid);
 extern void unregister_one_node(int nid);
 extern int register_cpu_under_node(unsigned int cpu, unsigned int nid);
 extern int unregister_cpu_under_node(unsigned int cpu, unsigned int nid);
+extern int register_mem_sect_under_node(struct memory_block *mem_blk,
+						int nid);
+extern int unregister_mem_sect_under_nodes(struct memory_block *mem_blk);
 #else
 static inline int register_one_node(int nid)
 {
@@ -52,6 +56,15 @@ static inline int unregister_cpu_under_node(unsigned int cpu, unsigned int nid)
 {
 	return 0;
 }
+static inline int register_mem_sect_under_node(struct memory_block *mem_blk,
+							int nid)
+{
+	return 0;
+}
+static inline int unregister_mem_sect_under_nodes(struct memory_block *mem_blk)
+{
+	return 0;
+}
 #endif
 
 #define to_node(sys_device) container_of(sys_device, struct node, sysdev)
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index b17371185468..2ba38bc07b47 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -216,7 +216,8 @@ static int __meminit __add_zone(struct zone *zone, unsigned long phys_start_pfn)
 	return 0;
 }
 
-static int __meminit __add_section(struct zone *zone, unsigned long phys_start_pfn)
+static int __meminit __add_section(int nid, struct zone *zone,
+					unsigned long phys_start_pfn)
 {
 	int nr_pages = PAGES_PER_SECTION;
 	int ret;
@@ -234,7 +235,7 @@ static int __meminit __add_section(struct zone *zone, unsigned long phys_start_p
 	if (ret < 0)
 		return ret;
 
-	return register_new_memory(__pfn_to_section(phys_start_pfn));
+	return register_new_memory(nid, __pfn_to_section(phys_start_pfn));
 }
 
 #ifdef CONFIG_SPARSEMEM_VMEMMAP
@@ -273,8 +274,8 @@ static int __remove_section(struct zone *zone, struct mem_section *ms)
  * call this function after deciding the zone to which to
  * add the new pages.
  */
-int __ref __add_pages(struct zone *zone, unsigned long phys_start_pfn,
-		 unsigned long nr_pages)
+int __ref __add_pages(int nid, struct zone *zone, unsigned long phys_start_pfn,
+			unsigned long nr_pages)
 {
 	unsigned long i;
 	int err = 0;
@@ -284,7 +285,7 @@ int __ref __add_pages(struct zone *zone, unsigned long phys_start_pfn,
 	end_sec = pfn_to_section_nr(phys_start_pfn + nr_pages - 1);
 
 	for (i = start_sec; i <= end_sec; i++) {
-		err = __add_section(zone, i << PFN_SECTION_SHIFT);
+		err = __add_section(nid, zone, i << PFN_SECTION_SHIFT);
 
 		/*
 		 * EEXIST is finally dealt with by ioresource collision
-- 
cgit v1.2.3


From 1b0bd118862cd9fe9ac2872137a1b8107e83ff9d Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Tue, 6 Jan 2009 14:39:15 -0800
Subject: mm: get rid of pagevec_release_nonlru()

speculative page references patch (commit:
e286781d5f2e9c846e012a39653a166e9d31777d) removed last
pagevec_release_nonlru() caller.

So this function can be removed now.

This patch doesn't have any functional change.

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/pagevec.h |  7 -------
 mm/swap.c               | 22 ----------------------
 2 files changed, 29 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/pagevec.h b/include/linux/pagevec.h
index e90a2cb02915..7b2886fa7fdc 100644
--- a/include/linux/pagevec.h
+++ b/include/linux/pagevec.h
@@ -21,7 +21,6 @@ struct pagevec {
 };
 
 void __pagevec_release(struct pagevec *pvec);
-void __pagevec_release_nonlru(struct pagevec *pvec);
 void __pagevec_free(struct pagevec *pvec);
 void ____pagevec_lru_add(struct pagevec *pvec, enum lru_list lru);
 void pagevec_strip(struct pagevec *pvec);
@@ -69,12 +68,6 @@ static inline void pagevec_release(struct pagevec *pvec)
 		__pagevec_release(pvec);
 }
 
-static inline void pagevec_release_nonlru(struct pagevec *pvec)
-{
-	if (pagevec_count(pvec))
-		__pagevec_release_nonlru(pvec);
-}
-
 static inline void pagevec_free(struct pagevec *pvec)
 {
 	if (pagevec_count(pvec))
diff --git a/mm/swap.c b/mm/swap.c
index b135ec90cdeb..21a566fc4570 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -397,28 +397,6 @@ void __pagevec_release(struct pagevec *pvec)
 
 EXPORT_SYMBOL(__pagevec_release);
 
-/*
- * pagevec_release() for pages which are known to not be on the LRU
- *
- * This function reinitialises the caller's pagevec.
- */
-void __pagevec_release_nonlru(struct pagevec *pvec)
-{
-	int i;
-	struct pagevec pages_to_free;
-
-	pagevec_init(&pages_to_free, pvec->cold);
-	for (i = 0; i < pagevec_count(pvec); i++) {
-		struct page *page = pvec->pages[i];
-
-		VM_BUG_ON(PageLRU(page));
-		if (put_page_testzero(page))
-			pagevec_add(&pages_to_free, page);
-	}
-	pagevec_free(&pages_to_free);
-	pagevec_reinit(pvec);
-}
-
 /*
  * Add the passed pages to the LRU, then drop the caller's refcount
  * on them.  Reinitialises the caller's pagevec.
-- 
cgit v1.2.3


From 64cdd548ffe26849d4cd113ac640f60606063b14 Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Tue, 6 Jan 2009 14:39:16 -0800
Subject: mm: cleanup: remove #ifdef CONFIG_MIGRATION

#ifdef in *.c file decrease source readability a bit.  removing is better.

This patch doesn't have any functional change.

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/migrate.h |  4 ++++
 mm/mprotect.c           |  6 ++----
 mm/rmap.c               | 10 +++-------
 3 files changed, 9 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/migrate.h b/include/linux/migrate.h
index 3f34005068d4..527602cdea1c 100644
--- a/include/linux/migrate.h
+++ b/include/linux/migrate.h
@@ -7,6 +7,8 @@
 typedef struct page *new_page_t(struct page *, unsigned long private, int **);
 
 #ifdef CONFIG_MIGRATION
+#define PAGE_MIGRATION 1
+
 extern int putback_lru_pages(struct list_head *l);
 extern int migrate_page(struct address_space *,
 			struct page *, struct page *);
@@ -20,6 +22,8 @@ extern int migrate_vmas(struct mm_struct *mm,
 		const nodemask_t *from, const nodemask_t *to,
 		unsigned long flags);
 #else
+#define PAGE_MIGRATION 0
+
 static inline int putback_lru_pages(struct list_head *l) { return 0; }
 static inline int migrate_pages(struct list_head *l, new_page_t x,
 		unsigned long private) { return -ENOSYS; }
diff --git a/mm/mprotect.c b/mm/mprotect.c
index cfb4c4852062..d0f6e7ce09f1 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -22,6 +22,7 @@
 #include <linux/swap.h>
 #include <linux/swapops.h>
 #include <linux/mmu_notifier.h>
+#include <linux/migrate.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 #include <asm/cacheflush.h>
@@ -59,8 +60,7 @@ static void change_pte_range(struct mm_struct *mm, pmd_t *pmd,
 				ptent = pte_mkwrite(ptent);
 
 			ptep_modify_prot_commit(mm, addr, pte, ptent);
-#ifdef CONFIG_MIGRATION
-		} else if (!pte_file(oldpte)) {
+		} else if (PAGE_MIGRATION && !pte_file(oldpte)) {
 			swp_entry_t entry = pte_to_swp_entry(oldpte);
 
 			if (is_write_migration_entry(entry)) {
@@ -72,9 +72,7 @@ static void change_pte_range(struct mm_struct *mm, pmd_t *pmd,
 				set_pte_at(mm, addr, pte,
 					swp_entry_to_pte(entry));
 			}
-#endif
 		}
-
 	} while (pte++, addr += PAGE_SIZE, addr != end);
 	arch_leave_lazy_mmu_mode();
 	pte_unmap_unlock(pte - 1, ptl);
diff --git a/mm/rmap.c b/mm/rmap.c
index 10993942d6c9..53c56dacd725 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -50,6 +50,7 @@
 #include <linux/kallsyms.h>
 #include <linux/memcontrol.h>
 #include <linux/mmu_notifier.h>
+#include <linux/migrate.h>
 
 #include <asm/tlbflush.h>
 
@@ -818,8 +819,7 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
 				spin_unlock(&mmlist_lock);
 			}
 			dec_mm_counter(mm, anon_rss);
-#ifdef CONFIG_MIGRATION
-		} else {
+		} else if (PAGE_MIGRATION) {
 			/*
 			 * Store the pfn of the page in a special migration
 			 * pte. do_swap_page() will wait until the migration
@@ -827,19 +827,15 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
 			 */
 			BUG_ON(!migration);
 			entry = make_migration_entry(page, pte_write(pteval));
-#endif
 		}
 		set_pte_at(mm, address, pte, swp_entry_to_pte(entry));
 		BUG_ON(pte_file(*pte));
-	} else
-#ifdef CONFIG_MIGRATION
-	if (migration) {
+	} else if (PAGE_MIGRATION && migration) {
 		/* Establish migration entry for a file page */
 		swp_entry_t entry;
 		entry = make_migration_entry(page, pte_write(pteval));
 		set_pte_at(mm, address, pte, swp_entry_to_pte(entry));
 	} else
-#endif
 		dec_mm_counter(mm, file_rss);
 
 
-- 
cgit v1.2.3


From e5991371ee0d1c0ce19e133c6f9075b49c5b4ae8 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:22 -0800
Subject: mm: remove cgroup_mm_owner_callbacks

cgroup_mm_owner_callbacks() was brought in to support the memrlimit
controller, but sneaked into mainline ahead of it.  That controller has
now been shelved, and the mm_owner_changed() args were inadequate for it
anyway (they needed an mm pointer instead of a task pointer).

Remove the dead code, and restore mm_update_next_owner() locking to how it
was before: taking mmap_sem there does nothing for memcontrol.c, now the
only user of mm->owner.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Paul Menage <menage@google.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cgroup.h | 14 +-------------
 kernel/cgroup.c        | 33 ---------------------------------
 kernel/exit.c          | 16 ++++++----------
 3 files changed, 7 insertions(+), 56 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index 1164963c3a85..08b78c09b09a 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -329,13 +329,7 @@ struct cgroup_subsys {
 			struct cgroup *cgrp);
 	void (*post_clone)(struct cgroup_subsys *ss, struct cgroup *cgrp);
 	void (*bind)(struct cgroup_subsys *ss, struct cgroup *root);
-	/*
-	 * This routine is called with the task_lock of mm->owner held
-	 */
-	void (*mm_owner_changed)(struct cgroup_subsys *ss,
-					struct cgroup *old,
-					struct cgroup *new,
-					struct task_struct *p);
+
 	int subsys_id;
 	int active;
 	int disabled;
@@ -400,9 +394,6 @@ void cgroup_iter_end(struct cgroup *cgrp, struct cgroup_iter *it);
 int cgroup_scan_tasks(struct cgroup_scanner *scan);
 int cgroup_attach_task(struct cgroup *, struct task_struct *);
 
-void cgroup_mm_owner_callbacks(struct task_struct *old,
-			       struct task_struct *new);
-
 #else /* !CONFIG_CGROUPS */
 
 static inline int cgroup_init_early(void) { return 0; }
@@ -420,9 +411,6 @@ static inline int cgroupstats_build(struct cgroupstats *stats,
 	return -EINVAL;
 }
 
-static inline void cgroup_mm_owner_callbacks(struct task_struct *old,
-					     struct task_struct *new) {}
-
 #endif /* !CONFIG_CGROUPS */
 
 #endif /* _LINUX_CGROUP_H */
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 87bb0258fd27..f221446aa02d 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -116,7 +116,6 @@ static int root_count;
  * be called.
  */
 static int need_forkexit_callback __read_mostly;
-static int need_mm_owner_callback __read_mostly;
 
 /* convenient tests for these bits */
 inline int cgroup_is_removed(const struct cgroup *cgrp)
@@ -2539,7 +2538,6 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss)
 	init_css_set.subsys[ss->subsys_id] = dummytop->subsys[ss->subsys_id];
 
 	need_forkexit_callback |= ss->fork || ss->exit;
-	need_mm_owner_callback |= !!ss->mm_owner_changed;
 
 	/* At system boot, before all subsystems have been
 	 * registered, no tasks have been forked, so we don't
@@ -2789,37 +2787,6 @@ void cgroup_fork_callbacks(struct task_struct *child)
 	}
 }
 
-#ifdef CONFIG_MM_OWNER
-/**
- * cgroup_mm_owner_callbacks - run callbacks when the mm->owner changes
- * @p: the new owner
- *
- * Called on every change to mm->owner. mm_init_owner() does not
- * invoke this routine, since it assigns the mm->owner the first time
- * and does not change it.
- *
- * The callbacks are invoked with mmap_sem held in read mode.
- */
-void cgroup_mm_owner_callbacks(struct task_struct *old, struct task_struct *new)
-{
-	struct cgroup *oldcgrp, *newcgrp = NULL;
-
-	if (need_mm_owner_callback) {
-		int i;
-		for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) {
-			struct cgroup_subsys *ss = subsys[i];
-			oldcgrp = task_cgroup(old, ss->subsys_id);
-			if (new)
-				newcgrp = task_cgroup(new, ss->subsys_id);
-			if (oldcgrp == newcgrp)
-				continue;
-			if (ss->mm_owner_changed)
-				ss->mm_owner_changed(ss, oldcgrp, newcgrp, new);
-		}
-	}
-}
-#endif /* CONFIG_MM_OWNER */
-
 /**
  * cgroup_post_fork - called on a new task after adding it to the task list
  * @child: the task in question
diff --git a/kernel/exit.c b/kernel/exit.c
index c9e5a1c14e08..f923724ab3c9 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -642,35 +642,31 @@ retry:
 	/*
 	 * We found no owner yet mm_users > 1: this implies that we are
 	 * most likely racing with swapoff (try_to_unuse()) or /proc or
-	 * ptrace or page migration (get_task_mm()).  Mark owner as NULL,
-	 * so that subsystems can understand the callback and take action.
+	 * ptrace or page migration (get_task_mm()).  Mark owner as NULL.
 	 */
-	down_write(&mm->mmap_sem);
-	cgroup_mm_owner_callbacks(mm->owner, NULL);
 	mm->owner = NULL;
-	up_write(&mm->mmap_sem);
 	return;
 
 assign_new_owner:
 	BUG_ON(c == p);
 	get_task_struct(c);
-	read_unlock(&tasklist_lock);
-	down_write(&mm->mmap_sem);
 	/*
 	 * The task_lock protects c->mm from changing.
 	 * We always want mm->owner->mm == mm
 	 */
 	task_lock(c);
+	/*
+	 * Delay read_unlock() till we have the task_lock()
+	 * to ensure that c does not slip away underneath us
+	 */
+	read_unlock(&tasklist_lock);
 	if (c->mm != mm) {
 		task_unlock(c);
-		up_write(&mm->mmap_sem);
 		put_task_struct(c);
 		goto retry;
 	}
-	cgroup_mm_owner_callbacks(mm->owner, c);
 	mm->owner = c;
 	task_unlock(c);
-	up_write(&mm->mmap_sem);
 	put_task_struct(c);
 }
 #endif /* CONFIG_MM_OWNER */
-- 
cgit v1.2.3


From 3c1d43787b48c798f44dc32a6e6deb5ca2da3e68 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:23 -0800
Subject: mm: remove GFP_HIGHUSER_PAGECACHE

GFP_HIGHUSER_PAGECACHE is just an alias for GFP_HIGHUSER_MOVABLE, making
that harder to track down: remove it, and its out-of-work brothers
GFP_NOFS_PAGECACHE and GFP_USER_PAGECACHE.

Since we're making that improvement to hotremove_migrate_alloc(), I think
we can now also remove one of the "o"s from its comment.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/inode.c          | 4 ++--
 include/linux/gfp.h | 6 ------
 mm/memory_hotplug.c | 9 +++------
 3 files changed, 5 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/fs/inode.c b/fs/inode.c
index bd48e5e6d3e8..a903860bc5ac 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -166,7 +166,7 @@ struct inode *inode_init_always(struct super_block *sb, struct inode *inode)
 	mapping->a_ops = &empty_aops;
 	mapping->host = inode;
 	mapping->flags = 0;
-	mapping_set_gfp_mask(mapping, GFP_HIGHUSER_PAGECACHE);
+	mapping_set_gfp_mask(mapping, GFP_HIGHUSER_MOVABLE);
 	mapping->assoc_mapping = NULL;
 	mapping->backing_dev_info = &default_backing_dev_info;
 	mapping->writeback_index = 0;
@@ -601,7 +601,7 @@ EXPORT_SYMBOL_GPL(inode_add_to_lists);
  *	@sb: superblock
  *
  *	Allocates a new inode for given superblock. The default gfp_mask
- *	for allocations related to inode->i_mapping is GFP_HIGHUSER_PAGECACHE.
+ *	for allocations related to inode->i_mapping is GFP_HIGHUSER_MOVABLE.
  *	If HIGHMEM pages are unsuitable or it is known that pages allocated
  *	for the page cache are not reclaimable or migratable,
  *	mapping_set_gfp_mask() must be called with suitable flags on the
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index e8003afeffba..dd20cd78faa8 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -69,12 +69,6 @@ struct vm_area_struct;
 #define GFP_HIGHUSER_MOVABLE	(__GFP_WAIT | __GFP_IO | __GFP_FS | \
 				 __GFP_HARDWALL | __GFP_HIGHMEM | \
 				 __GFP_MOVABLE)
-#define GFP_NOFS_PAGECACHE	(__GFP_WAIT | __GFP_IO | __GFP_MOVABLE)
-#define GFP_USER_PAGECACHE	(__GFP_WAIT | __GFP_IO | __GFP_FS | \
-				 __GFP_HARDWALL | __GFP_MOVABLE)
-#define GFP_HIGHUSER_PAGECACHE	(__GFP_WAIT | __GFP_IO | __GFP_FS | \
-				 __GFP_HARDWALL | __GFP_HIGHMEM | \
-				 __GFP_MOVABLE)
 
 #ifdef CONFIG_NUMA
 #define GFP_THISNODE	(__GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY)
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 2ba38bc07b47..c083cf5fd6df 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -627,15 +627,12 @@ int scan_lru_pages(unsigned long start, unsigned long end)
 }
 
 static struct page *
-hotremove_migrate_alloc(struct page *page,
-			unsigned long private,
-			int **x)
+hotremove_migrate_alloc(struct page *page, unsigned long private, int **x)
 {
-	/* This should be improoooooved!! */
-	return alloc_page(GFP_HIGHUSER_PAGECACHE);
+	/* This should be improooooved!! */
+	return alloc_page(GFP_HIGHUSER_MOVABLE);
 }
 
-
 #define NR_OFFLINE_AT_ONCE_PAGES	(256)
 static int
 do_migrate_range(unsigned long start_pfn, unsigned long end_pfn)
-- 
cgit v1.2.3


From 6d91add09f4bad5f4d4233b13faa392f0c4b16be Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:24 -0800
Subject: mm: add Set,ClearPageSwapCache stubs

If we add NOOP stubs for SetPageSwapCache() and ClearPageSwapCache(), then
we can remove the #ifdef CONFIG_SWAPs from mm/migrate.c.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Acked-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/page-flags.h | 1 +
 mm/migrate.c               | 4 ----
 2 files changed, 1 insertion(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index b12f93a3c345..628ec0802492 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -228,6 +228,7 @@ PAGEFLAG_FALSE(HighMem)
 PAGEFLAG(SwapCache, swapcache)
 #else
 PAGEFLAG_FALSE(SwapCache)
+	SETPAGEFLAG_NOOP(SwapCache) CLEARPAGEFLAG_NOOP(SwapCache)
 #endif
 
 #ifdef CONFIG_UNEVICTABLE_LRU
diff --git a/mm/migrate.c b/mm/migrate.c
index 60510306c44a..55373983c9c6 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -300,12 +300,10 @@ static int migrate_page_move_mapping(struct address_space *mapping,
 	 * Now we know that no one else is looking at the page.
 	 */
 	get_page(newpage);	/* add cache reference */
-#ifdef CONFIG_SWAP
 	if (PageSwapCache(page)) {
 		SetPageSwapCache(newpage);
 		set_page_private(newpage, page_private(page));
 	}
-#endif
 
 	radix_tree_replace_slot(pslot, newpage);
 
@@ -373,9 +371,7 @@ static void migrate_page_copy(struct page *newpage, struct page *page)
 
 	mlock_migrate_page(newpage, page);
 
-#ifdef CONFIG_SWAP
 	ClearPageSwapCache(page);
-#endif
 	ClearPagePrivate(page);
 	set_page_private(page, 0);
 	/* page->mapping contains a flag for PageAnon() */
-- 
cgit v1.2.3


From b5934c531849ff4a51ce0f290141efe564290e40 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:25 -0800
Subject: mm: add_active_or_unevictable into rmap

lru_cache_add_active_or_unevictable() and page_add_new_anon_rmap() always
appear together.  Save some symbol table space and some jumping around by
removing lru_cache_add_active_or_unevictable(), folding its code into
page_add_new_anon_rmap(): like how we add file pages to lru just after
adding them to page cache.

Remove the nearby "TODO: is this safe?" comments (yes, it is safe), and
change page_add_new_anon_rmap()'s address BUG_ON to VM_BUG_ON as
originally intended.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h |  2 --
 mm/memory.c          |  6 ------
 mm/rmap.c            |  7 ++++++-
 mm/swap.c            | 19 -------------------
 4 files changed, 6 insertions(+), 28 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index a3af95b2cb6d..48f309dc5a0c 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -174,8 +174,6 @@ extern unsigned int nr_free_pagecache_pages(void);
 /* linux/mm/swap.c */
 extern void __lru_cache_add(struct page *, enum lru_list lru);
 extern void lru_cache_add_lru(struct page *, enum lru_list lru);
-extern void lru_cache_add_active_or_unevictable(struct page *,
-					struct vm_area_struct *);
 extern void activate_page(struct page *);
 extern void mark_page_accessed(struct page *);
 extern void lru_add_drain(void);
diff --git a/mm/memory.c b/mm/memory.c
index b5af358b8b22..a138c50dc39a 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1949,10 +1949,7 @@ gotten:
 		 */
 		ptep_clear_flush_notify(vma, address, page_table);
 		SetPageSwapBacked(new_page);
-		lru_cache_add_active_or_unevictable(new_page, vma);
 		page_add_new_anon_rmap(new_page, vma, address);
-
-//TODO:  is this safe?  do_anonymous_page() does it this way.
 		set_pte_at(mm, address, page_table, entry);
 		update_mmu_cache(vma, address, entry);
 		if (old_page) {
@@ -2448,7 +2445,6 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		goto release;
 	inc_mm_counter(mm, anon_rss);
 	SetPageSwapBacked(page);
-	lru_cache_add_active_or_unevictable(page, vma);
 	page_add_new_anon_rmap(page, vma, address);
 	set_pte_at(mm, address, page_table, entry);
 
@@ -2597,7 +2593,6 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 		if (anon) {
 			inc_mm_counter(mm, anon_rss);
 			SetPageSwapBacked(page);
-			lru_cache_add_active_or_unevictable(page, vma);
 			page_add_new_anon_rmap(page, vma, address);
 		} else {
 			inc_mm_counter(mm, file_rss);
@@ -2607,7 +2602,6 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 				get_page(dirty_page);
 			}
 		}
-//TODO:  is this safe?  do_anonymous_page() does it this way.
 		set_pte_at(mm, address, page_table, entry);
 
 		/* no need to invalidate: a not-present page won't be cached */
diff --git a/mm/rmap.c b/mm/rmap.c
index f01e92244c53..10da68202b10 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -47,6 +47,7 @@
 #include <linux/rmap.h>
 #include <linux/rcupdate.h>
 #include <linux/module.h>
+#include <linux/mm_inline.h>
 #include <linux/kallsyms.h>
 #include <linux/memcontrol.h>
 #include <linux/mmu_notifier.h>
@@ -671,9 +672,13 @@ void page_add_anon_rmap(struct page *page,
 void page_add_new_anon_rmap(struct page *page,
 	struct vm_area_struct *vma, unsigned long address)
 {
-	BUG_ON(address < vma->vm_start || address >= vma->vm_end);
+	VM_BUG_ON(address < vma->vm_start || address >= vma->vm_end);
 	atomic_set(&page->_mapcount, 0); /* elevate count by 1 (starts at -1) */
 	__page_set_anon_rmap(page, vma, address);
+	if (page_evictable(page, vma))
+		lru_cache_add_lru(page, LRU_ACTIVE + page_is_file_cache(page));
+	else
+		add_page_to_unevictable_list(page);
 }
 
 /**
diff --git a/mm/swap.c b/mm/swap.c
index 21a566fc4570..ff0b290475fd 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -246,25 +246,6 @@ void add_page_to_unevictable_list(struct page *page)
 	spin_unlock_irq(&zone->lru_lock);
 }
 
-/**
- * lru_cache_add_active_or_unevictable
- * @page:  the page to be added to LRU
- * @vma:   vma in which page is mapped for determining reclaimability
- *
- * place @page on active or unevictable LRU list, depending on
- * page_evictable().  Note that if the page is not evictable,
- * it goes directly back onto it's zone's unevictable list.  It does
- * NOT use a per cpu pagevec.
- */
-void lru_cache_add_active_or_unevictable(struct page *page,
-					struct vm_area_struct *vma)
-{
-	if (page_evictable(page, vma))
-		lru_cache_add_lru(page, LRU_ACTIVE + page_is_file_cache(page));
-	else
-		add_page_to_unevictable_list(page);
-}
-
 /*
  * Drain pages out of the cpu's pagevecs.
  * Either "cpu" is the current CPU, and preemption has already been
-- 
cgit v1.2.3


From 2afd1c928f1132b8d0099866e75ce8ad713a1180 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:26 -0800
Subject: mm: make page_lock_anon_vma() static

page_lock_anon_vma() and page_unlock_anon_vma() were made available to
show_page_path() in vmscan.c; but now that has been removed, make them
static in rmap.c again, they're better kept private if possible.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/rmap.h | 3 ---
 mm/rmap.c            | 4 ++--
 2 files changed, 2 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index 89f0564b10c8..3593b18a07dd 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -63,9 +63,6 @@ void anon_vma_unlink(struct vm_area_struct *);
 void anon_vma_link(struct vm_area_struct *);
 void __anon_vma_link(struct vm_area_struct *);
 
-extern struct anon_vma *page_lock_anon_vma(struct page *page);
-extern void page_unlock_anon_vma(struct anon_vma *anon_vma);
-
 /*
  * rmap interfaces called when adding or removing pte of page
  */
diff --git a/mm/rmap.c b/mm/rmap.c
index 10da68202b10..892e1877366b 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -193,7 +193,7 @@ void __init anon_vma_init(void)
  * Getting a lock on a stable anon_vma from a page off the LRU is
  * tricky: page_lock_anon_vma rely on RCU to guard against the races.
  */
-struct anon_vma *page_lock_anon_vma(struct page *page)
+static struct anon_vma *page_lock_anon_vma(struct page *page)
 {
 	struct anon_vma *anon_vma;
 	unsigned long anon_mapping;
@@ -213,7 +213,7 @@ out:
 	return NULL;
 }
 
-void page_unlock_anon_vma(struct anon_vma *anon_vma)
+static void page_unlock_anon_vma(struct anon_vma *anon_vma)
 {
 	spin_unlock(&anon_vma->lock);
 	rcu_read_unlock();
-- 
cgit v1.2.3


From 364aeb2849789b51bf4b9af2ddd02fee7285c54e Mon Sep 17 00:00:00 2001
From: David Rientjes <rientjes@google.com>
Date: Tue, 6 Jan 2009 14:39:29 -0800
Subject: mm: change dirty limit type specifiers to unsigned long

The background dirty and dirty limits are better defined with type
specifiers of unsigned long since negative writeback thresholds are not
possible.

These values, as returned by get_dirty_limits(), are normally compared
with ZVC values to determine whether writeback shall commence or be
throttled.  Such page counts cannot be negative, so declaring the page
limits as signed is unnecessary.

Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Andrea Righi <righi.andrea@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/writeback.h |  4 ++--
 mm/backing-dev.c          |  6 +++---
 mm/page-writeback.c       | 22 +++++++++++-----------
 3 files changed, 16 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index e585657e9831..259e9ea58cab 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -125,8 +125,8 @@ struct file;
 int dirty_writeback_centisecs_handler(struct ctl_table *, int, struct file *,
 				      void __user *, size_t *, loff_t *);
 
-void get_dirty_limits(long *pbackground, long *pdirty, long *pbdi_dirty,
-		 struct backing_dev_info *bdi);
+void get_dirty_limits(unsigned long *pbackground, unsigned long *pdirty,
+		      unsigned long *pbdi_dirty, struct backing_dev_info *bdi);
 
 void page_writeback_init(void);
 void balance_dirty_pages_ratelimited_nr(struct address_space *mapping,
diff --git a/mm/backing-dev.c b/mm/backing-dev.c
index 801c08b046e6..6f80beddd8a4 100644
--- a/mm/backing-dev.c
+++ b/mm/backing-dev.c
@@ -24,9 +24,9 @@ static void bdi_debug_init(void)
 static int bdi_debug_stats_show(struct seq_file *m, void *v)
 {
 	struct backing_dev_info *bdi = m->private;
-	long background_thresh;
-	long dirty_thresh;
-	long bdi_thresh;
+	unsigned long background_thresh;
+	unsigned long dirty_thresh;
+	unsigned long bdi_thresh;
 
 	get_dirty_limits(&background_thresh, &dirty_thresh, &bdi_thresh, bdi);
 
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 08d2b960b294..4d4074cff300 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -362,13 +362,13 @@ unsigned long determine_dirtyable_memory(void)
 }
 
 void
-get_dirty_limits(long *pbackground, long *pdirty, long *pbdi_dirty,
-		 struct backing_dev_info *bdi)
+get_dirty_limits(unsigned long *pbackground, unsigned long *pdirty,
+		 unsigned long *pbdi_dirty, struct backing_dev_info *bdi)
 {
 	int background_ratio;		/* Percentages */
 	int dirty_ratio;
-	long background;
-	long dirty;
+	unsigned long background;
+	unsigned long dirty;
 	unsigned long available_memory = determine_dirtyable_memory();
 	struct task_struct *tsk;
 
@@ -423,9 +423,9 @@ static void balance_dirty_pages(struct address_space *mapping)
 {
 	long nr_reclaimable, bdi_nr_reclaimable;
 	long nr_writeback, bdi_nr_writeback;
-	long background_thresh;
-	long dirty_thresh;
-	long bdi_thresh;
+	unsigned long background_thresh;
+	unsigned long dirty_thresh;
+	unsigned long bdi_thresh;
 	unsigned long pages_written = 0;
 	unsigned long write_chunk = sync_writeback_pages();
 
@@ -580,8 +580,8 @@ EXPORT_SYMBOL(balance_dirty_pages_ratelimited_nr);
 
 void throttle_vm_writeout(gfp_t gfp_mask)
 {
-	long background_thresh;
-	long dirty_thresh;
+	unsigned long background_thresh;
+	unsigned long dirty_thresh;
 
         for ( ; ; ) {
 		get_dirty_limits(&background_thresh, &dirty_thresh, NULL, NULL);
@@ -624,8 +624,8 @@ static void background_writeout(unsigned long _min_pages)
 	};
 
 	for ( ; ; ) {
-		long background_thresh;
-		long dirty_thresh;
+		unsigned long background_thresh;
+		unsigned long dirty_thresh;
 
 		get_dirty_limits(&background_thresh, &dirty_thresh, NULL, NULL);
 		if (global_page_state(NR_FILE_DIRTY) +
-- 
cgit v1.2.3


From 2da02997e08d3efe8174c7a47696e6f7cbe69ba9 Mon Sep 17 00:00:00 2001
From: David Rientjes <rientjes@google.com>
Date: Tue, 6 Jan 2009 14:39:31 -0800
Subject: mm: add dirty_background_bytes and dirty_bytes sysctls

This change introduces two new sysctls to /proc/sys/vm:
dirty_background_bytes and dirty_bytes.

dirty_background_bytes is the counterpart to dirty_background_ratio and
dirty_bytes is the counterpart to dirty_ratio.

With growing memory capacities of individual machines, it's no longer
sufficient to specify dirty thresholds as a percentage of the amount of
dirtyable memory over the entire system.

dirty_background_bytes and dirty_bytes specify quantities of memory, in
bytes, that represent the dirty limits for the entire system.  If either
of these values is set, its value represents the amount of dirty memory
that is needed to commence either background or direct writeback.

When a `bytes' or `ratio' file is written, its counterpart becomes a
function of the written value.  For example, if dirty_bytes is written to
be 8096, 8K of memory is required to commence direct writeback.
dirty_ratio is then functionally equivalent to 8K / the amount of
dirtyable memory:

	dirtyable_memory = free pages + mapped pages + file cache

	dirty_background_bytes = dirty_background_ratio * dirtyable_memory
		-or-
	dirty_background_ratio = dirty_background_bytes / dirtyable_memory

		AND

	dirty_bytes = dirty_ratio * dirtyable_memory
		-or-
	dirty_ratio = dirty_bytes / dirtyable_memory

Only one of dirty_background_bytes and dirty_background_ratio may be
specified at a time, and only one of dirty_bytes and dirty_ratio may be
specified.  When one sysctl is written, the other appears as 0 when read.

The `bytes' files operate on a page size granularity since dirty limits
are compared with ZVC values, which are in page units.

Prior to this change, the minimum dirty_ratio was 5 as implemented by
get_dirty_limits() although /proc/sys/vm/dirty_ratio would show any user
written value between 0 and 100.  This restriction is maintained, but
dirty_bytes has a lower limit of only one page.

Also prior to this change, the dirty_background_ratio could not equal or
exceed dirty_ratio.  This restriction is maintained in addition to
restricting dirty_background_bytes.  If either background threshold equals
or exceeds that of the dirty threshold, it is implicitly set to half the
dirty threshold.

Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Andrea Righi <righi.andrea@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/filesystems/proc.txt |  26 +++++++++-
 Documentation/sysctl/vm.txt        |   3 +-
 include/linux/writeback.h          |  11 ++++
 kernel/sysctl.c                    |  27 ++++++++--
 mm/page-writeback.c                | 102 +++++++++++++++++++++++++++++++------
 5 files changed, 146 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt
index 71df353e367c..32e94635484f 100644
--- a/Documentation/filesystems/proc.txt
+++ b/Documentation/filesystems/proc.txt
@@ -1385,6 +1385,15 @@ swapcache reclaim.  Decreasing vfs_cache_pressure causes the kernel to prefer
 to retain dentry and inode caches.  Increasing vfs_cache_pressure beyond 100
 causes the kernel to prefer to reclaim dentries and inodes.
 
+dirty_background_bytes
+----------------------
+
+Contains the amount of dirty memory at which the pdflush background writeback
+daemon will start writeback.
+
+If dirty_background_bytes is written, dirty_background_ratio becomes a function
+of its value (dirty_background_bytes / the amount of dirtyable system memory).
+
 dirty_background_ratio
 ----------------------
 
@@ -1393,14 +1402,29 @@ pages + file cache, not including locked pages and HugePages), the number of
 pages at which the pdflush background writeback daemon will start writing out
 dirty data.
 
+If dirty_background_ratio is written, dirty_background_bytes becomes a function
+of its value (dirty_background_ratio * the amount of dirtyable system memory).
+
+dirty_bytes
+-----------
+
+Contains the amount of dirty memory at which a process generating disk writes
+will itself start writeback.
+
+If dirty_bytes is written, dirty_ratio becomes a function of its value
+(dirty_bytes / the amount of dirtyable system memory).
+
 dirty_ratio
------------------
+-----------
 
 Contains, as a percentage of the dirtyable system memory (free pages + mapped
 pages + file cache, not including locked pages and HugePages), the number of
 pages at which a process which is generating disk writes will itself start
 writing out dirty data.
 
+If dirty_ratio is written, dirty_bytes becomes a function of its value
+(dirty_ratio * the amount of dirtyable system memory).
+
 dirty_writeback_centisecs
 -------------------------
 
diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
index d79eeda7a699..cd05994a49e6 100644
--- a/Documentation/sysctl/vm.txt
+++ b/Documentation/sysctl/vm.txt
@@ -41,7 +41,8 @@ Currently, these files are in /proc/sys/vm:
 
 ==============================================================
 
-dirty_ratio, dirty_background_ratio, dirty_expire_centisecs,
+dirty_bytes, dirty_ratio, dirty_background_bytes,
+dirty_background_ratio, dirty_expire_centisecs,
 dirty_writeback_centisecs, highmem_is_dirtyable,
 vfs_cache_pressure, laptop_mode, block_dump, swap_token_timeout,
 drop-caches, hugepages_treat_as_movable:
diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index 259e9ea58cab..bb28c975c1d7 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -107,7 +107,9 @@ void throttle_vm_writeout(gfp_t gfp_mask);
 
 /* These are exported to sysctl. */
 extern int dirty_background_ratio;
+extern unsigned long dirty_background_bytes;
 extern int vm_dirty_ratio;
+extern unsigned long vm_dirty_bytes;
 extern int dirty_writeback_interval;
 extern int dirty_expire_interval;
 extern int vm_highmem_is_dirtyable;
@@ -116,9 +118,18 @@ extern int laptop_mode;
 
 extern unsigned long determine_dirtyable_memory(void);
 
+extern int dirty_background_ratio_handler(struct ctl_table *table, int write,
+		struct file *filp, void __user *buffer, size_t *lenp,
+		loff_t *ppos);
+extern int dirty_background_bytes_handler(struct ctl_table *table, int write,
+		struct file *filp, void __user *buffer, size_t *lenp,
+		loff_t *ppos);
 extern int dirty_ratio_handler(struct ctl_table *table, int write,
 		struct file *filp, void __user *buffer, size_t *lenp,
 		loff_t *ppos);
+extern int dirty_bytes_handler(struct ctl_table *table, int write,
+		struct file *filp, void __user *buffer, size_t *lenp,
+		loff_t *ppos);
 
 struct ctl_table;
 struct file;
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index ff6d45c7626f..92f6e5bc3c24 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -87,10 +87,6 @@ extern int rcutorture_runnable;
 #endif /* #ifdef CONFIG_RCU_TORTURE_TEST */
 
 /* Constants used for minimum and  maximum */
-#if defined(CONFIG_HIGHMEM) || defined(CONFIG_DETECT_SOFTLOCKUP)
-static int one = 1;
-#endif
-
 #ifdef CONFIG_DETECT_SOFTLOCKUP
 static int sixty = 60;
 static int neg_one = -1;
@@ -101,6 +97,7 @@ static int two = 2;
 #endif
 
 static int zero;
+static int one = 1;
 static int one_hundred = 100;
 
 /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */
@@ -952,11 +949,21 @@ static struct ctl_table vm_table[] = {
 		.data		= &dirty_background_ratio,
 		.maxlen		= sizeof(dirty_background_ratio),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_minmax,
+		.proc_handler	= &dirty_background_ratio_handler,
 		.strategy	= &sysctl_intvec,
 		.extra1		= &zero,
 		.extra2		= &one_hundred,
 	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "dirty_background_bytes",
+		.data		= &dirty_background_bytes,
+		.maxlen		= sizeof(dirty_background_bytes),
+		.mode		= 0644,
+		.proc_handler	= &dirty_background_bytes_handler,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &one,
+	},
 	{
 		.ctl_name	= VM_DIRTY_RATIO,
 		.procname	= "dirty_ratio",
@@ -968,6 +975,16 @@ static struct ctl_table vm_table[] = {
 		.extra1		= &zero,
 		.extra2		= &one_hundred,
 	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "dirty_bytes",
+		.data		= &vm_dirty_bytes,
+		.maxlen		= sizeof(vm_dirty_bytes),
+		.mode		= 0644,
+		.proc_handler	= &dirty_bytes_handler,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &one,
+	},
 	{
 		.procname	= "dirty_writeback_centisecs",
 		.data		= &dirty_writeback_interval,
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 4d4074cff300..b493db7841dc 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -68,6 +68,12 @@ static inline long sync_writeback_pages(void)
  */
 int dirty_background_ratio = 5;
 
+/*
+ * dirty_background_bytes starts at 0 (disabled) so that it is a function of
+ * dirty_background_ratio * the amount of dirtyable memory
+ */
+unsigned long dirty_background_bytes;
+
 /*
  * free highmem will not be subtracted from the total free memory
  * for calculating free ratios if vm_highmem_is_dirtyable is true
@@ -79,6 +85,12 @@ int vm_highmem_is_dirtyable;
  */
 int vm_dirty_ratio = 10;
 
+/*
+ * vm_dirty_bytes starts at 0 (disabled) so that it is a function of
+ * vm_dirty_ratio * the amount of dirtyable memory
+ */
+unsigned long vm_dirty_bytes;
+
 /*
  * The interval between `kupdate'-style writebacks, in jiffies
  */
@@ -135,23 +147,75 @@ static int calc_period_shift(void)
 {
 	unsigned long dirty_total;
 
-	dirty_total = (vm_dirty_ratio * determine_dirtyable_memory()) / 100;
+	if (vm_dirty_bytes)
+		dirty_total = vm_dirty_bytes / PAGE_SIZE;
+	else
+		dirty_total = (vm_dirty_ratio * determine_dirtyable_memory()) /
+				100;
 	return 2 + ilog2(dirty_total - 1);
 }
 
 /*
- * update the period when the dirty ratio changes.
+ * update the period when the dirty threshold changes.
  */
+static void update_completion_period(void)
+{
+	int shift = calc_period_shift();
+	prop_change_shift(&vm_completions, shift);
+	prop_change_shift(&vm_dirties, shift);
+}
+
+int dirty_background_ratio_handler(struct ctl_table *table, int write,
+		struct file *filp, void __user *buffer, size_t *lenp,
+		loff_t *ppos)
+{
+	int ret;
+
+	ret = proc_dointvec_minmax(table, write, filp, buffer, lenp, ppos);
+	if (ret == 0 && write)
+		dirty_background_bytes = 0;
+	return ret;
+}
+
+int dirty_background_bytes_handler(struct ctl_table *table, int write,
+		struct file *filp, void __user *buffer, size_t *lenp,
+		loff_t *ppos)
+{
+	int ret;
+
+	ret = proc_doulongvec_minmax(table, write, filp, buffer, lenp, ppos);
+	if (ret == 0 && write)
+		dirty_background_ratio = 0;
+	return ret;
+}
+
 int dirty_ratio_handler(struct ctl_table *table, int write,
 		struct file *filp, void __user *buffer, size_t *lenp,
 		loff_t *ppos)
 {
 	int old_ratio = vm_dirty_ratio;
-	int ret = proc_dointvec_minmax(table, write, filp, buffer, lenp, ppos);
+	int ret;
+
+	ret = proc_dointvec_minmax(table, write, filp, buffer, lenp, ppos);
 	if (ret == 0 && write && vm_dirty_ratio != old_ratio) {
-		int shift = calc_period_shift();
-		prop_change_shift(&vm_completions, shift);
-		prop_change_shift(&vm_dirties, shift);
+		update_completion_period();
+		vm_dirty_bytes = 0;
+	}
+	return ret;
+}
+
+
+int dirty_bytes_handler(struct ctl_table *table, int write,
+		struct file *filp, void __user *buffer, size_t *lenp,
+		loff_t *ppos)
+{
+	int old_bytes = vm_dirty_bytes;
+	int ret;
+
+	ret = proc_doulongvec_minmax(table, write, filp, buffer, lenp, ppos);
+	if (ret == 0 && write && vm_dirty_bytes != old_bytes) {
+		update_completion_period();
+		vm_dirty_ratio = 0;
 	}
 	return ret;
 }
@@ -365,23 +429,29 @@ void
 get_dirty_limits(unsigned long *pbackground, unsigned long *pdirty,
 		 unsigned long *pbdi_dirty, struct backing_dev_info *bdi)
 {
-	int background_ratio;		/* Percentages */
-	int dirty_ratio;
 	unsigned long background;
 	unsigned long dirty;
 	unsigned long available_memory = determine_dirtyable_memory();
 	struct task_struct *tsk;
 
-	dirty_ratio = vm_dirty_ratio;
-	if (dirty_ratio < 5)
-		dirty_ratio = 5;
+	if (vm_dirty_bytes)
+		dirty = DIV_ROUND_UP(vm_dirty_bytes, PAGE_SIZE);
+	else {
+		int dirty_ratio;
 
-	background_ratio = dirty_background_ratio;
-	if (background_ratio >= dirty_ratio)
-		background_ratio = dirty_ratio / 2;
+		dirty_ratio = vm_dirty_ratio;
+		if (dirty_ratio < 5)
+			dirty_ratio = 5;
+		dirty = (dirty_ratio * available_memory) / 100;
+	}
+
+	if (dirty_background_bytes)
+		background = DIV_ROUND_UP(dirty_background_bytes, PAGE_SIZE);
+	else
+		background = (dirty_background_ratio * available_memory) / 100;
 
-	background = (background_ratio * available_memory) / 100;
-	dirty = (dirty_ratio * available_memory) / 100;
+	if (background >= dirty)
+		background = dirty / 2;
 	tsk = current;
 	if (tsk->flags & PF_LESS_THROTTLE || rt_task(tsk)) {
 		background += background / 4;
-- 
cgit v1.2.3


From 7b1fe59793e61f826bef053107b57b23954833bb Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:34 -0800
Subject: mm: reuse_swap_page replaces can_share_swap_page

A good place to free up old swap is where do_wp_page(), or do_swap_page(),
is about to redirty the page: the data on disk is then stale and won't be
read again; and if we do decide to write the page out later, using the
previous swap location makes an unnecessary disk seek very likely.

So give can_share_swap_page() the side-effect of delete_from_swap_cache()
when it safely can.  And can_share_swap_page() was always a misleading
name, the more so if it has a side-effect: rename it reuse_swap_page().

Irrelevant cleanup nearby: remove swap_token_default_timeout definition
from swap.h: it's used nowhere.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Robin Holt <holt@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h |  6 ++----
 mm/memory.c          |  4 ++--
 mm/swapfile.c        | 15 +++++++++++----
 3 files changed, 15 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 48f309dc5a0c..366556c5b148 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -304,7 +304,7 @@ extern unsigned int count_swap_pages(int, int);
 extern sector_t map_swap_page(struct swap_info_struct *, pgoff_t);
 extern sector_t swapdev_block(int, pgoff_t);
 extern struct swap_info_struct *get_swap_info_struct(unsigned);
-extern int can_share_swap_page(struct page *);
+extern int reuse_swap_page(struct page *);
 extern int remove_exclusive_swap_page(struct page *);
 extern int remove_exclusive_swap_page_ref(struct page *);
 struct backing_dev_info;
@@ -372,8 +372,6 @@ static inline struct page *lookup_swap_cache(swp_entry_t swp)
 	return NULL;
 }
 
-#define can_share_swap_page(p)			(page_mapcount(p) == 1)
-
 static inline int add_to_swap_cache(struct page *page, swp_entry_t entry,
 							gfp_t gfp_mask)
 {
@@ -388,7 +386,7 @@ static inline void delete_from_swap_cache(struct page *page)
 {
 }
 
-#define swap_token_default_timeout		0
+#define reuse_swap_page(page)	(page_mapcount(page) == 1)
 
 static inline int remove_exclusive_swap_page(struct page *p)
 {
diff --git a/mm/memory.c b/mm/memory.c
index 3922ffcf3dff..8f471edcb985 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1861,7 +1861,7 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
 			}
 			page_cache_release(old_page);
 		}
-		reuse = can_share_swap_page(old_page);
+		reuse = reuse_swap_page(old_page);
 		unlock_page(old_page);
 	} else if (unlikely((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
 					(VM_WRITE|VM_SHARED))) {
@@ -2392,7 +2392,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 
 	inc_mm_counter(mm, anon_rss);
 	pte = mk_pte(page, vma->vm_page_prot);
-	if (write_access && can_share_swap_page(page)) {
+	if (write_access && reuse_swap_page(page)) {
 		pte = maybe_mkwrite(pte_mkdirty(pte), vma);
 		write_access = 0;
 	}
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 214e90b94946..bfd4ee59cb88 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -326,17 +326,24 @@ static inline int page_swapcount(struct page *page)
 }
 
 /*
- * We can use this swap cache entry directly
- * if there are no other references to it.
+ * We can write to an anon page without COW if there are no other references
+ * to it.  And as a side-effect, free up its swap: because the old content
+ * on disk will never be read, and seeking back there to write new content
+ * later would only waste time away from clustering.
  */
-int can_share_swap_page(struct page *page)
+int reuse_swap_page(struct page *page)
 {
 	int count;
 
 	VM_BUG_ON(!PageLocked(page));
 	count = page_mapcount(page);
-	if (count <= 1 && PageSwapCache(page))
+	if (count <= 1 && PageSwapCache(page)) {
 		count += page_swapcount(page);
+		if (count == 1 && !PageWriteback(page)) {
+			delete_from_swap_cache(page);
+			SetPageDirty(page);
+		}
+	}
 	return count == 1;
 }
 
-- 
cgit v1.2.3


From a2c43eed8334e878702fca713b212ae2a11d84b9 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:36 -0800
Subject: mm: try_to_free_swap replaces remove_exclusive_swap_page

remove_exclusive_swap_page(): its problem is in living up to its name.

It doesn't matter if someone else has a reference to the page (raised
page_count); it doesn't matter if the page is mapped into userspace
(raised page_mapcount - though that hints it may be worth keeping the
swap): all that matters is that there be no more references to the swap
(and no writeback in progress).

swapoff (try_to_unuse) has been removing pages from swapcache for years,
with no concern for page count or page mapcount, and we used to have a
comment in lookup_swap_cache() recognizing that: if you go for a page of
swapcache, you'll get the right page, but it could have been removed from
swapcache by the time you get page lock.

So, give up asking for exclusivity: get rid of
remove_exclusive_swap_page(), and remove_exclusive_swap_page_ref() and
remove_exclusive_swap_page_count() which were spawned for the recent LRU
work: replace them by the simpler try_to_free_swap() which just checks
page_swapcount().

Similarly, remove the page_count limitation from free_swap_and_count(),
but assume that it's worth holding on to the swap if page is mapped and
swap nowhere near full.  Add a vm_swap_full() test in free_swap_cache()?
It would be consistent, but I think we probably have enough for now.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Robin Holt <holt@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h | 10 ++------
 mm/memory.c          |  2 +-
 mm/page_io.c         |  2 +-
 mm/swap.c            |  3 +--
 mm/swap_state.c      |  8 +++---
 mm/swapfile.c        | 70 +++++++++-------------------------------------------
 mm/vmscan.c          |  2 +-
 7 files changed, 22 insertions(+), 75 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 366556c5b148..c3ecd478840e 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -305,8 +305,7 @@ extern sector_t map_swap_page(struct swap_info_struct *, pgoff_t);
 extern sector_t swapdev_block(int, pgoff_t);
 extern struct swap_info_struct *get_swap_info_struct(unsigned);
 extern int reuse_swap_page(struct page *);
-extern int remove_exclusive_swap_page(struct page *);
-extern int remove_exclusive_swap_page_ref(struct page *);
+extern int try_to_free_swap(struct page *);
 struct backing_dev_info;
 
 /* linux/mm/thrash.c */
@@ -388,12 +387,7 @@ static inline void delete_from_swap_cache(struct page *page)
 
 #define reuse_swap_page(page)	(page_mapcount(page) == 1)
 
-static inline int remove_exclusive_swap_page(struct page *p)
-{
-	return 0;
-}
-
-static inline int remove_exclusive_swap_page_ref(struct page *page)
+static inline int try_to_free_swap(struct page *page)
 {
 	return 0;
 }
diff --git a/mm/memory.c b/mm/memory.c
index 8f471edcb985..1a83fe5339a9 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2403,7 +2403,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 
 	swap_free(entry);
 	if (vm_swap_full() || (vma->vm_flags & VM_LOCKED) || PageMlocked(page))
-		remove_exclusive_swap_page(page);
+		try_to_free_swap(page);
 	unlock_page(page);
 
 	if (write_access) {
diff --git a/mm/page_io.c b/mm/page_io.c
index d277a80efa71..dc6ce0afbded 100644
--- a/mm/page_io.c
+++ b/mm/page_io.c
@@ -98,7 +98,7 @@ int swap_writepage(struct page *page, struct writeback_control *wbc)
 	struct bio *bio;
 	int ret = 0, rw = WRITE;
 
-	if (remove_exclusive_swap_page(page)) {
+	if (try_to_free_swap(page)) {
 		unlock_page(page);
 		goto out;
 	}
diff --git a/mm/swap.c b/mm/swap.c
index ff0b290475fd..ba2c0e8b8b54 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -454,8 +454,7 @@ void pagevec_swap_free(struct pagevec *pvec)
 		struct page *page = pvec->pages[i];
 
 		if (PageSwapCache(page) && trylock_page(page)) {
-			if (PageSwapCache(page))
-				remove_exclusive_swap_page_ref(page);
+			try_to_free_swap(page);
 			unlock_page(page);
 		}
 	}
diff --git a/mm/swap_state.c b/mm/swap_state.c
index e793fdea275d..bcb472769299 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -195,14 +195,14 @@ void delete_from_swap_cache(struct page *page)
  * If we are the only user, then try to free up the swap cache. 
  * 
  * Its ok to check for PageSwapCache without the page lock
- * here because we are going to recheck again inside 
- * exclusive_swap_page() _with_ the lock. 
+ * here because we are going to recheck again inside
+ * try_to_free_swap() _with_ the lock.
  * 					- Marcelo
  */
 static inline void free_swap_cache(struct page *page)
 {
-	if (PageSwapCache(page) && trylock_page(page)) {
-		remove_exclusive_swap_page(page);
+	if (PageSwapCache(page) && !page_mapped(page) && trylock_page(page)) {
+		try_to_free_swap(page);
 		unlock_page(page);
 	}
 }
diff --git a/mm/swapfile.c b/mm/swapfile.c
index bfd4ee59cb88..f43601827607 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -348,68 +348,23 @@ int reuse_swap_page(struct page *page)
 }
 
 /*
- * Work out if there are any other processes sharing this
- * swap cache page. Free it if you can. Return success.
+ * If swap is getting full, or if there are no more mappings of this page,
+ * then try_to_free_swap is called to free its swap space.
  */
-static int remove_exclusive_swap_page_count(struct page *page, int count)
+int try_to_free_swap(struct page *page)
 {
-	int retval;
-	struct swap_info_struct * p;
-	swp_entry_t entry;
-
 	VM_BUG_ON(!PageLocked(page));
 
 	if (!PageSwapCache(page))
 		return 0;
 	if (PageWriteback(page))
 		return 0;
-	if (page_count(page) != count) /* us + cache + ptes */
-		return 0;
-
-	entry.val = page_private(page);
-	p = swap_info_get(entry);
-	if (!p)
+	if (page_swapcount(page))
 		return 0;
 
-	/* Is the only swap cache user the cache itself? */
-	retval = 0;
-	if (p->swap_map[swp_offset(entry)] == 1) {
-		/* Recheck the page count with the swapcache lock held.. */
-		spin_lock_irq(&swapper_space.tree_lock);
-		if ((page_count(page) == count) && !PageWriteback(page)) {
-			__delete_from_swap_cache(page);
-			SetPageDirty(page);
-			retval = 1;
-		}
-		spin_unlock_irq(&swapper_space.tree_lock);
-	}
-	spin_unlock(&swap_lock);
-
-	if (retval) {
-		swap_free(entry);
-		page_cache_release(page);
-	}
-
-	return retval;
-}
-
-/*
- * Most of the time the page should have two references: one for the
- * process and one for the swap cache.
- */
-int remove_exclusive_swap_page(struct page *page)
-{
-	return remove_exclusive_swap_page_count(page, 2);
-}
-
-/*
- * The pageout code holds an extra reference to the page.  That raises
- * the reference count to test for to 2 for a page that is only in the
- * swap cache plus 1 for each process that maps the page.
- */
-int remove_exclusive_swap_page_ref(struct page *page)
-{
-	return remove_exclusive_swap_page_count(page, 2 + page_mapcount(page));
+	delete_from_swap_cache(page);
+	SetPageDirty(page);
+	return 1;
 }
 
 /*
@@ -436,13 +391,12 @@ void free_swap_and_cache(swp_entry_t entry)
 		spin_unlock(&swap_lock);
 	}
 	if (page) {
-		int one_user;
-
-		one_user = (page_count(page) == 2);
-		/* Only cache user (+us), or swap space full? Free it! */
-		/* Also recheck PageSwapCache after page is locked (above) */
+		/*
+		 * Not mapped elsewhere, or swap space full? Free it!
+		 * Also recheck PageSwapCache now page is locked (above).
+		 */
 		if (PageSwapCache(page) && !PageWriteback(page) &&
-					(one_user || vm_swap_full())) {
+				(!page_mapped(page) || vm_swap_full())) {
 			delete_from_swap_cache(page);
 			SetPageDirty(page);
 		}
diff --git a/mm/vmscan.c b/mm/vmscan.c
index d196f46c8808..c8601dd36603 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -759,7 +759,7 @@ cull_mlocked:
 activate_locked:
 		/* Not a candidate for swapping, so reclaim swap space. */
 		if (PageSwapCache(page) && vm_swap_full())
-			remove_exclusive_swap_page_ref(page);
+			try_to_free_swap(page);
 		VM_BUG_ON(PageActive(page));
 		SetPageActive(page);
 		pgactivate++;
-- 
cgit v1.2.3


From ac47b003d03c2a4f28aef1d505b66d24ad191c4f Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:39 -0800
Subject: mm: remove gfp_mask from add_to_swap

Remove gfp_mask argument from add_to_swap(): it's misleading because its
only caller, shrink_page_list(), is not atomic at that point; and in due
course (implementing discard) we'll sometimes want to allocate some memory
with GFP_NOIO (as is used in swap_writepage) when allocating swap.

No change to the gfp_mask passed down to add_to_swap_cache(): still use
__GFP_HIGH without __GFP_WAIT (with nomemalloc and nowarn as before):
though it's not obvious if that's the best combination to ask for here.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Robin Holt <holt@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h | 2 +-
 mm/swap_state.c      | 4 ++--
 mm/vmscan.c          | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index c3ecd478840e..c38bd157695b 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -278,7 +278,7 @@ extern void end_swap_bio_read(struct bio *bio, int err);
 extern struct address_space swapper_space;
 #define total_swapcache_pages  swapper_space.nrpages
 extern void show_swap_cache_info(void);
-extern int add_to_swap(struct page *, gfp_t);
+extern int add_to_swap(struct page *);
 extern int add_to_swap_cache(struct page *, swp_entry_t, gfp_t);
 extern void __delete_from_swap_cache(struct page *);
 extern void delete_from_swap_cache(struct page *);
diff --git a/mm/swap_state.c b/mm/swap_state.c
index bcb472769299..81c825f67a7f 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -128,7 +128,7 @@ void __delete_from_swap_cache(struct page *page)
  * Allocate swap space for the page and add the page to the
  * swap cache.  Caller needs to hold the page lock. 
  */
-int add_to_swap(struct page * page, gfp_t gfp_mask)
+int add_to_swap(struct page *page)
 {
 	swp_entry_t entry;
 	int err;
@@ -153,7 +153,7 @@ int add_to_swap(struct page * page, gfp_t gfp_mask)
 		 * Add it to the swap cache and mark it dirty
 		 */
 		err = add_to_swap_cache(page, entry,
-				gfp_mask|__GFP_NOMEMALLOC|__GFP_NOWARN);
+				__GFP_HIGH|__GFP_NOMEMALLOC|__GFP_NOWARN);
 
 		switch (err) {
 		case 0:				/* Success */
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 74f875733e2b..cc7401571cba 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -625,7 +625,7 @@ static unsigned long shrink_page_list(struct list_head *page_list,
 		if (PageAnon(page) && !PageSwapCache(page)) {
 			if (!(sc->gfp_mask & __GFP_IO))
 				goto keep_locked;
-			if (!add_to_swap(page, GFP_ATOMIC))
+			if (!add_to_swap(page))
 				goto activate_locked;
 			may_enter_fs = 1;
 		}
-- 
cgit v1.2.3


From 60371d971a3d01afd102f0bbf2681f32ecc31d78 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:40 -0800
Subject: mm: add add_to_swap stub

If we add a failing stub for add_to_swap(), then we can remove the #ifdef
CONFIG_SWAP from mm/vmscan.c.

This was intended as a source cleanup, but looking more closely, it turns
out that the !CONFIG_SWAP case was going to keep_locked for an anonymous
page, whereas now it goes to the more suitable activate_locked, like the
CONFIG_SWAP nr_swap_pages 0 case.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Robin Holt <holt@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h | 5 +++++
 mm/vmscan.c          | 2 --
 2 files changed, 5 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index c38bd157695b..c0d23ac710d5 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -371,6 +371,11 @@ static inline struct page *lookup_swap_cache(swp_entry_t swp)
 	return NULL;
 }
 
+static inline int add_to_swap(struct page *page)
+{
+	return 0;
+}
+
 static inline int add_to_swap_cache(struct page *page, swp_entry_t entry,
 							gfp_t gfp_mask)
 {
diff --git a/mm/vmscan.c b/mm/vmscan.c
index cc7401571cba..f350523a8eee 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -617,7 +617,6 @@ static unsigned long shrink_page_list(struct list_head *page_list,
 					referenced && page_mapping_inuse(page))
 			goto activate_locked;
 
-#ifdef CONFIG_SWAP
 		/*
 		 * Anonymous process memory has backing store?
 		 * Try to allocate it some swap space here.
@@ -629,7 +628,6 @@ static unsigned long shrink_page_list(struct list_head *page_list,
 				goto activate_locked;
 			may_enter_fs = 1;
 		}
-#endif /* CONFIG_SWAP */
 
 		mapping = page_mapping(page);
 
-- 
cgit v1.2.3


From b962716b459505a8d83aea313fea0abe76749f42 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:41 -0800
Subject: mm: optimize get_scan_ratio for no swap

Rik suggests a simplified get_scan_ratio() for !CONFIG_SWAP.  Yes, the gcc
optimizer gives us that, when nr_swap_pages is #defined as 0L.  Move usual
declaration to swapfile.c: it never belonged in page_alloc.c.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Robin Holt <holt@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h |  5 +++--
 mm/page_alloc.c      |  1 -
 mm/swapfile.c        |  1 +
 mm/vmscan.c          | 12 ++++++------
 4 files changed, 10 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index c0d23ac710d5..3a31cc25bd2c 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -163,7 +163,6 @@ struct swap_list_t {
 /* linux/mm/page_alloc.c */
 extern unsigned long totalram_pages;
 extern unsigned long totalreserve_pages;
-extern long nr_swap_pages;
 extern unsigned int nr_free_buffer_pages(void);
 extern unsigned int nr_free_pagecache_pages(void);
 
@@ -291,6 +290,7 @@ extern struct page *swapin_readahead(swp_entry_t, gfp_t,
 			struct vm_area_struct *vma, unsigned long addr);
 
 /* linux/mm/swapfile.c */
+extern long nr_swap_pages;
 extern long total_swap_pages;
 extern void si_swapinfo(struct sysinfo *);
 extern swp_entry_t get_swap_page(void);
@@ -331,7 +331,8 @@ static inline void disable_swap_token(void)
 
 #else /* CONFIG_SWAP */
 
-#define total_swap_pages			0
+#define nr_swap_pages				0L
+#define total_swap_pages			0L
 #define total_swapcache_pages			0UL
 
 #define si_swapinfo(val) \
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index ce2a026219bc..65133436fe3d 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -69,7 +69,6 @@ EXPORT_SYMBOL(node_states);
 
 unsigned long totalram_pages __read_mostly;
 unsigned long totalreserve_pages __read_mostly;
-long nr_swap_pages;
 int percpu_pagelist_fraction;
 
 #ifdef CONFIG_HUGETLB_PAGE_SIZE_VARIABLE
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 9ce7f81c8abc..725e56c362de 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -35,6 +35,7 @@
 
 static DEFINE_SPINLOCK(swap_lock);
 static unsigned int nr_swapfiles;
+long nr_swap_pages;
 long total_swap_pages;
 static int swap_overflow;
 static int least_priority;
diff --git a/mm/vmscan.c b/mm/vmscan.c
index f350523a8eee..d500b906746c 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1327,12 +1327,6 @@ static void get_scan_ratio(struct zone *zone, struct scan_control *sc,
 	unsigned long anon_prio, file_prio;
 	unsigned long ap, fp;
 
-	anon  = zone_page_state(zone, NR_ACTIVE_ANON) +
-		zone_page_state(zone, NR_INACTIVE_ANON);
-	file  = zone_page_state(zone, NR_ACTIVE_FILE) +
-		zone_page_state(zone, NR_INACTIVE_FILE);
-	free  = zone_page_state(zone, NR_FREE_PAGES);
-
 	/* If we have no swap space, do not bother scanning anon pages. */
 	if (nr_swap_pages <= 0) {
 		percent[0] = 0;
@@ -1340,6 +1334,12 @@ static void get_scan_ratio(struct zone *zone, struct scan_control *sc,
 		return;
 	}
 
+	anon  = zone_page_state(zone, NR_ACTIVE_ANON) +
+		zone_page_state(zone, NR_INACTIVE_ANON);
+	file  = zone_page_state(zone, NR_ACTIVE_FILE) +
+		zone_page_state(zone, NR_INACTIVE_FILE);
+	free  = zone_page_state(zone, NR_FREE_PAGES);
+
 	/* If we have very few page cache pages, force-scan anon pages. */
 	if (unlikely(file + free <= zone->pages_high)) {
 		percent[0] = 100;
-- 
cgit v1.2.3


From 69beeb1d3428424fbc7546f85e5cd7ac4119c09d Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Tue, 6 Jan 2009 14:39:46 -0800
Subject: mm: make vread() and vwrite() declaration

Sparse output following warnings.

mm/vmalloc.c:1436:6: warning: symbol 'vread' was not declared. Should it be static?
mm/vmalloc.c:1474:6: warning: symbol 'vwrite' was not declared. Should it be static?

However, it is used by /dev/kmem. fixed here.

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/mem.c      | 3 ---
 include/linux/vmalloc.h | 4 ++++
 2 files changed, 4 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/char/mem.c b/drivers/char/mem.c
index 6431f6921a67..3586b3b3df3f 100644
--- a/drivers/char/mem.c
+++ b/drivers/char/mem.c
@@ -425,9 +425,6 @@ static ssize_t read_oldmem(struct file *file, char __user *buf,
 }
 #endif
 
-extern long vread(char *buf, char *addr, unsigned long count);
-extern long vwrite(char *buf, char *addr, unsigned long count);
-
 #ifdef CONFIG_DEVKMEM
 /*
  * This function reads the *virtual* memory as seen by the kernel.
diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h
index 307b88577eaa..506e7620a986 100644
--- a/include/linux/vmalloc.h
+++ b/include/linux/vmalloc.h
@@ -97,6 +97,10 @@ extern void unmap_kernel_range(unsigned long addr, unsigned long size);
 extern struct vm_struct *alloc_vm_area(size_t size);
 extern void free_vm_area(struct vm_struct *area);
 
+/* for /dev/kmem */
+extern long vread(char *buf, char *addr, unsigned long count);
+extern long vwrite(char *buf, char *addr, unsigned long count);
+
 /*
  *	Internals.  Dont't use..
  */
-- 
cgit v1.2.3


From 22c6f8fdb31993cf49bdd4a47b64a7002391e1c7 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:48 -0800
Subject: swapfile: remove SWP_ACTIVE mask

Remove the SWP_ACTIVE mask: it just obscures the SWP_WRITEOK flag.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h | 1 -
 mm/swapfile.c        | 4 ++--
 2 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 3a31cc25bd2c..410c8e473727 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -120,7 +120,6 @@ struct swap_extent {
 enum {
 	SWP_USED	= (1 << 0),	/* is slot in swap_info[] used? */
 	SWP_WRITEOK	= (1 << 1),	/* ok to write to this swap?	*/
-	SWP_ACTIVE	= (SWP_USED | SWP_WRITEOK),
 					/* add others here before... */
 	SWP_SCANNING	= (1 << 8),	/* refcount in scan_swap_map */
 };
diff --git a/mm/swapfile.c b/mm/swapfile.c
index e2adc8eb9317..915cb3fc43d7 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -1222,7 +1222,7 @@ asmlinkage long sys_swapoff(const char __user * specialfile)
 	spin_lock(&swap_lock);
 	for (type = swap_list.head; type >= 0; type = swap_info[type].next) {
 		p = swap_info + type;
-		if ((p->flags & SWP_ACTIVE) == SWP_ACTIVE) {
+		if (p->flags & SWP_WRITEOK) {
 			if (p->swap_file->f_mapping == mapping)
 				break;
 		}
@@ -1674,7 +1674,7 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags)
 	else
 		p->prio = --least_priority;
 	p->swap_map = swap_map;
-	p->flags = SWP_ACTIVE;
+	p->flags |= SWP_WRITEOK;
 	nr_swap_pages += nr_good_pages;
 	total_swap_pages += nr_good_pages;
 
-- 
cgit v1.2.3


From ebebbbe904634b0ca1c674457b399f68db5e05b1 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:50 -0800
Subject: swapfile: rearrange scan and swap_info

Before making functional changes, rearrange scan_swap_map() to simplify
subsequent diffs.  Actually, there is one functional change in there:
leave cluster_nr negative while scanning for a new cluster - resetting it
early increased the likelihood that when we have difficulty finding a free
cluster, another task may come in and try doing exactly the same - just a
waste of cpu.

Before making functional changes, rearrange struct swap_info_struct
slightly: flags will be needed as an unsigned long (for wait_on_bit), next
is a good int to pair with prio, old_block_size is uninteresting so shift
it to the end.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h |  8 +++----
 mm/swapfile.c        | 66 +++++++++++++++++++++++++++++-----------------------
 2 files changed, 41 insertions(+), 33 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 410c8e473727..9cabb8b21aba 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -133,14 +133,14 @@ enum {
  * The in-memory structure used to track swap areas.
  */
 struct swap_info_struct {
-	unsigned int flags;
+	unsigned long flags;
 	int prio;			/* swap priority */
+	int next;			/* next entry on swap list */
 	struct file *swap_file;
 	struct block_device *bdev;
 	struct list_head extent_list;
 	struct swap_extent *curr_swap_extent;
-	unsigned old_block_size;
-	unsigned short * swap_map;
+	unsigned short *swap_map;
 	unsigned int lowest_bit;
 	unsigned int highest_bit;
 	unsigned int cluster_next;
@@ -148,7 +148,7 @@ struct swap_info_struct {
 	unsigned int pages;
 	unsigned int max;
 	unsigned int inuse_pages;
-	int next;			/* next entry on swap list */
+	unsigned int old_block_size;
 };
 
 struct swap_list_t {
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 85ff603385c3..4d9855f86e7d 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -89,7 +89,8 @@ void swap_unplug_io_fn(struct backing_dev_info *unused_bdi, struct page *page)
 
 static inline unsigned long scan_swap_map(struct swap_info_struct *si)
 {
-	unsigned long offset, last_in_cluster;
+	unsigned long offset;
+	unsigned long last_in_cluster;
 	int latency_ration = LATENCY_LIMIT;
 
 	/*
@@ -103,10 +104,13 @@ static inline unsigned long scan_swap_map(struct swap_info_struct *si)
 	 */
 
 	si->flags += SWP_SCANNING;
-	if (unlikely(!si->cluster_nr)) {
-		si->cluster_nr = SWAPFILE_CLUSTER - 1;
-		if (si->pages - si->inuse_pages < SWAPFILE_CLUSTER)
-			goto lowest;
+	offset = si->cluster_next;
+
+	if (unlikely(!si->cluster_nr--)) {
+		if (si->pages - si->inuse_pages < SWAPFILE_CLUSTER) {
+			si->cluster_nr = SWAPFILE_CLUSTER - 1;
+			goto checks;
+		}
 		spin_unlock(&swap_lock);
 
 		offset = si->lowest_bit;
@@ -118,43 +122,47 @@ static inline unsigned long scan_swap_map(struct swap_info_struct *si)
 				last_in_cluster = offset + SWAPFILE_CLUSTER;
 			else if (offset == last_in_cluster) {
 				spin_lock(&swap_lock);
-				si->cluster_next = offset-SWAPFILE_CLUSTER+1;
-				goto cluster;
+				offset -= SWAPFILE_CLUSTER - 1;
+				si->cluster_next = offset;
+				si->cluster_nr = SWAPFILE_CLUSTER - 1;
+				goto checks;
 			}
 			if (unlikely(--latency_ration < 0)) {
 				cond_resched();
 				latency_ration = LATENCY_LIMIT;
 			}
 		}
+
+		offset = si->lowest_bit;
 		spin_lock(&swap_lock);
-		goto lowest;
+		si->cluster_nr = SWAPFILE_CLUSTER - 1;
 	}
 
-	si->cluster_nr--;
-cluster:
-	offset = si->cluster_next;
-	if (offset > si->highest_bit)
-lowest:		offset = si->lowest_bit;
-checks:	if (!(si->flags & SWP_WRITEOK))
+checks:
+	if (!(si->flags & SWP_WRITEOK))
 		goto no_page;
 	if (!si->highest_bit)
 		goto no_page;
-	if (!si->swap_map[offset]) {
-		if (offset == si->lowest_bit)
-			si->lowest_bit++;
-		if (offset == si->highest_bit)
-			si->highest_bit--;
-		si->inuse_pages++;
-		if (si->inuse_pages == si->pages) {
-			si->lowest_bit = si->max;
-			si->highest_bit = 0;
-		}
-		si->swap_map[offset] = 1;
-		si->cluster_next = offset + 1;
-		si->flags -= SWP_SCANNING;
-		return offset;
+	if (offset > si->highest_bit)
+		offset = si->lowest_bit;
+	if (si->swap_map[offset])
+		goto scan;
+
+	if (offset == si->lowest_bit)
+		si->lowest_bit++;
+	if (offset == si->highest_bit)
+		si->highest_bit--;
+	si->inuse_pages++;
+	if (si->inuse_pages == si->pages) {
+		si->lowest_bit = si->max;
+		si->highest_bit = 0;
 	}
+	si->swap_map[offset] = 1;
+	si->cluster_next = offset + 1;
+	si->flags -= SWP_SCANNING;
+	return offset;
 
+scan:
 	spin_unlock(&swap_lock);
 	while (++offset <= si->highest_bit) {
 		if (!si->swap_map[offset]) {
@@ -167,7 +175,7 @@ checks:	if (!(si->flags & SWP_WRITEOK))
 		}
 	}
 	spin_lock(&swap_lock);
-	goto lowest;
+	goto checks;
 
 no_page:
 	si->flags -= SWP_SCANNING;
-- 
cgit v1.2.3


From 6a6ba83175c029c7820765bae44692266b29e67a Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:51 -0800
Subject: swapfile: swapon use discard (trim)

When adding swap, all the old data on swap can be forgotten: sys_swapon()
discard all but the header page of the swap partition (or every extent but
the header of the swap file), to give a solidstate swap device the
opportunity to optimize its wear-levelling.

If that succeeds, note SWP_DISCARDABLE for later use, and report it with a
"D" at the right end of the kernel's "Adding ...  swap" message.  Perhaps
something should be shown in /proc/swaps (swapon -s), but we have to be
more cautious before making any addition to that format.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Joern Engel <joern@logfs.org>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Donjun Shin <djshin90@gmail.com>
Cc: Tejun Heo <teheo@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h |  1 +
 mm/swapfile.c        | 39 +++++++++++++++++++++++++++++++++++++--
 2 files changed, 38 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 9cabb8b21aba..0b9210ea96c7 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -120,6 +120,7 @@ struct swap_extent {
 enum {
 	SWP_USED	= (1 << 0),	/* is slot in swap_info[] used? */
 	SWP_WRITEOK	= (1 << 1),	/* ok to write to this swap?	*/
+	SWP_DISCARDABLE = (1 << 2),	/* blkdev supports discard */
 					/* add others here before... */
 	SWP_SCANNING	= (1 << 8),	/* refcount in scan_swap_map */
 };
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 4d9855f86e7d..fbeb4bb8eb50 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -84,6 +84,37 @@ void swap_unplug_io_fn(struct backing_dev_info *unused_bdi, struct page *page)
 	up_read(&swap_unplug_sem);
 }
 
+/*
+ * swapon tell device that all the old swap contents can be discarded,
+ * to allow the swap device to optimize its wear-levelling.
+ */
+static int discard_swap(struct swap_info_struct *si)
+{
+	struct swap_extent *se;
+	int err = 0;
+
+	list_for_each_entry(se, &si->extent_list, list) {
+		sector_t start_block = se->start_block << (PAGE_SHIFT - 9);
+		pgoff_t nr_blocks = se->nr_pages << (PAGE_SHIFT - 9);
+
+		if (se->start_page == 0) {
+			/* Do not discard the swap header page! */
+			start_block += 1 << (PAGE_SHIFT - 9);
+			nr_blocks -= 1 << (PAGE_SHIFT - 9);
+			if (!nr_blocks)
+				continue;
+		}
+
+		err = blkdev_issue_discard(si->bdev, start_block,
+						nr_blocks, GFP_KERNEL);
+		if (err)
+			break;
+
+		cond_resched();
+	}
+	return err;		/* That will often be -EOPNOTSUPP */
+}
+
 #define SWAPFILE_CLUSTER	256
 #define LATENCY_LIMIT		256
 
@@ -1658,6 +1689,9 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags)
 		goto bad_swap;
 	}
 
+	if (discard_swap(p) == 0)
+		p->flags |= SWP_DISCARDABLE;
+
 	mutex_lock(&swapon_mutex);
 	spin_lock(&swap_lock);
 	if (swap_flags & SWAP_FLAG_PREFER)
@@ -1671,9 +1705,10 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags)
 	total_swap_pages += nr_good_pages;
 
 	printk(KERN_INFO "Adding %uk swap on %s.  "
-			"Priority:%d extents:%d across:%lluk\n",
+			"Priority:%d extents:%d across:%lluk%s\n",
 		nr_good_pages<<(PAGE_SHIFT-10), name, p->prio,
-		nr_extents, (unsigned long long)span<<(PAGE_SHIFT-10));
+		nr_extents, (unsigned long long)span<<(PAGE_SHIFT-10),
+		(p->flags & SWP_DISCARDABLE) ? " D" : "");
 
 	/* insert swap space into swap_list: */
 	prev = -1;
-- 
cgit v1.2.3


From 7992fde72ce06c73280a1939b7a1e903bc95ef85 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:53 -0800
Subject: swapfile: swap allocation use discard

When scan_swap_map() finds a free cluster of swap pages to allocate,
discard the old contents of the cluster if the device supports discard.
But don't bother when swap is so fragmented that we allocate single pages.

Be careful about racing allocations made while we're scanning for a
cluster; and hold up allocations made while we're discarding.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Joern Engel <joern@logfs.org>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Donjun Shin <djshin90@gmail.com>
Cc: Tejun Heo <teheo@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h |   3 ++
 mm/swapfile.c        | 119 ++++++++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 121 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 0b9210ea96c7..fe79f44c858e 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -121,6 +121,7 @@ enum {
 	SWP_USED	= (1 << 0),	/* is slot in swap_info[] used? */
 	SWP_WRITEOK	= (1 << 1),	/* ok to write to this swap?	*/
 	SWP_DISCARDABLE = (1 << 2),	/* blkdev supports discard */
+	SWP_DISCARDING	= (1 << 3),	/* now discarding a free cluster */
 					/* add others here before... */
 	SWP_SCANNING	= (1 << 8),	/* refcount in scan_swap_map */
 };
@@ -144,6 +145,8 @@ struct swap_info_struct {
 	unsigned short *swap_map;
 	unsigned int lowest_bit;
 	unsigned int highest_bit;
+	unsigned int lowest_alloc;	/* while preparing discard cluster */
+	unsigned int highest_alloc;	/* while preparing discard cluster */
 	unsigned int cluster_next;
 	unsigned int cluster_nr;
 	unsigned int pages;
diff --git a/mm/swapfile.c b/mm/swapfile.c
index fbeb4bb8eb50..ca75b9e7c09f 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -115,14 +115,62 @@ static int discard_swap(struct swap_info_struct *si)
 	return err;		/* That will often be -EOPNOTSUPP */
 }
 
+/*
+ * swap allocation tell device that a cluster of swap can now be discarded,
+ * to allow the swap device to optimize its wear-levelling.
+ */
+static void discard_swap_cluster(struct swap_info_struct *si,
+				 pgoff_t start_page, pgoff_t nr_pages)
+{
+	struct swap_extent *se = si->curr_swap_extent;
+	int found_extent = 0;
+
+	while (nr_pages) {
+		struct list_head *lh;
+
+		if (se->start_page <= start_page &&
+		    start_page < se->start_page + se->nr_pages) {
+			pgoff_t offset = start_page - se->start_page;
+			sector_t start_block = se->start_block + offset;
+			pgoff_t nr_blocks = se->nr_pages - offset;
+
+			if (nr_blocks > nr_pages)
+				nr_blocks = nr_pages;
+			start_page += nr_blocks;
+			nr_pages -= nr_blocks;
+
+			if (!found_extent++)
+				si->curr_swap_extent = se;
+
+			start_block <<= PAGE_SHIFT - 9;
+			nr_blocks <<= PAGE_SHIFT - 9;
+			if (blkdev_issue_discard(si->bdev, start_block,
+							nr_blocks, GFP_NOIO))
+				break;
+		}
+
+		lh = se->list.next;
+		if (lh == &si->extent_list)
+			lh = lh->next;
+		se = list_entry(lh, struct swap_extent, list);
+	}
+}
+
+static int wait_for_discard(void *word)
+{
+	schedule();
+	return 0;
+}
+
 #define SWAPFILE_CLUSTER	256
 #define LATENCY_LIMIT		256
 
 static inline unsigned long scan_swap_map(struct swap_info_struct *si)
 {
 	unsigned long offset;
-	unsigned long last_in_cluster;
+	unsigned long last_in_cluster = 0;
 	int latency_ration = LATENCY_LIMIT;
+	int found_free_cluster = 0;
 
 	/*
 	 * We try to cluster swap pages by allocating them sequentially
@@ -142,6 +190,19 @@ static inline unsigned long scan_swap_map(struct swap_info_struct *si)
 			si->cluster_nr = SWAPFILE_CLUSTER - 1;
 			goto checks;
 		}
+		if (si->flags & SWP_DISCARDABLE) {
+			/*
+			 * Start range check on racing allocations, in case
+			 * they overlap the cluster we eventually decide on
+			 * (we scan without swap_lock to allow preemption).
+			 * It's hardly conceivable that cluster_nr could be
+			 * wrapped during our scan, but don't depend on it.
+			 */
+			if (si->lowest_alloc)
+				goto checks;
+			si->lowest_alloc = si->max;
+			si->highest_alloc = 0;
+		}
 		spin_unlock(&swap_lock);
 
 		offset = si->lowest_bit;
@@ -156,6 +217,7 @@ static inline unsigned long scan_swap_map(struct swap_info_struct *si)
 				offset -= SWAPFILE_CLUSTER - 1;
 				si->cluster_next = offset;
 				si->cluster_nr = SWAPFILE_CLUSTER - 1;
+				found_free_cluster = 1;
 				goto checks;
 			}
 			if (unlikely(--latency_ration < 0)) {
@@ -167,6 +229,7 @@ static inline unsigned long scan_swap_map(struct swap_info_struct *si)
 		offset = si->lowest_bit;
 		spin_lock(&swap_lock);
 		si->cluster_nr = SWAPFILE_CLUSTER - 1;
+		si->lowest_alloc = 0;
 	}
 
 checks:
@@ -191,6 +254,60 @@ checks:
 	si->swap_map[offset] = 1;
 	si->cluster_next = offset + 1;
 	si->flags -= SWP_SCANNING;
+
+	if (si->lowest_alloc) {
+		/*
+		 * Only set when SWP_DISCARDABLE, and there's a scan
+		 * for a free cluster in progress or just completed.
+		 */
+		if (found_free_cluster) {
+			/*
+			 * To optimize wear-levelling, discard the
+			 * old data of the cluster, taking care not to
+			 * discard any of its pages that have already
+			 * been allocated by racing tasks (offset has
+			 * already stepped over any at the beginning).
+			 */
+			if (offset < si->highest_alloc &&
+			    si->lowest_alloc <= last_in_cluster)
+				last_in_cluster = si->lowest_alloc - 1;
+			si->flags |= SWP_DISCARDING;
+			spin_unlock(&swap_lock);
+
+			if (offset < last_in_cluster)
+				discard_swap_cluster(si, offset,
+					last_in_cluster - offset + 1);
+
+			spin_lock(&swap_lock);
+			si->lowest_alloc = 0;
+			si->flags &= ~SWP_DISCARDING;
+
+			smp_mb();	/* wake_up_bit advises this */
+			wake_up_bit(&si->flags, ilog2(SWP_DISCARDING));
+
+		} else if (si->flags & SWP_DISCARDING) {
+			/*
+			 * Delay using pages allocated by racing tasks
+			 * until the whole discard has been issued. We
+			 * could defer that delay until swap_writepage,
+			 * but it's easier to keep this self-contained.
+			 */
+			spin_unlock(&swap_lock);
+			wait_on_bit(&si->flags, ilog2(SWP_DISCARDING),
+				wait_for_discard, TASK_UNINTERRUPTIBLE);
+			spin_lock(&swap_lock);
+		} else {
+			/*
+			 * Note pages allocated by racing tasks while
+			 * scan for a free cluster is in progress, so
+			 * that its final discard can exclude them.
+			 */
+			if (offset < si->lowest_alloc)
+				si->lowest_alloc = offset;
+			if (offset > si->highest_alloc)
+				si->highest_alloc = offset;
+		}
+	}
 	return offset;
 
 scan:
-- 
cgit v1.2.3


From 20137a490f397d9c01fc9fadd83a8d198bda4477 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:39:54 -0800
Subject: swapfile: swapon randomize if nonrot

Swap allocation has always started from the beginning of the swap area;
but if we're dealing with a solidstate swap device which can only remap
blocks within limited zones, that would sooner wear out the first zone.

Therefore sys_swapon() test whether blk_queue is non-rotational, and if so
randomize the cluster_next starting position for allocation.

If blk_queue is nonrot, note SWP_SOLIDSTATE for later use, and report it
with an "SS" at the right end of the kernel's "Adding ...  swap" message
(so that if it's both nonrot and discardable, "SSD" will be shown there).
Perhaps something should be shown in /proc/swaps (swapon -s), but we have
to be more cautious before making any addition to that format.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Joern Engel <joern@logfs.org>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Donjun Shin <djshin90@gmail.com>
Cc: Tejun Heo <teheo@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h |  1 +
 mm/swapfile.c        | 11 +++++++++--
 2 files changed, 10 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index fe79f44c858e..cbf7fbed3dfd 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -122,6 +122,7 @@ enum {
 	SWP_WRITEOK	= (1 << 1),	/* ok to write to this swap?	*/
 	SWP_DISCARDABLE = (1 << 2),	/* blkdev supports discard */
 	SWP_DISCARDING	= (1 << 3),	/* now discarding a free cluster */
+	SWP_SOLIDSTATE	= (1 << 4),	/* blkdev seeks are cheap */
 					/* add others here before... */
 	SWP_SCANNING	= (1 << 8),	/* refcount in scan_swap_map */
 };
diff --git a/mm/swapfile.c b/mm/swapfile.c
index ca75b9e7c09f..b0f56603b9be 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -16,6 +16,7 @@
 #include <linux/namei.h>
 #include <linux/shm.h>
 #include <linux/blkdev.h>
+#include <linux/random.h>
 #include <linux/writeback.h>
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
@@ -1806,6 +1807,11 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags)
 		goto bad_swap;
 	}
 
+	if (blk_queue_nonrot(bdev_get_queue(p->bdev))) {
+		p->flags |= SWP_SOLIDSTATE;
+		srandom32((u32)get_seconds());
+		p->cluster_next = 1 + (random32() % p->highest_bit);
+	}
 	if (discard_swap(p) == 0)
 		p->flags |= SWP_DISCARDABLE;
 
@@ -1822,10 +1828,11 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags)
 	total_swap_pages += nr_good_pages;
 
 	printk(KERN_INFO "Adding %uk swap on %s.  "
-			"Priority:%d extents:%d across:%lluk%s\n",
+			"Priority:%d extents:%d across:%lluk %s%s\n",
 		nr_good_pages<<(PAGE_SHIFT-10), name, p->prio,
 		nr_extents, (unsigned long long)span<<(PAGE_SHIFT-10),
-		(p->flags & SWP_DISCARDABLE) ? " D" : "");
+		(p->flags & SWP_SOLIDSTATE) ? "SS" : "",
+		(p->flags & SWP_DISCARDABLE) ? "D" : "");
 
 	/* insert swap space into swap_list: */
 	prev = -1;
-- 
cgit v1.2.3


From 79f4b7bf393e67bbffec807cc68caaefc72b82ee Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:40:05 -0800
Subject: badpage: simplify page_alloc flag check+clear

Simplify the PAGE_FLAGS checking and clearing when freeing and allocating
a page: check the same flags as before when freeing, clear ALL the flags
(unless PageReserved) when freeing, check ALL flags off when allocating.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/page-flags.h | 25 ++++++++-----------------
 mm/page_alloc.c            | 19 ++++++-------------
 2 files changed, 14 insertions(+), 30 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 628ec0802492..219a523ecdb0 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -373,31 +373,22 @@ static inline void __ClearPageTail(struct page *page)
 #define __PG_MLOCKED		0
 #endif
 
-#define PAGE_FLAGS	(1 << PG_lru   | 1 << PG_private   | 1 << PG_locked | \
-			 1 << PG_buddy | 1 << PG_writeback | \
-			 1 << PG_slab  | 1 << PG_swapcache | 1 << PG_active | \
-			 __PG_UNEVICTABLE | __PG_MLOCKED)
-
-/*
- * Flags checked in bad_page().  Pages on the free list should not have
- * these flags set.  It they are, there is a problem.
- */
-#define PAGE_FLAGS_CLEAR_WHEN_BAD (PAGE_FLAGS | \
-		1 << PG_reclaim | 1 << PG_dirty | 1 << PG_swapbacked)
-
 /*
  * Flags checked when a page is freed.  Pages being freed should not have
  * these flags set.  It they are, there is a problem.
  */
-#define PAGE_FLAGS_CHECK_AT_FREE (PAGE_FLAGS | 1 << PG_reserved)
+#define PAGE_FLAGS_CHECK_AT_FREE \
+	(1 << PG_lru   | 1 << PG_private   | 1 << PG_locked | \
+	 1 << PG_buddy | 1 << PG_writeback | 1 << PG_reserved | \
+	 1 << PG_slab  | 1 << PG_swapcache | 1 << PG_active | \
+	 __PG_UNEVICTABLE | __PG_MLOCKED)
 
 /*
  * Flags checked when a page is prepped for return by the page allocator.
- * Pages being prepped should not have these flags set.  It they are, there
- * is a problem.
+ * Pages being prepped should not have any flags set.  It they are set,
+ * there has been a kernel bug or struct page corruption.
  */
-#define PAGE_FLAGS_CHECK_AT_PREP (PAGE_FLAGS | \
-		1 << PG_reserved | 1 << PG_dirty | 1 << PG_swapbacked)
+#define PAGE_FLAGS_CHECK_AT_PREP	((1 << NR_PAGEFLAGS) - 1)
 
 #endif /* !__GENERATING_BOUNDS_H */
 #endif	/* PAGE_FLAGS_H */
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 31c512410a99..b90a74d28485 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -231,7 +231,6 @@ static void bad_page(struct page *page)
 	printk(KERN_EMERG "Trying to fix it up, but a reboot is needed\n"
 		KERN_EMERG "Backtrace:\n");
 	dump_stack();
-	page->flags &= ~PAGE_FLAGS_CLEAR_WHEN_BAD;
 	set_page_count(page, 0);
 	reset_page_mapcount(page);
 	page->mapping = NULL;
@@ -468,16 +467,16 @@ static inline int free_pages_check(struct page *page)
 		(page_count(page) != 0)  |
 		(page->flags & PAGE_FLAGS_CHECK_AT_FREE)))
 		bad_page(page);
-	if (PageDirty(page))
-		__ClearPageDirty(page);
-	if (PageSwapBacked(page))
-		__ClearPageSwapBacked(page);
 	/*
 	 * For now, we report if PG_reserved was found set, but do not
 	 * clear it, and do not free the page.  But we shall soon need
 	 * to do more, for when the ZERO_PAGE count wraps negative.
 	 */
-	return PageReserved(page);
+	if (PageReserved(page))
+		return 1;
+	if (page->flags & PAGE_FLAGS_CHECK_AT_PREP)
+		page->flags &= ~PAGE_FLAGS_CHECK_AT_PREP;
+	return 0;
 }
 
 /*
@@ -621,13 +620,7 @@ static int prep_new_page(struct page *page, int order, gfp_t gfp_flags)
 	if (PageReserved(page))
 		return 1;
 
-	page->flags &= ~(1 << PG_uptodate | 1 << PG_error | 1 << PG_reclaim |
-			1 << PG_referenced | 1 << PG_arch_1 |
-			1 << PG_owner_priv_1 | 1 << PG_mappedtodisk
-#ifdef CONFIG_UNEVICTABLE_LRU
-			| 1 << PG_mlocked
-#endif
-			);
+	page->flags &= ~PAGE_FLAGS_CHECK_AT_PREP;
 	set_page_private(page, 0);
 	set_page_refcounted(page);
 
-- 
cgit v1.2.3


From 2509ef26db4699a5d9fa876e90ddfc107afcab84 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:40:10 -0800
Subject: badpage: zap print_bad_pte on swap and file

Complete zap_pte_range()'s coverage of bad pagetable entries by calling
print_bad_pte() on a pte_file in a linear vma and on a bad swap entry.
That needs free_swap_and_cache() to tell it, which will also have shown
one of those "swap_free" errors (but with much less information).

Similar checks in fork's copy_one_pte()?  No, that would be more noisy
than helpful: we'll see them when parent and child exec or exit.

Where do_nonlinear_fault() calls print_bad_pte(): omit !VM_CAN_NONLINEAR
case, that could only be a bug in sys_remap_file_pages(), not a bad pte.
VM_FAULT_OOM rather than VM_FAULT_SIGBUS?  Well, okay, that is consistent
with what happens if do_swap_page() operates a bad swap entry; but don't
we have patches to be more careful about killing when VM_FAULT_OOM?

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h | 12 +++---------
 mm/memory.c          | 11 +++++++----
 mm/swapfile.c        |  7 ++++---
 3 files changed, 14 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/swap.h b/include/linux/swap.h
index cbf7fbed3dfd..91dee50fe260 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -302,7 +302,7 @@ extern swp_entry_t get_swap_page_of_type(int);
 extern int swap_duplicate(swp_entry_t);
 extern int valid_swaphandles(swp_entry_t, unsigned long *);
 extern void swap_free(swp_entry_t);
-extern void free_swap_and_cache(swp_entry_t);
+extern int free_swap_and_cache(swp_entry_t);
 extern int swap_type_of(dev_t, sector_t, struct block_device **);
 extern unsigned int count_swap_pages(int, int);
 extern sector_t map_swap_page(struct swap_info_struct *, pgoff_t);
@@ -352,14 +352,8 @@ static inline void show_swap_cache_info(void)
 {
 }
 
-static inline void free_swap_and_cache(swp_entry_t swp)
-{
-}
-
-static inline int swap_duplicate(swp_entry_t swp)
-{
-	return 0;
-}
+#define free_swap_and_cache(swp)	is_migration_entry(swp)
+#define swap_duplicate(swp)		is_migration_entry(swp)
 
 static inline void swap_free(swp_entry_t swp)
 {
diff --git a/mm/memory.c b/mm/memory.c
index 890095f5f36d..b273cc12b15d 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -810,8 +810,12 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb,
 		 */
 		if (unlikely(details))
 			continue;
-		if (!pte_file(ptent))
-			free_swap_and_cache(pte_to_swp_entry(ptent));
+		if (pte_file(ptent)) {
+			if (unlikely(!(vma->vm_flags & VM_NONLINEAR)))
+				print_bad_pte(vma, addr, ptent, NULL);
+		} else if
+		  (unlikely(!free_swap_and_cache(pte_to_swp_entry(ptent))))
+			print_bad_pte(vma, addr, ptent, NULL);
 		pte_clear_not_present_full(mm, addr, pte, tlb->fullmm);
 	} while (pte++, addr += PAGE_SIZE, (addr != end && *zap_work > 0));
 
@@ -2707,8 +2711,7 @@ static int do_nonlinear_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 	if (!pte_unmap_same(mm, pmd, page_table, orig_pte))
 		return 0;
 
-	if (unlikely(!(vma->vm_flags & VM_NONLINEAR) ||
-			!(vma->vm_flags & VM_CAN_NONLINEAR))) {
+	if (unlikely(!(vma->vm_flags & VM_NONLINEAR))) {
 		/*
 		 * Page table corrupted: show pte and kill process.
 		 */
diff --git a/mm/swapfile.c b/mm/swapfile.c
index d00523601913..f28745855772 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -571,13 +571,13 @@ int try_to_free_swap(struct page *page)
  * Free the swap entry like above, but also try to
  * free the page cache entry if it is the last user.
  */
-void free_swap_and_cache(swp_entry_t entry)
+int free_swap_and_cache(swp_entry_t entry)
 {
-	struct swap_info_struct * p;
+	struct swap_info_struct *p;
 	struct page *page = NULL;
 
 	if (is_migration_entry(entry))
-		return;
+		return 1;
 
 	p = swap_info_get(entry);
 	if (p) {
@@ -603,6 +603,7 @@ void free_swap_and_cache(swp_entry_t entry)
 		unlock_page(page);
 		page_cache_release(page);
 	}
+	return p != NULL;
 }
 
 #ifdef CONFIG_HIBERNATION
-- 
cgit v1.2.3


From edc315fd222497ae4f4b959a9e31ada1e68a4755 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 6 Jan 2009 14:40:11 -0800
Subject: badpage: remove vma from page_remove_rmap

Remove page_remove_rmap()'s vma arg, which was only for the Eeek message.
And remove the BUG_ON(page_mapcount(page) == 0) from CONFIG_DEBUG_VM's
page_dup_rmap(): we're trying to be more resilient about that than BUGs.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/rmap.h | 2 +-
 mm/filemap_xip.c     | 2 +-
 mm/fremap.c          | 2 +-
 mm/memory.c          | 4 ++--
 mm/rmap.c            | 8 +++-----
 5 files changed, 8 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index 3593b18a07dd..b35bc0e19cd9 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -69,7 +69,7 @@ void __anon_vma_link(struct vm_area_struct *);
 void page_add_anon_rmap(struct page *, struct vm_area_struct *, unsigned long);
 void page_add_new_anon_rmap(struct page *, struct vm_area_struct *, unsigned long);
 void page_add_file_rmap(struct page *);
-void page_remove_rmap(struct page *, struct vm_area_struct *);
+void page_remove_rmap(struct page *);
 
 #ifdef CONFIG_DEBUG_VM
 void page_dup_rmap(struct page *page, struct vm_area_struct *vma, unsigned long address);
diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c
index b5167dfb2f2d..0c04615651b7 100644
--- a/mm/filemap_xip.c
+++ b/mm/filemap_xip.c
@@ -193,7 +193,7 @@ retry:
 			/* Nuke the page table entry. */
 			flush_cache_page(vma, address, pte_pfn(*pte));
 			pteval = ptep_clear_flush_notify(vma, address, pte);
-			page_remove_rmap(page, vma);
+			page_remove_rmap(page);
 			dec_mm_counter(mm, file_rss);
 			BUG_ON(pte_dirty(pteval));
 			pte_unmap_unlock(pte, ptl);
diff --git a/mm/fremap.c b/mm/fremap.c
index 7d12ca70ef7b..62d5bbda921a 100644
--- a/mm/fremap.c
+++ b/mm/fremap.c
@@ -37,7 +37,7 @@ static void zap_pte(struct mm_struct *mm, struct vm_area_struct *vma,
 		if (page) {
 			if (pte_dirty(pte))
 				set_page_dirty(page);
-			page_remove_rmap(page, vma);
+			page_remove_rmap(page);
 			page_cache_release(page);
 			update_hiwater_rss(mm);
 			dec_mm_counter(mm, file_rss);
diff --git a/mm/memory.c b/mm/memory.c
index b273cc12b15d..0f9abbaf618c 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -798,7 +798,7 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb,
 					mark_page_accessed(page);
 				file_rss--;
 			}
-			page_remove_rmap(page, vma);
+			page_remove_rmap(page);
 			if (unlikely(page_mapcount(page) < 0))
 				print_bad_pte(vma, addr, ptent, page);
 			tlb_remove_page(tlb, page);
@@ -2023,7 +2023,7 @@ gotten:
 			 * mapcount is visible. So transitively, TLBs to
 			 * old page will be flushed before it can be reused.
 			 */
-			page_remove_rmap(old_page, vma);
+			page_remove_rmap(old_page);
 		}
 
 		/* Free the old page.. */
diff --git a/mm/rmap.c b/mm/rmap.c
index 32098255082e..ac4af8cffbf9 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -707,7 +707,6 @@ void page_add_file_rmap(struct page *page)
  */
 void page_dup_rmap(struct page *page, struct vm_area_struct *vma, unsigned long address)
 {
-	BUG_ON(page_mapcount(page) == 0);
 	if (PageAnon(page))
 		__page_check_anon_rmap(page, vma, address);
 	atomic_inc(&page->_mapcount);
@@ -717,11 +716,10 @@ void page_dup_rmap(struct page *page, struct vm_area_struct *vma, unsigned long
 /**
  * page_remove_rmap - take down pte mapping from a page
  * @page: page to remove mapping from
- * @vma: the vm area in which the mapping is removed
  *
  * The caller needs to hold the pte lock.
  */
-void page_remove_rmap(struct page *page, struct vm_area_struct *vma)
+void page_remove_rmap(struct page *page)
 {
 	if (atomic_add_negative(-1, &page->_mapcount)) {
 		/*
@@ -837,7 +835,7 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
 		dec_mm_counter(mm, file_rss);
 
 
-	page_remove_rmap(page, vma);
+	page_remove_rmap(page);
 	page_cache_release(page);
 
 out_unmap:
@@ -952,7 +950,7 @@ static int try_to_unmap_cluster(unsigned long cursor, unsigned int *mapcount,
 		if (pte_dirty(pteval))
 			set_page_dirty(page);
 
-		page_remove_rmap(page, vma);
+		page_remove_rmap(page);
 		page_cache_release(page);
 		dec_mm_counter(mm, file_rss);
 		(*mapcount)--;
-- 
cgit v1.2.3


From 4f5a99d64c17470a784a6c68064207d82e3e74a5 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Tue, 6 Jan 2009 14:40:25 -0800
Subject: fs: remove WB_SYNC_HOLD

Remove WB_SYNC_HOLD.  The primary motiviation is the design of my
anti-starvation code for fsync.  It requires taking an inode lock over the
sync operation, so we could run into lock ordering problems with multiple
inodes.  It is possible to take a single global lock to solve the ordering
problem, but then that would prevent a future nice implementation of "sync
multiple inodes" based on lock order via inode address.

Seems like a backward step to remove this, but actually it is busted
anyway: we can't use the inode lists for data integrity wait: an inode can
be taken off the dirty lists but still be under writeback.  In order to
satisfy data integrity semantics, we should wait for it to finish
writeback, but if we only search the dirty lists, we'll miss it.

It would be possible to have a "writeback" list, for sys_sync, I suppose.
But why complicate things by prematurely optimise?  For unmounting, we
could avoid the "livelock avoidance" code, which would be easier, but
again premature IMO.

Fixing the existing data integrity problem will come next.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/fs-writeback.c         | 12 ++----------
 include/linux/writeback.h |  1 -
 2 files changed, 2 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index d0ff0b8cf309..d99601af9e48 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -421,9 +421,6 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc)
  * If we're a pdlfush thread, then implement pdflush collision avoidance
  * against the entire list.
  *
- * WB_SYNC_HOLD is a hack for sys_sync(): reattach the inode to sb->s_dirty so
- * that it can be located for waiting on in __writeback_single_inode().
- *
  * If `bdi' is non-zero then we're being asked to writeback a specific queue.
  * This function assumes that the blockdev superblock's inodes are backed by
  * a variety of queues, so all inodes are searched.  For other superblocks,
@@ -499,10 +496,6 @@ void generic_sync_sb_inodes(struct super_block *sb,
 		__iget(inode);
 		pages_skipped = wbc->pages_skipped;
 		__writeback_single_inode(inode, wbc);
-		if (wbc->sync_mode == WB_SYNC_HOLD) {
-			inode->dirtied_when = jiffies;
-			list_move(&inode->i_list, &sb->s_dirty);
-		}
 		if (current_is_pdflush())
 			writeback_release(bdi);
 		if (wbc->pages_skipped != pages_skipped) {
@@ -588,8 +581,7 @@ restart:
 
 /*
  * writeback and wait upon the filesystem's dirty inodes.  The caller will
- * do this in two passes - one to write, and one to wait.  WB_SYNC_HOLD is
- * used to park the written inodes on sb->s_dirty for the wait pass.
+ * do this in two passes - one to write, and one to wait.
  *
  * A finite limit is set on the number of pages which will be written.
  * To prevent infinite livelock of sys_sync().
@@ -600,7 +592,7 @@ restart:
 void sync_inodes_sb(struct super_block *sb, int wait)
 {
 	struct writeback_control wbc = {
-		.sync_mode	= wait ? WB_SYNC_ALL : WB_SYNC_HOLD,
+		.sync_mode	= wait ? WB_SYNC_ALL : WB_SYNC_NONE,
 		.range_start	= 0,
 		.range_end	= LLONG_MAX,
 	};
diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index bb28c975c1d7..7300ecdc480c 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -30,7 +30,6 @@ static inline int task_is_pdflush(struct task_struct *task)
 enum writeback_sync_modes {
 	WB_SYNC_NONE,	/* Don't wait on anything */
 	WB_SYNC_ALL,	/* Wait on every mapping */
-	WB_SYNC_HOLD,	/* Hold the inode on sb_dirty for sys_sync() */
 };
 
 /*
-- 
cgit v1.2.3


From 856bf4d717feb8c55d4e2f817b71ebb70cfbc67b Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Tue, 6 Jan 2009 14:40:26 -0800
Subject: fs: sys_sync fix

s_syncing livelock avoidance was breaking data integrity guarantee of
sys_sync, by allowing sys_sync to skip writing or waiting for superblocks
if there is a concurrent sys_sync happening.

This livelock avoidance is much less important now that we don't have the
get_super_to_sync() call after every sb that we sync.  This was replaced
by __put_super_and_need_restart.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/fs-writeback.c  | 20 +-------------------
 include/linux/fs.h |  1 -
 2 files changed, 1 insertion(+), 20 deletions(-)

(limited to 'include/linux')

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index a9ee474f9691..e5eaa62fd17f 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -652,18 +652,6 @@ void sync_inodes_sb(struct super_block *sb, int wait)
 	sync_sb_inodes(sb, &wbc);
 }
 
-/*
- * Rather lame livelock avoidance.
- */
-static void set_sb_syncing(int val)
-{
-	struct super_block *sb;
-	spin_lock(&sb_lock);
-	list_for_each_entry_reverse(sb, &super_blocks, s_list)
-		sb->s_syncing = val;
-	spin_unlock(&sb_lock);
-}
-
 /**
  * sync_inodes - writes all inodes to disk
  * @wait: wait for completion
@@ -690,9 +678,6 @@ static void __sync_inodes(int wait)
 	spin_lock(&sb_lock);
 restart:
 	list_for_each_entry(sb, &super_blocks, s_list) {
-		if (sb->s_syncing)
-			continue;
-		sb->s_syncing = 1;
 		sb->s_count++;
 		spin_unlock(&sb_lock);
 		down_read(&sb->s_umount);
@@ -710,13 +695,10 @@ restart:
 
 void sync_inodes(int wait)
 {
-	set_sb_syncing(0);
 	__sync_inodes(0);
 
-	if (wait) {
-		set_sb_syncing(0);
+	if (wait)
 		__sync_inodes(1);
-	}
 }
 
 /**
diff --git a/include/linux/fs.h b/include/linux/fs.h
index fb59673c60b1..d7eba77f666e 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1133,7 +1133,6 @@ struct super_block {
 	struct rw_semaphore	s_umount;
 	struct mutex		s_lock;
 	int			s_count;
-	int			s_syncing;
 	int			s_need_sync_fs;
 	atomic_t		s_active;
 #ifdef CONFIG_SECURITY
-- 
cgit v1.2.3


From 901608d9045146aec6f14a7777ea4b1501c379f0 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@redhat.com>
Date: Tue, 6 Jan 2009 14:40:29 -0800
Subject: mm: introduce get_mm_hiwater_xxx(), fix taskstats->hiwater_xxx
 accounting

xacct_add_tsk() relies on do_exit()->update_hiwater_xxx() and uses
mm->hiwater_xxx directly, this leads to 2 problems:

- taskstats_user_cmd() can call fill_pid()->xacct_add_tsk() at any
  moment before the task exits, so we should check the current values of
  rss/vm anyway.

- do_exit()->update_hiwater_xxx() calls are racy.  An exiting thread can
  be preempted right before mm->hiwater_xxx = new_val, and another thread
  can use A_LOT of memory and exit in between.  When the first thread
  resumes it can be the last thread in the thread group, in that case we
  report the wrong hiwater_xxx values which do not take A_LOT into
  account.

Introduce get_mm_hiwater_rss() and get_mm_hiwater_vm() helpers and change
xacct_add_tsk() to use them.  The first helper will also be used by
rusage->ru_maxrss accounting.

Kill do_exit()->update_hiwater_xxx() calls.  Unless we are going to
decrease rss/vm there is no point to update mm->hiwater_xxx, and nobody
can look at this mm_struct when exit_mmap() actually unmaps the memory.

Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Hugh Dickins <hugh@veritas.com>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/sched.h | 3 +++
 kernel/exit.c         | 5 +----
 kernel/tsacct.c       | 4 ++--
 mm/mmap.c             | 2 +-
 4 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 38a3f4b15394..ea415136ac9e 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -386,6 +386,9 @@ extern void arch_unmap_area_topdown(struct mm_struct *, unsigned long);
 		(mm)->hiwater_vm = (mm)->total_vm;	\
 } while (0)
 
+#define get_mm_hiwater_rss(mm)	max((mm)->hiwater_rss, get_mm_rss(mm))
+#define get_mm_hiwater_vm(mm)	max((mm)->hiwater_vm, (mm)->total_vm)
+
 extern void set_dumpable(struct mm_struct *mm, int value);
 extern int get_dumpable(struct mm_struct *mm);
 
diff --git a/kernel/exit.c b/kernel/exit.c
index f923724ab3c9..c7740fa3252c 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -1051,10 +1051,7 @@ NORET_TYPE void do_exit(long code)
 				preempt_count());
 
 	acct_update_integrals(tsk);
-	if (tsk->mm) {
-		update_hiwater_rss(tsk->mm);
-		update_hiwater_vm(tsk->mm);
-	}
+
 	group_dead = atomic_dec_and_test(&tsk->signal->live);
 	if (group_dead) {
 		hrtimer_cancel(&tsk->signal->real_timer);
diff --git a/kernel/tsacct.c b/kernel/tsacct.c
index 2dc06ab35716..43f891b05a4b 100644
--- a/kernel/tsacct.c
+++ b/kernel/tsacct.c
@@ -92,8 +92,8 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p)
 	mm = get_task_mm(p);
 	if (mm) {
 		/* adjust to KB unit */
-		stats->hiwater_rss   = mm->hiwater_rss * PAGE_SIZE / KB;
-		stats->hiwater_vm    = mm->hiwater_vm * PAGE_SIZE / KB;
+		stats->hiwater_rss   = get_mm_hiwater_rss(mm) * PAGE_SIZE / KB;
+		stats->hiwater_vm    = get_mm_hiwater_vm(mm)  * PAGE_SIZE / KB;
 		mmput(mm);
 	}
 	stats->read_char	= p->ioac.rchar;
diff --git a/mm/mmap.c b/mm/mmap.c
index e4507b23e620..1f97d8aa9b05 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -2102,7 +2102,7 @@ void exit_mmap(struct mm_struct *mm)
 	lru_add_drain();
 	flush_cache_mm(mm);
 	tlb = tlb_gather_mmu(mm, 1);
-	/* Don't update_hiwater_rss(mm) here, do_exit already did */
+	/* update_hiwater_rss(mm) here? but nobody should be looking */
 	/* Use -1 here to ensure all VMAs in the mm are unmapped */
 	end = unmap_vmas(&tlb, vma, 0, -1, &nr_accounted, NULL);
 	vm_unacct_memory(nr_accounted);
-- 
cgit v1.2.3


From ea435467500612636f8f4fb639ff6e76b2496e4b Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Tue, 6 Jan 2009 14:40:39 -0800
Subject: atomic_t: unify all arch definitions

The atomic_t type cannot currently be used in some header files because it
would create an include loop with asm/atomic.h.  Move the type definition
to linux/types.h to break the loop.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/alpha/include/asm/atomic.h     |  9 +--------
 arch/arm/include/asm/atomic.h       |  3 +--
 arch/avr32/include/asm/atomic.h     |  2 +-
 arch/blackfin/include/asm/atomic.h  |  4 +---
 arch/cris/include/asm/atomic.h      |  4 +---
 arch/h8300/include/asm/atomic.h     |  3 ++-
 arch/ia64/include/asm/atomic.h      |  6 ------
 arch/m68knommu/include/asm/atomic.h |  2 +-
 arch/mips/include/asm/atomic.h      |  5 +----
 arch/parisc/include/asm/atomic.h    | 11 +++--------
 arch/powerpc/include/asm/atomic.h   |  4 +---
 arch/s390/include/asm/atomic.h      |  7 +------
 arch/sh/include/asm/atomic.h        |  7 +++----
 arch/sparc/include/asm/atomic_32.h  |  2 --
 arch/sparc/include/asm/atomic_64.h  |  3 ---
 arch/x86/include/asm/atomic_32.h    | 10 +---------
 arch/x86/include/asm/atomic_64.h    | 18 ++----------------
 include/asm-frv/atomic.h            |  4 ----
 include/asm-m32r/atomic.h           |  8 +-------
 include/asm-m68k/atomic.h           |  3 +--
 include/asm-mn10300/atomic.h        |  9 ---------
 include/asm-xtensa/atomic.h         |  3 +--
 include/linux/types.h               | 10 ++++++++++
 23 files changed, 33 insertions(+), 104 deletions(-)

(limited to 'include/linux')

diff --git a/arch/alpha/include/asm/atomic.h b/arch/alpha/include/asm/atomic.h
index ca88e54dec93..62b363584b2b 100644
--- a/arch/alpha/include/asm/atomic.h
+++ b/arch/alpha/include/asm/atomic.h
@@ -1,6 +1,7 @@
 #ifndef _ALPHA_ATOMIC_H
 #define _ALPHA_ATOMIC_H
 
+#include <linux/types.h>
 #include <asm/barrier.h>
 #include <asm/system.h>
 
@@ -13,14 +14,6 @@
  */
 
 
-/*
- * Counter is volatile to make sure gcc doesn't try to be clever
- * and move things around on us. We need to use _exactly_ the address
- * the user gave us, not some alias that contains the same information.
- */
-typedef struct { volatile int counter; } atomic_t;
-typedef struct { volatile long counter; } atomic64_t;
-
 #define ATOMIC_INIT(i)		( (atomic_t) { (i) } )
 #define ATOMIC64_INIT(i)	( (atomic64_t) { (i) } )
 
diff --git a/arch/arm/include/asm/atomic.h b/arch/arm/include/asm/atomic.h
index 325f881ccb50..ee99723b3a6c 100644
--- a/arch/arm/include/asm/atomic.h
+++ b/arch/arm/include/asm/atomic.h
@@ -12,10 +12,9 @@
 #define __ASM_ARM_ATOMIC_H
 
 #include <linux/compiler.h>
+#include <linux/types.h>
 #include <asm/system.h>
 
-typedef struct { volatile int counter; } atomic_t;
-
 #define ATOMIC_INIT(i)	{ (i) }
 
 #ifdef __KERNEL__
diff --git a/arch/avr32/include/asm/atomic.h b/arch/avr32/include/asm/atomic.h
index 7ef3862a73d0..318815107748 100644
--- a/arch/avr32/include/asm/atomic.h
+++ b/arch/avr32/include/asm/atomic.h
@@ -14,9 +14,9 @@
 #ifndef __ASM_AVR32_ATOMIC_H
 #define __ASM_AVR32_ATOMIC_H
 
+#include <linux/types.h>
 #include <asm/system.h>
 
-typedef struct { volatile int counter; } atomic_t;
 #define ATOMIC_INIT(i)  { (i) }
 
 #define atomic_read(v)		((v)->counter)
diff --git a/arch/blackfin/include/asm/atomic.h b/arch/blackfin/include/asm/atomic.h
index 7cf508718605..25776c19064b 100644
--- a/arch/blackfin/include/asm/atomic.h
+++ b/arch/blackfin/include/asm/atomic.h
@@ -1,6 +1,7 @@
 #ifndef __ARCH_BLACKFIN_ATOMIC__
 #define __ARCH_BLACKFIN_ATOMIC__
 
+#include <linux/types.h>
 #include <asm/system.h>	/* local_irq_XXX() */
 
 /*
@@ -13,9 +14,6 @@
  * Tony Kou (tonyko@lineo.ca)   Lineo Inc.   2001
  */
 
-typedef struct {
-	int counter;
-} atomic_t;
 #define ATOMIC_INIT(i)	{ (i) }
 
 #define atomic_read(v)		((v)->counter)
diff --git a/arch/cris/include/asm/atomic.h b/arch/cris/include/asm/atomic.h
index f71ea686a2ea..5718dd8902a1 100644
--- a/arch/cris/include/asm/atomic.h
+++ b/arch/cris/include/asm/atomic.h
@@ -4,7 +4,7 @@
 #define __ASM_CRIS_ATOMIC__
 
 #include <linux/compiler.h>
-
+#include <linux/types.h>
 #include <asm/system.h>
 #include <arch/atomic.h>
 
@@ -13,8 +13,6 @@
  * resource counting etc..
  */
 
-typedef struct { volatile int counter; } atomic_t;
-
 #define ATOMIC_INIT(i)  { (i) }
 
 #define atomic_read(v) ((v)->counter)
diff --git a/arch/h8300/include/asm/atomic.h b/arch/h8300/include/asm/atomic.h
index b4cf0ea97ede..833186c8dc3b 100644
--- a/arch/h8300/include/asm/atomic.h
+++ b/arch/h8300/include/asm/atomic.h
@@ -1,12 +1,13 @@
 #ifndef __ARCH_H8300_ATOMIC__
 #define __ARCH_H8300_ATOMIC__
 
+#include <linux/types.h>
+
 /*
  * Atomic operations that C can't guarantee us.  Useful for
  * resource counting etc..
  */
 
-typedef struct { int counter; } atomic_t;
 #define ATOMIC_INIT(i)	{ (i) }
 
 #define atomic_read(v)		((v)->counter)
diff --git a/arch/ia64/include/asm/atomic.h b/arch/ia64/include/asm/atomic.h
index 50c2b83fd5a0..d37292bd9875 100644
--- a/arch/ia64/include/asm/atomic.h
+++ b/arch/ia64/include/asm/atomic.h
@@ -17,12 +17,6 @@
 #include <asm/intrinsics.h>
 #include <asm/system.h>
 
-/*
- * On IA-64, counter must always be volatile to ensure that that the
- * memory accesses are ordered.
- */
-typedef struct { volatile __s32 counter; } atomic_t;
-typedef struct { volatile __s64 counter; } atomic64_t;
 
 #define ATOMIC_INIT(i)		((atomic_t) { (i) })
 #define ATOMIC64_INIT(i)	((atomic64_t) { (i) })
diff --git a/arch/m68knommu/include/asm/atomic.h b/arch/m68knommu/include/asm/atomic.h
index d5632a305dae..6bb674855a3f 100644
--- a/arch/m68knommu/include/asm/atomic.h
+++ b/arch/m68knommu/include/asm/atomic.h
@@ -1,6 +1,7 @@
 #ifndef __ARCH_M68KNOMMU_ATOMIC__
 #define __ARCH_M68KNOMMU_ATOMIC__
 
+#include <linux/types.h>
 #include <asm/system.h>
 
 /*
@@ -12,7 +13,6 @@
  * We do not have SMP m68k systems, so we don't have to deal with that.
  */
 
-typedef struct { int counter; } atomic_t;
 #define ATOMIC_INIT(i)	{ (i) }
 
 #define atomic_read(v)		((v)->counter)
diff --git a/arch/mips/include/asm/atomic.h b/arch/mips/include/asm/atomic.h
index 1232be3885b0..c996c3b4d074 100644
--- a/arch/mips/include/asm/atomic.h
+++ b/arch/mips/include/asm/atomic.h
@@ -15,13 +15,12 @@
 #define _ASM_ATOMIC_H
 
 #include <linux/irqflags.h>
+#include <linux/types.h>
 #include <asm/barrier.h>
 #include <asm/cpu-features.h>
 #include <asm/war.h>
 #include <asm/system.h>
 
-typedef struct { volatile int counter; } atomic_t;
-
 #define ATOMIC_INIT(i)    { (i) }
 
 /*
@@ -404,8 +403,6 @@ static __inline__ int atomic_add_unless(atomic_t *v, int a, int u)
 
 #ifdef CONFIG_64BIT
 
-typedef struct { volatile long counter; } atomic64_t;
-
 #define ATOMIC64_INIT(i)    { (i) }
 
 /*
diff --git a/arch/parisc/include/asm/atomic.h b/arch/parisc/include/asm/atomic.h
index 57fcc4a5ebb4..edbfe25c5fc1 100644
--- a/arch/parisc/include/asm/atomic.h
+++ b/arch/parisc/include/asm/atomic.h
@@ -155,14 +155,11 @@ static inline unsigned long __cmpxchg_local(volatile void *ptr,
 #define cmpxchg64_local(ptr, o, n) __cmpxchg64_local_generic((ptr), (o), (n))
 #endif
 
-/* Note that we need not lock read accesses - aligned word writes/reads
- * are atomic, so a reader never sees unconsistent values.
- *
- * Cache-line alignment would conflict with, for example, linux/module.h
+/*
+ * Note that we need not lock read accesses - aligned word writes/reads
+ * are atomic, so a reader never sees inconsistent values.
  */
 
-typedef struct { volatile int counter; } atomic_t;
-
 /* It's possible to reduce all atomic operations to either
  * __atomic_add_return, atomic_set and atomic_read (the latter
  * is there only for consistency).
@@ -260,8 +257,6 @@ static __inline__ int atomic_add_unless(atomic_t *v, int a, int u)
 
 #ifdef CONFIG_64BIT
 
-typedef struct { volatile s64 counter; } atomic64_t;
-
 #define ATOMIC64_INIT(i) ((atomic64_t) { (i) })
 
 static __inline__ int
diff --git a/arch/powerpc/include/asm/atomic.h b/arch/powerpc/include/asm/atomic.h
index 499be5bdd6fa..b401950f5259 100644
--- a/arch/powerpc/include/asm/atomic.h
+++ b/arch/powerpc/include/asm/atomic.h
@@ -5,7 +5,7 @@
  * PowerPC atomic operations
  */
 
-typedef struct { int counter; } atomic_t;
+#include <linux/types.h>
 
 #ifdef __KERNEL__
 #include <linux/compiler.h>
@@ -251,8 +251,6 @@ static __inline__ int atomic_dec_if_positive(atomic_t *v)
 
 #ifdef __powerpc64__
 
-typedef struct { long counter; } atomic64_t;
-
 #define ATOMIC64_INIT(i)	{ (i) }
 
 static __inline__ long atomic64_read(const atomic64_t *v)
diff --git a/arch/s390/include/asm/atomic.h b/arch/s390/include/asm/atomic.h
index 2d184655bc5d..de432f2de2d2 100644
--- a/arch/s390/include/asm/atomic.h
+++ b/arch/s390/include/asm/atomic.h
@@ -2,6 +2,7 @@
 #define __ARCH_S390_ATOMIC__
 
 #include <linux/compiler.h>
+#include <linux/types.h>
 
 /*
  *  include/asm-s390/atomic.h
@@ -23,9 +24,6 @@
  * S390 uses 'Compare And Swap' for atomicity in SMP enviroment
  */
 
-typedef struct {
-	int counter;
-} __attribute__ ((aligned (4))) atomic_t;
 #define ATOMIC_INIT(i)  { (i) }
 
 #ifdef __KERNEL__
@@ -149,9 +147,6 @@ static __inline__ int atomic_add_unless(atomic_t *v, int a, int u)
 #undef __CS_LOOP
 
 #ifdef __s390x__
-typedef struct {
-	long long counter;
-} __attribute__ ((aligned (8))) atomic64_t;
 #define ATOMIC64_INIT(i)  { (i) }
 
 #if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
diff --git a/arch/sh/include/asm/atomic.h b/arch/sh/include/asm/atomic.h
index c043ef003028..6327ffbb1992 100644
--- a/arch/sh/include/asm/atomic.h
+++ b/arch/sh/include/asm/atomic.h
@@ -7,16 +7,15 @@
  *
  */
 
-typedef struct { volatile int counter; } atomic_t;
+#include <linux/compiler.h>
+#include <linux/types.h>
+#include <asm/system.h>
 
 #define ATOMIC_INIT(i)	( (atomic_t) { (i) } )
 
 #define atomic_read(v)		((v)->counter)
 #define atomic_set(v,i)		((v)->counter = (i))
 
-#include <linux/compiler.h>
-#include <asm/system.h>
-
 #if defined(CONFIG_GUSA_RB)
 #include <asm/atomic-grb.h>
 #elif defined(CONFIG_CPU_SH4A)
diff --git a/arch/sparc/include/asm/atomic_32.h b/arch/sparc/include/asm/atomic_32.h
index 5c944b5a8040..ce465975a6a5 100644
--- a/arch/sparc/include/asm/atomic_32.h
+++ b/arch/sparc/include/asm/atomic_32.h
@@ -13,8 +13,6 @@
 
 #include <linux/types.h>
 
-typedef struct { volatile int counter; } atomic_t;
-
 #ifdef __KERNEL__
 
 #define ATOMIC_INIT(i)  { (i) }
diff --git a/arch/sparc/include/asm/atomic_64.h b/arch/sparc/include/asm/atomic_64.h
index 5982c5ae7f07..a0a706492696 100644
--- a/arch/sparc/include/asm/atomic_64.h
+++ b/arch/sparc/include/asm/atomic_64.h
@@ -10,9 +10,6 @@
 #include <linux/types.h>
 #include <asm/system.h>
 
-typedef struct { volatile int counter; } atomic_t;
-typedef struct { volatile __s64 counter; } atomic64_t;
-
 #define ATOMIC_INIT(i)		{ (i) }
 #define ATOMIC64_INIT(i)	{ (i) }
 
diff --git a/arch/x86/include/asm/atomic_32.h b/arch/x86/include/asm/atomic_32.h
index ad5b9f6ecddf..85b46fba4229 100644
--- a/arch/x86/include/asm/atomic_32.h
+++ b/arch/x86/include/asm/atomic_32.h
@@ -2,6 +2,7 @@
 #define _ASM_X86_ATOMIC_32_H
 
 #include <linux/compiler.h>
+#include <linux/types.h>
 #include <asm/processor.h>
 #include <asm/cmpxchg.h>
 
@@ -10,15 +11,6 @@
  * resource counting etc..
  */
 
-/*
- * Make sure gcc doesn't try to be clever and move things around
- * on us. We need to use _exactly_ the address the user gave us,
- * not some alias that contains the same information.
- */
-typedef struct {
-	int counter;
-} atomic_t;
-
 #define ATOMIC_INIT(i)	{ (i) }
 
 /**
diff --git a/arch/x86/include/asm/atomic_64.h b/arch/x86/include/asm/atomic_64.h
index 279d2a731f3f..8c21731984da 100644
--- a/arch/x86/include/asm/atomic_64.h
+++ b/arch/x86/include/asm/atomic_64.h
@@ -1,25 +1,15 @@
 #ifndef _ASM_X86_ATOMIC_64_H
 #define _ASM_X86_ATOMIC_64_H
 
+#include <linux/types.h>
 #include <asm/alternative.h>
 #include <asm/cmpxchg.h>
 
-/* atomic_t should be 32 bit signed type */
-
 /*
  * Atomic operations that C can't guarantee us.  Useful for
  * resource counting etc..
  */
 
-/*
- * Make sure gcc doesn't try to be clever and move things around
- * on us. We need to use _exactly_ the address the user gave us,
- * not some alias that contains the same information.
- */
-typedef struct {
-	int counter;
-} atomic_t;
-
 #define ATOMIC_INIT(i)	{ (i) }
 
 /**
@@ -191,11 +181,7 @@ static inline int atomic_sub_return(int i, atomic_t *v)
 #define atomic_inc_return(v)  (atomic_add_return(1, v))
 #define atomic_dec_return(v)  (atomic_sub_return(1, v))
 
-/* An 64bit atomic type */
-
-typedef struct {
-	long counter;
-} atomic64_t;
+/* The 64-bit atomic type */
 
 #define ATOMIC64_INIT(i)	{ (i) }
 
diff --git a/include/asm-frv/atomic.h b/include/asm-frv/atomic.h
index 46d696b331e7..296c35cfb207 100644
--- a/include/asm-frv/atomic.h
+++ b/include/asm-frv/atomic.h
@@ -35,10 +35,6 @@
 #define smp_mb__before_atomic_inc()	barrier()
 #define smp_mb__after_atomic_inc()	barrier()
 
-typedef struct {
-	int counter;
-} atomic_t;
-
 #define ATOMIC_INIT(i)		{ (i) }
 #define atomic_read(v)		((v)->counter)
 #define atomic_set(v, i)	(((v)->counter) = (i))
diff --git a/include/asm-m32r/atomic.h b/include/asm-m32r/atomic.h
index 3a38ffe4a4f4..2eed30f84080 100644
--- a/include/asm-m32r/atomic.h
+++ b/include/asm-m32r/atomic.h
@@ -9,6 +9,7 @@
  *    Copyright (C) 2004  Hirokazu Takata <takata at linux-m32r.org>
  */
 
+#include <linux/types.h>
 #include <asm/assembler.h>
 #include <asm/system.h>
 
@@ -17,13 +18,6 @@
  * resource counting etc..
  */
 
-/*
- * Make sure gcc doesn't try to be clever and move things around
- * on us. We need to use _exactly_ the address the user gave us,
- * not some alias that contains the same information.
- */
-typedef struct { volatile int counter; } atomic_t;
-
 #define ATOMIC_INIT(i)	{ (i) }
 
 /**
diff --git a/include/asm-m68k/atomic.h b/include/asm-m68k/atomic.h
index 4915294fea63..eb0ab9d4ee77 100644
--- a/include/asm-m68k/atomic.h
+++ b/include/asm-m68k/atomic.h
@@ -1,7 +1,7 @@
 #ifndef __ARCH_M68K_ATOMIC__
 #define __ARCH_M68K_ATOMIC__
 
-
+#include <linux/types.h>
 #include <asm/system.h>
 
 /*
@@ -13,7 +13,6 @@
  * We do not have SMP m68k systems, so we don't have to deal with that.
  */
 
-typedef struct { int counter; } atomic_t;
 #define ATOMIC_INIT(i)	{ (i) }
 
 #define atomic_read(v)		((v)->counter)
diff --git a/include/asm-mn10300/atomic.h b/include/asm-mn10300/atomic.h
index 27c9690b9574..bc064825f9b1 100644
--- a/include/asm-mn10300/atomic.h
+++ b/include/asm-mn10300/atomic.h
@@ -20,15 +20,6 @@
  * resource counting etc..
  */
 
-/*
- * Make sure gcc doesn't try to be clever and move things around
- * on us. We need to use _exactly_ the address the user gave us,
- * not some alias that contains the same information.
- */
-typedef struct {
-	int	counter;
-} atomic_t;
-
 #define ATOMIC_INIT(i)	{ (i) }
 
 #ifdef __KERNEL__
diff --git a/include/asm-xtensa/atomic.h b/include/asm-xtensa/atomic.h
index b3b23540f14d..67ad67bed8c1 100644
--- a/include/asm-xtensa/atomic.h
+++ b/include/asm-xtensa/atomic.h
@@ -14,8 +14,7 @@
 #define _XTENSA_ATOMIC_H
 
 #include <linux/stringify.h>
-
-typedef struct { volatile int counter; } atomic_t;
+#include <linux/types.h>
 
 #ifdef __KERNEL__
 #include <asm/processor.h>
diff --git a/include/linux/types.h b/include/linux/types.h
index 121f349cb7ec..3b864f2d9560 100644
--- a/include/linux/types.h
+++ b/include/linux/types.h
@@ -195,6 +195,16 @@ typedef u32 phys_addr_t;
 
 typedef phys_addr_t resource_size_t;
 
+typedef struct {
+	volatile int counter;
+} atomic_t;
+
+#ifdef CONFIG_64BIT
+typedef struct {
+	volatile long counter;
+} atomic64_t;
+#endif
+
 struct ustat {
 	__kernel_daddr_t	f_tfree;
 	__kernel_ino_t		f_tinode;
-- 
cgit v1.2.3


From f1883f86dea84fe47a71a39fc1afccc005915ed8 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Tue, 6 Jan 2009 14:40:45 -0800
Subject: Remove remaining unwinder code

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Gabor Gombas <gombasg@sztaki.hu>
Cc: Jan Beulich <jbeulich@novell.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Ingo Molnar <mingo@elte.hu>,
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/include/asm/unwind.h | 13 ---------
 arch/x86/kernel/traps.c       |  2 --
 include/linux/module.h        |  3 --
 include/linux/unwind.h        | 68 -------------------------------------------
 init/main.c                   |  3 --
 kernel/module.c               | 15 ----------
 lib/fault-inject.c            |  1 -
 7 files changed, 105 deletions(-)
 delete mode 100644 arch/x86/include/asm/unwind.h
 delete mode 100644 include/linux/unwind.h

(limited to 'include/linux')

diff --git a/arch/x86/include/asm/unwind.h b/arch/x86/include/asm/unwind.h
deleted file mode 100644
index 8b064bd9c553..000000000000
--- a/arch/x86/include/asm/unwind.h
+++ /dev/null
@@ -1,13 +0,0 @@
-#ifndef _ASM_X86_UNWIND_H
-#define _ASM_X86_UNWIND_H
-
-#define UNW_PC(frame) ((void)(frame), 0UL)
-#define UNW_SP(frame) ((void)(frame), 0UL)
-#define UNW_FP(frame) ((void)(frame), 0UL)
-
-static inline int arch_unw_user_mode(const void *info)
-{
-	return 0;
-}
-
-#endif /* _ASM_X86_UNWIND_H */
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index ce6650eb64e9..c9a666cdd3db 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -20,7 +20,6 @@
 #include <linux/module.h>
 #include <linux/ptrace.h>
 #include <linux/string.h>
-#include <linux/unwind.h>
 #include <linux/delay.h>
 #include <linux/errno.h>
 #include <linux/kexec.h>
@@ -51,7 +50,6 @@
 #include <asm/debugreg.h>
 #include <asm/atomic.h>
 #include <asm/system.h>
-#include <asm/unwind.h>
 #include <asm/traps.h>
 #include <asm/desc.h>
 #include <asm/i387.h>
diff --git a/include/linux/module.h b/include/linux/module.h
index 3bfed013350b..03cb93d1865a 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -294,9 +294,6 @@ struct module
 	/* The size of the executable code in each section.  */
 	unsigned int init_text_size, core_text_size;
 
-	/* The handle returned from unwind_add_table. */
-	void *unwind_info;
-
 	/* Arch-specific module values */
 	struct mod_arch_specific arch;
 
diff --git a/include/linux/unwind.h b/include/linux/unwind.h
deleted file mode 100644
index 7760860fa170..000000000000
--- a/include/linux/unwind.h
+++ /dev/null
@@ -1,68 +0,0 @@
-#ifndef _LINUX_UNWIND_H
-#define _LINUX_UNWIND_H
-
-/*
- * Copyright (C) 2002-2006 Novell, Inc.
- *	Jan Beulich <jbeulich@novell.com>
- * This code is released under version 2 of the GNU GPL.
- *
- * A simple API for unwinding kernel stacks.  This is used for
- * debugging and error reporting purposes.  The kernel doesn't need
- * full-blown stack unwinding with all the bells and whistles, so there
- * is not much point in implementing the full Dwarf2 unwind API.
- */
-
-struct module;
-
-struct unwind_frame_info {};
-
-static inline void unwind_init(void) {}
-static inline void unwind_setup(void) {}
-
-#ifdef CONFIG_MODULES
-
-static inline void *unwind_add_table(struct module *mod,
-                                     const void *table_start,
-                                     unsigned long table_size)
-{
-	return NULL;
-}
-
-static inline void unwind_remove_table(void *handle, int init_only)
-{
-}
-
-#endif
-
-static inline int unwind_init_frame_info(struct unwind_frame_info *info,
-                                         struct task_struct *tsk,
-                                         const struct pt_regs *regs)
-{
-	return -ENOSYS;
-}
-
-static inline int unwind_init_blocked(struct unwind_frame_info *info,
-                                      struct task_struct *tsk)
-{
-	return -ENOSYS;
-}
-
-static inline int unwind_init_running(struct unwind_frame_info *info,
-                                      asmlinkage int (*cb)(struct unwind_frame_info *,
-                                                           void *arg),
-                                      void *arg)
-{
-	return -ENOSYS;
-}
-
-static inline int unwind(struct unwind_frame_info *info)
-{
-	return -ENOSYS;
-}
-
-static inline int unwind_to_user(struct unwind_frame_info *info)
-{
-	return -ENOSYS;
-}
-
-#endif /* _LINUX_UNWIND_H */
diff --git a/init/main.c b/init/main.c
index 90926dadc20d..e119dd28dd7d 100644
--- a/init/main.c
+++ b/init/main.c
@@ -50,7 +50,6 @@
 #include <linux/rmap.h>
 #include <linux/mempolicy.h>
 #include <linux/key.h>
-#include <linux/unwind.h>
 #include <linux/buffer_head.h>
 #include <linux/page_cgroup.h>
 #include <linux/debug_locks.h>
@@ -537,7 +536,6 @@ asmlinkage void __init start_kernel(void)
 	 * Need to run as early as possible, to initialize the
 	 * lockdep hash:
 	 */
-	unwind_init();
 	lockdep_init();
 	debug_objects_early_init();
 	cgroup_init_early();
@@ -559,7 +557,6 @@ asmlinkage void __init start_kernel(void)
 	setup_arch(&command_line);
 	mm_init_owner(&init_mm, &init_task);
 	setup_command_line(command_line);
-	unwind_setup();
 	setup_per_cpu_areas();
 	setup_nr_cpu_ids();
 	smp_prepare_boot_cpu();	/* arch-specific boot-cpu hooks */
diff --git a/kernel/module.c b/kernel/module.c
index f47cce910f25..34b56cf06615 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -43,7 +43,6 @@
 #include <linux/device.h>
 #include <linux/string.h>
 #include <linux/mutex.h>
-#include <linux/unwind.h>
 #include <linux/rculist.h>
 #include <asm/uaccess.h>
 #include <asm/cacheflush.h>
@@ -1449,8 +1448,6 @@ static void free_module(struct module *mod)
 	remove_sect_attrs(mod);
 	mod_kobject_remove(mod);
 
-	unwind_remove_table(mod->unwind_info, 0);
-
 	/* Arch-specific cleanup. */
 	module_arch_cleanup(mod);
 
@@ -1867,7 +1864,6 @@ static noinline struct module *load_module(void __user *umod,
 	unsigned int symindex = 0;
 	unsigned int strindex = 0;
 	unsigned int modindex, versindex, infoindex, pcpuindex;
-	unsigned int unwindex = 0;
 	unsigned int num_kp, num_mcount;
 	struct kernel_param *kp;
 	struct module *mod;
@@ -1957,9 +1953,6 @@ static noinline struct module *load_module(void __user *umod,
 	versindex = find_sec(hdr, sechdrs, secstrings, "__versions");
 	infoindex = find_sec(hdr, sechdrs, secstrings, ".modinfo");
 	pcpuindex = find_pcpusec(hdr, sechdrs, secstrings);
-#ifdef ARCH_UNWIND_SECTION_NAME
-	unwindex = find_sec(hdr, sechdrs, secstrings, ARCH_UNWIND_SECTION_NAME);
-#endif
 
 	/* Don't keep modinfo and version sections. */
 	sechdrs[infoindex].sh_flags &= ~(unsigned long)SHF_ALLOC;
@@ -1969,8 +1962,6 @@ static noinline struct module *load_module(void __user *umod,
 	sechdrs[symindex].sh_flags |= SHF_ALLOC;
 	sechdrs[strindex].sh_flags |= SHF_ALLOC;
 #endif
-	if (unwindex)
-		sechdrs[unwindex].sh_flags |= SHF_ALLOC;
 
 	/* Check module struct version now, before we try to use module. */
 	if (!check_modstruct_version(sechdrs, versindex, mod)) {
@@ -2267,11 +2258,6 @@ static noinline struct module *load_module(void __user *umod,
 	add_sect_attrs(mod, hdr->e_shnum, secstrings, sechdrs);
 	add_notes_attrs(mod, hdr->e_shnum, secstrings, sechdrs);
 
-	/* Size of section 0 is 0, so this works well if no unwind info. */
-	mod->unwind_info = unwind_add_table(mod,
-					    (void *)sechdrs[unwindex].sh_addr,
-					    sechdrs[unwindex].sh_size);
-
 	/* Get rid of temporary copy */
 	vfree(hdr);
 
@@ -2370,7 +2356,6 @@ sys_init_module(void __user *umod,
 	mutex_lock(&module_mutex);
 	/* Drop initial reference. */
 	module_put(mod);
-	unwind_remove_table(mod->unwind_info, 1);
 	module_free(mod, mod->module_init);
 	mod->module_init = NULL;
 	mod->init_size = 0;
diff --git a/lib/fault-inject.c b/lib/fault-inject.c
index a50a311554cc..f97af55bdd96 100644
--- a/lib/fault-inject.c
+++ b/lib/fault-inject.c
@@ -6,7 +6,6 @@
 #include <linux/fs.h>
 #include <linux/module.h>
 #include <linux/interrupt.h>
-#include <linux/unwind.h>
 #include <linux/stacktrace.h>
 #include <linux/kallsyms.h>
 #include <linux/fault-inject.h>
-- 
cgit v1.2.3


From 9fe06081ef145d6582c39e18203b5fffe6f3abc2 Mon Sep 17 00:00:00 2001
From: "Darrick J. Wong" <djwong@us.ibm.com>
Date: Tue, 6 Jan 2009 14:40:51 -0800
Subject: Create a DIV_ROUND_CLOSEST macro to do division with rounding

Create a helper macro to divide two numbers and round the result to the
nearest whole number.  This is a helper macro for hwmon drivers that want
to convert incoming sysfs values per standard hwmon practice, though the
macro itself can be used by anyone.

Signed-off-by: Darrick J. Wong <djwong@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/kernel.h | 6 ++++++
 1 file changed, 6 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index ca9ff6411dfa..721984844c94 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -48,6 +48,12 @@ extern const char linux_proc_banner[];
 #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f))
 #define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
 #define roundup(x, y) ((((x) + ((y) - 1)) / (y)) * (y))
+#define DIV_ROUND_CLOSEST(x, divisor)(			\
+{							\
+	typeof(divisor) __divisor = divisor;		\
+	(((x) + ((__divisor) / 2)) / (__divisor));	\
+}							\
+)
 
 #define _RET_IP_		(unsigned long)__builtin_return_address(0)
 #define _THIS_IP_  ({ __label__ __here; __here: (unsigned long)&&__here; })
-- 
cgit v1.2.3


From 5f820f648c92a5ecc771a96b3c29aa6e90013bba Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Tue, 6 Jan 2009 14:40:59 -0800
Subject: poll: allow f_op->poll to sleep

f_op->poll is the only vfs operation which is not allowed to sleep.  It's
because poll and select implementation used task state to synchronize
against wake ups, which doesn't have to be the case anymore as wait/wake
interface can now use custom wake up functions.  The non-sleep restriction
can be a bit tricky because ->poll is not called from an atomic context
and the result of accidentally sleeping in ->poll only shows up as
temporary busy looping when the timing is right or rather wrong.

This patch converts poll/select to use custom wake up function and use
separate triggered variable to synchronize against wake up events.  The
only added overhead is an extra function call during wake up and
negligible.

This patch removes the one non-sleep exception from vfs locking rules and
is beneficial to userland filesystem implementations like FUSE, 9p or
peculiar fs like spufs as it's very difficult for those to implement
non-sleeping poll method.

While at it, make the following cosmetic changes to make poll.h and
select.c checkpatch friendly.

* s/type * symbol/type *symbol/		   : three places in poll.h
* remove blank line before EXPORT_SYMBOL() : two places in select.c

Oleg: spotted missing barrier in poll_schedule_timeout()
Davide: spotted missing write barrier in pollwake()

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Eric Van Hensbergen <ericvh@gmail.com>
Cc: Ron Minnich <rminnich@sandia.gov>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Cc: Davide Libenzi <davidel@xmailserver.org>
Cc: Brad Boyer <flar@allandria.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Roland McGrath <roland@redhat.com>
Cc: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Davide Libenzi <davidel@xmailserver.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/filesystems/Locking |  2 +-
 drivers/media/video/v4l1-compat.c |  4 +--
 fs/select.c                       | 76 +++++++++++++++++++++++++++++++--------
 include/linux/poll.h              | 15 ++++++--
 4 files changed, 76 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking
index ccec55394380..cfbfa15a46ba 100644
--- a/Documentation/filesystems/Locking
+++ b/Documentation/filesystems/Locking
@@ -397,7 +397,7 @@ prototypes:
 };
 
 locking rules:
-	All except ->poll() may block.
+	All may block.
 			BKL
 llseek:			no	(see below)
 read:			no
diff --git a/drivers/media/video/v4l1-compat.c b/drivers/media/video/v4l1-compat.c
index d450cab20be4..b617bf05e2d7 100644
--- a/drivers/media/video/v4l1-compat.c
+++ b/drivers/media/video/v4l1-compat.c
@@ -203,7 +203,6 @@ static int poll_one(struct file *file, struct poll_wqueues *pwq)
 	table = &pwq->pt;
 	for (;;) {
 		int mask;
-		set_current_state(TASK_INTERRUPTIBLE);
 		mask = file->f_op->poll(file, table);
 		if (mask & POLLIN)
 			break;
@@ -212,9 +211,8 @@ static int poll_one(struct file *file, struct poll_wqueues *pwq)
 			retval = -ERESTARTSYS;
 			break;
 		}
-		schedule();
+		poll_schedule(pwq, TASK_INTERRUPTIBLE);
 	}
-	set_current_state(TASK_RUNNING);
 	poll_freewait(pwq);
 	return retval;
 }
diff --git a/fs/select.c b/fs/select.c
index 87df51eadcf2..08b91beed806 100644
--- a/fs/select.c
+++ b/fs/select.c
@@ -109,11 +109,11 @@ static void __pollwait(struct file *filp, wait_queue_head_t *wait_address,
 void poll_initwait(struct poll_wqueues *pwq)
 {
 	init_poll_funcptr(&pwq->pt, __pollwait);
+	pwq->polling_task = current;
 	pwq->error = 0;
 	pwq->table = NULL;
 	pwq->inline_index = 0;
 }
-
 EXPORT_SYMBOL(poll_initwait);
 
 static void free_poll_entry(struct poll_table_entry *entry)
@@ -142,12 +142,10 @@ void poll_freewait(struct poll_wqueues *pwq)
 		free_page((unsigned long) old);
 	}
 }
-
 EXPORT_SYMBOL(poll_freewait);
 
-static struct poll_table_entry *poll_get_entry(poll_table *_p)
+static struct poll_table_entry *poll_get_entry(struct poll_wqueues *p)
 {
-	struct poll_wqueues *p = container_of(_p, struct poll_wqueues, pt);
 	struct poll_table_page *table = p->table;
 
 	if (p->inline_index < N_INLINE_POLL_ENTRIES)
@@ -159,7 +157,6 @@ static struct poll_table_entry *poll_get_entry(poll_table *_p)
 		new_table = (struct poll_table_page *) __get_free_page(GFP_KERNEL);
 		if (!new_table) {
 			p->error = -ENOMEM;
-			__set_current_state(TASK_RUNNING);
 			return NULL;
 		}
 		new_table->entry = new_table->entries;
@@ -171,20 +168,75 @@ static struct poll_table_entry *poll_get_entry(poll_table *_p)
 	return table->entry++;
 }
 
+static int pollwake(wait_queue_t *wait, unsigned mode, int sync, void *key)
+{
+	struct poll_wqueues *pwq = wait->private;
+	DECLARE_WAITQUEUE(dummy_wait, pwq->polling_task);
+
+	/*
+	 * Although this function is called under waitqueue lock, LOCK
+	 * doesn't imply write barrier and the users expect write
+	 * barrier semantics on wakeup functions.  The following
+	 * smp_wmb() is equivalent to smp_wmb() in try_to_wake_up()
+	 * and is paired with set_mb() in poll_schedule_timeout.
+	 */
+	smp_wmb();
+	pwq->triggered = 1;
+
+	/*
+	 * Perform the default wake up operation using a dummy
+	 * waitqueue.
+	 *
+	 * TODO: This is hacky but there currently is no interface to
+	 * pass in @sync.  @sync is scheduled to be removed and once
+	 * that happens, wake_up_process() can be used directly.
+	 */
+	return default_wake_function(&dummy_wait, mode, sync, key);
+}
+
 /* Add a new entry */
 static void __pollwait(struct file *filp, wait_queue_head_t *wait_address,
 				poll_table *p)
 {
-	struct poll_table_entry *entry = poll_get_entry(p);
+	struct poll_wqueues *pwq = container_of(p, struct poll_wqueues, pt);
+	struct poll_table_entry *entry = poll_get_entry(pwq);
 	if (!entry)
 		return;
 	get_file(filp);
 	entry->filp = filp;
 	entry->wait_address = wait_address;
-	init_waitqueue_entry(&entry->wait, current);
+	init_waitqueue_func_entry(&entry->wait, pollwake);
+	entry->wait.private = pwq;
 	add_wait_queue(wait_address, &entry->wait);
 }
 
+int poll_schedule_timeout(struct poll_wqueues *pwq, int state,
+			  ktime_t *expires, unsigned long slack)
+{
+	int rc = -EINTR;
+
+	set_current_state(state);
+	if (!pwq->triggered)
+		rc = schedule_hrtimeout_range(expires, slack, HRTIMER_MODE_ABS);
+	__set_current_state(TASK_RUNNING);
+
+	/*
+	 * Prepare for the next iteration.
+	 *
+	 * The following set_mb() serves two purposes.  First, it's
+	 * the counterpart rmb of the wmb in pollwake() such that data
+	 * written before wake up is always visible after wake up.
+	 * Second, the full barrier guarantees that triggered clearing
+	 * doesn't pass event check of the next iteration.  Note that
+	 * this problem doesn't exist for the first iteration as
+	 * add_wait_queue() has full barrier semantics.
+	 */
+	set_mb(pwq->triggered, 0);
+
+	return rc;
+}
+EXPORT_SYMBOL(poll_schedule_timeout);
+
 /**
  * poll_select_set_timeout - helper function to setup the timeout value
  * @to:		pointer to timespec variable for the final timeout
@@ -340,8 +392,6 @@ int do_select(int n, fd_set_bits *fds, struct timespec *end_time)
 	for (;;) {
 		unsigned long *rinp, *routp, *rexp, *inp, *outp, *exp;
 
-		set_current_state(TASK_INTERRUPTIBLE);
-
 		inp = fds->in; outp = fds->out; exp = fds->ex;
 		rinp = fds->res_in; routp = fds->res_out; rexp = fds->res_ex;
 
@@ -411,10 +461,10 @@ int do_select(int n, fd_set_bits *fds, struct timespec *end_time)
 			to = &expire;
 		}
 
-		if (!schedule_hrtimeout_range(to, slack, HRTIMER_MODE_ABS))
+		if (!poll_schedule_timeout(&table, TASK_INTERRUPTIBLE,
+					   to, slack))
 			timed_out = 1;
 	}
-	__set_current_state(TASK_RUNNING);
 
 	poll_freewait(&table);
 
@@ -666,7 +716,6 @@ static int do_poll(unsigned int nfds,  struct poll_list *list,
 	for (;;) {
 		struct poll_list *walk;
 
-		set_current_state(TASK_INTERRUPTIBLE);
 		for (walk = list; walk != NULL; walk = walk->next) {
 			struct pollfd * pfd, * pfd_end;
 
@@ -709,10 +758,9 @@ static int do_poll(unsigned int nfds,  struct poll_list *list,
 			to = &expire;
 		}
 
-		if (!schedule_hrtimeout_range(to, slack, HRTIMER_MODE_ABS))
+		if (!poll_schedule_timeout(wait, TASK_INTERRUPTIBLE, to, slack))
 			timed_out = 1;
 	}
-	__set_current_state(TASK_RUNNING);
 	return count;
 }
 
diff --git a/include/linux/poll.h b/include/linux/poll.h
index badd98ab06f6..8c24ef8d9976 100644
--- a/include/linux/poll.h
+++ b/include/linux/poll.h
@@ -46,9 +46,9 @@ static inline void init_poll_funcptr(poll_table *pt, poll_queue_proc qproc)
 }
 
 struct poll_table_entry {
-	struct file * filp;
+	struct file *filp;
 	wait_queue_t wait;
-	wait_queue_head_t * wait_address;
+	wait_queue_head_t *wait_address;
 };
 
 /*
@@ -56,7 +56,9 @@ struct poll_table_entry {
  */
 struct poll_wqueues {
 	poll_table pt;
-	struct poll_table_page * table;
+	struct poll_table_page *table;
+	struct task_struct *polling_task;
+	int triggered;
 	int error;
 	int inline_index;
 	struct poll_table_entry inline_entries[N_INLINE_POLL_ENTRIES];
@@ -64,6 +66,13 @@ struct poll_wqueues {
 
 extern void poll_initwait(struct poll_wqueues *pwq);
 extern void poll_freewait(struct poll_wqueues *pwq);
+extern int poll_schedule_timeout(struct poll_wqueues *pwq, int state,
+				 ktime_t *expires, unsigned long slack);
+
+static inline int poll_schedule(struct poll_wqueues *pwq, int state)
+{
+	return poll_schedule_timeout(pwq, state, NULL, 0);
+}
 
 /*
  * Scaleable version of the fd_set.
-- 
cgit v1.2.3


From 179f7ebff6be45738c6e2fa68c8d2cc5c2c6308e Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Tue, 6 Jan 2009 14:41:04 -0800
Subject: percpu_counter: FBC_BATCH should be a variable

For NR_CPUS >= 16 values, FBC_BATCH is 2*NR_CPUS

Considering more and more distros are using high NR_CPUS values, it makes
sense to use a more sensible value for FBC_BATCH, and get rid of NR_CPUS.

A sensible value is 2*num_online_cpus(), with a minimum value of 32 (This
minimum value helps branch prediction in __percpu_counter_add())

We already have a hotcpu notifier, so we can adjust FBC_BATCH dynamically.

We rename FBC_BATCH to percpu_counter_batch since its not a constant
anymore.

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: David S. Miller <davem@davemloft.net>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext4/ext4.h                 |  6 +++---
 fs/ext4/inode.c                |  2 +-
 include/linux/percpu_counter.h |  8 ++------
 lib/percpu_counter.c           | 18 ++++++++++++++----
 4 files changed, 20 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index b0537c827024..6c46c648430d 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -1225,11 +1225,11 @@ do {								\
 } while (0)
 
 #ifdef CONFIG_SMP
-/* Each CPU can accumulate FBC_BATCH blocks in their local
+/* Each CPU can accumulate percpu_counter_batch blocks in their local
  * counters. So we need to make sure we have free blocks more
- * than FBC_BATCH  * nr_cpu_ids. Also add a window of 4 times.
+ * than percpu_counter_batch  * nr_cpu_ids. Also add a window of 4 times.
  */
-#define EXT4_FREEBLOCKS_WATERMARK (4 * (FBC_BATCH * nr_cpu_ids))
+#define EXT4_FREEBLOCKS_WATERMARK (4 * (percpu_counter_batch * nr_cpu_ids))
 #else
 #define EXT4_FREEBLOCKS_WATERMARK 0
 #endif
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 6702a49992a6..98d3fe7057ef 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2498,7 +2498,7 @@ static int ext4_nonda_switch(struct super_block *sb)
 	/*
 	 * switch to non delalloc mode if we are running low
 	 * on free block. The free block accounting via percpu
-	 * counters can get slightly wrong with FBC_BATCH getting
+	 * counters can get slightly wrong with percpu_counter_batch getting
 	 * accumulated on each CPU without updating global counters
 	 * Delalloc need an accurate free block accounting. So switch
 	 * to non delalloc when we are near to error range.
diff --git a/include/linux/percpu_counter.h b/include/linux/percpu_counter.h
index 9007ccdfc112..99de7a31bab8 100644
--- a/include/linux/percpu_counter.h
+++ b/include/linux/percpu_counter.h
@@ -24,11 +24,7 @@ struct percpu_counter {
 	s32 *counters;
 };
 
-#if NR_CPUS >= 16
-#define FBC_BATCH	(NR_CPUS*2)
-#else
-#define FBC_BATCH	(NR_CPUS*4)
-#endif
+extern int percpu_counter_batch;
 
 int percpu_counter_init(struct percpu_counter *fbc, s64 amount);
 int percpu_counter_init_irq(struct percpu_counter *fbc, s64 amount);
@@ -39,7 +35,7 @@ s64 __percpu_counter_sum(struct percpu_counter *fbc);
 
 static inline void percpu_counter_add(struct percpu_counter *fbc, s64 amount)
 {
-	__percpu_counter_add(fbc, amount, FBC_BATCH);
+	__percpu_counter_add(fbc, amount, percpu_counter_batch);
 }
 
 static inline s64 percpu_counter_sum_positive(struct percpu_counter *fbc)
diff --git a/lib/percpu_counter.c b/lib/percpu_counter.c
index b255b939bc1b..a60bd8046095 100644
--- a/lib/percpu_counter.c
+++ b/lib/percpu_counter.c
@@ -9,10 +9,8 @@
 #include <linux/cpu.h>
 #include <linux/module.h>
 
-#ifdef CONFIG_HOTPLUG_CPU
 static LIST_HEAD(percpu_counters);
 static DEFINE_MUTEX(percpu_counters_lock);
-#endif
 
 void percpu_counter_set(struct percpu_counter *fbc, s64 amount)
 {
@@ -111,13 +109,24 @@ void percpu_counter_destroy(struct percpu_counter *fbc)
 }
 EXPORT_SYMBOL(percpu_counter_destroy);
 
-#ifdef CONFIG_HOTPLUG_CPU
+int percpu_counter_batch __read_mostly = 32;
+EXPORT_SYMBOL(percpu_counter_batch);
+
+static void compute_batch_value(void)
+{
+	int nr = num_online_cpus();
+
+	percpu_counter_batch = max(32, nr*2);
+}
+
 static int __cpuinit percpu_counter_hotcpu_callback(struct notifier_block *nb,
 					unsigned long action, void *hcpu)
 {
+#ifdef CONFIG_HOTPLUG_CPU
 	unsigned int cpu;
 	struct percpu_counter *fbc;
 
+	compute_batch_value();
 	if (action != CPU_DEAD)
 		return NOTIFY_OK;
 
@@ -134,13 +143,14 @@ static int __cpuinit percpu_counter_hotcpu_callback(struct notifier_block *nb,
 		spin_unlock_irqrestore(&fbc->lock, flags);
 	}
 	mutex_unlock(&percpu_counters_lock);
+#endif
 	return NOTIFY_OK;
 }
 
 static int __init percpu_counter_startup(void)
 {
+	compute_batch_value();
 	hotcpu_notifier(percpu_counter_hotcpu_callback, 0);
 	return 0;
 }
 module_init(percpu_counter_startup);
-#endif
-- 
cgit v1.2.3


From 8c3659347efb43857b2c2d7bc63a9c7d68d1a608 Mon Sep 17 00:00:00 2001
From: Jesper Juhl <jj@chaosbits.net>
Date: Tue, 6 Jan 2009 14:41:14 -0800
Subject: include/linux/interrupt.h: do not include linux/irqnr.h twice

Signed-off-by: Jesper Juhl <jj@chaosbits.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/interrupt.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index 0702c4d7bdf0..af886b26c9d1 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -14,7 +14,6 @@
 #include <linux/irqflags.h>
 #include <linux/smp.h>
 #include <linux/percpu.h>
-#include <linux/irqnr.h>
 
 #include <asm/atomic.h>
 #include <asm/ptrace.h>
-- 
cgit v1.2.3


From 5cf0cc4e670b8da2231a3375db87ec3b6cb84432 Mon Sep 17 00:00:00 2001
From: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Date: Tue, 6 Jan 2009 14:41:38 -0800
Subject: binfmts.h: include list.h

linux_binfmt uses list_head, so list.h is needed.

[akpm@linux-foundation.org: fix `make headerscheck']
Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/binfmts.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h
index 6cbfbe297180..0d0150b4901e 100644
--- a/include/linux/binfmts.h
+++ b/include/linux/binfmts.h
@@ -18,6 +18,7 @@ struct pt_regs;
 #define BINPRM_BUF_SIZE 128
 
 #ifdef __KERNEL__
+#include <linux/list.h>
 
 #define CORENAME_MAX_SIZE 128
 
-- 
cgit v1.2.3


From d29389de0b0ee1715333bafc6ac3f22a75aa4313 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Tue, 6 Jan 2009 14:41:41 -0800
Subject: spi_gpio driver

Generalize the old at91rm9200 "bootstrap" bitbanging SPI master driver as
"spi_gpio", so it works with arbitrary GPIOs and can be configured through
platform_data.  Such SPI masters support:

 - any number of bus instances (bus_num is the platform_device.id)
 - any number of chipselects (one GPIO per spi_device)
 - all four SPI_MODE values, and SPI_CS_HIGH
 - i/o word sizes from 1 to 32 bits;
 - devices configured as with any other spi_master controller

When configured using platform_data, this provides relatively low clock
rates.  On platforms that support inlined GPIO calls, significantly
improved transfer speeds are also possible with a semi-custom driver.
(It's still painful when accessing flash memory, but less so.)

Sanity checked by using this version to replace both native controllers on
a board with six different SPI slaves, relying on three different
SPI_MODE_* values and both SPI_CS_HIGH settings for correct operation.

[akpm@linux-foundation.org: cleanups]
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Acked-by: Magnus Damm <damm@igel.co.jp>
Tested-by: Magnus Damm <damm@igel.co.jp>
Cc: Torgil Svensson <torgil.svensson@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/spi/Kconfig          |  18 ++-
 drivers/spi/Makefile         |   1 +
 drivers/spi/spi_gpio.c       | 360 +++++++++++++++++++++++++++++++++++++++++++
 include/linux/spi/spi_gpio.h |  60 ++++++++
 4 files changed, 438 insertions(+), 1 deletion(-)
 create mode 100644 drivers/spi/spi_gpio.c
 create mode 100644 include/linux/spi/spi_gpio.h

(limited to 'include/linux')

diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index b9d0efb6803f..4a6fe01831a8 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -78,7 +78,7 @@ config SPI_AU1550
 	  will be called au1550_spi.
 
 config SPI_BITBANG
-	tristate "Bitbanging SPI master"
+	tristate "Utilities for Bitbanging SPI masters"
 	help
 	  With a few GPIO pins, your system can bitbang the SPI protocol.
 	  Select this to get SPI support through I/O pins (GPIO, parallel
@@ -100,6 +100,22 @@ config SPI_BUTTERFLY
 	  inexpensive battery powered microcontroller evaluation board.
 	  This same cable can be used to flash new firmware.
 
+config SPI_GPIO
+	tristate "GPIO-based bitbanging SPI Master"
+	depends on GENERIC_GPIO
+	select SPI_BITBANG
+	help
+	  This simple GPIO bitbanging SPI master uses the arch-neutral GPIO
+	  interface to manage MOSI, MISO, SCK, and chipselect signals.  SPI
+	  slaves connected to a bus using this driver are configured as usual,
+	  except that the spi_board_info.controller_data holds the GPIO number
+	  for the chipselect used by this controller driver.
+
+	  Note that this driver often won't achieve even 1 Mbit/sec speeds,
+	  making it unusually slow for SPI.  If your platform can inline
+	  GPIO operations, you should be able to leverage that for better
+	  speed with a custom version of this driver; see the source code.
+
 config SPI_IMX
 	tristate "Freescale iMX SPI controller"
 	depends on ARCH_IMX && EXPERIMENTAL
diff --git a/drivers/spi/Makefile b/drivers/spi/Makefile
index ccf18de34e1e..5e9f521b8844 100644
--- a/drivers/spi/Makefile
+++ b/drivers/spi/Makefile
@@ -16,6 +16,7 @@ obj-$(CONFIG_SPI_BFIN)			+= spi_bfin5xx.o
 obj-$(CONFIG_SPI_BITBANG)		+= spi_bitbang.o
 obj-$(CONFIG_SPI_AU1550)		+= au1550_spi.o
 obj-$(CONFIG_SPI_BUTTERFLY)		+= spi_butterfly.o
+obj-$(CONFIG_SPI_GPIO)			+= spi_gpio.o
 obj-$(CONFIG_SPI_IMX)			+= spi_imx.o
 obj-$(CONFIG_SPI_LM70_LLP)		+= spi_lm70llp.o
 obj-$(CONFIG_SPI_PXA2XX)		+= pxa2xx_spi.o
diff --git a/drivers/spi/spi_gpio.c b/drivers/spi/spi_gpio.c
new file mode 100644
index 000000000000..49698cabc30d
--- /dev/null
+++ b/drivers/spi/spi_gpio.c
@@ -0,0 +1,360 @@
+/*
+ * spi_gpio.c - SPI master driver using generic bitbanged GPIO
+ *
+ * Copyright (C) 2006,2008 David Brownell
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/gpio.h>
+
+#include <linux/spi/spi.h>
+#include <linux/spi/spi_bitbang.h>
+#include <linux/spi/spi_gpio.h>
+
+
+/*
+ * This bitbanging SPI master driver should help make systems usable
+ * when a native hardware SPI engine is not available, perhaps because
+ * its driver isn't yet working or because the I/O pins it requires
+ * are used for other purposes.
+ *
+ * platform_device->driver_data ... points to spi_gpio
+ *
+ * spi->controller_state ... reserved for bitbang framework code
+ * spi->controller_data ... holds chipselect GPIO
+ *
+ * spi->master->dev.driver_data ... points to spi_gpio->bitbang
+ */
+
+struct spi_gpio {
+	struct spi_bitbang		bitbang;
+	struct spi_gpio_platform_data	pdata;
+	struct platform_device		*pdev;
+};
+
+/*----------------------------------------------------------------------*/
+
+/*
+ * Because the overhead of going through four GPIO procedure calls
+ * per transferred bit can make performance a problem, this code
+ * is set up so that you can use it in either of two ways:
+ *
+ *   - The slow generic way:  set up platform_data to hold the GPIO
+ *     numbers used for MISO/MOSI/SCK, and issue procedure calls for
+ *     each of them.  This driver can handle several such busses.
+ *
+ *   - The quicker inlined way:  only helps with platform GPIO code
+ *     that inlines operations for constant GPIOs.  This can give
+ *     you tight (fast!) inner loops, but each such bus needs a
+ *     new driver.  You'll define a new C file, with Makefile and
+ *     Kconfig support; the C code can be a total of six lines:
+ *
+ *		#define DRIVER_NAME	"myboard_spi2"
+ *		#define	SPI_MISO_GPIO	119
+ *		#define	SPI_MOSI_GPIO	120
+ *		#define	SPI_SCK_GPIO	121
+ *		#define	SPI_N_CHIPSEL	4
+ *		#include "spi_gpio.c"
+ */
+
+#ifndef DRIVER_NAME
+#define DRIVER_NAME	"spi_gpio"
+
+#define GENERIC_BITBANG	/* vs tight inlines */
+
+/* all functions referencing these symbols must define pdata */
+#define SPI_MISO_GPIO	((pdata)->miso)
+#define SPI_MOSI_GPIO	((pdata)->mosi)
+#define SPI_SCK_GPIO	((pdata)->sck)
+
+#define SPI_N_CHIPSEL	((pdata)->num_chipselect)
+
+#endif
+
+/*----------------------------------------------------------------------*/
+
+static inline const struct spi_gpio_platform_data * __pure
+spi_to_pdata(const struct spi_device *spi)
+{
+	const struct spi_bitbang	*bang;
+	const struct spi_gpio		*spi_gpio;
+
+	bang = spi_master_get_devdata(spi->master);
+	spi_gpio = container_of(bang, struct spi_gpio, bitbang);
+	return &spi_gpio->pdata;
+}
+
+/* this is #defined to avoid unused-variable warnings when inlining */
+#define pdata		spi_to_pdata(spi)
+
+static inline void setsck(const struct spi_device *spi, int is_on)
+{
+	gpio_set_value(SPI_SCK_GPIO, is_on);
+}
+
+static inline void setmosi(const struct spi_device *spi, int is_on)
+{
+	gpio_set_value(SPI_MOSI_GPIO, is_on);
+}
+
+static inline int getmiso(const struct spi_device *spi)
+{
+	return gpio_get_value(SPI_MISO_GPIO);
+}
+
+#undef pdata
+
+/*
+ * NOTE:  this clocks "as fast as we can".  It "should" be a function of the
+ * requested device clock.  Software overhead means we usually have trouble
+ * reaching even one Mbit/sec (except when we can inline bitops), so for now
+ * we'll just assume we never need additional per-bit slowdowns.
+ */
+#define spidelay(nsecs)	do {} while (0)
+
+#define	EXPAND_BITBANG_TXRX
+#include <linux/spi/spi_bitbang.h>
+
+/*
+ * These functions can leverage inline expansion of GPIO calls to shrink
+ * costs for a txrx bit, often by factors of around ten (by instruction
+ * count).  That is particularly visible for larger word sizes, but helps
+ * even with default 8-bit words.
+ *
+ * REVISIT overheads calling these functions for each word also have
+ * significant performance costs.  Having txrx_bufs() calls that inline
+ * the txrx_word() logic would help performance, e.g. on larger blocks
+ * used with flash storage or MMC/SD.  There should also be ways to make
+ * GCC be less stupid about reloading registers inside the I/O loops,
+ * even without inlined GPIO calls; __attribute__((hot)) on GCC 4.3?
+ */
+
+static u32 spi_gpio_txrx_word_mode0(struct spi_device *spi,
+		unsigned nsecs, u32 word, u8 bits)
+{
+	return bitbang_txrx_be_cpha0(spi, nsecs, 0, word, bits);
+}
+
+static u32 spi_gpio_txrx_word_mode1(struct spi_device *spi,
+		unsigned nsecs, u32 word, u8 bits)
+{
+	return bitbang_txrx_be_cpha1(spi, nsecs, 0, word, bits);
+}
+
+static u32 spi_gpio_txrx_word_mode2(struct spi_device *spi,
+		unsigned nsecs, u32 word, u8 bits)
+{
+	return bitbang_txrx_be_cpha0(spi, nsecs, 1, word, bits);
+}
+
+static u32 spi_gpio_txrx_word_mode3(struct spi_device *spi,
+		unsigned nsecs, u32 word, u8 bits)
+{
+	return bitbang_txrx_be_cpha1(spi, nsecs, 1, word, bits);
+}
+
+/*----------------------------------------------------------------------*/
+
+static void spi_gpio_chipselect(struct spi_device *spi, int is_active)
+{
+	unsigned long cs = (unsigned long) spi->controller_data;
+
+	/* set initial clock polarity */
+	if (is_active)
+		setsck(spi, spi->mode & SPI_CPOL);
+
+	/* SPI is normally active-low */
+	gpio_set_value(cs, (spi->mode & SPI_CS_HIGH) ? is_active : !is_active);
+}
+
+static int spi_gpio_setup(struct spi_device *spi)
+{
+	unsigned long	cs = (unsigned long) spi->controller_data;
+	int		status = 0;
+
+	if (spi->bits_per_word > 32)
+		return -EINVAL;
+
+	if (!spi->controller_state) {
+		status = gpio_request(cs, spi->dev.bus_id);
+		if (status)
+			return status;
+		status = gpio_direction_output(cs, spi->mode & SPI_CS_HIGH);
+	}
+	if (!status)
+		status = spi_bitbang_setup(spi);
+	if (status) {
+		if (!spi->controller_state)
+			gpio_free(cs);
+	}
+	return status;
+}
+
+static void spi_gpio_cleanup(struct spi_device *spi)
+{
+	unsigned long	cs = (unsigned long) spi->controller_data;
+
+	gpio_free(cs);
+	spi_bitbang_cleanup(spi);
+}
+
+static int __init spi_gpio_alloc(unsigned pin, const char *label, bool is_in)
+{
+	int value;
+
+	value = gpio_request(pin, label);
+	if (value == 0) {
+		if (is_in)
+			value = gpio_direction_input(pin);
+		else
+			value = gpio_direction_output(pin, 0);
+	}
+	return value;
+}
+
+static int __init
+spi_gpio_request(struct spi_gpio_platform_data *pdata, const char *label)
+{
+	int value;
+
+	/* NOTE:  SPI_*_GPIO symbols may reference "pdata" */
+
+	value = spi_gpio_alloc(SPI_MOSI_GPIO, label, false);
+	if (value)
+		goto done;
+
+	value = spi_gpio_alloc(SPI_MISO_GPIO, label, true);
+	if (value)
+		goto free_mosi;
+
+	value = spi_gpio_alloc(SPI_SCK_GPIO, label, false);
+	if (value)
+		goto free_miso;
+
+	goto done;
+
+free_miso:
+	gpio_free(SPI_MISO_GPIO);
+free_mosi:
+	gpio_free(SPI_MOSI_GPIO);
+done:
+	return value;
+}
+
+static int __init spi_gpio_probe(struct platform_device *pdev)
+{
+	int				status;
+	struct spi_master		*master;
+	struct spi_gpio			*spi_gpio;
+	struct spi_gpio_platform_data	*pdata;
+
+	pdata = pdev->dev.platform_data;
+#ifdef GENERIC_BITBANG
+	if (!pdata || !pdata->num_chipselect)
+		return -ENODEV;
+#endif
+
+	status = spi_gpio_request(pdata, dev_name(&pdev->dev));
+	if (status < 0)
+		return status;
+
+	master = spi_alloc_master(&pdev->dev, sizeof *spi_gpio);
+	if (!master) {
+		status = -ENOMEM;
+		goto gpio_free;
+	}
+	spi_gpio = spi_master_get_devdata(master);
+	platform_set_drvdata(pdev, spi_gpio);
+
+	spi_gpio->pdev = pdev;
+	if (pdata)
+		spi_gpio->pdata = *pdata;
+
+	master->bus_num = pdev->id;
+	master->num_chipselect = SPI_N_CHIPSEL;
+	master->setup = spi_gpio_setup;
+	master->cleanup = spi_gpio_cleanup;
+
+	spi_gpio->bitbang.master = spi_master_get(master);
+	spi_gpio->bitbang.chipselect = spi_gpio_chipselect;
+	spi_gpio->bitbang.txrx_word[SPI_MODE_0] = spi_gpio_txrx_word_mode0;
+	spi_gpio->bitbang.txrx_word[SPI_MODE_1] = spi_gpio_txrx_word_mode1;
+	spi_gpio->bitbang.txrx_word[SPI_MODE_2] = spi_gpio_txrx_word_mode2;
+	spi_gpio->bitbang.txrx_word[SPI_MODE_3] = spi_gpio_txrx_word_mode3;
+	spi_gpio->bitbang.setup_transfer = spi_bitbang_setup_transfer;
+	spi_gpio->bitbang.flags = SPI_CS_HIGH;
+
+	status = spi_bitbang_start(&spi_gpio->bitbang);
+	if (status < 0) {
+		spi_master_put(spi_gpio->bitbang.master);
+gpio_free:
+		gpio_free(SPI_MISO_GPIO);
+		gpio_free(SPI_MOSI_GPIO);
+		gpio_free(SPI_SCK_GPIO);
+		spi_master_put(master);
+	}
+
+	return status;
+}
+
+static int __exit spi_gpio_remove(struct platform_device *pdev)
+{
+	struct spi_gpio			*spi_gpio;
+	struct spi_gpio_platform_data	*pdata;
+	int				status;
+
+	spi_gpio = platform_get_drvdata(pdev);
+	pdata = pdev->dev.platform_data;
+
+	/* stop() unregisters child devices too */
+	status = spi_bitbang_stop(&spi_gpio->bitbang);
+	spi_master_put(spi_gpio->bitbang.master);
+
+	platform_set_drvdata(pdev, NULL);
+
+	gpio_free(SPI_MISO_GPIO);
+	gpio_free(SPI_MOSI_GPIO);
+	gpio_free(SPI_SCK_GPIO);
+
+	return status;
+}
+
+MODULE_ALIAS("platform:" DRIVER_NAME);
+
+static struct platform_driver spi_gpio_driver = {
+	.driver.name	= DRIVER_NAME,
+	.driver.owner	= THIS_MODULE,
+	.remove		= __exit_p(spi_gpio_remove),
+};
+
+static int __init spi_gpio_init(void)
+{
+	return platform_driver_probe(&spi_gpio_driver, spi_gpio_probe);
+}
+module_init(spi_gpio_init);
+
+static void __exit spi_gpio_exit(void)
+{
+	platform_driver_unregister(&spi_gpio_driver);
+}
+module_exit(spi_gpio_exit);
+
+
+MODULE_DESCRIPTION("SPI master driver using generic bitbanged GPIO ");
+MODULE_AUTHOR("David Brownell");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/spi/spi_gpio.h b/include/linux/spi/spi_gpio.h
new file mode 100644
index 000000000000..0f01a0f1f40c
--- /dev/null
+++ b/include/linux/spi/spi_gpio.h
@@ -0,0 +1,60 @@
+#ifndef __LINUX_SPI_GPIO_H
+#define __LINUX_SPI_GPIO_H
+
+/*
+ * For each bitbanged SPI bus, set up a platform_device node with:
+ *   - name "spi_gpio"
+ *   - id the same as the SPI bus number it implements
+ *   - dev.platform data pointing to a struct spi_gpio_platform_data
+ *
+ * Or, see the driver code for information about speedups that are
+ * possible on platforms that support inlined access for GPIOs (no
+ * spi_gpio_platform_data is used).
+ *
+ * Use spi_board_info with these busses in the usual way, being sure
+ * that the controller_data being the GPIO used for each device's
+ * chipselect:
+ *
+ *	static struct spi_board_info ... [] = {
+ *	...
+ *		// this slave uses GPIO 42 for its chipselect
+ *		.controller_data = (void *) 42,
+ *	...
+ *		// this one uses GPIO 86 for its chipselect
+ *		.controller_data = (void *) 86,
+ *	...
+ *	};
+ *
+ * If the bitbanged bus is later switched to a "native" controller,
+ * that platform_device and controller_data should be removed.
+ */
+
+/**
+ * struct spi_gpio_platform_data - parameter for bitbanged SPI master
+ * @sck: number of the GPIO used for clock output
+ * @mosi: number of the GPIO used for Master Output, Slave In (MOSI) data
+ * @miso: number of the GPIO used for Master Input, Slave Output (MISO) data
+ * @num_chipselect: how many slaves to allow
+ *
+ * All GPIO signals used with the SPI bus managed through this driver
+ * (chipselects, MOSI, MISO, SCK) must be configured as GPIOs, instead
+ * of some alternate function.
+ *
+ * It can be convenient to use this driver with pins that have alternate
+ * functions associated with a "native" SPI controller if a driver for that
+ * controller is not available, or is missing important functionality.
+ *
+ * On platforms which can do so, configure MISO with a weak pullup unless
+ * there's an external pullup on that signal.  That saves power by avoiding
+ * floating signals.  (A weak pulldown would save power too, but many
+ * drivers expect to see all-ones data as the no slave "response".)
+ */
+struct spi_gpio_platform_data {
+	unsigned	sck;
+	unsigned	mosi;
+	unsigned	miso;
+
+	u16		num_chipselect;
+};
+
+#endif /* __LINUX_SPI_GPIO_H */
-- 
cgit v1.2.3


From a06f6211ef9b1785922f9d0e8766d63ac4e66de1 Mon Sep 17 00:00:00 2001
From: Masami Hiramatsu <mhiramat@redhat.com>
Date: Tue, 6 Jan 2009 14:41:49 -0800
Subject: module: add within_module_core() and within_module_init()

This series of patches allows kprobes to probe module's __init and __exit
functions.  This means, you can probe driver initialization and
terminating.

Currently, kprobes can't probe __init function because these functions are
freed after module initialization.  And it also can't probe module __exit
functions because kprobe increments reference count of target module and
user can't unload it.  this means __exit functions never be called unless
removing probes from the module.

To solve both cases, this series of patches introduces GONE flag and sets
it when the target code is freed(for this purpose, kprobes hooks
MODULE_STATE_* events).  This also removes refcount incrementing for
allowing user to unload target module.  Users can check which probes are
GONE by debugfs interface.  For taking timing of freeing module's .init
text, these also include a patch which adds module's notifier of
MODULE_STATE_LIVE event.

This patch:

Add within_module_core() and within_module_init() for checking whether an
address is in the module .init.text section or .text section, and replace
within() local inline functions in kernel/module.c with them.

kprobes uses these functions to check where the kprobe is inserted.

Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/module.h | 12 ++++++++++++
 kernel/module.c        | 16 ++++++++--------
 2 files changed, 20 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/module.h b/include/linux/module.h
index 03cb93d1865a..4f7ea12463d3 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -365,6 +365,18 @@ struct module *module_text_address(unsigned long addr);
 struct module *__module_text_address(unsigned long addr);
 int is_module_address(unsigned long addr);
 
+static inline int within_module_core(unsigned long addr, struct module *mod)
+{
+	return (unsigned long)mod->module_core <= addr &&
+	       addr < (unsigned long)mod->module_core + mod->core_size;
+}
+
+static inline int within_module_init(unsigned long addr, struct module *mod)
+{
+	return (unsigned long)mod->module_init <= addr &&
+	       addr < (unsigned long)mod->module_init + mod->init_size;
+}
+
 /* Returns 0 and fills in value, defined and namebuf, or -ERANGE if
    symnum out of range. */
 int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type,
diff --git a/kernel/module.c b/kernel/module.c
index 34b56cf06615..cc79c942c572 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -2390,7 +2390,7 @@ static const char *get_ksymbol(struct module *mod,
 	unsigned long nextval;
 
 	/* At worse, next value is at end of module */
-	if (within(addr, mod->module_init, mod->init_size))
+	if (within_module_init(addr, mod))
 		nextval = (unsigned long)mod->module_init+mod->init_text_size;
 	else
 		nextval = (unsigned long)mod->module_core+mod->core_text_size;
@@ -2438,8 +2438,8 @@ const char *module_address_lookup(unsigned long addr,
 
 	preempt_disable();
 	list_for_each_entry_rcu(mod, &modules, list) {
-		if (within(addr, mod->module_init, mod->init_size)
-		    || within(addr, mod->module_core, mod->core_size)) {
+		if (within_module_init(addr, mod) ||
+		    within_module_core(addr, mod)) {
 			if (modname)
 				*modname = mod->name;
 			ret = get_ksymbol(mod, addr, size, offset);
@@ -2461,8 +2461,8 @@ int lookup_module_symbol_name(unsigned long addr, char *symname)
 
 	preempt_disable();
 	list_for_each_entry_rcu(mod, &modules, list) {
-		if (within(addr, mod->module_init, mod->init_size) ||
-		    within(addr, mod->module_core, mod->core_size)) {
+		if (within_module_init(addr, mod) ||
+		    within_module_core(addr, mod)) {
 			const char *sym;
 
 			sym = get_ksymbol(mod, addr, NULL, NULL);
@@ -2485,8 +2485,8 @@ int lookup_module_symbol_attrs(unsigned long addr, unsigned long *size,
 
 	preempt_disable();
 	list_for_each_entry_rcu(mod, &modules, list) {
-		if (within(addr, mod->module_init, mod->init_size) ||
-		    within(addr, mod->module_core, mod->core_size)) {
+		if (within_module_init(addr, mod) ||
+		    within_module_core(addr, mod)) {
 			const char *sym;
 
 			sym = get_ksymbol(mod, addr, size, offset);
@@ -2705,7 +2705,7 @@ int is_module_address(unsigned long addr)
 	preempt_disable();
 
 	list_for_each_entry_rcu(mod, &modules, list) {
-		if (within(addr, mod->module_core, mod->core_size)) {
+		if (within_module_core(addr, mod)) {
 			preempt_enable();
 			return 1;
 		}
-- 
cgit v1.2.3


From 129415607845d4daea11ddcba706005c69dcb942 Mon Sep 17 00:00:00 2001
From: Masami Hiramatsu <mhiramat@redhat.com>
Date: Tue, 6 Jan 2009 14:41:50 -0800
Subject: kprobes: add kprobe_insn_mutex and cleanup arch_remove_kprobe()

Add kprobe_insn_mutex for protecting kprobe_insn_pages hlist, and remove
kprobe_mutex from architecture dependent code.

This allows us to call arch_remove_kprobe() (and free_insn_slot) while
holding kprobe_mutex.

Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/arm/kernel/kprobes.c     |  2 --
 arch/ia64/kernel/kprobes.c    |  8 +++++---
 arch/powerpc/kernel/kprobes.c |  7 ++++---
 arch/s390/kernel/kprobes.c    |  7 ++++---
 arch/x86/kernel/kprobes.c     |  7 ++++---
 include/linux/kprobes.h       |  1 -
 kernel/kprobes.c              | 25 +++++++++++++++++++++----
 7 files changed, 38 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/kernel/kprobes.c b/arch/arm/kernel/kprobes.c
index 3f9abe0e9aff..f692efddd449 100644
--- a/arch/arm/kernel/kprobes.c
+++ b/arch/arm/kernel/kprobes.c
@@ -92,9 +92,7 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
 	if (p->ainsn.insn) {
-		mutex_lock(&kprobe_mutex);
 		free_insn_slot(p->ainsn.insn, 0);
-		mutex_unlock(&kprobe_mutex);
 		p->ainsn.insn = NULL;
 	}
 }
diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index f07688da947c..097b84d54e73 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -670,9 +670,11 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 
 void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
-	mutex_lock(&kprobe_mutex);
-	free_insn_slot(p->ainsn.insn, p->ainsn.inst_flag & INST_FLAG_BOOSTABLE);
-	mutex_unlock(&kprobe_mutex);
+	if (p->ainsn.insn) {
+		free_insn_slot(p->ainsn.insn,
+			       p->ainsn.inst_flag & INST_FLAG_BOOSTABLE);
+		p->ainsn.insn = NULL;
+	}
 }
 /*
  * We are resuming execution after a single step fault, so the pt_regs
diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c
index de79915452c8..989edcdf0297 100644
--- a/arch/powerpc/kernel/kprobes.c
+++ b/arch/powerpc/kernel/kprobes.c
@@ -96,9 +96,10 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 
 void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
-	mutex_lock(&kprobe_mutex);
-	free_insn_slot(p->ainsn.insn, 0);
-	mutex_unlock(&kprobe_mutex);
+	if (p->ainsn.insn) {
+		free_insn_slot(p->ainsn.insn, 0);
+		p->ainsn.insn = NULL;
+	}
 }
 
 static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c
index 569079ec4ff0..9b92856632cf 100644
--- a/arch/s390/kernel/kprobes.c
+++ b/arch/s390/kernel/kprobes.c
@@ -218,9 +218,10 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 
 void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
-	mutex_lock(&kprobe_mutex);
-	free_insn_slot(p->ainsn.insn, 0);
-	mutex_unlock(&kprobe_mutex);
+	if (p->ainsn.insn) {
+		free_insn_slot(p->ainsn.insn, 0);
+		p->ainsn.insn = NULL;
+	}
 }
 
 static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 6c27679ec6aa..eead6f8f9218 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -376,9 +376,10 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 
 void __kprobes arch_remove_kprobe(struct kprobe *p)
 {
-	mutex_lock(&kprobe_mutex);
-	free_insn_slot(p->ainsn.insn, (p->ainsn.boostable == 1));
-	mutex_unlock(&kprobe_mutex);
+	if (p->ainsn.insn) {
+		free_insn_slot(p->ainsn.insn, (p->ainsn.boostable == 1));
+		p->ainsn.insn = NULL;
+	}
 }
 
 static void __kprobes save_previous_kprobe(struct kprobe_ctlblk *kcb)
diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index 497b1d1f7a05..b93e44ce2284 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -201,7 +201,6 @@ static inline int init_test_probes(void)
 }
 #endif /* CONFIG_KPROBES_SANITY_TEST */
 
-extern struct mutex kprobe_mutex;
 extern int arch_prepare_kprobe(struct kprobe *p);
 extern void arch_arm_kprobe(struct kprobe *p);
 extern void arch_disarm_kprobe(struct kprobe *p);
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 3afd354c46f1..29e87921437d 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -69,7 +69,7 @@ static struct hlist_head kretprobe_inst_table[KPROBE_TABLE_SIZE];
 /* NOTE: change this value only with kprobe_mutex held */
 static bool kprobe_enabled;
 
-DEFINE_MUTEX(kprobe_mutex);		/* Protects kprobe_table */
+static DEFINE_MUTEX(kprobe_mutex);	/* Protects kprobe_table */
 static DEFINE_PER_CPU(struct kprobe *, kprobe_instance) = NULL;
 static struct {
 	spinlock_t lock ____cacheline_aligned_in_smp;
@@ -115,6 +115,7 @@ enum kprobe_slot_state {
 	SLOT_USED = 2,
 };
 
+static DEFINE_MUTEX(kprobe_insn_mutex);	/* Protects kprobe_insn_pages */
 static struct hlist_head kprobe_insn_pages;
 static int kprobe_garbage_slots;
 static int collect_garbage_slots(void);
@@ -144,10 +145,10 @@ loop_end:
 }
 
 /**
- * get_insn_slot() - Find a slot on an executable page for an instruction.
+ * __get_insn_slot() - Find a slot on an executable page for an instruction.
  * We allocate an executable page if there's no room on existing ones.
  */
-kprobe_opcode_t __kprobes *get_insn_slot(void)
+static kprobe_opcode_t __kprobes *__get_insn_slot(void)
 {
 	struct kprobe_insn_page *kip;
 	struct hlist_node *pos;
@@ -196,6 +197,15 @@ kprobe_opcode_t __kprobes *get_insn_slot(void)
 	return kip->insns;
 }
 
+kprobe_opcode_t __kprobes *get_insn_slot(void)
+{
+	kprobe_opcode_t *ret;
+	mutex_lock(&kprobe_insn_mutex);
+	ret = __get_insn_slot();
+	mutex_unlock(&kprobe_insn_mutex);
+	return ret;
+}
+
 /* Return 1 if all garbages are collected, otherwise 0. */
 static int __kprobes collect_one_slot(struct kprobe_insn_page *kip, int idx)
 {
@@ -226,9 +236,13 @@ static int __kprobes collect_garbage_slots(void)
 {
 	struct kprobe_insn_page *kip;
 	struct hlist_node *pos, *next;
+	int safety;
 
 	/* Ensure no-one is preepmted on the garbages */
-	if (check_safety() != 0)
+	mutex_unlock(&kprobe_insn_mutex);
+	safety = check_safety();
+	mutex_lock(&kprobe_insn_mutex);
+	if (safety != 0)
 		return -EAGAIN;
 
 	hlist_for_each_entry_safe(kip, pos, next, &kprobe_insn_pages, hlist) {
@@ -251,6 +265,7 @@ void __kprobes free_insn_slot(kprobe_opcode_t * slot, int dirty)
 	struct kprobe_insn_page *kip;
 	struct hlist_node *pos;
 
+	mutex_lock(&kprobe_insn_mutex);
 	hlist_for_each_entry(kip, pos, &kprobe_insn_pages, hlist) {
 		if (kip->insns <= slot &&
 		    slot < kip->insns + (INSNS_PER_PAGE * MAX_INSN_SIZE)) {
@@ -267,6 +282,8 @@ void __kprobes free_insn_slot(kprobe_opcode_t * slot, int dirty)
 
 	if (dirty && ++kprobe_garbage_slots > INSNS_PER_PAGE)
 		collect_garbage_slots();
+
+	mutex_unlock(&kprobe_insn_mutex);
 }
 #endif
 
-- 
cgit v1.2.3


From e8386a0cb22f4a2d439384212c494ad0bda848fe Mon Sep 17 00:00:00 2001
From: Masami Hiramatsu <mhiramat@redhat.com>
Date: Tue, 6 Jan 2009 14:41:52 -0800
Subject: kprobes: support probing module __exit function

Allows kprobes to probe __exit routine.  This adds flags member to struct
kprobe.  When module is freed(kprobes hooks module_notifier to get this
event), kprobes which probe the functions in that module are set to "Gone"
flag to the flags member.  These "Gone" probes are never be enabled.
Users can check the GONE flag through debugfs.

This also removes mod_refcounted, because we couldn't free a module if
kprobe incremented the refcount of that module.

[akpm@linux-foundation.org: document some locking]
[mhiramat@redhat.com: bugfix: pass aggr_kprobe to arch_remove_kprobe]
[mhiramat@redhat.com: bugfix: release old_p's insn_slot before error return]
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/kprobes.txt |   5 +-
 include/linux/kprobes.h   |  14 +++-
 kernel/kprobes.c          | 159 ++++++++++++++++++++++++++++++++++------------
 3 files changed, 134 insertions(+), 44 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/kprobes.txt b/Documentation/kprobes.txt
index a79633d702bf..48b3de90eb1e 100644
--- a/Documentation/kprobes.txt
+++ b/Documentation/kprobes.txt
@@ -497,7 +497,10 @@ The first column provides the kernel address where the probe is inserted.
 The second column identifies the type of probe (k - kprobe, r - kretprobe
 and j - jprobe), while the third column specifies the symbol+offset of
 the probe. If the probed function belongs to a module, the module name
-is also specified.
+is also specified. Following columns show probe status. If the probe is on
+a virtual address that is no longer valid (module init sections, module
+virtual addresses that correspond to modules that've been unloaded),
+such probes are marked with [GONE].
 
 /debug/kprobes/enabled: Turn kprobes ON/OFF
 
diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index b93e44ce2284..d6ea19e314bb 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -69,9 +69,6 @@ struct kprobe {
 	/* list of kprobes for multi-handler support */
 	struct list_head list;
 
-	/* Indicates that the corresponding module has been ref counted */
-	unsigned int mod_refcounted;
-
 	/*count the number of times this probe was temporarily disarmed */
 	unsigned long nmissed;
 
@@ -103,8 +100,19 @@ struct kprobe {
 
 	/* copy of the original instruction */
 	struct arch_specific_insn ainsn;
+
+	/* Indicates various status flags.  Protected by kprobe_mutex. */
+	u32 flags;
 };
 
+/* Kprobe status flags */
+#define KPROBE_FLAG_GONE	1 /* breakpoint has already gone */
+
+static inline int kprobe_gone(struct kprobe *p)
+{
+	return p->flags & KPROBE_FLAG_GONE;
+}
+
 /*
  * Special probe type that uses setjmp-longjmp type tricks to resume
  * execution at a specified entry with a matching prototype corresponding
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index a1e233a19586..cb732a9aa55f 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -327,7 +327,7 @@ static int __kprobes aggr_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	struct kprobe *kp;
 
 	list_for_each_entry_rcu(kp, &p->list, list) {
-		if (kp->pre_handler) {
+		if (kp->pre_handler && !kprobe_gone(kp)) {
 			set_kprobe_instance(kp);
 			if (kp->pre_handler(kp, regs))
 				return 1;
@@ -343,7 +343,7 @@ static void __kprobes aggr_post_handler(struct kprobe *p, struct pt_regs *regs,
 	struct kprobe *kp;
 
 	list_for_each_entry_rcu(kp, &p->list, list) {
-		if (kp->post_handler) {
+		if (kp->post_handler && !kprobe_gone(kp)) {
 			set_kprobe_instance(kp);
 			kp->post_handler(kp, regs, flags);
 			reset_kprobe_instance();
@@ -545,9 +545,10 @@ static inline void add_aggr_kprobe(struct kprobe *ap, struct kprobe *p)
 	ap->addr = p->addr;
 	ap->pre_handler = aggr_pre_handler;
 	ap->fault_handler = aggr_fault_handler;
-	if (p->post_handler)
+	/* We don't care the kprobe which has gone. */
+	if (p->post_handler && !kprobe_gone(p))
 		ap->post_handler = aggr_post_handler;
-	if (p->break_handler)
+	if (p->break_handler && !kprobe_gone(p))
 		ap->break_handler = aggr_break_handler;
 
 	INIT_LIST_HEAD(&ap->list);
@@ -566,17 +567,41 @@ static int __kprobes register_aggr_kprobe(struct kprobe *old_p,
 	int ret = 0;
 	struct kprobe *ap;
 
+	if (kprobe_gone(old_p)) {
+		/*
+		 * Attempting to insert new probe at the same location that
+		 * had a probe in the module vaddr area which already
+		 * freed. So, the instruction slot has already been
+		 * released. We need a new slot for the new probe.
+		 */
+		ret = arch_prepare_kprobe(old_p);
+		if (ret)
+			return ret;
+	}
 	if (old_p->pre_handler == aggr_pre_handler) {
 		copy_kprobe(old_p, p);
 		ret = add_new_kprobe(old_p, p);
+		ap = old_p;
 	} else {
 		ap = kzalloc(sizeof(struct kprobe), GFP_KERNEL);
-		if (!ap)
+		if (!ap) {
+			if (kprobe_gone(old_p))
+				arch_remove_kprobe(old_p);
 			return -ENOMEM;
+		}
 		add_aggr_kprobe(ap, old_p);
 		copy_kprobe(ap, p);
 		ret = add_new_kprobe(ap, p);
 	}
+	if (kprobe_gone(old_p)) {
+		/*
+		 * If the old_p has gone, its breakpoint has been disarmed.
+		 * We have to arm it again after preparing real kprobes.
+		 */
+		ap->flags &= ~KPROBE_FLAG_GONE;
+		if (kprobe_enabled)
+			arch_arm_kprobe(ap);
+	}
 	return ret;
 }
 
@@ -639,8 +664,7 @@ static int __kprobes __register_kprobe(struct kprobe *p,
 		return -EINVAL;
 	}
 
-	p->mod_refcounted = 0;
-
+	p->flags = 0;
 	/*
 	 * Check if are we probing a module.
 	 */
@@ -649,16 +673,14 @@ static int __kprobes __register_kprobe(struct kprobe *p,
 		struct module *calling_mod;
 		calling_mod = __module_text_address(called_from);
 		/*
-		 * We must allow modules to probe themself and in this case
-		 * avoid incrementing the module refcount, so as to allow
-		 * unloading of self probing modules.
+		 * We must hold a refcount of the probed module while updating
+		 * its code to prohibit unexpected unloading.
 		 */
 		if (calling_mod != probed_mod) {
 			if (unlikely(!try_module_get(probed_mod))) {
 				preempt_enable();
 				return -EINVAL;
 			}
-			p->mod_refcounted = 1;
 		} else
 			probed_mod = NULL;
 	}
@@ -687,8 +709,9 @@ static int __kprobes __register_kprobe(struct kprobe *p,
 out:
 	mutex_unlock(&kprobe_mutex);
 
-	if (ret && probed_mod)
+	if (probed_mod)
 		module_put(probed_mod);
+
 	return ret;
 }
 
@@ -716,16 +739,16 @@ valid_p:
 	     list_is_singular(&old_p->list))) {
 		/*
 		 * Only probe on the hash list. Disarm only if kprobes are
-		 * enabled - otherwise, the breakpoint would already have
-		 * been removed. We save on flushing icache.
+		 * enabled and not gone - otherwise, the breakpoint would
+		 * already have been removed. We save on flushing icache.
 		 */
-		if (kprobe_enabled)
+		if (kprobe_enabled && !kprobe_gone(old_p))
 			arch_disarm_kprobe(p);
 		hlist_del_rcu(&old_p->hlist);
 	} else {
-		if (p->break_handler)
+		if (p->break_handler && !kprobe_gone(p))
 			old_p->break_handler = NULL;
-		if (p->post_handler) {
+		if (p->post_handler && !kprobe_gone(p)) {
 			list_for_each_entry_rcu(list_p, &old_p->list, list) {
 				if ((list_p != p) && (list_p->post_handler))
 					goto noclean;
@@ -740,27 +763,16 @@ noclean:
 
 static void __kprobes __unregister_kprobe_bottom(struct kprobe *p)
 {
-	struct module *mod;
 	struct kprobe *old_p;
 
-	if (p->mod_refcounted) {
-		/*
-		 * Since we've already incremented refcount,
-		 * we don't need to disable preemption.
-		 */
-		mod = module_text_address((unsigned long)p->addr);
-		if (mod)
-			module_put(mod);
-	}
-
-	if (list_empty(&p->list) || list_is_singular(&p->list)) {
-		if (!list_empty(&p->list)) {
-			/* "p" is the last child of an aggr_kprobe */
-			old_p = list_entry(p->list.next, struct kprobe, list);
-			list_del(&p->list);
-			kfree(old_p);
-		}
+	if (list_empty(&p->list))
 		arch_remove_kprobe(p);
+	else if (list_is_singular(&p->list)) {
+		/* "p" is the last child of an aggr_kprobe */
+		old_p = list_entry(p->list.next, struct kprobe, list);
+		list_del(&p->list);
+		arch_remove_kprobe(old_p);
+		kfree(old_p);
 	}
 }
 
@@ -1074,6 +1086,67 @@ static int __kprobes pre_handler_kretprobe(struct kprobe *p,
 
 #endif /* CONFIG_KRETPROBES */
 
+/* Set the kprobe gone and remove its instruction buffer. */
+static void __kprobes kill_kprobe(struct kprobe *p)
+{
+	struct kprobe *kp;
+	p->flags |= KPROBE_FLAG_GONE;
+	if (p->pre_handler == aggr_pre_handler) {
+		/*
+		 * If this is an aggr_kprobe, we have to list all the
+		 * chained probes and mark them GONE.
+		 */
+		list_for_each_entry_rcu(kp, &p->list, list)
+			kp->flags |= KPROBE_FLAG_GONE;
+		p->post_handler = NULL;
+		p->break_handler = NULL;
+	}
+	/*
+	 * Here, we can remove insn_slot safely, because no thread calls
+	 * the original probed function (which will be freed soon) any more.
+	 */
+	arch_remove_kprobe(p);
+}
+
+/* Module notifier call back, checking kprobes on the module */
+static int __kprobes kprobes_module_callback(struct notifier_block *nb,
+					     unsigned long val, void *data)
+{
+	struct module *mod = data;
+	struct hlist_head *head;
+	struct hlist_node *node;
+	struct kprobe *p;
+	unsigned int i;
+
+	if (val != MODULE_STATE_GOING)
+		return NOTIFY_DONE;
+
+	/*
+	 * module .text section will be freed. We need to
+	 * disable kprobes which have been inserted in the section.
+	 */
+	mutex_lock(&kprobe_mutex);
+	for (i = 0; i < KPROBE_TABLE_SIZE; i++) {
+		head = &kprobe_table[i];
+		hlist_for_each_entry_rcu(p, node, head, hlist)
+			if (within_module_core((unsigned long)p->addr, mod)) {
+				/*
+				 * The vaddr this probe is installed will soon
+				 * be vfreed buy not synced to disk. Hence,
+				 * disarming the breakpoint isn't needed.
+				 */
+				kill_kprobe(p);
+			}
+	}
+	mutex_unlock(&kprobe_mutex);
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block kprobe_module_nb = {
+	.notifier_call = kprobes_module_callback,
+	.priority = 0
+};
+
 static int __init init_kprobes(void)
 {
 	int i, err = 0;
@@ -1130,6 +1203,9 @@ static int __init init_kprobes(void)
 	err = arch_init_kprobes();
 	if (!err)
 		err = register_die_notifier(&kprobe_exceptions_nb);
+	if (!err)
+		err = register_module_notifier(&kprobe_module_nb);
+
 	kprobes_initialized = (err == 0);
 
 	if (!err)
@@ -1150,10 +1226,12 @@ static void __kprobes report_probe(struct seq_file *pi, struct kprobe *p,
 	else
 		kprobe_type = "k";
 	if (sym)
-		seq_printf(pi, "%p  %s  %s+0x%x  %s\n", p->addr, kprobe_type,
-			sym, offset, (modname ? modname : " "));
+		seq_printf(pi, "%p  %s  %s+0x%x  %s %s\n", p->addr, kprobe_type,
+			sym, offset, (modname ? modname : " "),
+			(kprobe_gone(p) ? "[GONE]" : ""));
 	else
-		seq_printf(pi, "%p  %s  %p\n", p->addr, kprobe_type, p->addr);
+		seq_printf(pi, "%p  %s  %p %s\n", p->addr, kprobe_type, p->addr,
+			(kprobe_gone(p) ? "[GONE]" : ""));
 }
 
 static void __kprobes *kprobe_seq_start(struct seq_file *f, loff_t *pos)
@@ -1234,7 +1312,8 @@ static void __kprobes enable_all_kprobes(void)
 	for (i = 0; i < KPROBE_TABLE_SIZE; i++) {
 		head = &kprobe_table[i];
 		hlist_for_each_entry_rcu(p, node, head, hlist)
-			arch_arm_kprobe(p);
+			if (!kprobe_gone(p))
+				arch_arm_kprobe(p);
 	}
 
 	kprobe_enabled = true;
@@ -1263,7 +1342,7 @@ static void __kprobes disable_all_kprobes(void)
 	for (i = 0; i < KPROBE_TABLE_SIZE; i++) {
 		head = &kprobe_table[i];
 		hlist_for_each_entry_rcu(p, node, head, hlist) {
-			if (!arch_trampoline_kprobe(p))
+			if (!arch_trampoline_kprobe(p) && !kprobe_gone(p))
 				arch_disarm_kprobe(p);
 		}
 	}
-- 
cgit v1.2.3


From 730c9eeca9808fc2cfb506cc68c90aa330da17b0 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Tue, 6 Jan 2009 14:42:06 -0800
Subject: autofs4: improve parameter usage

The parameter usage in the device node ioctl code uses arg1 and arg2 as
parameter names.  This patch redefines the parameter names to reflect what
they actually are in an effort to make the code more readable.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/dev-ioctl.c         | 54 ++++++++++++++++--------------
 include/linux/auto_dev-ioctl.h | 75 +++++++++++++++++++++++++++++++++++++++---
 2 files changed, 100 insertions(+), 29 deletions(-)

(limited to 'include/linux')

diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c
index 63b7c7afe8df..054d6d9ad9ba 100644
--- a/fs/autofs4/dev-ioctl.c
+++ b/fs/autofs4/dev-ioctl.c
@@ -180,7 +180,7 @@ static int autofs_dev_ioctl_protover(struct file *fp,
 				     struct autofs_sb_info *sbi,
 				     struct autofs_dev_ioctl *param)
 {
-	param->arg1 = sbi->version;
+	param->protover.version = sbi->version;
 	return 0;
 }
 
@@ -189,7 +189,7 @@ static int autofs_dev_ioctl_protosubver(struct file *fp,
 					struct autofs_sb_info *sbi,
 					struct autofs_dev_ioctl *param)
 {
-	param->arg1 = sbi->sub_version;
+	param->protosubver.sub_version = sbi->sub_version;
 	return 0;
 }
 
@@ -335,13 +335,13 @@ static int autofs_dev_ioctl_openmount(struct file *fp,
 	int err, fd;
 
 	/* param->path has already been checked */
-	if (!param->arg1)
+	if (!param->openmount.devid)
 		return -EINVAL;
 
 	param->ioctlfd = -1;
 
 	path = param->path;
-	devid = param->arg1;
+	devid = param->openmount.devid;
 
 	err = 0;
 	fd = autofs_dev_ioctl_open_mountpoint(path, devid);
@@ -373,7 +373,7 @@ static int autofs_dev_ioctl_ready(struct file *fp,
 {
 	autofs_wqt_t token;
 
-	token = (autofs_wqt_t) param->arg1;
+	token = (autofs_wqt_t) param->ready.token;
 	return autofs4_wait_release(sbi, token, 0);
 }
 
@@ -388,8 +388,8 @@ static int autofs_dev_ioctl_fail(struct file *fp,
 	autofs_wqt_t token;
 	int status;
 
-	token = (autofs_wqt_t) param->arg1;
-	status = param->arg2 ? param->arg2 : -ENOENT;
+	token = (autofs_wqt_t) param->fail.token;
+	status = param->fail.status ? param->fail.status : -ENOENT;
 	return autofs4_wait_release(sbi, token, status);
 }
 
@@ -412,10 +412,10 @@ static int autofs_dev_ioctl_setpipefd(struct file *fp,
 	int pipefd;
 	int err = 0;
 
-	if (param->arg1 == -1)
+	if (param->setpipefd.pipefd == -1)
 		return -EINVAL;
 
-	pipefd = param->arg1;
+	pipefd = param->setpipefd.pipefd;
 
 	mutex_lock(&sbi->wq_mutex);
 	if (!sbi->catatonic) {
@@ -457,8 +457,8 @@ static int autofs_dev_ioctl_timeout(struct file *fp,
 {
 	unsigned long timeout;
 
-	timeout = param->arg1;
-	param->arg1 = sbi->exp_timeout / HZ;
+	timeout = param->timeout.timeout;
+	param->timeout.timeout = sbi->exp_timeout / HZ;
 	sbi->exp_timeout = timeout * HZ;
 	return 0;
 }
@@ -489,7 +489,7 @@ static int autofs_dev_ioctl_requester(struct file *fp,
 	path = param->path;
 	devid = sbi->sb->s_dev;
 
-	param->arg1 = param->arg2 = -1;
+	param->requester.uid = param->requester.gid = -1;
 
 	/* Get nameidata of the parent directory */
 	err = path_lookup(path, LOOKUP_PARENT, &nd);
@@ -505,8 +505,8 @@ static int autofs_dev_ioctl_requester(struct file *fp,
 		err = 0;
 		autofs4_expire_wait(nd.path.dentry);
 		spin_lock(&sbi->fs_lock);
-		param->arg1 = ino->uid;
-		param->arg2 = ino->gid;
+		param->requester.uid = ino->uid;
+		param->requester.gid = ino->gid;
 		spin_unlock(&sbi->fs_lock);
 	}
 
@@ -529,7 +529,7 @@ static int autofs_dev_ioctl_expire(struct file *fp,
 	int err = -EAGAIN;
 	int how;
 
-	how = param->arg1;
+	how = param->expire.how;
 	mnt = fp->f_path.mnt;
 
 	if (sbi->type & AUTOFS_TYPE_TRIGGER)
@@ -565,9 +565,9 @@ static int autofs_dev_ioctl_askumount(struct file *fp,
 				      struct autofs_sb_info *sbi,
 				      struct autofs_dev_ioctl *param)
 {
-	param->arg1 = 0;
+	param->askumount.may_umount = 0;
 	if (may_umount(fp->f_path.mnt))
-		param->arg1 = 1;
+		param->askumount.may_umount = 1;
 	return 0;
 }
 
@@ -600,6 +600,7 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp,
 	struct nameidata nd;
 	const char *path;
 	unsigned int type;
+	unsigned int devid, magic;
 	int err = -ENOENT;
 
 	if (param->size <= sizeof(*param)) {
@@ -608,10 +609,10 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp,
 	}
 
 	path = param->path;
-	type = param->arg1;
+	type = param->ismountpoint.in.type;
 
-	param->arg1 = 0;
-	param->arg2 = 0;
+	param->ismountpoint.out.devid = devid = 0;
+	param->ismountpoint.out.magic = magic = 0;
 
 	if (!fp || param->ioctlfd == -1) {
 		if (type == AUTOFS_TYPE_ANY) {
@@ -622,7 +623,7 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp,
 				goto out;
 
 			sb = nd.path.dentry->d_sb;
-			param->arg1 = new_encode_dev(sb->s_dev);
+			devid = new_encode_dev(sb->s_dev);
 		} else {
 			struct autofs_info *ino;
 
@@ -635,14 +636,14 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp,
 				goto out_release;
 
 			ino = autofs4_dentry_ino(nd.path.dentry);
-			param->arg1 = autofs4_get_dev(ino->sbi);
+			devid = autofs4_get_dev(ino->sbi);
 		}
 
 		err = 0;
 		if (nd.path.dentry->d_inode &&
 		    nd.path.mnt->mnt_root == nd.path.dentry) {
 			err = 1;
-			param->arg2 = nd.path.dentry->d_inode->i_sb->s_magic;
+			magic = nd.path.dentry->d_inode->i_sb->s_magic;
 		}
 	} else {
 		dev_t devid = new_encode_dev(sbi->sb->s_dev);
@@ -655,18 +656,21 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp,
 		if (err)
 			goto out_release;
 
-		param->arg1 = autofs4_get_dev(sbi);
+		devid = autofs4_get_dev(sbi);
 
 		err = have_submounts(nd.path.dentry);
 
 		if (nd.path.mnt->mnt_mountpoint != nd.path.mnt->mnt_root) {
 			if (follow_down(&nd.path.mnt, &nd.path.dentry)) {
 				struct inode *inode = nd.path.dentry->d_inode;
-				param->arg2 = inode->i_sb->s_magic;
+				magic = inode->i_sb->s_magic;
 			}
 		}
 	}
 
+	param->ismountpoint.out.devid = devid;
+	param->ismountpoint.out.magic = magic;
+
 out_release:
 	path_put(&nd.path);
 out:
diff --git a/include/linux/auto_dev-ioctl.h b/include/linux/auto_dev-ioctl.h
index f4d05ccd731f..91a773993a5c 100644
--- a/include/linux/auto_dev-ioctl.h
+++ b/include/linux/auto_dev-ioctl.h
@@ -10,6 +10,7 @@
 #ifndef _LINUX_AUTO_DEV_IOCTL_H
 #define _LINUX_AUTO_DEV_IOCTL_H
 
+#include <linux/string.h>
 #include <linux/types.h>
 
 #define AUTOFS_DEVICE_NAME		"autofs"
@@ -25,6 +26,60 @@
  * An ioctl interface for autofs mount point control.
  */
 
+struct args_protover {
+	__u32	version;
+};
+
+struct args_protosubver {
+	__u32	sub_version;
+};
+
+struct args_openmount {
+	__u32	devid;
+};
+
+struct args_ready {
+	__u32	token;
+};
+
+struct args_fail {
+	__u32	token;
+	__s32	status;
+};
+
+struct args_setpipefd {
+	__s32	pipefd;
+};
+
+struct args_timeout {
+	__u64	timeout;
+};
+
+struct args_requester {
+	__u32	uid;
+	__u32	gid;
+};
+
+struct args_expire {
+	__u32	how;
+};
+
+struct args_askumount {
+	__u32	may_umount;
+};
+
+struct args_ismountpoint {
+	union {
+		struct args_in {
+			__u32	type;
+		} in;
+		struct args_out {
+			__u32	devid;
+			__u32	magic;
+		} out;
+	};
+};
+
 /*
  * All the ioctls use this structure.
  * When sending a path size must account for the total length
@@ -39,20 +94,32 @@ struct autofs_dev_ioctl {
 				 * including this struct */
 	__s32 ioctlfd;		/* automount command fd */
 
-	__u32 arg1;		/* Command parameters */
-	__u32 arg2;
+	/* Command parameters */
+
+	union {
+		struct args_protover		protover;
+		struct args_protosubver		protosubver;
+		struct args_openmount		openmount;
+		struct args_ready		ready;
+		struct args_fail		fail;
+		struct args_setpipefd		setpipefd;
+		struct args_timeout		timeout;
+		struct args_requester		requester;
+		struct args_expire		expire;
+		struct args_askumount		askumount;
+		struct args_ismountpoint	ismountpoint;
+	};
 
 	char path[0];
 };
 
 static inline void init_autofs_dev_ioctl(struct autofs_dev_ioctl *in)
 {
+	memset(in, 0, sizeof(struct autofs_dev_ioctl));
 	in->ver_major = AUTOFS_DEV_IOCTL_VERSION_MAJOR;
 	in->ver_minor = AUTOFS_DEV_IOCTL_VERSION_MINOR;
 	in->size = sizeof(struct autofs_dev_ioctl);
 	in->ioctlfd = -1;
-	in->arg1 = 0;
-	in->arg2 = 0;
 	return;
 }
 
-- 
cgit v1.2.3


From a92daf6ba1f9ace8584edc8eb557a77aa7c2c71d Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Tue, 6 Jan 2009 14:42:08 -0800
Subject: autofs4: make autofs type usage explicit

- the type assigned at mount when no type is given is changed
  from 0 to AUTOFS_TYPE_INDIRECT. This was done because 0 and
  AUTOFS_TYPE_INDIRECT were being treated implicitly as the same
  type.

- previously, an offset mount had it's type set to
  AUTOFS_TYPE_DIRECT|AUTOFS_TYPE_OFFSET but the mount control
  re-implementation needs to be able distinguish all three types.
  So this was changed to make the type setting explicit.

- a type AUTOFS_TYPE_ANY was added for use by the re-implementation
  when checking if a given path is a mountpoint. It's not really a
  type as we use this to ask if a given path is a mountpoint in the
  autofs_dev_ioctl_ismountpoint() function.

- functions to set and test the autofs mount types have been added to
  improve readability and make the type usage explicit.

- the mount type is used from user space for the mount control
  re-implementtion so, for consistency, all the definitions have
  been moved to the user space include file include/linux/auto_fs4.h.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/autofs_i.h    |  2 --
 fs/autofs4/dev-ioctl.c   |  4 ++--
 fs/autofs4/expire.c      |  4 ++--
 fs/autofs4/inode.c       | 14 +++++------
 fs/autofs4/waitq.c       |  8 +++----
 include/linux/auto_fs4.h | 62 ++++++++++++++++++++++++++++++++++++++++++++----
 6 files changed, 73 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index e0f16da00e54..a76803108d06 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -25,8 +25,6 @@
 #define AUTOFS_DEV_IOCTL_IOC_FIRST	(AUTOFS_DEV_IOCTL_VERSION)
 #define AUTOFS_DEV_IOCTL_IOC_COUNT	(AUTOFS_IOC_COUNT - 11)
 
-#define AUTOFS_TYPE_TRIGGER	(AUTOFS_TYPE_DIRECT|AUTOFS_TYPE_OFFSET)
-
 #include <linux/kernel.h>
 #include <linux/slab.h>
 #include <linux/time.h>
diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c
index 0566ff8db4cd..aa3f47293bb8 100644
--- a/fs/autofs4/dev-ioctl.c
+++ b/fs/autofs4/dev-ioctl.c
@@ -532,7 +532,7 @@ static int autofs_dev_ioctl_expire(struct file *fp,
 	how = param->expire.how;
 	mnt = fp->f_path.mnt;
 
-	if (sbi->type & AUTOFS_TYPE_TRIGGER)
+	if (autofs_type_trigger(sbi->type))
 		dentry = autofs4_expire_direct(sbi->sb, mnt, sbi, how);
 	else
 		dentry = autofs4_expire_indirect(sbi->sb, mnt, sbi, how);
@@ -615,7 +615,7 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp,
 	param->ismountpoint.out.magic = magic = 0;
 
 	if (!fp || param->ioctlfd == -1) {
-		if (type == AUTOFS_TYPE_ANY) {
+		if (autofs_type_any(type)) {
 			struct super_block *sb;
 
 			err = path_lookup(path, LOOKUP_FOLLOW, &nd);
diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c
index 4b6fb3f628c0..e3bd50776f9e 100644
--- a/fs/autofs4/expire.c
+++ b/fs/autofs4/expire.c
@@ -63,7 +63,7 @@ static int autofs4_mount_busy(struct vfsmount *mnt, struct dentry *dentry)
 		struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
 
 		/* This is an autofs submount, we can't expire it */
-		if (sbi->type == AUTOFS_TYPE_INDIRECT)
+		if (autofs_type_indirect(sbi->type))
 			goto done;
 
 		/*
@@ -490,7 +490,7 @@ int autofs4_expire_multi(struct super_block *sb, struct vfsmount *mnt,
 	if (arg && get_user(do_now, arg))
 		return -EFAULT;
 
-	if (sbi->type & AUTOFS_TYPE_TRIGGER)
+	if (autofs_type_trigger(sbi->type))
 		dentry = autofs4_expire_direct(sb, mnt, sbi, do_now);
 	else
 		dentry = autofs4_expire_indirect(sb, mnt, sbi, do_now);
diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
index cfc23e53b6f4..716e12b627b2 100644
--- a/fs/autofs4/inode.c
+++ b/fs/autofs4/inode.c
@@ -197,9 +197,9 @@ static int autofs4_show_options(struct seq_file *m, struct vfsmount *mnt)
 	seq_printf(m, ",minproto=%d", sbi->min_proto);
 	seq_printf(m, ",maxproto=%d", sbi->max_proto);
 
-	if (sbi->type & AUTOFS_TYPE_OFFSET)
+	if (autofs_type_offset(sbi->type))
 		seq_printf(m, ",offset");
-	else if (sbi->type & AUTOFS_TYPE_DIRECT)
+	else if (autofs_type_direct(sbi->type))
 		seq_printf(m, ",direct");
 	else
 		seq_printf(m, ",indirect");
@@ -284,13 +284,13 @@ static int parse_options(char *options, int *pipefd, uid_t *uid, gid_t *gid,
 			*maxproto = option;
 			break;
 		case Opt_indirect:
-			*type = AUTOFS_TYPE_INDIRECT;
+			set_autofs_type_indirect(type);
 			break;
 		case Opt_direct:
-			*type = AUTOFS_TYPE_DIRECT;
+			set_autofs_type_direct(type);
 			break;
 		case Opt_offset:
-			*type = AUTOFS_TYPE_OFFSET;
+			set_autofs_type_offset(type);
 			break;
 		default:
 			return 1;
@@ -338,7 +338,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent)
 	sbi->sb = s;
 	sbi->version = 0;
 	sbi->sub_version = 0;
-	sbi->type = AUTOFS_TYPE_INDIRECT;
+	set_autofs_type_indirect(&sbi->type);
 	sbi->min_proto = 0;
 	sbi->max_proto = 0;
 	mutex_init(&sbi->wq_mutex);
@@ -380,7 +380,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent)
 	}
 
 	root_inode->i_fop = &autofs4_root_operations;
-	root_inode->i_op = sbi->type & AUTOFS_TYPE_TRIGGER ?
+	root_inode->i_op = autofs_type_trigger(sbi->type) ?
 			&autofs4_direct_root_inode_operations :
 			&autofs4_indirect_root_inode_operations;
 
diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c
index e02cc8ae5eb3..eeb246845909 100644
--- a/fs/autofs4/waitq.c
+++ b/fs/autofs4/waitq.c
@@ -337,7 +337,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		 * is very similar for indirect mounts except only dentrys
 		 * in the root of the autofs file system may be negative.
 		 */
-		if (sbi->type & AUTOFS_TYPE_TRIGGER)
+		if (autofs_type_trigger(sbi->type))
 			return -ENOENT;
 		else if (!IS_ROOT(dentry->d_parent))
 			return -ENOENT;
@@ -348,7 +348,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		return -ENOMEM;
 
 	/* If this is a direct mount request create a dummy name */
-	if (IS_ROOT(dentry) && sbi->type & AUTOFS_TYPE_TRIGGER)
+	if (IS_ROOT(dentry) && autofs_type_trigger(sbi->type))
 		qstr.len = sprintf(name, "%p", dentry);
 	else {
 		qstr.len = autofs4_getpath(sbi, dentry, &name);
@@ -406,11 +406,11 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 				type = autofs_ptype_expire_multi;
 		} else {
 			if (notify == NFY_MOUNT)
-				type = (sbi->type & AUTOFS_TYPE_TRIGGER) ?
+				type = autofs_type_trigger(sbi->type) ?
 					autofs_ptype_missing_direct :
 					 autofs_ptype_missing_indirect;
 			else
-				type = (sbi->type & AUTOFS_TYPE_TRIGGER) ?
+				type = autofs_type_trigger(sbi->type) ?
 					autofs_ptype_expire_direct :
 					autofs_ptype_expire_indirect;
 		}
diff --git a/include/linux/auto_fs4.h b/include/linux/auto_fs4.h
index 2253716d4b92..55fa478bd639 100644
--- a/include/linux/auto_fs4.h
+++ b/include/linux/auto_fs4.h
@@ -29,10 +29,64 @@
 #define AUTOFS_EXP_IMMEDIATE		1
 #define AUTOFS_EXP_LEAVES		2
 
-#define AUTOFS_TYPE_ANY			0x0000
-#define AUTOFS_TYPE_INDIRECT		0x0001
-#define AUTOFS_TYPE_DIRECT		0x0002
-#define AUTOFS_TYPE_OFFSET		0x0004
+#define AUTOFS_TYPE_ANY			0U
+#define AUTOFS_TYPE_INDIRECT		1U
+#define AUTOFS_TYPE_DIRECT		2U
+#define AUTOFS_TYPE_OFFSET		4U
+
+static inline void set_autofs_type_indirect(unsigned int *type)
+{
+	*type = AUTOFS_TYPE_INDIRECT;
+	return;
+}
+
+static inline unsigned int autofs_type_indirect(unsigned int type)
+{
+	return (type == AUTOFS_TYPE_INDIRECT);
+}
+
+static inline void set_autofs_type_direct(unsigned int *type)
+{
+	*type = AUTOFS_TYPE_DIRECT;
+	return;
+}
+
+static inline unsigned int autofs_type_direct(unsigned int type)
+{
+	return (type == AUTOFS_TYPE_DIRECT);
+}
+
+static inline void set_autofs_type_offset(unsigned int *type)
+{
+	*type = AUTOFS_TYPE_OFFSET;
+	return;
+}
+
+static inline unsigned int autofs_type_offset(unsigned int type)
+{
+	return (type == AUTOFS_TYPE_OFFSET);
+}
+
+static inline unsigned int autofs_type_trigger(unsigned int type)
+{
+	return (type == AUTOFS_TYPE_DIRECT || type == AUTOFS_TYPE_OFFSET);
+}
+
+/*
+ * This isn't really a type as we use it to say "no type set" to
+ * indicate we want to search for "any" mount in the
+ * autofs_dev_ioctl_ismountpoint() device ioctl function.
+ */
+static inline void set_autofs_type_any(unsigned int *type)
+{
+	*type = AUTOFS_TYPE_ANY;
+	return;
+}
+
+static inline unsigned int autofs_type_any(unsigned int type)
+{
+	return (type == AUTOFS_TYPE_ANY);
+}
 
 /* Daemon notification packet types */
 enum autofs_notify {
-- 
cgit v1.2.3


From cabb3fc4bd1628c37c37e054960eb3e4bf30dc26 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Tue, 6 Jan 2009 14:42:26 -0800
Subject: twl4030-gpio: cleanup debounce

Provide a static debounce configuration mechanism for twl4030 GPIOs,
replacing the previous dynamic one.  The single user of that mechanism was
for MMC card detect debouncing.

Boards can provide a bitmask saying which GPIOs to debounce (30 msec).
It's always enabled for pins with the MMC card-detect/VMMCx link active,
so most boards won't need to set the debounce mask.

This is a net code shrink, including runtime footprint.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/gpio/twl4030-gpio.c | 54 +++++++++++++++++++--------------------------
 include/linux/i2c/twl4030.h |  9 +++-----
 2 files changed, 26 insertions(+), 37 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/gpio/twl4030-gpio.c b/drivers/gpio/twl4030-gpio.c
index 37d3eec8730a..afad14792141 100644
--- a/drivers/gpio/twl4030-gpio.c
+++ b/drivers/gpio/twl4030-gpio.c
@@ -202,37 +202,6 @@ static int twl4030_get_gpio_datain(int gpio)
 	return ret;
 }
 
-/*
- * Configure debounce timing value for a GPIO pin on TWL4030
- */
-int twl4030_set_gpio_debounce(int gpio, int enable)
-{
-	u8 d_bnk = gpio >> 3;
-	u8 d_msk = BIT(gpio & 0x7);
-	u8 reg = 0;
-	u8 base = 0;
-	int ret = 0;
-
-	if (unlikely((gpio >= TWL4030_GPIO_MAX)
-		|| !(gpio_usage_count & BIT(gpio))))
-		return -EPERM;
-
-	base = REG_GPIO_DEBEN1 + d_bnk;
-	mutex_lock(&gpio_lock);
-	ret = gpio_twl4030_read(base);
-	if (ret >= 0) {
-		if (enable)
-			reg = ret | d_msk;
-		else
-			reg = ret & ~d_msk;
-
-		ret = gpio_twl4030_write(base, reg);
-	}
-	mutex_unlock(&gpio_lock);
-	return ret;
-}
-EXPORT_SYMBOL(twl4030_set_gpio_debounce);
-
 /*----------------------------------------------------------------------*/
 
 static int twl_request(struct gpio_chip *chip, unsigned offset)
@@ -405,6 +374,23 @@ static int __devinit gpio_twl4030_pulls(u32 ups, u32 downs)
 				REG_GPIOPUPDCTR1, 5);
 }
 
+static int __devinit gpio_twl4030_debounce(u32 debounce, u8 mmc_cd)
+{
+	u8		message[4];
+
+	/* 30 msec of debouncing is always used for MMC card detect,
+	 * and is optional for everything else.
+	 */
+	message[1] = (debounce & 0xff) | (mmc_cd & 0x03);
+	debounce >>= 8;
+	message[2] = (debounce & 0xff);
+	debounce >>= 8;
+	message[3] = (debounce & 0x03);
+
+	return twl4030_i2c_write(TWL4030_MODULE_GPIO, message,
+				REG_GPIO_DEBEN1, 3);
+}
+
 static int gpio_twl4030_remove(struct platform_device *pdev);
 
 static int __devinit gpio_twl4030_probe(struct platform_device *pdev)
@@ -439,6 +425,12 @@ no_irqs:
 				pdata->pullups, pdata->pulldowns,
 				ret);
 
+	ret = gpio_twl4030_debounce(pdata->debounce, pdata->mmc_cd);
+	if (ret)
+		dev_dbg(&pdev->dev, "debounce %.03x %.01x --> %d\n",
+				pdata->debounce, pdata->mmc_cd,
+				ret);
+
 	twl_gpiochip.base = pdata->gpio_base;
 	twl_gpiochip.ngpio = TWL4030_GPIO_MAX;
 	twl_gpiochip.dev = &pdev->dev;
diff --git a/include/linux/i2c/twl4030.h b/include/linux/i2c/twl4030.h
index a8f84c01f82e..8137f660a5cc 100644
--- a/include/linux/i2c/twl4030.h
+++ b/include/linux/i2c/twl4030.h
@@ -234,6 +234,9 @@ struct twl4030_gpio_platform_data {
 	/* gpio-n should control VMMC(n+1) if BIT(n) in mmc_cd is set */
 	u8		mmc_cd;
 
+	/* if BIT(N) is set, or VMMC(n+1) is linked, debounce GPIO-N */
+	u32		debounce;
+
 	/* For gpio-N, bit (1 << N) in "pullups" is set if that pullup
 	 * should be enabled.  Else, if that bit is set in "pulldowns",
 	 * that pulldown is enabled.  Don't waste power by letting any
@@ -307,12 +310,6 @@ int twl4030_sih_setup(int module);
 #define TWL4030_VAUX3_DEV_GRP		0x1F
 #define TWL4030_VAUX3_DEDICATED		0x22
 
-/*
- * Exported TWL4030 GPIO APIs
- *
- * WARNING -- use standard GPIO and IRQ calls instead; these will vanish.
- */
-int twl4030_set_gpio_debounce(int gpio, int enable);
 
 #if defined(CONFIG_TWL4030_BCI_BATTERY) || \
 	defined(CONFIG_TWL4030_BCI_BATTERY_MODULE)
-- 
cgit v1.2.3


From d3635abfee0c55ad9dcd6b6172a0c6a5455900c9 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Tue, 6 Jan 2009 14:42:41 -0800
Subject: rapidio: remove excess kernel-doc notation

Remove excess kernel-doc notation from rio header and driver:

Warning(include/linux/rio_drv.h:399): Excess function parameter or struct member 'buffer' description in 'rio_get_inb_message'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rapidio/rio-driver.c | 1 -
 include/linux/rio_drv.h      | 1 -
 2 files changed, 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/rapidio/rio-driver.c b/drivers/rapidio/rio-driver.c
index 956d3e79f6aa..addb87cf44d9 100644
--- a/drivers/rapidio/rio-driver.c
+++ b/drivers/rapidio/rio-driver.c
@@ -79,7 +79,6 @@ void rio_dev_put(struct rio_dev *rdev)
 
 /**
  *  rio_device_probe - Tell if a RIO device structure has a matching RIO device id structure
- *  @id: the RIO device id structure to match against
  *  @dev: the RIO device structure to match against
  *
  * return 0 and set rio_dev->driver when drv claims rio_dev, else error
diff --git a/include/linux/rio_drv.h b/include/linux/rio_drv.h
index 32c0547ffafc..c93a58a40033 100644
--- a/include/linux/rio_drv.h
+++ b/include/linux/rio_drv.h
@@ -391,7 +391,6 @@ static inline int rio_add_inb_buffer(struct rio_mport *mport, int mbox,
  * rio_get_inb_message - Get A RIO message from an inbound mailbox queue
  * @mport: Master port containing the inbound mailbox
  * @mbox: The inbound mailbox number
- * @buffer: Pointer to the message buffer
  *
  * Get a RIO message from an inbound mailbox queue. Returns 0 on success.
  */
-- 
cgit v1.2.3


From 8cd3ac3aca3f2afe8570708066d64d893da468e8 Mon Sep 17 00:00:00 2001
From: WANG Cong <wangcong@zeuux.org>
Date: Tue, 6 Jan 2009 14:42:48 -0800
Subject: fs/exec.c: make do_coredump() void

No one cares do_coredump()'s return value, and also it seems that it
is also not necessary. So make it void.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: WANG Cong <wangcong@zeuux.org>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c               | 4 ++--
 include/linux/binfmts.h | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/fs/exec.c b/fs/exec.c
index 6b09d6fa4f7e..71a6efe5d8bd 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1686,7 +1686,7 @@ int get_dumpable(struct mm_struct *mm)
 	return (ret >= 2) ? 2 : ret;
 }
 
-int do_coredump(long signr, int exit_code, struct pt_regs * regs)
+void do_coredump(long signr, int exit_code, struct pt_regs *regs)
 {
 	struct core_state core_state;
 	char corename[CORENAME_MAX_SIZE + 1];
@@ -1842,5 +1842,5 @@ fail_unlock:
 	put_cred(cred);
 	coredump_finish(mm);
 fail:
-	return retval;
+	return;
 }
diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h
index 0d0150b4901e..77b4a9e46004 100644
--- a/include/linux/binfmts.h
+++ b/include/linux/binfmts.h
@@ -107,7 +107,7 @@ extern int setup_arg_pages(struct linux_binprm * bprm,
 extern int bprm_mm_init(struct linux_binprm *bprm);
 extern int copy_strings_kernel(int argc,char ** argv,struct linux_binprm *bprm);
 extern void install_exec_creds(struct linux_binprm *bprm);
-extern int do_coredump(long signr, int exit_code, struct pt_regs * regs);
+extern void do_coredump(long signr, int exit_code, struct pt_regs *regs);
 extern int set_binfmt(struct linux_binfmt *new);
 extern void free_bprm(struct linux_binprm *);
 
-- 
cgit v1.2.3


From 991c0e6d1ae3df59f0ddfe05edecec8319e35a1b Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Tue, 6 Jan 2009 14:56:21 -0800
Subject: byteorder: only use linux/swab.h

The first step to make swab.h a regular header that will
include an asm/swab.h with arch overrides.

Avoid the gratuitous differences introduced in the new
linux/swab.h by naming the ___constant_swabXX bits and
__fswabXX bits exactly as found in the old implementation
in byteorder/swab[b].h

Use this new swab.h in byteorder/[big|little]_endian.h and
remove the two old swab headers.

Although the inclusion of asm/byteorder.h looks strange in
linux/swab.h, this will allow each arch to move the actual
arch overrides for the swab bits in an asm file and then
the includes can be cleaned up without requiring a flag day
for all arches at once.

Keep providing __fswabXX in case some userspace was using them
directly, but the revised __swabXX should be used instead in
any new code and will always do constant folding not dependent
on the optimization level, which means the __constant versions
can be phased out in-kernel.

Arches that use the old-style arch macros will lose their
optimized versions until they move to the new style, but at
least they will still compile.  Many arches have already moved
and the patches to move the remaining arches are trivial.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/byteorder/Kbuild          |   2 -
 include/linux/byteorder/big_endian.h    |   3 +-
 include/linux/byteorder/little_endian.h |   3 +-
 include/linux/byteorder/swab.h          | 222 --------------------------------
 include/linux/byteorder/swabb.h         | 135 -------------------
 include/linux/swab.h                    |  50 +++----
 6 files changed, 27 insertions(+), 388 deletions(-)
 delete mode 100644 include/linux/byteorder/swab.h
 delete mode 100644 include/linux/byteorder/swabb.h

(limited to 'include/linux')

diff --git a/include/linux/byteorder/Kbuild b/include/linux/byteorder/Kbuild
index fbaa7f9cee32..38437225b092 100644
--- a/include/linux/byteorder/Kbuild
+++ b/include/linux/byteorder/Kbuild
@@ -1,4 +1,2 @@
 unifdef-y += big_endian.h
 unifdef-y += little_endian.h
-unifdef-y += swab.h
-unifdef-y += swabb.h
diff --git a/include/linux/byteorder/big_endian.h b/include/linux/byteorder/big_endian.h
index 1cba3f3efe5f..3c80fd7e8b56 100644
--- a/include/linux/byteorder/big_endian.h
+++ b/include/linux/byteorder/big_endian.h
@@ -9,8 +9,7 @@
 #endif
 
 #include <linux/types.h>
-#include <linux/byteorder/swab.h>
-#include <linux/byteorder/swabb.h>
+#include <linux/swab.h>
 
 #define __constant_htonl(x) ((__force __be32)(__u32)(x))
 #define __constant_ntohl(x) ((__force __u32)(__be32)(x))
diff --git a/include/linux/byteorder/little_endian.h b/include/linux/byteorder/little_endian.h
index cedc1b5a289c..83195fb82962 100644
--- a/include/linux/byteorder/little_endian.h
+++ b/include/linux/byteorder/little_endian.h
@@ -9,8 +9,7 @@
 #endif
 
 #include <linux/types.h>
-#include <linux/byteorder/swab.h>
-#include <linux/byteorder/swabb.h>
+#include <linux/swab.h>
 
 #define __constant_htonl(x) ((__force __be32)___constant_swab32((x)))
 #define __constant_ntohl(x) ___constant_swab32((__force __be32)(x))
diff --git a/include/linux/byteorder/swab.h b/include/linux/byteorder/swab.h
deleted file mode 100644
index 142134ff1645..000000000000
--- a/include/linux/byteorder/swab.h
+++ /dev/null
@@ -1,222 +0,0 @@
-#ifndef _LINUX_BYTEORDER_SWAB_H
-#define _LINUX_BYTEORDER_SWAB_H
-
-/*
- * linux/byteorder/swab.h
- * Byte-swapping, independently from CPU endianness
- *	swabXX[ps]?(foo)
- *
- * Francois-Rene Rideau <fare@tunes.org> 19971205
- *    separated swab functions from cpu_to_XX,
- *    to clean up support for bizarre-endian architectures.
- *
- * Trent Piepho <xyzzy@speakeasy.org> 2007114
- *    make constant-folding work, provide C versions that
- *    gcc can optimize better, explain different versions
- *
- * See asm-i386/byteorder.h and suches for examples of how to provide
- * architecture-dependent optimized versions
- *
- */
-
-#include <linux/compiler.h>
-
-/* Functions/macros defined, there are a lot:
- *
- * ___swabXX
- *    Generic C versions of the swab functions.
- *
- * ___constant_swabXX
- *    C versions that gcc can fold into a compile-time constant when
- *    the argument is a compile-time constant.
- *
- * __arch__swabXX[sp]?
- *    Architecture optimized versions of all the swab functions
- *    (including the s and p versions).  These can be defined in
- *    asm-arch/byteorder.h.  Any which are not, are defined here.
- *    __arch__swabXXs() is defined in terms of __arch__swabXXp(), which
- *    is defined in terms of __arch__swabXX(), which is in turn defined
- *    in terms of ___swabXX(x).
- *    These must be macros.  They may be unsafe for arguments with
- *    side-effects.
- *
- * __fswabXX
- *    Inline function versions of the __arch__ macros.  These _are_ safe
- *    if the arguments have side-effects.  Note there are no s and p
- *    versions of these.
- *
- * __swabXX[sb]
- *    There are the ones you should actually use.  The __swabXX versions
- *    will be a constant given a constant argument and use the arch
- *    specific code (if any) for non-constant arguments.  The s and p
- *    versions always use the arch specific code (constant folding
- *    doesn't apply).  They are safe to use with arguments with
- *    side-effects.
- *
- * swabXX[sb]
- *    Nicknames for __swabXX[sb] to use in the kernel.
- */
-
-/* casts are necessary for constants, because we never know how for sure
- * how U/UL/ULL map to __u16, __u32, __u64. At least not in a portable way.
- */
-
-static __inline__ __attribute_const__ __u16 ___swab16(__u16 x)
-{
-	return x<<8 | x>>8;
-}
-static __inline__ __attribute_const__ __u32 ___swab32(__u32 x)
-{
-	return x<<24 | x>>24 |
-		(x & (__u32)0x0000ff00UL)<<8 |
-		(x & (__u32)0x00ff0000UL)>>8;
-}
-static __inline__ __attribute_const__ __u64 ___swab64(__u64 x)
-{
-	return x<<56 | x>>56 |
-		(x & (__u64)0x000000000000ff00ULL)<<40 |
-		(x & (__u64)0x0000000000ff0000ULL)<<24 |
-		(x & (__u64)0x00000000ff000000ULL)<< 8 |
-	        (x & (__u64)0x000000ff00000000ULL)>> 8 |
-		(x & (__u64)0x0000ff0000000000ULL)>>24 |
-		(x & (__u64)0x00ff000000000000ULL)>>40;
-}
-
-#define ___constant_swab16(x) \
-	((__u16)( \
-		(((__u16)(x) & (__u16)0x00ffU) << 8) | \
-		(((__u16)(x) & (__u16)0xff00U) >> 8) ))
-#define ___constant_swab32(x) \
-	((__u32)( \
-		(((__u32)(x) & (__u32)0x000000ffUL) << 24) | \
-		(((__u32)(x) & (__u32)0x0000ff00UL) <<  8) | \
-		(((__u32)(x) & (__u32)0x00ff0000UL) >>  8) | \
-		(((__u32)(x) & (__u32)0xff000000UL) >> 24) ))
-#define ___constant_swab64(x) \
-	((__u64)( \
-		(__u64)(((__u64)(x) & (__u64)0x00000000000000ffULL) << 56) | \
-		(__u64)(((__u64)(x) & (__u64)0x000000000000ff00ULL) << 40) | \
-		(__u64)(((__u64)(x) & (__u64)0x0000000000ff0000ULL) << 24) | \
-		(__u64)(((__u64)(x) & (__u64)0x00000000ff000000ULL) <<  8) | \
-	        (__u64)(((__u64)(x) & (__u64)0x000000ff00000000ULL) >>  8) | \
-		(__u64)(((__u64)(x) & (__u64)0x0000ff0000000000ULL) >> 24) | \
-		(__u64)(((__u64)(x) & (__u64)0x00ff000000000000ULL) >> 40) | \
-		(__u64)(((__u64)(x) & (__u64)0xff00000000000000ULL) >> 56) ))
-
-/*
- * provide defaults when no architecture-specific optimization is detected
- */
-#ifndef __arch__swab16
-#  define __arch__swab16(x) ___swab16(x)
-#endif
-#ifndef __arch__swab32
-#  define __arch__swab32(x) ___swab32(x)
-#endif
-#ifndef __arch__swab64
-#  define __arch__swab64(x) ___swab64(x)
-#endif
-
-#ifndef __arch__swab16p
-#  define __arch__swab16p(x) __arch__swab16(*(x))
-#endif
-#ifndef __arch__swab32p
-#  define __arch__swab32p(x) __arch__swab32(*(x))
-#endif
-#ifndef __arch__swab64p
-#  define __arch__swab64p(x) __arch__swab64(*(x))
-#endif
-
-#ifndef __arch__swab16s
-#  define __arch__swab16s(x) ((void)(*(x) = __arch__swab16p(x)))
-#endif
-#ifndef __arch__swab32s
-#  define __arch__swab32s(x) ((void)(*(x) = __arch__swab32p(x)))
-#endif
-#ifndef __arch__swab64s
-#  define __arch__swab64s(x) ((void)(*(x) = __arch__swab64p(x)))
-#endif
-
-
-/*
- * Allow constant folding
- */
-#if defined(__GNUC__) && defined(__OPTIMIZE__)
-#  define __swab16(x) \
-(__builtin_constant_p((__u16)(x)) ? \
- ___constant_swab16((x)) : \
- __fswab16((x)))
-#  define __swab32(x) \
-(__builtin_constant_p((__u32)(x)) ? \
- ___constant_swab32((x)) : \
- __fswab32((x)))
-#  define __swab64(x) \
-(__builtin_constant_p((__u64)(x)) ? \
- ___constant_swab64((x)) : \
- __fswab64((x)))
-#else
-#  define __swab16(x) __fswab16(x)
-#  define __swab32(x) __fswab32(x)
-#  define __swab64(x) __fswab64(x)
-#endif /* OPTIMIZE */
-
-
-static __inline__ __attribute_const__ __u16 __fswab16(__u16 x)
-{
-	return __arch__swab16(x);
-}
-static __inline__ __u16 __swab16p(const __u16 *x)
-{
-	return __arch__swab16p(x);
-}
-static __inline__ void __swab16s(__u16 *addr)
-{
-	__arch__swab16s(addr);
-}
-
-static __inline__ __attribute_const__ __u32 __fswab32(__u32 x)
-{
-	return __arch__swab32(x);
-}
-static __inline__ __u32 __swab32p(const __u32 *x)
-{
-	return __arch__swab32p(x);
-}
-static __inline__ void __swab32s(__u32 *addr)
-{
-	__arch__swab32s(addr);
-}
-
-#ifdef __BYTEORDER_HAS_U64__
-static __inline__ __attribute_const__ __u64 __fswab64(__u64 x)
-{
-#  ifdef __SWAB_64_THRU_32__
-	__u32 h = x >> 32;
-        __u32 l = x & ((1ULL<<32)-1);
-        return (((__u64)__swab32(l)) << 32) | ((__u64)(__swab32(h)));
-#  else
-	return __arch__swab64(x);
-#  endif
-}
-static __inline__ __u64 __swab64p(const __u64 *x)
-{
-	return __arch__swab64p(x);
-}
-static __inline__ void __swab64s(__u64 *addr)
-{
-	__arch__swab64s(addr);
-}
-#endif /* __BYTEORDER_HAS_U64__ */
-
-#if defined(__KERNEL__)
-#define swab16 __swab16
-#define swab32 __swab32
-#define swab64 __swab64
-#define swab16p __swab16p
-#define swab32p __swab32p
-#define swab64p __swab64p
-#define swab16s __swab16s
-#define swab32s __swab32s
-#define swab64s __swab64s
-#endif
-
-#endif /* _LINUX_BYTEORDER_SWAB_H */
diff --git a/include/linux/byteorder/swabb.h b/include/linux/byteorder/swabb.h
deleted file mode 100644
index 8c780c7d779e..000000000000
--- a/include/linux/byteorder/swabb.h
+++ /dev/null
@@ -1,135 +0,0 @@
-#ifndef _LINUX_BYTEORDER_SWABB_H
-#define _LINUX_BYTEORDER_SWABB_H
-
-/*
- * linux/byteorder/swabb.h
- * SWAp Bytes Bizarrely
- *	swaHHXX[ps]?(foo)
- *
- * Support for obNUXIous pdp-endian and other bizarre architectures.
- * Will Linux ever run on such ancient beasts? if not, this file
- * will be but a programming pearl. Still, it's a reminder that we
- * shouldn't be making too many assumptions when trying to be portable.
- *
- */
-
-/*
- * Meaning of the names I chose (vaxlinux people feel free to correct them):
- * swahw32	swap 16-bit half-words in a 32-bit word
- * swahb32	swap 8-bit halves of each 16-bit half-word in a 32-bit word
- *
- * No 64-bit support yet. I don't know NUXI conventions for long longs.
- * I guarantee it will be a mess when it's there, though :->
- * It will be even worse if there are conflicting 64-bit conventions.
- * Hopefully, no one ever used 64-bit objects on NUXI machines.
- *
- */
-
-#include <linux/types.h>
-
-#define ___swahw32(x) \
-({ \
-	__u32 __x = (x); \
-	((__u32)( \
-		(((__u32)(__x) & (__u32)0x0000ffffUL) << 16) | \
-		(((__u32)(__x) & (__u32)0xffff0000UL) >> 16) )); \
-})
-#define ___swahb32(x) \
-({ \
-	__u32 __x = (x); \
-	((__u32)( \
-		(((__u32)(__x) & (__u32)0x00ff00ffUL) << 8) | \
-		(((__u32)(__x) & (__u32)0xff00ff00UL) >> 8) )); \
-})
-
-#define ___constant_swahw32(x) \
-	((__u32)( \
-		(((__u32)(x) & (__u32)0x0000ffffUL) << 16) | \
-		(((__u32)(x) & (__u32)0xffff0000UL) >> 16) ))
-#define ___constant_swahb32(x) \
-	((__u32)( \
-		(((__u32)(x) & (__u32)0x00ff00ffUL) << 8) | \
-		(((__u32)(x) & (__u32)0xff00ff00UL) >> 8) ))
-
-/*
- * provide defaults when no architecture-specific optimization is detected
- */
-#ifndef __arch__swahw32
-#  define __arch__swahw32(x) ___swahw32(x)
-#endif
-#ifndef __arch__swahb32
-#  define __arch__swahb32(x) ___swahb32(x)
-#endif
-
-#ifndef __arch__swahw32p
-#  define __arch__swahw32p(x) __swahw32(*(x))
-#endif
-#ifndef __arch__swahb32p
-#  define __arch__swahb32p(x) __swahb32(*(x))
-#endif
-
-#ifndef __arch__swahw32s
-#  define __arch__swahw32s(x) do { *(x) = __swahw32p((x)); } while (0)
-#endif
-#ifndef __arch__swahb32s
-#  define __arch__swahb32s(x) do { *(x) = __swahb32p((x)); } while (0)
-#endif
-
-
-/*
- * Allow constant folding
- */
-#define __swahw32(x) \
-(__builtin_constant_p((__u32)(x)) ? \
- ___swahw32((x)) : \
- __fswahw32((x)))
-#define __swahb32(x) \
-(__builtin_constant_p((__u32)(x)) ? \
- ___swahb32((x)) : \
- __fswahb32((x)))
-
-
-static inline __u32 __fswahw32(__u32 x)
-{
-	return __arch__swahw32(x);
-}
-
-static inline __u32 __swahw32p(__u32 *x)
-{
-	return __arch__swahw32p(x);
-}
-
-static inline void __swahw32s(__u32 *addr)
-{
-	__arch__swahw32s(addr);
-}
-
-static inline __u32 __fswahb32(__u32 x)
-{
-	return __arch__swahb32(x);
-}
-
-static inline __u32 __swahb32p(__u32 *x)
-{
-	return __arch__swahb32p(x);
-}
-
-static inline void __swahb32s(__u32 *addr)
-{
-	__arch__swahb32s(addr);
-}
-
-#ifdef __BYTEORDER_HAS_U64__
-/*
- * Not supported yet
- */
-#endif /* __BYTEORDER_HAS_U64__ */
-
-#define swahw32 __swahw32
-#define swahb32 __swahb32
-#define swahw32p __swahw32p
-#define swahb32p __swahb32p
-#define swahw32s __swahw32s
-#define swahb32s __swahb32s
-
-#endif /* _LINUX_BYTEORDER_SWABB_H */
diff --git a/include/linux/swab.h b/include/linux/swab.h
index bbed279f3b32..9a2d33e0a98a 100644
--- a/include/linux/swab.h
+++ b/include/linux/swab.h
@@ -9,17 +9,17 @@
  * casts are necessary for constants, because we never know how for sure
  * how U/UL/ULL map to __u16, __u32, __u64. At least not in a portable way.
  */
-#define __const_swab16(x) ((__u16)(				\
+#define ___constant_swab16(x) ((__u16)(				\
 	(((__u16)(x) & (__u16)0x00ffU) << 8) |			\
 	(((__u16)(x) & (__u16)0xff00U) >> 8)))
 
-#define __const_swab32(x) ((__u32)(				\
+#define ___constant_swab32(x) ((__u32)(				\
 	(((__u32)(x) & (__u32)0x000000ffUL) << 24) |		\
 	(((__u32)(x) & (__u32)0x0000ff00UL) <<  8) |		\
 	(((__u32)(x) & (__u32)0x00ff0000UL) >>  8) |		\
 	(((__u32)(x) & (__u32)0xff000000UL) >> 24)))
 
-#define __const_swab64(x) ((__u64)(				\
+#define ___constant_swab64(x) ((__u64)(				\
 	(((__u64)(x) & (__u64)0x00000000000000ffULL) << 56) |	\
 	(((__u64)(x) & (__u64)0x000000000000ff00ULL) << 40) |	\
 	(((__u64)(x) & (__u64)0x0000000000ff0000ULL) << 24) |	\
@@ -29,11 +29,11 @@
 	(((__u64)(x) & (__u64)0x00ff000000000000ULL) >> 40) |	\
 	(((__u64)(x) & (__u64)0xff00000000000000ULL) >> 56)))
 
-#define __const_swahw32(x) ((__u32)(				\
+#define ___constant_swahw32(x) ((__u32)(			\
 	(((__u32)(x) & (__u32)0x0000ffffUL) << 16) |		\
 	(((__u32)(x) & (__u32)0xffff0000UL) >> 16)))
 
-#define __const_swahb32(x) ((__u32)(				\
+#define ___constant_swahb32(x) ((__u32)(			\
 	(((__u32)(x) & (__u32)0x00ff00ffUL) << 8) |		\
 	(((__u32)(x) & (__u32)0xff00ff00UL) >> 8)))
 
@@ -43,25 +43,25 @@
  * ___swab16, ___swab32, ___swab64, ___swahw32, ___swahb32
  */
 
-static inline __attribute_const__ __u16 ___swab16(__u16 val)
+static inline __attribute_const__ __u16 __fswab16(__u16 val)
 {
 #ifdef __arch_swab16
 	return __arch_swab16(val);
 #else
-	return __const_swab16(val);
+	return ___constant_swab16(val);
 #endif
 }
 
-static inline __attribute_const__ __u32 ___swab32(__u32 val)
+static inline __attribute_const__ __u32 __fswab32(__u32 val)
 {
 #ifdef __arch_swab32
 	return __arch_swab32(val);
 #else
-	return __const_swab32(val);
+	return ___constant_swab32(val);
 #endif
 }
 
-static inline __attribute_const__ __u64 ___swab64(__u64 val)
+static inline __attribute_const__ __u64 __fswab64(__u64 val)
 {
 #ifdef __arch_swab64
 	return __arch_swab64(val);
@@ -70,25 +70,25 @@ static inline __attribute_const__ __u64 ___swab64(__u64 val)
 	__u32 l = val & ((1ULL << 32) - 1);
 	return (((__u64)___swab32(l)) << 32) | ((__u64)(___swab32(h)));
 #else
-	return __const_swab64(val);
+	return ___constant_swab64(val);
 #endif
 }
 
-static inline __attribute_const__ __u32 ___swahw32(__u32 val)
+static inline __attribute_const__ __u32 __fswahw32(__u32 val)
 {
 #ifdef __arch_swahw32
 	return __arch_swahw32(val);
 #else
-	return __const_swahw32(val);
+	return ___constant_swahw32(val);
 #endif
 }
 
-static inline __attribute_const__ __u32 ___swahb32(__u32 val)
+static inline __attribute_const__ __u32 __fswahb32(__u32 val)
 {
 #ifdef __arch_swahb32
 	return __arch_swahb32(val);
 #else
-	return __const_swahb32(val);
+	return ___constant_swahb32(val);
 #endif
 }
 
@@ -98,8 +98,8 @@ static inline __attribute_const__ __u32 ___swahb32(__u32 val)
  */
 #define __swab16(x)				\
 	(__builtin_constant_p((__u16)(x)) ?	\
-	__const_swab16((x)) :			\
-	___swab16((x)))
+	___constant_swab16(x) :			\
+	__fswab16(x))
 
 /**
  * __swab32 - return a byteswapped 32-bit value
@@ -107,8 +107,8 @@ static inline __attribute_const__ __u32 ___swahb32(__u32 val)
  */
 #define __swab32(x)				\
 	(__builtin_constant_p((__u32)(x)) ?	\
-	__const_swab32((x)) :			\
-	___swab32((x)))
+	___constant_swab32(x) :			\
+	__fswab32(x))
 
 /**
  * __swab64 - return a byteswapped 64-bit value
@@ -116,8 +116,8 @@ static inline __attribute_const__ __u32 ___swahb32(__u32 val)
  */
 #define __swab64(x)				\
 	(__builtin_constant_p((__u64)(x)) ?	\
-	__const_swab64((x)) :			\
-	___swab64((x)))
+	___constant_swab64(x) :			\
+	__fswab64(x))
 
 /**
  * __swahw32 - return a word-swapped 32-bit value
@@ -127,8 +127,8 @@ static inline __attribute_const__ __u32 ___swahb32(__u32 val)
  */
 #define __swahw32(x)				\
 	(__builtin_constant_p((__u32)(x)) ?	\
-	__const_swahw32((x)) :			\
-	___swahw32((x)))
+	___constant_swahw32(x) :		\
+	__fswahw32(x))
 
 /**
  * __swahb32 - return a high and low byte-swapped 32-bit value
@@ -138,8 +138,8 @@ static inline __attribute_const__ __u32 ___swahb32(__u32 val)
  */
 #define __swahb32(x)				\
 	(__builtin_constant_p((__u32)(x)) ?	\
-	__const_swahb32((x)) :			\
-	___swahb32((x)))
+	___constant_swahb32(x) :		\
+	__fswahb32(x))
 
 /**
  * __swab16p - return a byteswapped 16-bit value from a pointer
-- 
cgit v1.2.3


From 637b180c23313f2964e0ef20f1ee375203866968 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Tue, 6 Jan 2009 13:30:58 -0800
Subject: byteorder: remove the now unused byteorder.h

This implementation caused problems in userspace which can, and does
define _both_ __LITTLE_ENDIAN and __BIG_ENDIAN.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/Kbuild      |   1 -
 include/linux/byteorder.h | 372 ----------------------------------------------
 2 files changed, 373 deletions(-)
 delete mode 100644 include/linux/byteorder.h

(limited to 'include/linux')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 39da666067b9..a3323f337e4d 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -179,7 +179,6 @@ unifdef-y += auto_fs.h
 unifdef-y += auxvec.h
 unifdef-y += binfmts.h
 unifdef-y += blktrace_api.h
-unifdef-y += byteorder.h
 unifdef-y += capability.h
 unifdef-y += capi.h
 unifdef-y += cciss_ioctl.h
diff --git a/include/linux/byteorder.h b/include/linux/byteorder.h
deleted file mode 100644
index 29f002d73d98..000000000000
--- a/include/linux/byteorder.h
+++ /dev/null
@@ -1,372 +0,0 @@
-#ifndef _LINUX_BYTEORDER_H
-#define _LINUX_BYTEORDER_H
-
-#include <linux/types.h>
-#include <linux/swab.h>
-
-#if defined(__LITTLE_ENDIAN) && defined(__BIG_ENDIAN)
-# error Fix asm/byteorder.h to define one endianness
-#endif
-
-#if !defined(__LITTLE_ENDIAN) && !defined(__BIG_ENDIAN)
-# error Fix asm/byteorder.h to define arch endianness
-#endif
-
-#ifdef __LITTLE_ENDIAN
-# undef __LITTLE_ENDIAN
-# define __LITTLE_ENDIAN 1234
-#endif
-
-#ifdef __BIG_ENDIAN
-# undef __BIG_ENDIAN
-# define __BIG_ENDIAN 4321
-#endif
-
-#if defined(__LITTLE_ENDIAN) && !defined(__LITTLE_ENDIAN_BITFIELD)
-# define __LITTLE_ENDIAN_BITFIELD
-#endif
-
-#if defined(__BIG_ENDIAN) && !defined(__BIG_ENDIAN_BITFIELD)
-# define __BIG_ENDIAN_BITFIELD
-#endif
-
-#ifdef __LITTLE_ENDIAN
-# define __le16_to_cpu(x) ((__force __u16)(__le16)(x))
-# define __le32_to_cpu(x) ((__force __u32)(__le32)(x))
-# define __le64_to_cpu(x) ((__force __u64)(__le64)(x))
-# define __cpu_to_le16(x) ((__force __le16)(__u16)(x))
-# define __cpu_to_le32(x) ((__force __le32)(__u32)(x))
-# define __cpu_to_le64(x) ((__force __le64)(__u64)(x))
-
-# define __be16_to_cpu(x) __swab16((__force __u16)(__be16)(x))
-# define __be32_to_cpu(x) __swab32((__force __u32)(__be32)(x))
-# define __be64_to_cpu(x) __swab64((__force __u64)(__be64)(x))
-# define __cpu_to_be16(x) ((__force __be16)__swab16(x))
-# define __cpu_to_be32(x) ((__force __be32)__swab32(x))
-# define __cpu_to_be64(x) ((__force __be64)__swab64(x))
-#endif
-
-#ifdef __BIG_ENDIAN
-# define __be16_to_cpu(x) ((__force __u16)(__be16)(x))
-# define __be32_to_cpu(x) ((__force __u32)(__be32)(x))
-# define __be64_to_cpu(x) ((__force __u64)(__be64)(x))
-# define __cpu_to_be16(x) ((__force __be16)(__u16)(x))
-# define __cpu_to_be32(x) ((__force __be32)(__u32)(x))
-# define __cpu_to_be64(x) ((__force __be64)(__u64)(x))
-
-# define __le16_to_cpu(x) __swab16((__force __u16)(__le16)(x))
-# define __le32_to_cpu(x) __swab32((__force __u32)(__le32)(x))
-# define __le64_to_cpu(x) __swab64((__force __u64)(__le64)(x))
-# define __cpu_to_le16(x) ((__force __le16)__swab16(x))
-# define __cpu_to_le32(x) ((__force __le32)__swab32(x))
-# define __cpu_to_le64(x) ((__force __le64)__swab64(x))
-#endif
-
-/*
- * These helpers could be phased out over time as the base version
- * handles constant folding.
- */
-#define __constant_htonl(x) __cpu_to_be32(x)
-#define __constant_ntohl(x) __be32_to_cpu(x)
-#define __constant_htons(x) __cpu_to_be16(x)
-#define __constant_ntohs(x) __be16_to_cpu(x)
-
-#define __constant_le16_to_cpu(x) __le16_to_cpu(x)
-#define __constant_le32_to_cpu(x) __le32_to_cpu(x)
-#define __constant_le64_to_cpu(x) __le64_to_cpu(x)
-#define __constant_be16_to_cpu(x) __be16_to_cpu(x)
-#define __constant_be32_to_cpu(x) __be32_to_cpu(x)
-#define __constant_be64_to_cpu(x) __be64_to_cpu(x)
-
-#define __constant_cpu_to_le16(x) __cpu_to_le16(x)
-#define __constant_cpu_to_le32(x) __cpu_to_le32(x)
-#define __constant_cpu_to_le64(x) __cpu_to_le64(x)
-#define __constant_cpu_to_be16(x) __cpu_to_be16(x)
-#define __constant_cpu_to_be32(x) __cpu_to_be32(x)
-#define __constant_cpu_to_be64(x) __cpu_to_be64(x)
-
-static inline void __le16_to_cpus(__u16 *p)
-{
-#ifdef __BIG_ENDIAN
-	__swab16s(p);
-#endif
-}
-
-static inline void __cpu_to_le16s(__u16 *p)
-{
-#ifdef __BIG_ENDIAN
-	__swab16s(p);
-#endif
-}
-
-static inline void __le32_to_cpus(__u32 *p)
-{
-#ifdef __BIG_ENDIAN
-	__swab32s(p);
-#endif
-}
-
-static inline void __cpu_to_le32s(__u32 *p)
-{
-#ifdef __BIG_ENDIAN
-	__swab32s(p);
-#endif
-}
-
-static inline void __le64_to_cpus(__u64 *p)
-{
-#ifdef __BIG_ENDIAN
-	__swab64s(p);
-#endif
-}
-
-static inline void __cpu_to_le64s(__u64 *p)
-{
-#ifdef __BIG_ENDIAN
-	__swab64s(p);
-#endif
-}
-
-static inline void __be16_to_cpus(__u16 *p)
-{
-#ifdef __LITTLE_ENDIAN
-	__swab16s(p);
-#endif
-}
-
-static inline void __cpu_to_be16s(__u16 *p)
-{
-#ifdef __LITTLE_ENDIAN
-	__swab16s(p);
-#endif
-}
-
-static inline void __be32_to_cpus(__u32 *p)
-{
-#ifdef __LITTLE_ENDIAN
-	__swab32s(p);
-#endif
-}
-
-static inline void __cpu_to_be32s(__u32 *p)
-{
-#ifdef __LITTLE_ENDIAN
-	__swab32s(p);
-#endif
-}
-
-static inline void __be64_to_cpus(__u64 *p)
-{
-#ifdef __LITTLE_ENDIAN
-	__swab64s(p);
-#endif
-}
-
-static inline void __cpu_to_be64s(__u64 *p)
-{
-#ifdef __LITTLE_ENDIAN
-	__swab64s(p);
-#endif
-}
-
-static inline __u16 __le16_to_cpup(const __le16 *p)
-{
-#ifdef __LITTLE_ENDIAN
-	return (__force __u16)*p;
-#else
-	return __swab16p((__force __u16 *)p);
-#endif
-}
-
-static inline __u32 __le32_to_cpup(const __le32 *p)
-{
-#ifdef __LITTLE_ENDIAN
-	return (__force __u32)*p;
-#else
-	return __swab32p((__force __u32 *)p);
-#endif
-}
-
-static inline __u64 __le64_to_cpup(const __le64 *p)
-{
-#ifdef __LITTLE_ENDIAN
-	return (__force __u64)*p;
-#else
-	return __swab64p((__force __u64 *)p);
-#endif
-}
-
-static inline __le16 __cpu_to_le16p(const __u16 *p)
-{
-#ifdef __LITTLE_ENDIAN
-	return (__force __le16)*p;
-#else
-	return (__force __le16)__swab16p(p);
-#endif
-}
-
-static inline __le32 __cpu_to_le32p(const __u32 *p)
-{
-#ifdef __LITTLE_ENDIAN
-	return (__force __le32)*p;
-#else
-	return (__force __le32)__swab32p(p);
-#endif
-}
-
-static inline __le64 __cpu_to_le64p(const __u64 *p)
-{
-#ifdef __LITTLE_ENDIAN
-	return (__force __le64)*p;
-#else
-	return (__force __le64)__swab64p(p);
-#endif
-}
-
-static inline __u16 __be16_to_cpup(const __be16 *p)
-{
-#ifdef __BIG_ENDIAN
-	return (__force __u16)*p;
-#else
-	return __swab16p((__force __u16 *)p);
-#endif
-}
-
-static inline __u32 __be32_to_cpup(const __be32 *p)
-{
-#ifdef __BIG_ENDIAN
-	return (__force __u32)*p;
-#else
-	return __swab32p((__force __u32 *)p);
-#endif
-}
-
-static inline __u64 __be64_to_cpup(const __be64 *p)
-{
-#ifdef __BIG_ENDIAN
-	return (__force __u64)*p;
-#else
-	return __swab64p((__force __u64 *)p);
-#endif
-}
-
-static inline __be16 __cpu_to_be16p(const __u16 *p)
-{
-#ifdef __BIG_ENDIAN
-	return (__force __be16)*p;
-#else
-	return (__force __be16)__swab16p(p);
-#endif
-}
-
-static inline __be32 __cpu_to_be32p(const __u32 *p)
-{
-#ifdef __BIG_ENDIAN
-	return (__force __be32)*p;
-#else
-	return (__force __be32)__swab32p(p);
-#endif
-}
-
-static inline __be64 __cpu_to_be64p(const __u64 *p)
-{
-#ifdef __BIG_ENDIAN
-	return (__force __be64)*p;
-#else
-	return (__force __be64)__swab64p(p);
-#endif
-}
-
-#ifdef __KERNEL__
-
-# define le16_to_cpu __le16_to_cpu
-# define le32_to_cpu __le32_to_cpu
-# define le64_to_cpu __le64_to_cpu
-# define be16_to_cpu __be16_to_cpu
-# define be32_to_cpu __be32_to_cpu
-# define be64_to_cpu __be64_to_cpu
-# define cpu_to_le16 __cpu_to_le16
-# define cpu_to_le32 __cpu_to_le32
-# define cpu_to_le64 __cpu_to_le64
-# define cpu_to_be16 __cpu_to_be16
-# define cpu_to_be32 __cpu_to_be32
-# define cpu_to_be64 __cpu_to_be64
-
-# define le16_to_cpup __le16_to_cpup
-# define le32_to_cpup __le32_to_cpup
-# define le64_to_cpup __le64_to_cpup
-# define be16_to_cpup __be16_to_cpup
-# define be32_to_cpup __be32_to_cpup
-# define be64_to_cpup __be64_to_cpup
-# define cpu_to_le16p __cpu_to_le16p
-# define cpu_to_le32p __cpu_to_le32p
-# define cpu_to_le64p __cpu_to_le64p
-# define cpu_to_be16p __cpu_to_be16p
-# define cpu_to_be32p __cpu_to_be32p
-# define cpu_to_be64p __cpu_to_be64p
-
-# define le16_to_cpus __le16_to_cpus
-# define le32_to_cpus __le32_to_cpus
-# define le64_to_cpus __le64_to_cpus
-# define be16_to_cpus __be16_to_cpus
-# define be32_to_cpus __be32_to_cpus
-# define be64_to_cpus __be64_to_cpus
-# define cpu_to_le16s __cpu_to_le16s
-# define cpu_to_le32s __cpu_to_le32s
-# define cpu_to_le64s __cpu_to_le64s
-# define cpu_to_be16s __cpu_to_be16s
-# define cpu_to_be32s __cpu_to_be32s
-# define cpu_to_be64s __cpu_to_be64s
-
-/*
- * They have to be macros in order to do the constant folding
- * correctly - if the argument passed into a inline function
- * it is no longer constant according to gcc..
- */
-# undef ntohl
-# undef ntohs
-# undef htonl
-# undef htons
-
-# define ___htonl(x) __cpu_to_be32(x)
-# define ___htons(x) __cpu_to_be16(x)
-# define ___ntohl(x) __be32_to_cpu(x)
-# define ___ntohs(x) __be16_to_cpu(x)
-
-# define htonl(x) ___htonl(x)
-# define ntohl(x) ___ntohl(x)
-# define htons(x) ___htons(x)
-# define ntohs(x) ___ntohs(x)
-
-static inline void le16_add_cpu(__le16 *var, u16 val)
-{
-	*var = cpu_to_le16(le16_to_cpup(var) + val);
-}
-
-static inline void le32_add_cpu(__le32 *var, u32 val)
-{
-	*var = cpu_to_le32(le32_to_cpup(var) + val);
-}
-
-static inline void le64_add_cpu(__le64 *var, u64 val)
-{
-	*var = cpu_to_le64(le64_to_cpup(var) + val);
-}
-
-static inline void be16_add_cpu(__be16 *var, u16 val)
-{
-	*var = cpu_to_be16(be16_to_cpup(var) + val);
-}
-
-static inline void be32_add_cpu(__be32 *var, u32 val)
-{
-	*var = cpu_to_be32(be32_to_cpup(var) + val);
-}
-
-static inline void be64_add_cpu(__be64 *var, u64 val)
-{
-	*var = cpu_to_be64(be64_to_cpup(var) + val);
-}
-
-#endif /* __KERNEL__ */
-#endif /* _LINUX_BYTEORDER_H */
-- 
cgit v1.2.3


From ede6f5aea054d3fb67c78857f7abdee602302043 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Tue, 6 Jan 2009 21:17:57 -0800
Subject: Fix up 64-bit byte swaps for most 32-bit architectures

The __SWAB_64_THRU_32__ case of a 64-bit byte swap was depending on the
no-longer-existant ___swab32() method (three underscores).  We got rid
of some of the worst indirection and complexity, and now it should just
use the 32-bit swab function that was defined right above it.

Reported-and-tested-by: Nicolas Pitre <nico@cam.org>
Reported-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swab.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/swab.h b/include/linux/swab.h
index 9a2d33e0a98a..be5284d4a053 100644
--- a/include/linux/swab.h
+++ b/include/linux/swab.h
@@ -68,7 +68,7 @@ static inline __attribute_const__ __u64 __fswab64(__u64 val)
 #elif defined(__SWAB_64_THRU_32__)
 	__u32 h = val >> 32;
 	__u32 l = val & ((1ULL << 32) - 1);
-	return (((__u64)___swab32(l)) << 32) | ((__u64)(___swab32(h)));
+	return (((__u64)__fswab32(l)) << 32) | ((__u64)(__fswab32(h)));
 #else
 	return ___constant_swab64(val);
 #endif
-- 
cgit v1.2.3


From e1995f65be0786ca201f466f049dad1e2e4c3421 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Wed, 7 Jan 2009 14:29:16 +0100
Subject: i2c: Drop I2C_CLASS_ALL

I2C_CLASS_ALL is almost never what bus driver authors really want.
These i2c classes are really only about which devices must be probed,
not what devices can be present. As device drivers get converted to the
new i2c device driver model, only a few device types will keep relying
on probing.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Sonic Zhang <sonic.zhang@analog.com>
---
 drivers/i2c/busses/i2c-bfin-twi.c | 2 +-
 drivers/i2c/busses/i2c-sh7760.c   | 2 +-
 include/linux/i2c.h               | 1 -
 3 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/i2c/busses/i2c-bfin-twi.c b/drivers/i2c/busses/i2c-bfin-twi.c
index 3c855ff2992f..3fd2c417c1e0 100644
--- a/drivers/i2c/busses/i2c-bfin-twi.c
+++ b/drivers/i2c/busses/i2c-bfin-twi.c
@@ -656,7 +656,7 @@ static int i2c_bfin_twi_probe(struct platform_device *pdev)
 	strlcpy(p_adap->name, pdev->name, sizeof(p_adap->name));
 	p_adap->algo = &bfin_twi_algorithm;
 	p_adap->algo_data = iface;
-	p_adap->class = I2C_CLASS_ALL;
+	p_adap->class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	p_adap->dev.parent = &pdev->dev;
 
 	rc = peripheral_request_list(pin_req[pdev->id], "i2c-bfin-twi");
diff --git a/drivers/i2c/busses/i2c-sh7760.c b/drivers/i2c/busses/i2c-sh7760.c
index 5e0e254976de..baa28b73ae42 100644
--- a/drivers/i2c/busses/i2c-sh7760.c
+++ b/drivers/i2c/busses/i2c-sh7760.c
@@ -475,7 +475,7 @@ static int __devinit sh7760_i2c_probe(struct platform_device *pdev)
 
 	id->adap.nr = pdev->id;
 	id->adap.algo = &sh7760_i2c_algo;
-	id->adap.class = I2C_CLASS_ALL;
+	id->adap.class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	id->adap.retries = 3;
 	id->adap.algo_data = id;
 	id->adap.dev.parent = &pdev->dev;
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 33a5992d4936..0184de4050b8 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -397,7 +397,6 @@ static inline void i2c_set_adapdata(struct i2c_adapter *dev, void *data)
 #define I2C_CLASS_CAM_DIGITAL	(1<<5)	/* most webcams */
 #define I2C_CLASS_SOUND		(1<<6)	/* sound devices */
 #define I2C_CLASS_SPD		(1<<7)	/* SPD EEPROMs and similar */
-#define I2C_CLASS_ALL		(UINT_MAX) /* all of the above */
 
 /* i2c_client_address_data is the struct for holding default client
  * addresses for a driver and for the parameters supplied on the
-- 
cgit v1.2.3


From 994a075f0f2e8cdb919d8e495f98211651e3c461 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Wed, 7 Jan 2009 14:29:17 +0100
Subject: i2c: Drop I2C_CLASS_CAM_ANALOG and I2C_CLASS_SOUND

There are no users left of these two i2c probe class flags so we can
drop the now.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 include/linux/i2c.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 0184de4050b8..4fd79d2e4021 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -393,9 +393,7 @@ static inline void i2c_set_adapdata(struct i2c_adapter *dev, void *data)
 #define I2C_CLASS_TV_ANALOG	(1<<1)	/* bttv + friends */
 #define I2C_CLASS_TV_DIGITAL	(1<<2)	/* dvb cards */
 #define I2C_CLASS_DDC		(1<<3)	/* DDC bus on graphics adapters */
-#define I2C_CLASS_CAM_ANALOG	(1<<4)	/* camera with analog CCD */
 #define I2C_CLASS_CAM_DIGITAL	(1<<5)	/* most webcams */
-#define I2C_CLASS_SOUND		(1<<6)	/* sound devices */
 #define I2C_CLASS_SPD		(1<<7)	/* SPD EEPROMs and similar */
 
 /* i2c_client_address_data is the struct for holding default client
-- 
cgit v1.2.3


From b305271861219f0ce162eb565f0f28f4c781299d Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Wed, 7 Jan 2009 14:29:17 +0100
Subject: i2c: Drop I2C_CLASS_CAM_DIGITAL

There are a number of drivers which set their i2c bus class to
I2C_CLASS_CAM_DIGITAL, however no chip driver actually checks for this
flag, so we might as well drop it now.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 drivers/media/video/cafe_ccic.c                | 1 -
 drivers/media/video/ov7670.c                   | 1 -
 drivers/media/video/ovcamchip/ovcamchip_core.c | 1 -
 drivers/media/video/w9968cf.c                  | 1 -
 include/linux/i2c.h                            | 1 -
 5 files changed, 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/media/video/cafe_ccic.c b/drivers/media/video/cafe_ccic.c
index 34a39d2e4703..46fd573a4f15 100644
--- a/drivers/media/video/cafe_ccic.c
+++ b/drivers/media/video/cafe_ccic.c
@@ -569,7 +569,6 @@ static int cafe_smbus_setup(struct cafe_camera *cam)
 
 	cafe_smbus_enable_irq(cam);
 	adap->id = I2C_HW_SMBUS_CAFE;
-	adap->class = I2C_CLASS_CAM_DIGITAL;
 	adap->owner = THIS_MODULE;
 	adap->client_register = cafe_smbus_attach;
 	adap->client_unregister = cafe_smbus_detach;
diff --git a/drivers/media/video/ov7670.c b/drivers/media/video/ov7670.c
index ca26b0c50cf2..05c14a29375a 100644
--- a/drivers/media/video/ov7670.c
+++ b/drivers/media/video/ov7670.c
@@ -1347,7 +1347,6 @@ static struct i2c_driver ov7670_driver = {
 		.name = "ov7670",
 	},
 	.id 		= I2C_DRIVERID_OV7670,
-	.class 		= I2C_CLASS_CAM_DIGITAL,
 	.attach_adapter = ov7670_attach,
 	.detach_client	= ov7670_detach,
 	.command	= ov7670_command,
diff --git a/drivers/media/video/ovcamchip/ovcamchip_core.c b/drivers/media/video/ovcamchip/ovcamchip_core.c
index 2c4acbf5a4fe..c841f4e4fbe4 100644
--- a/drivers/media/video/ovcamchip/ovcamchip_core.c
+++ b/drivers/media/video/ovcamchip/ovcamchip_core.c
@@ -405,7 +405,6 @@ static struct i2c_driver driver = {
 		.name =		"ovcamchip",
 	},
 	.id =			I2C_DRIVERID_OVCAMCHIP,
-	.class =		I2C_CLASS_CAM_DIGITAL,
 	.attach_adapter =	ovcamchip_attach,
 	.detach_client =	ovcamchip_detach,
 	.command =		ovcamchip_command,
diff --git a/drivers/media/video/w9968cf.c b/drivers/media/video/w9968cf.c
index a3997b7d4366..105a832531f2 100644
--- a/drivers/media/video/w9968cf.c
+++ b/drivers/media/video/w9968cf.c
@@ -1553,7 +1553,6 @@ static int w9968cf_i2c_init(struct w9968cf_device* cam)
 
 	static struct i2c_adapter adap = {
 		.id =                I2C_HW_SMBUS_W9968CF,
-		.class =             I2C_CLASS_CAM_DIGITAL,
 		.owner =             THIS_MODULE,
 		.client_register =   w9968cf_i2c_attach_inform,
 		.client_unregister = w9968cf_i2c_detach_inform,
diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 4fd79d2e4021..20873d402467 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -393,7 +393,6 @@ static inline void i2c_set_adapdata(struct i2c_adapter *dev, void *data)
 #define I2C_CLASS_TV_ANALOG	(1<<1)	/* bttv + friends */
 #define I2C_CLASS_TV_DIGITAL	(1<<2)	/* dvb cards */
 #define I2C_CLASS_DDC		(1<<3)	/* DDC bus on graphics adapters */
-#define I2C_CLASS_CAM_DIGITAL	(1<<5)	/* most webcams */
 #define I2C_CLASS_SPD		(1<<7)	/* SPD EEPROMs and similar */
 
 /* i2c_client_address_data is the struct for holding default client
-- 
cgit v1.2.3


From 22a9d645677feefd402befd02edd59b122289ef1 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Wed, 7 Jan 2009 08:45:46 -0800
Subject: async: Asynchronous function calls to speed up kernel boot

Right now, most of the kernel boot is strictly synchronous, such that
various hardware delays are done sequentially.

In order to make the kernel boot faster, this patch introduces
infrastructure to allow doing some of the initialization steps
asynchronously, which will hide significant portions of the hardware delays
in practice.

In order to not change device order and other similar observables, this
patch does NOT do full parallel initialization.

Rather, it operates more in the way an out of order CPU does; the work may
be done out of order and asynchronous, but the observable effects
(instruction retiring for the CPU) are still done in the original sequence.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
---
 include/linux/async.h  |  25 ++++
 init/do_mounts.c       |   2 +
 init/main.c            |   5 +-
 kernel/Makefile        |   3 +-
 kernel/async.c         | 321 +++++++++++++++++++++++++++++++++++++++++++++++++
 kernel/irq/autoprobe.c |   5 +
 kernel/module.c        |   2 +
 7 files changed, 361 insertions(+), 2 deletions(-)
 create mode 100644 include/linux/async.h
 create mode 100644 kernel/async.c

(limited to 'include/linux')

diff --git a/include/linux/async.h b/include/linux/async.h
new file mode 100644
index 000000000000..c4ecacd0b327
--- /dev/null
+++ b/include/linux/async.h
@@ -0,0 +1,25 @@
+/*
+ * async.h: Asynchronous function calls for boot performance
+ *
+ * (C) Copyright 2009 Intel Corporation
+ * Author: Arjan van de Ven <arjan@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/types.h>
+#include <linux/list.h>
+
+typedef u64 async_cookie_t;
+typedef void (async_func_ptr) (void *data, async_cookie_t cookie);
+
+extern async_cookie_t async_schedule(async_func_ptr *ptr, void *data);
+extern async_cookie_t async_schedule_special(async_func_ptr *ptr, void *data, struct list_head *list);
+extern void async_synchronize_full(void);
+extern void async_synchronize_full_special(struct list_head *list);
+extern void async_synchronize_cookie(async_cookie_t cookie);
+extern void async_synchronize_cookie_special(async_cookie_t cookie, struct list_head *list);
+
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 5efca73b39f9..708105e163df 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -13,6 +13,7 @@
 #include <linux/init.h>
 #include <linux/fs.h>
 #include <linux/initrd.h>
+#include <linux/async.h>
 
 #include <linux/nfs_fs.h>
 #include <linux/nfs_fs_sb.h>
@@ -372,6 +373,7 @@ void __init prepare_namespace(void)
 	/* wait for the known devices to complete their probing */
 	while (driver_probe_done() != 0)
 		msleep(100);
+	async_synchronize_full();
 
 	md_run_setup();
 
diff --git a/init/main.c b/init/main.c
index b5a892c68375..f66715d8a853 100644
--- a/init/main.c
+++ b/init/main.c
@@ -62,6 +62,7 @@
 #include <linux/signal.h>
 #include <linux/idr.h>
 #include <linux/ftrace.h>
+#include <linux/async.h>
 #include <trace/boot.h>
 
 #include <asm/io.h>
@@ -684,7 +685,7 @@ asmlinkage void __init start_kernel(void)
 	rest_init();
 }
 
-static int initcall_debug;
+int initcall_debug;
 core_param(initcall_debug, initcall_debug, bool, 0644);
 
 int do_one_initcall(initcall_t fn)
@@ -785,6 +786,8 @@ static void run_init_process(char *init_filename)
  */
 static noinline int init_post(void)
 {
+	/* need to finish all async __init code before freeing the memory */
+	async_synchronize_full();
 	free_initmem();
 	unlock_kernel();
 	mark_rodata_ro();
diff --git a/kernel/Makefile b/kernel/Makefile
index e1c5bf3365c0..2921d90ce32f 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -9,7 +9,8 @@ obj-y     = sched.o fork.o exec_domain.o panic.o printk.o \
 	    rcupdate.o extable.o params.o posix-timers.o \
 	    kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \
 	    hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \
-	    notifier.o ksysfs.o pm_qos_params.o sched_clock.o cred.o
+	    notifier.o ksysfs.o pm_qos_params.o sched_clock.o cred.o \
+	    async.o
 
 ifdef CONFIG_FUNCTION_TRACER
 # Do not trace debug files and internal ftrace files
diff --git a/kernel/async.c b/kernel/async.c
new file mode 100644
index 000000000000..afaa8a653d5a
--- /dev/null
+++ b/kernel/async.c
@@ -0,0 +1,321 @@
+/*
+ * async.c: Asynchronous function calls for boot performance
+ *
+ * (C) Copyright 2009 Intel Corporation
+ * Author: Arjan van de Ven <arjan@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+
+/*
+
+Goals and Theory of Operation
+
+The primary goal of this feature is to reduce the kernel boot time,
+by doing various independent hardware delays and discovery operations
+decoupled and not strictly serialized.
+
+More specifically, the asynchronous function call concept allows
+certain operations (primarily during system boot) to happen
+asynchronously, out of order, while these operations still
+have their externally visible parts happen sequentially and in-order.
+(not unlike how out-of-order CPUs retire their instructions in order)
+
+Key to the asynchronous function call implementation is the concept of
+a "sequence cookie" (which, although it has an abstracted type, can be
+thought of as a monotonically incrementing number).
+
+The async core will assign each scheduled event such a sequence cookie and
+pass this to the called functions.
+
+The asynchronously called function should before doing a globally visible
+operation, such as registering device numbers, call the
+async_synchronize_cookie() function and pass in its own cookie. The
+async_synchronize_cookie() function will make sure that all asynchronous
+operations that were scheduled prior to the operation corresponding with the
+cookie have completed.
+
+Subsystem/driver initialization code that scheduled asynchronous probe
+functions, but which shares global resources with other drivers/subsystems
+that do not use the asynchronous call feature, need to do a full
+synchronization with the async_synchronize_full() function, before returning
+from their init function. This is to maintain strict ordering between the
+asynchronous and synchronous parts of the kernel.
+
+*/
+
+#include <linux/async.h>
+#include <linux/module.h>
+#include <linux/wait.h>
+#include <linux/sched.h>
+#include <linux/init.h>
+#include <linux/kthread.h>
+#include <asm/atomic.h>
+
+static async_cookie_t next_cookie = 1;
+
+#define MAX_THREADS	256
+#define MAX_WORK	32768
+
+static LIST_HEAD(async_pending);
+static LIST_HEAD(async_running);
+static DEFINE_SPINLOCK(async_lock);
+
+struct async_entry {
+	struct list_head list;
+	async_cookie_t   cookie;
+	async_func_ptr	 *func;
+	void             *data;
+	struct list_head *running;
+};
+
+static DECLARE_WAIT_QUEUE_HEAD(async_done);
+static DECLARE_WAIT_QUEUE_HEAD(async_new);
+
+static atomic_t entry_count;
+static atomic_t thread_count;
+
+extern int initcall_debug;
+
+
+/*
+ * MUST be called with the lock held!
+ */
+static async_cookie_t  __lowest_in_progress(struct list_head *running)
+{
+	struct async_entry *entry;
+	if (!list_empty(&async_pending)) {
+		entry = list_first_entry(&async_pending,
+			struct async_entry, list);
+		return entry->cookie;
+	} else if (!list_empty(running)) {
+		entry = list_first_entry(running,
+			struct async_entry, list);
+		return entry->cookie;
+	} else {
+		/* nothing in progress... next_cookie is "infinity" */
+		return next_cookie;
+	}
+
+}
+/*
+ * pick the first pending entry and run it
+ */
+static void run_one_entry(void)
+{
+	unsigned long flags;
+	struct async_entry *entry;
+	ktime_t calltime, delta, rettime;
+
+	/* 1) pick one task from the pending queue */
+
+	spin_lock_irqsave(&async_lock, flags);
+	if (list_empty(&async_pending))
+		goto out;
+	entry = list_first_entry(&async_pending, struct async_entry, list);
+
+	/* 2) move it to the running queue */
+	list_del(&entry->list);
+	list_add_tail(&entry->list, &async_running);
+	spin_unlock_irqrestore(&async_lock, flags);
+
+	/* 3) run it (and print duration)*/
+	if (initcall_debug) {
+		printk("calling  %lli_%pF @ %i\n", entry->cookie, entry->func, task_pid_nr(current));
+		calltime = ktime_get();
+	}
+	entry->func(entry->data, entry->cookie);
+	if (initcall_debug) {
+		rettime = ktime_get();
+		delta = ktime_sub(rettime, calltime);
+		printk("initcall %lli_%pF returned 0 after %lld usecs\n", entry->cookie,
+			entry->func, ktime_to_ns(delta) >> 10);
+	}
+
+	/* 4) remove it from the running queue */
+	spin_lock_irqsave(&async_lock, flags);
+	list_del(&entry->list);
+
+	/* 5) free the entry  */
+	kfree(entry);
+	atomic_dec(&entry_count);
+
+	spin_unlock_irqrestore(&async_lock, flags);
+
+	/* 6) wake up any waiters. */
+	wake_up(&async_done);
+	return;
+
+out:
+	spin_unlock_irqrestore(&async_lock, flags);
+}
+
+
+static async_cookie_t __async_schedule(async_func_ptr *ptr, void *data, struct list_head *running)
+{
+	struct async_entry *entry;
+	unsigned long flags;
+	async_cookie_t newcookie;
+	
+
+	/* allow irq-off callers */
+	entry = kzalloc(sizeof(struct async_entry), GFP_ATOMIC);
+
+	/*
+	 * If we're out of memory or if there's too much work
+	 * pending already, we execute synchronously.
+	 */
+	if (!entry || atomic_read(&entry_count) > MAX_WORK) {
+		kfree(entry);
+		spin_lock_irqsave(&async_lock, flags);
+		newcookie = next_cookie++;
+		spin_unlock_irqrestore(&async_lock, flags);
+
+		/* low on memory.. run synchronously */
+		ptr(data, newcookie);
+		return newcookie;
+	}
+	entry->func = ptr;
+	entry->data = data;
+	entry->running = running;
+
+	spin_lock_irqsave(&async_lock, flags);
+	newcookie = entry->cookie = next_cookie++;
+	list_add_tail(&entry->list, &async_pending);
+	atomic_inc(&entry_count);
+	spin_unlock_irqrestore(&async_lock, flags);
+	wake_up(&async_new);
+	return newcookie;
+}
+
+async_cookie_t async_schedule(async_func_ptr *ptr, void *data)
+{
+	return __async_schedule(ptr, data, &async_pending);
+}
+EXPORT_SYMBOL_GPL(async_schedule);
+
+async_cookie_t async_schedule_special(async_func_ptr *ptr, void *data, struct list_head *running)
+{
+	return __async_schedule(ptr, data, running);
+}
+EXPORT_SYMBOL_GPL(async_schedule_special);
+
+void async_synchronize_full(void)
+{
+	async_synchronize_cookie(next_cookie);
+}
+EXPORT_SYMBOL_GPL(async_synchronize_full);
+
+void async_synchronize_full_special(struct list_head *list)
+{
+	async_synchronize_cookie_special(next_cookie, list);
+}
+EXPORT_SYMBOL_GPL(async_synchronize_full_special);
+
+void async_synchronize_cookie_special(async_cookie_t cookie, struct list_head *running)
+{
+	ktime_t starttime, delta, endtime;
+
+	if (initcall_debug) {
+		printk("async_waiting @ %i\n", task_pid_nr(current));
+		starttime = ktime_get();
+	}
+
+	wait_event(async_done, __lowest_in_progress(running) >= cookie);
+
+	if (initcall_debug) {
+		endtime = ktime_get();
+		delta = ktime_sub(endtime, starttime);
+
+		printk("async_continuing @ %i after %lli usec\n",
+			task_pid_nr(current), ktime_to_ns(delta) >> 10);
+	}
+}
+EXPORT_SYMBOL_GPL(async_synchronize_cookie_special);
+
+void async_synchronize_cookie(async_cookie_t cookie)
+{
+	async_synchronize_cookie_special(cookie, &async_running);
+}
+EXPORT_SYMBOL_GPL(async_synchronize_cookie);
+
+
+static int async_thread(void *unused)
+{
+	DECLARE_WAITQUEUE(wq, current);
+	add_wait_queue(&async_new, &wq);
+
+	while (!kthread_should_stop()) {
+		int ret = HZ;
+		set_current_state(TASK_INTERRUPTIBLE);
+		/*
+		 * check the list head without lock.. false positives
+		 * are dealt with inside run_one_entry() while holding
+		 * the lock.
+		 */
+		rmb();
+		if (!list_empty(&async_pending))
+			run_one_entry();
+		else
+			ret = schedule_timeout(HZ);
+
+		if (ret == 0) {
+			/*
+			 * we timed out, this means we as thread are redundant.
+			 * we sign off and die, but we to avoid any races there
+			 * is a last-straw check to see if work snuck in.
+			 */
+			atomic_dec(&thread_count);
+			wmb(); /* manager must see our departure first */
+			if (list_empty(&async_pending))
+				break;
+			/*
+			 * woops work came in between us timing out and us
+			 * signing off; we need to stay alive and keep working.
+			 */
+			atomic_inc(&thread_count);
+		}
+	}
+	remove_wait_queue(&async_new, &wq);
+
+	return 0;
+}
+
+static int async_manager_thread(void *unused)
+{
+	DECLARE_WAITQUEUE(wq, current);
+	add_wait_queue(&async_new, &wq);
+
+	while (!kthread_should_stop()) {
+		int tc, ec;
+
+		set_current_state(TASK_INTERRUPTIBLE);
+
+		tc = atomic_read(&thread_count);
+		rmb();
+		ec = atomic_read(&entry_count);
+
+		while (tc < ec && tc < MAX_THREADS) {
+			kthread_run(async_thread, NULL, "async/%i", tc);
+			atomic_inc(&thread_count);
+			tc++;
+		}
+
+		schedule();
+	}
+	remove_wait_queue(&async_new, &wq);
+
+	return 0;
+}
+
+static int __init async_init(void)
+{
+	kthread_run(async_manager_thread, NULL, "async/mgr");
+	return 0;
+}
+
+core_initcall(async_init);
diff --git a/kernel/irq/autoprobe.c b/kernel/irq/autoprobe.c
index cc0f7321b8ce..1de9700f416e 100644
--- a/kernel/irq/autoprobe.c
+++ b/kernel/irq/autoprobe.c
@@ -10,6 +10,7 @@
 #include <linux/module.h>
 #include <linux/interrupt.h>
 #include <linux/delay.h>
+#include <linux/async.h>
 
 #include "internals.h"
 
@@ -34,6 +35,10 @@ unsigned long probe_irq_on(void)
 	unsigned int status;
 	int i;
 
+	/*
+	 * quiesce the kernel, or at least the asynchronous portion
+	 */
+	async_synchronize_full();
 	mutex_lock(&probing_active);
 	/*
 	 * something may have generated an irq long ago and we want to
diff --git a/kernel/module.c b/kernel/module.c
index 496dcb57b608..c9332c90d5a0 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -50,6 +50,7 @@
 #include <asm/sections.h>
 #include <linux/tracepoint.h>
 #include <linux/ftrace.h>
+#include <linux/async.h>
 
 #if 0
 #define DEBUGP printk
@@ -816,6 +817,7 @@ sys_delete_module(const char __user *name_user, unsigned int flags)
 		mod->exit();
 	blocking_notifier_call_chain(&module_notify_list,
 				     MODULE_STATE_GOING, mod);
+	async_synchronize_full();
 	mutex_lock(&module_mutex);
 	/* Store the name of the last unloaded module for diagnostic purposes */
 	strlcpy(last_unloaded_module, mod->name, sizeof(last_unloaded_module));
-- 
cgit v1.2.3


From efaee192063a54749c56b7383803e16fe553630e Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Tue, 6 Jan 2009 07:20:54 -0800
Subject: async: make the final inode deletion an asynchronous event

this makes "rm -rf" on a (names cached) kernel tree go from
11.6 to 8.6 seconds on an ext3 filesystem

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
---
 fs/inode.c         | 20 +++++++++++++-------
 fs/super.c         | 10 ++++++++++
 include/linux/fs.h |  5 +++++
 3 files changed, 28 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/fs/inode.c b/fs/inode.c
index 7a6e8c2ff7b1..0013ac1af8e7 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -22,6 +22,7 @@
 #include <linux/bootmem.h>
 #include <linux/inotify.h>
 #include <linux/mount.h>
+#include <linux/async.h>
 
 /*
  * This is needed for the following functions:
@@ -1138,16 +1139,11 @@ EXPORT_SYMBOL(remove_inode_hash);
  * I_FREEING is set so that no-one will take a new reference to the inode while
  * it is being deleted.
  */
-void generic_delete_inode(struct inode *inode)
+static void generic_delete_inode_async(void *data, async_cookie_t cookie)
 {
+	struct inode *inode = data;
 	const struct super_operations *op = inode->i_sb->s_op;
 
-	list_del_init(&inode->i_list);
-	list_del_init(&inode->i_sb_list);
-	inode->i_state |= I_FREEING;
-	inodes_stat.nr_inodes--;
-	spin_unlock(&inode_lock);
-
 	security_inode_delete(inode);
 
 	if (op->delete_inode) {
@@ -1171,6 +1167,16 @@ void generic_delete_inode(struct inode *inode)
 	destroy_inode(inode);
 }
 
+void generic_delete_inode(struct inode *inode)
+{
+	list_del_init(&inode->i_list);
+	list_del_init(&inode->i_sb_list);
+	inode->i_state |= I_FREEING;
+	inodes_stat.nr_inodes--;
+	spin_unlock(&inode_lock);
+	async_schedule_special(generic_delete_inode_async, inode, &inode->i_sb->s_async_list);
+}
+
 EXPORT_SYMBOL(generic_delete_inode);
 
 static void generic_forget_inode(struct inode *inode)
diff --git a/fs/super.c b/fs/super.c
index ddba069d7a99..cb20744ec789 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -38,6 +38,7 @@
 #include <linux/kobject.h>
 #include <linux/mutex.h>
 #include <linux/file.h>
+#include <linux/async.h>
 #include <asm/uaccess.h>
 #include "internal.h"
 
@@ -71,6 +72,7 @@ static struct super_block *alloc_super(struct file_system_type *type)
 		INIT_HLIST_HEAD(&s->s_anon);
 		INIT_LIST_HEAD(&s->s_inodes);
 		INIT_LIST_HEAD(&s->s_dentry_lru);
+		INIT_LIST_HEAD(&s->s_async_list);
 		init_rwsem(&s->s_umount);
 		mutex_init(&s->s_lock);
 		lockdep_set_class(&s->s_umount, &type->s_umount_key);
@@ -289,11 +291,18 @@ void generic_shutdown_super(struct super_block *sb)
 {
 	const struct super_operations *sop = sb->s_op;
 
+
 	if (sb->s_root) {
 		shrink_dcache_for_umount(sb);
 		fsync_super(sb);
 		lock_super(sb);
 		sb->s_flags &= ~MS_ACTIVE;
+
+		/*
+		 * wait for asynchronous fs operations to finish before going further
+		 */
+		async_synchronize_full_special(&sb->s_async_list);
+
 		/* bad name - it should be evict_inodes() */
 		invalidate_inodes(sb);
 		lock_kernel();
@@ -449,6 +458,7 @@ void sync_filesystems(int wait)
 		if (sb->s_flags & MS_RDONLY)
 			continue;
 		sb->s_need_sync_fs = 1;
+		async_synchronize_full_special(&sb->s_async_list);
 	}
 
 restart:
diff --git a/include/linux/fs.h b/include/linux/fs.h
index d7eba77f666e..e38a64d71eff 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1184,6 +1184,11 @@ struct super_block {
 	 * generic_show_options()
 	 */
 	char *s_options;
+
+	/*
+	 * storage for asynchronous operations
+	 */
+	struct list_head s_async_list;
 };
 
 extern struct timespec current_fs_time(struct super_block *sb);
-- 
cgit v1.2.3


From 490dea45d00f01847ebebd007685d564aaf2cd98 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <peterz@infradead.org>
Date: Mon, 24 Nov 2008 17:06:57 +0100
Subject: itimers: remove the per-cpu-ish-ness

Either we bounce once cacheline per cpu per tick, yielding n^2 bounces
or we just bounce a single..

Also, using per-cpu allocations for the thread-groups complicates the
per-cpu allocator in that its currently aimed to be a fixed sized
allocator and the only possible extention to that would be vmap based,
which is seriously constrained on 32 bit archs.

So making the per-cpu memory requirement depend on the number of
processes is an issue.

Lastly, it didn't deal with cpu-hotplug, although admittedly that might
be fixable.

Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/init_task.h |  6 ++++
 include/linux/sched.h     | 29 ++++++++++++--------
 kernel/fork.c             | 15 +++++-----
 kernel/posix-cpu-timers.c | 70 -----------------------------------------------
 kernel/sched_stats.h      | 33 ++++++++++------------
 5 files changed, 46 insertions(+), 107 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 2f3c2d4ef73b..ea0ea1a4c36f 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -48,6 +48,12 @@ extern struct fs_struct init_fs;
 	.posix_timers	 = LIST_HEAD_INIT(sig.posix_timers),		\
 	.cpu_timers	= INIT_CPU_TIMERS(sig.cpu_timers),		\
 	.rlim		= INIT_RLIMITS,					\
+	.cputime	= { .totals = {					\
+		.utime = cputime_zero,					\
+		.stime = cputime_zero,					\
+		.sum_exec_runtime = 0,					\
+		.lock = __SPIN_LOCK_UNLOCKED(sig.cputime.totals.lock),	\
+	}, },								\
 }
 
 extern struct nsproxy init_nsproxy;
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4cae9b81a1f8..c20943eabb4c 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -450,6 +450,7 @@ struct task_cputime {
 	cputime_t utime;
 	cputime_t stime;
 	unsigned long long sum_exec_runtime;
+	spinlock_t lock;
 };
 /* Alternate field names when used to cache expirations. */
 #define prof_exp	stime
@@ -465,7 +466,7 @@ struct task_cputime {
  * used for thread group CPU clock calculations.
  */
 struct thread_group_cputime {
-	struct task_cputime *totals;
+	struct task_cputime totals;
 };
 
 /*
@@ -2180,24 +2181,30 @@ static inline int spin_needbreak(spinlock_t *lock)
  * Thread group CPU time accounting.
  */
 
-extern int thread_group_cputime_alloc(struct task_struct *);
-extern void thread_group_cputime(struct task_struct *, struct task_cputime *);
-
-static inline void thread_group_cputime_init(struct signal_struct *sig)
+static inline
+void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times)
 {
-	sig->cputime.totals = NULL;
+	struct task_cputime *totals = &tsk->signal->cputime.totals;
+	unsigned long flags;
+
+	spin_lock_irqsave(&totals->lock, flags);
+	*times = *totals;
+	spin_unlock_irqrestore(&totals->lock, flags);
 }
 
-static inline int thread_group_cputime_clone_thread(struct task_struct *curr)
+static inline void thread_group_cputime_init(struct signal_struct *sig)
 {
-	if (curr->signal->cputime.totals)
-		return 0;
-	return thread_group_cputime_alloc(curr);
+	sig->cputime.totals = (struct task_cputime){
+		.utime = cputime_zero,
+		.stime = cputime_zero,
+		.sum_exec_runtime = 0,
+	};
+
+	spin_lock_init(&sig->cputime.totals.lock);
 }
 
 static inline void thread_group_cputime_free(struct signal_struct *sig)
 {
-	free_percpu(sig->cputime.totals);
 }
 
 /*
diff --git a/kernel/fork.c b/kernel/fork.c
index 7b8f2a78be3d..7087d8c0e5e2 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -820,14 +820,15 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
 	int ret;
 
 	if (clone_flags & CLONE_THREAD) {
-		ret = thread_group_cputime_clone_thread(current);
-		if (likely(!ret)) {
-			atomic_inc(&current->signal->count);
-			atomic_inc(&current->signal->live);
-		}
-		return ret;
+		atomic_inc(&current->signal->count);
+		atomic_inc(&current->signal->live);
+		return 0;
 	}
 	sig = kmem_cache_alloc(signal_cachep, GFP_KERNEL);
+
+	if (sig)
+		posix_cpu_timers_init_group(sig);
+
 	tsk->signal = sig;
 	if (!sig)
 		return -ENOMEM;
@@ -864,8 +865,6 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
 	memcpy(sig->rlim, current->signal->rlim, sizeof sig->rlim);
 	task_unlock(current->group_leader);
 
-	posix_cpu_timers_init_group(sig);
-
 	acct_init_pacct(&sig->pacct);
 
 	tty_audit_fork(sig);
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index 157de3a47832..fa07da94d7be 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -9,76 +9,6 @@
 #include <asm/uaccess.h>
 #include <linux/kernel_stat.h>
 
-/*
- * Allocate the thread_group_cputime structure appropriately and fill in the
- * current values of the fields.  Called from copy_signal() via
- * thread_group_cputime_clone_thread() when adding a second or subsequent
- * thread to a thread group.  Assumes interrupts are enabled when called.
- */
-int thread_group_cputime_alloc(struct task_struct *tsk)
-{
-	struct signal_struct *sig = tsk->signal;
-	struct task_cputime *cputime;
-
-	/*
-	 * If we have multiple threads and we don't already have a
-	 * per-CPU task_cputime struct (checked in the caller), allocate
-	 * one and fill it in with the times accumulated so far.  We may
-	 * race with another thread so recheck after we pick up the sighand
-	 * lock.
-	 */
-	cputime = alloc_percpu(struct task_cputime);
-	if (cputime == NULL)
-		return -ENOMEM;
-	spin_lock_irq(&tsk->sighand->siglock);
-	if (sig->cputime.totals) {
-		spin_unlock_irq(&tsk->sighand->siglock);
-		free_percpu(cputime);
-		return 0;
-	}
-	sig->cputime.totals = cputime;
-	cputime = per_cpu_ptr(sig->cputime.totals, smp_processor_id());
-	cputime->utime = tsk->utime;
-	cputime->stime = tsk->stime;
-	cputime->sum_exec_runtime = tsk->se.sum_exec_runtime;
-	spin_unlock_irq(&tsk->sighand->siglock);
-	return 0;
-}
-
-/**
- * thread_group_cputime - Sum the thread group time fields across all CPUs.
- *
- * @tsk:	The task we use to identify the thread group.
- * @times:	task_cputime structure in which we return the summed fields.
- *
- * Walk the list of CPUs to sum the per-CPU time fields in the thread group
- * time structure.
- */
-void thread_group_cputime(
-	struct task_struct *tsk,
-	struct task_cputime *times)
-{
-	struct task_cputime *totals, *tot;
-	int i;
-
-	totals = tsk->signal->cputime.totals;
-	if (!totals) {
-		times->utime = tsk->utime;
-		times->stime = tsk->stime;
-		times->sum_exec_runtime = tsk->se.sum_exec_runtime;
-		return;
-	}
-
-	times->stime = times->utime = cputime_zero;
-	times->sum_exec_runtime = 0;
-	for_each_possible_cpu(i) {
-		tot = per_cpu_ptr(totals, i);
-		times->utime = cputime_add(times->utime, tot->utime);
-		times->stime = cputime_add(times->stime, tot->stime);
-		times->sum_exec_runtime += tot->sum_exec_runtime;
-	}
-}
-
 /*
  * Called after updating RLIMIT_CPU to set timer expiration if necessary.
  */
diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h
index f2773b5d1226..8ab0cef8ecab 100644
--- a/kernel/sched_stats.h
+++ b/kernel/sched_stats.h
@@ -296,6 +296,7 @@ sched_info_switch(struct task_struct *prev, struct task_struct *next)
 static inline void account_group_user_time(struct task_struct *tsk,
 					   cputime_t cputime)
 {
+	struct task_cputime *times;
 	struct signal_struct *sig;
 
 	/* tsk == current, ensure it is safe to use ->signal */
@@ -303,13 +304,11 @@ static inline void account_group_user_time(struct task_struct *tsk,
 		return;
 
 	sig = tsk->signal;
-	if (sig->cputime.totals) {
-		struct task_cputime *times;
+	times = &sig->cputime.totals;
 
-		times = per_cpu_ptr(sig->cputime.totals, get_cpu());
-		times->utime = cputime_add(times->utime, cputime);
-		put_cpu_no_resched();
-	}
+	spin_lock(&times->lock);
+	times->utime = cputime_add(times->utime, cputime);
+	spin_unlock(&times->lock);
 }
 
 /**
@@ -325,6 +324,7 @@ static inline void account_group_user_time(struct task_struct *tsk,
 static inline void account_group_system_time(struct task_struct *tsk,
 					     cputime_t cputime)
 {
+	struct task_cputime *times;
 	struct signal_struct *sig;
 
 	/* tsk == current, ensure it is safe to use ->signal */
@@ -332,13 +332,11 @@ static inline void account_group_system_time(struct task_struct *tsk,
 		return;
 
 	sig = tsk->signal;
-	if (sig->cputime.totals) {
-		struct task_cputime *times;
+	times = &sig->cputime.totals;
 
-		times = per_cpu_ptr(sig->cputime.totals, get_cpu());
-		times->stime = cputime_add(times->stime, cputime);
-		put_cpu_no_resched();
-	}
+	spin_lock(&times->lock);
+	times->stime = cputime_add(times->stime, cputime);
+	spin_unlock(&times->lock);
 }
 
 /**
@@ -354,6 +352,7 @@ static inline void account_group_system_time(struct task_struct *tsk,
 static inline void account_group_exec_runtime(struct task_struct *tsk,
 					      unsigned long long ns)
 {
+	struct task_cputime *times;
 	struct signal_struct *sig;
 
 	sig = tsk->signal;
@@ -362,11 +361,9 @@ static inline void account_group_exec_runtime(struct task_struct *tsk,
 	if (unlikely(!sig))
 		return;
 
-	if (sig->cputime.totals) {
-		struct task_cputime *times;
+	times = &sig->cputime.totals;
 
-		times = per_cpu_ptr(sig->cputime.totals, get_cpu());
-		times->sum_exec_runtime += ns;
-		put_cpu_no_resched();
-	}
+	spin_lock(&times->lock);
+	times->sum_exec_runtime += ns;
+	spin_unlock(&times->lock);
 }
-- 
cgit v1.2.3


From b92a78e582b1a45649143dc86e526f5824092478 Mon Sep 17 00:00:00 2001
From: Rodolfo Giometti <giometti@linux.it>
Date: Thu, 23 Oct 2008 10:08:07 +0200
Subject: usb host: Oxford OXU210HP HCD driver.

This driver implements the support for Oxford OXU210HP USB high-speed host,
no peripheral nor OTG.

Signed-off-by: Rodolfo Giometti <giometti@linux.it>
Cc: Kan Liu <kan.k.liu@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/host/Kconfig        |   13 +
 drivers/usb/host/Makefile       |    1 +
 drivers/usb/host/oxu210hp-hcd.c | 3985 +++++++++++++++++++++++++++++++++++++++
 drivers/usb/host/oxu210hp.h     |  447 +++++
 include/linux/oxu210hp.h        |    7 +
 5 files changed, 4453 insertions(+)
 create mode 100644 drivers/usb/host/oxu210hp-hcd.c
 create mode 100644 drivers/usb/host/oxu210hp.h
 create mode 100644 include/linux/oxu210hp.h

(limited to 'include/linux')

diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
index f3a75a929e0a..2b476b6b3d4d 100644
--- a/drivers/usb/host/Kconfig
+++ b/drivers/usb/host/Kconfig
@@ -96,6 +96,19 @@ config USB_EHCI_HCD_PPC_OF
 	  Enables support for the USB controller present on the PowerPC
 	  OpenFirmware platform bus.
 
+config USB_OXU210HP_HCD
+	tristate "OXU210HP HCD support"
+	depends on USB
+	---help---
+	  The OXU210HP is an USB host/OTG/device controller. Enable this
+	  option if your board has this chip. If unsure, say N.
+
+	  This driver does not support isochronous transfers and doesn't
+	  implement OTG nor USB device controllers.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called oxu210hp-hcd.
+
 config USB_ISP116X_HCD
 	tristate "ISP116X HCD support"
 	depends on USB
diff --git a/drivers/usb/host/Makefile b/drivers/usb/host/Makefile
index 23be22224044..e5f3f20787e4 100644
--- a/drivers/usb/host/Makefile
+++ b/drivers/usb/host/Makefile
@@ -13,6 +13,7 @@ obj-$(CONFIG_USB_WHCI_HCD)	+= whci/
 obj-$(CONFIG_PCI)		+= pci-quirks.o
 
 obj-$(CONFIG_USB_EHCI_HCD)	+= ehci-hcd.o
+obj-$(CONFIG_USB_OXU210HP_HCD)	+= oxu210hp-hcd.o
 obj-$(CONFIG_USB_ISP116X_HCD)	+= isp116x-hcd.o
 obj-$(CONFIG_USB_OHCI_HCD)	+= ohci-hcd.o
 obj-$(CONFIG_USB_UHCI_HCD)	+= uhci-hcd.o
diff --git a/drivers/usb/host/oxu210hp-hcd.c b/drivers/usb/host/oxu210hp-hcd.c
new file mode 100644
index 000000000000..251123c29d81
--- /dev/null
+++ b/drivers/usb/host/oxu210hp-hcd.c
@@ -0,0 +1,3985 @@
+/*
+ * Copyright (c) 2008 Rodolfo Giometti <giometti@linux.it>
+ * Copyright (c) 2008 Eurotech S.p.A. <info@eurtech.it>
+ *
+ * This code is *strongly* based on EHCI-HCD code by David Brownell since
+ * the chip is a quasi-EHCI compatible.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software Foundation,
+ * Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/dmapool.h>
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <linux/ioport.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/timer.h>
+#include <linux/list.h>
+#include <linux/interrupt.h>
+#include <linux/reboot.h>
+#include <linux/usb.h>
+#include <linux/moduleparam.h>
+#include <linux/dma-mapping.h>
+#include <linux/io.h>
+
+#include "../core/hcd.h"
+
+#include <asm/irq.h>
+#include <asm/system.h>
+#include <asm/unaligned.h>
+
+#include <linux/irq.h>
+#include <linux/platform_device.h>
+
+#include "oxu210hp.h"
+
+#define DRIVER_VERSION "0.0.50"
+
+/*
+ * Main defines
+ */
+
+#define oxu_dbg(oxu, fmt, args...) \
+		dev_dbg(oxu_to_hcd(oxu)->self.controller , fmt , ## args)
+#define oxu_err(oxu, fmt, args...) \
+		dev_err(oxu_to_hcd(oxu)->self.controller , fmt , ## args)
+#define oxu_info(oxu, fmt, args...) \
+		dev_info(oxu_to_hcd(oxu)->self.controller , fmt , ## args)
+
+static inline struct usb_hcd *oxu_to_hcd(struct oxu_hcd *oxu)
+{
+	return container_of((void *) oxu, struct usb_hcd, hcd_priv);
+}
+
+static inline struct oxu_hcd *hcd_to_oxu(struct usb_hcd *hcd)
+{
+	return (struct oxu_hcd *) (hcd->hcd_priv);
+}
+
+/*
+ * Debug stuff
+ */
+
+#undef OXU_URB_TRACE
+#undef OXU_VERBOSE_DEBUG
+
+#ifdef OXU_VERBOSE_DEBUG
+#define oxu_vdbg			oxu_dbg
+#else
+#define oxu_vdbg(oxu, fmt, args...)	/* Nop */
+#endif
+
+#ifdef DEBUG
+
+static int __attribute__((__unused__))
+dbg_status_buf(char *buf, unsigned len, const char *label, u32 status)
+{
+	return scnprintf(buf, len, "%s%sstatus %04x%s%s%s%s%s%s%s%s%s%s",
+		label, label[0] ? " " : "", status,
+		(status & STS_ASS) ? " Async" : "",
+		(status & STS_PSS) ? " Periodic" : "",
+		(status & STS_RECL) ? " Recl" : "",
+		(status & STS_HALT) ? " Halt" : "",
+		(status & STS_IAA) ? " IAA" : "",
+		(status & STS_FATAL) ? " FATAL" : "",
+		(status & STS_FLR) ? " FLR" : "",
+		(status & STS_PCD) ? " PCD" : "",
+		(status & STS_ERR) ? " ERR" : "",
+		(status & STS_INT) ? " INT" : ""
+		);
+}
+
+static int __attribute__((__unused__))
+dbg_intr_buf(char *buf, unsigned len, const char *label, u32 enable)
+{
+	return scnprintf(buf, len, "%s%sintrenable %02x%s%s%s%s%s%s",
+		label, label[0] ? " " : "", enable,
+		(enable & STS_IAA) ? " IAA" : "",
+		(enable & STS_FATAL) ? " FATAL" : "",
+		(enable & STS_FLR) ? " FLR" : "",
+		(enable & STS_PCD) ? " PCD" : "",
+		(enable & STS_ERR) ? " ERR" : "",
+		(enable & STS_INT) ? " INT" : ""
+		);
+}
+
+static const char *const fls_strings[] =
+    { "1024", "512", "256", "??" };
+
+static int dbg_command_buf(char *buf, unsigned len,
+				const char *label, u32 command)
+{
+	return scnprintf(buf, len,
+		"%s%scommand %06x %s=%d ithresh=%d%s%s%s%s period=%s%s %s",
+		label, label[0] ? " " : "", command,
+		(command & CMD_PARK) ? "park" : "(park)",
+		CMD_PARK_CNT(command),
+		(command >> 16) & 0x3f,
+		(command & CMD_LRESET) ? " LReset" : "",
+		(command & CMD_IAAD) ? " IAAD" : "",
+		(command & CMD_ASE) ? " Async" : "",
+		(command & CMD_PSE) ? " Periodic" : "",
+		fls_strings[(command >> 2) & 0x3],
+		(command & CMD_RESET) ? " Reset" : "",
+		(command & CMD_RUN) ? "RUN" : "HALT"
+		);
+}
+
+static int dbg_port_buf(char *buf, unsigned len, const char *label,
+				int port, u32 status)
+{
+	char	*sig;
+
+	/* signaling state */
+	switch (status & (3 << 10)) {
+	case 0 << 10:
+		sig = "se0";
+		break;
+	case 1 << 10:
+		sig = "k";	/* low speed */
+		break;
+	case 2 << 10:
+		sig = "j";
+		break;
+	default:
+		sig = "?";
+		break;
+	}
+
+	return scnprintf(buf, len,
+		"%s%sport %d status %06x%s%s sig=%s%s%s%s%s%s%s%s%s%s",
+		label, label[0] ? " " : "", port, status,
+		(status & PORT_POWER) ? " POWER" : "",
+		(status & PORT_OWNER) ? " OWNER" : "",
+		sig,
+		(status & PORT_RESET) ? " RESET" : "",
+		(status & PORT_SUSPEND) ? " SUSPEND" : "",
+		(status & PORT_RESUME) ? " RESUME" : "",
+		(status & PORT_OCC) ? " OCC" : "",
+		(status & PORT_OC) ? " OC" : "",
+		(status & PORT_PEC) ? " PEC" : "",
+		(status & PORT_PE) ? " PE" : "",
+		(status & PORT_CSC) ? " CSC" : "",
+		(status & PORT_CONNECT) ? " CONNECT" : ""
+	    );
+}
+
+#else
+
+static inline int __attribute__((__unused__))
+dbg_status_buf(char *buf, unsigned len, const char *label, u32 status)
+{ return 0; }
+
+static inline int __attribute__((__unused__))
+dbg_command_buf(char *buf, unsigned len, const char *label, u32 command)
+{ return 0; }
+
+static inline int __attribute__((__unused__))
+dbg_intr_buf(char *buf, unsigned len, const char *label, u32 enable)
+{ return 0; }
+
+static inline int __attribute__((__unused__))
+dbg_port_buf(char *buf, unsigned len, const char *label, int port, u32 status)
+{ return 0; }
+
+#endif /* DEBUG */
+
+/* functions have the "wrong" filename when they're output... */
+#define dbg_status(oxu, label, status) { \
+	char _buf[80]; \
+	dbg_status_buf(_buf, sizeof _buf, label, status); \
+	oxu_dbg(oxu, "%s\n", _buf); \
+}
+
+#define dbg_cmd(oxu, label, command) { \
+	char _buf[80]; \
+	dbg_command_buf(_buf, sizeof _buf, label, command); \
+	oxu_dbg(oxu, "%s\n", _buf); \
+}
+
+#define dbg_port(oxu, label, port, status) { \
+	char _buf[80]; \
+	dbg_port_buf(_buf, sizeof _buf, label, port, status); \
+	oxu_dbg(oxu, "%s\n", _buf); \
+}
+
+/*
+ * Module parameters
+ */
+
+/* Initial IRQ latency: faster than hw default */
+static int log2_irq_thresh;			/* 0 to 6 */
+module_param(log2_irq_thresh, int, S_IRUGO);
+MODULE_PARM_DESC(log2_irq_thresh, "log2 IRQ latency, 1-64 microframes");
+
+/* Initial park setting: slower than hw default */
+static unsigned park;
+module_param(park, uint, S_IRUGO);
+MODULE_PARM_DESC(park, "park setting; 1-3 back-to-back async packets");
+
+/* For flakey hardware, ignore overcurrent indicators */
+static int ignore_oc;
+module_param(ignore_oc, bool, S_IRUGO);
+MODULE_PARM_DESC(ignore_oc, "ignore bogus hardware overcurrent indications");
+
+
+static void ehci_work(struct oxu_hcd *oxu);
+static int oxu_hub_control(struct usb_hcd *hcd,
+				u16 typeReq, u16 wValue, u16 wIndex,
+				char *buf, u16 wLength);
+
+/*
+ * Local functions
+ */
+
+/* Low level read/write registers functions */
+static inline u32 oxu_readl(void *base, u32 reg)
+{
+	return readl(base + reg);
+}
+
+static inline void oxu_writel(void *base, u32 reg, u32 val)
+{
+	writel(val, base + reg);
+}
+
+static inline void timer_action_done(struct oxu_hcd *oxu,
+					enum ehci_timer_action action)
+{
+	clear_bit(action, &oxu->actions);
+}
+
+static inline void timer_action(struct oxu_hcd *oxu,
+					enum ehci_timer_action action)
+{
+	if (!test_and_set_bit(action, &oxu->actions)) {
+		unsigned long t;
+
+		switch (action) {
+		case TIMER_IAA_WATCHDOG:
+			t = EHCI_IAA_JIFFIES;
+			break;
+		case TIMER_IO_WATCHDOG:
+			t = EHCI_IO_JIFFIES;
+			break;
+		case TIMER_ASYNC_OFF:
+			t = EHCI_ASYNC_JIFFIES;
+			break;
+		case TIMER_ASYNC_SHRINK:
+		default:
+			t = EHCI_SHRINK_JIFFIES;
+			break;
+		}
+		t += jiffies;
+		/* all timings except IAA watchdog can be overridden.
+		 * async queue SHRINK often precedes IAA.  while it's ready
+		 * to go OFF neither can matter, and afterwards the IO
+		 * watchdog stops unless there's still periodic traffic.
+		 */
+		if (action != TIMER_IAA_WATCHDOG
+				&& t > oxu->watchdog.expires
+				&& timer_pending(&oxu->watchdog))
+			return;
+		mod_timer(&oxu->watchdog, t);
+	}
+}
+
+/*
+ * handshake - spin reading hc until handshake completes or fails
+ * @ptr: address of hc register to be read
+ * @mask: bits to look at in result of read
+ * @done: value of those bits when handshake succeeds
+ * @usec: timeout in microseconds
+ *
+ * Returns negative errno, or zero on success
+ *
+ * Success happens when the "mask" bits have the specified value (hardware
+ * handshake done).  There are two failure modes:  "usec" have passed (major
+ * hardware flakeout), or the register reads as all-ones (hardware removed).
+ *
+ * That last failure should_only happen in cases like physical cardbus eject
+ * before driver shutdown. But it also seems to be caused by bugs in cardbus
+ * bridge shutdown:  shutting down the bridge before the devices using it.
+ */
+static int handshake(struct oxu_hcd *oxu, void __iomem *ptr,
+					u32 mask, u32 done, int usec)
+{
+	u32 result;
+
+	do {
+		result = readl(ptr);
+		if (result == ~(u32)0)		/* card removed */
+			return -ENODEV;
+		result &= mask;
+		if (result == done)
+			return 0;
+		udelay(1);
+		usec--;
+	} while (usec > 0);
+	return -ETIMEDOUT;
+}
+
+/* Force HC to halt state from unknown (EHCI spec section 2.3) */
+static int ehci_halt(struct oxu_hcd *oxu)
+{
+	u32	temp = readl(&oxu->regs->status);
+
+	/* disable any irqs left enabled by previous code */
+	writel(0, &oxu->regs->intr_enable);
+
+	if ((temp & STS_HALT) != 0)
+		return 0;
+
+	temp = readl(&oxu->regs->command);
+	temp &= ~CMD_RUN;
+	writel(temp, &oxu->regs->command);
+	return handshake(oxu, &oxu->regs->status,
+			  STS_HALT, STS_HALT, 16 * 125);
+}
+
+/* Put TDI/ARC silicon into EHCI mode */
+static void tdi_reset(struct oxu_hcd *oxu)
+{
+	u32 __iomem *reg_ptr;
+	u32 tmp;
+
+	reg_ptr = (u32 __iomem *)(((u8 __iomem *)oxu->regs) + 0x68);
+	tmp = readl(reg_ptr);
+	tmp |= 0x3;
+	writel(tmp, reg_ptr);
+}
+
+/* Reset a non-running (STS_HALT == 1) controller */
+static int ehci_reset(struct oxu_hcd *oxu)
+{
+	int	retval;
+	u32	command = readl(&oxu->regs->command);
+
+	command |= CMD_RESET;
+	dbg_cmd(oxu, "reset", command);
+	writel(command, &oxu->regs->command);
+	oxu_to_hcd(oxu)->state = HC_STATE_HALT;
+	oxu->next_statechange = jiffies;
+	retval = handshake(oxu, &oxu->regs->command,
+			    CMD_RESET, 0, 250 * 1000);
+
+	if (retval)
+		return retval;
+
+	tdi_reset(oxu);
+
+	return retval;
+}
+
+/* Idle the controller (from running) */
+static void ehci_quiesce(struct oxu_hcd *oxu)
+{
+	u32	temp;
+
+#ifdef DEBUG
+	if (!HC_IS_RUNNING(oxu_to_hcd(oxu)->state))
+		BUG();
+#endif
+
+	/* wait for any schedule enables/disables to take effect */
+	temp = readl(&oxu->regs->command) << 10;
+	temp &= STS_ASS | STS_PSS;
+	if (handshake(oxu, &oxu->regs->status, STS_ASS | STS_PSS,
+				temp, 16 * 125) != 0) {
+		oxu_to_hcd(oxu)->state = HC_STATE_HALT;
+		return;
+	}
+
+	/* then disable anything that's still active */
+	temp = readl(&oxu->regs->command);
+	temp &= ~(CMD_ASE | CMD_IAAD | CMD_PSE);
+	writel(temp, &oxu->regs->command);
+
+	/* hardware can take 16 microframes to turn off ... */
+	if (handshake(oxu, &oxu->regs->status, STS_ASS | STS_PSS,
+				0, 16 * 125) != 0) {
+		oxu_to_hcd(oxu)->state = HC_STATE_HALT;
+		return;
+	}
+}
+
+static int check_reset_complete(struct oxu_hcd *oxu, int index,
+				u32 __iomem *status_reg, int port_status)
+{
+	if (!(port_status & PORT_CONNECT)) {
+		oxu->reset_done[index] = 0;
+		return port_status;
+	}
+
+	/* if reset finished and it's still not enabled -- handoff */
+	if (!(port_status & PORT_PE)) {
+		oxu_dbg(oxu, "Failed to enable port %d on root hub TT\n",
+				index+1);
+		return port_status;
+	} else
+		oxu_dbg(oxu, "port %d high speed\n", index + 1);
+
+	return port_status;
+}
+
+static void ehci_hub_descriptor(struct oxu_hcd *oxu,
+				struct usb_hub_descriptor *desc)
+{
+	int ports = HCS_N_PORTS(oxu->hcs_params);
+	u16 temp;
+
+	desc->bDescriptorType = 0x29;
+	desc->bPwrOn2PwrGood = 10;	/* oxu 1.0, 2.3.9 says 20ms max */
+	desc->bHubContrCurrent = 0;
+
+	desc->bNbrPorts = ports;
+	temp = 1 + (ports / 8);
+	desc->bDescLength = 7 + 2 * temp;
+
+	/* two bitmaps:  ports removable, and usb 1.0 legacy PortPwrCtrlMask */
+	memset(&desc->bitmap[0], 0, temp);
+	memset(&desc->bitmap[temp], 0xff, temp);
+
+	temp = 0x0008;			/* per-port overcurrent reporting */
+	if (HCS_PPC(oxu->hcs_params))
+		temp |= 0x0001;		/* per-port power control */
+	else
+		temp |= 0x0002;		/* no power switching */
+	desc->wHubCharacteristics = (__force __u16)cpu_to_le16(temp);
+}
+
+
+/* Allocate an OXU210HP on-chip memory data buffer
+ *
+ * An on-chip memory data buffer is required for each OXU210HP USB transfer.
+ * Each transfer descriptor has one or more on-chip memory data buffers.
+ *
+ * Data buffers are allocated from a fix sized pool of data blocks.
+ * To minimise fragmentation and give reasonable memory utlisation,
+ * data buffers are allocated with sizes the power of 2 multiples of
+ * the block size, starting on an address a multiple of the allocated size.
+ *
+ * FIXME: callers of this function require a buffer to be allocated for
+ * len=0. This is a waste of on-chip memory and should be fix. Then this
+ * function should be changed to not allocate a buffer for len=0.
+ */
+static int oxu_buf_alloc(struct oxu_hcd *oxu, struct ehci_qtd *qtd, int len)
+{
+	int n_blocks;	/* minium blocks needed to hold len */
+	int a_blocks;	/* blocks allocated */
+	int i, j;
+
+	/* Don't allocte bigger than supported */
+	if (len > BUFFER_SIZE * BUFFER_NUM) {
+		oxu_err(oxu, "buffer too big (%d)\n", len);
+		return -ENOMEM;
+	}
+
+	spin_lock(&oxu->mem_lock);
+
+	/* Number of blocks needed to hold len */
+	n_blocks = (len + BUFFER_SIZE - 1) / BUFFER_SIZE;
+
+	/* Round the number of blocks up to the power of 2 */
+	for (a_blocks = 1; a_blocks < n_blocks; a_blocks <<= 1)
+		;
+
+	/* Find a suitable available data buffer */
+	for (i = 0; i < BUFFER_NUM;
+			i += max(a_blocks, (int)oxu->db_used[i])) {
+
+		/* Check all the required blocks are available */
+		for (j = 0; j < a_blocks; j++)
+			if (oxu->db_used[i + j])
+				break;
+
+		if (j != a_blocks)
+			continue;
+
+		/* Allocate blocks found! */
+		qtd->buffer = (void *) &oxu->mem->db_pool[i];
+		qtd->buffer_dma = virt_to_phys(qtd->buffer);
+
+		qtd->qtd_buffer_len = BUFFER_SIZE * a_blocks;
+		oxu->db_used[i] = a_blocks;
+
+		spin_unlock(&oxu->mem_lock);
+
+		return 0;
+	}
+
+	/* Failed */
+
+	spin_unlock(&oxu->mem_lock);
+
+	return -ENOMEM;
+}
+
+static void oxu_buf_free(struct oxu_hcd *oxu, struct ehci_qtd *qtd)
+{
+	int index;
+
+	spin_lock(&oxu->mem_lock);
+
+	index = (qtd->buffer - (void *) &oxu->mem->db_pool[0])
+							 / BUFFER_SIZE;
+	oxu->db_used[index] = 0;
+	qtd->qtd_buffer_len = 0;
+	qtd->buffer_dma = 0;
+	qtd->buffer = NULL;
+
+	spin_unlock(&oxu->mem_lock);
+
+	return;
+}
+
+static inline void ehci_qtd_init(struct ehci_qtd *qtd, dma_addr_t dma)
+{
+	memset(qtd, 0, sizeof *qtd);
+	qtd->qtd_dma = dma;
+	qtd->hw_token = cpu_to_le32(QTD_STS_HALT);
+	qtd->hw_next = EHCI_LIST_END;
+	qtd->hw_alt_next = EHCI_LIST_END;
+	INIT_LIST_HEAD(&qtd->qtd_list);
+}
+
+static inline void oxu_qtd_free(struct oxu_hcd *oxu, struct ehci_qtd *qtd)
+{
+	int index;
+
+	if (qtd->buffer)
+		oxu_buf_free(oxu, qtd);
+
+	spin_lock(&oxu->mem_lock);
+
+	index = qtd - &oxu->mem->qtd_pool[0];
+	oxu->qtd_used[index] = 0;
+
+	spin_unlock(&oxu->mem_lock);
+
+	return;
+}
+
+static struct ehci_qtd *ehci_qtd_alloc(struct oxu_hcd *oxu)
+{
+	int i;
+	struct ehci_qtd *qtd = NULL;
+
+	spin_lock(&oxu->mem_lock);
+
+	for (i = 0; i < QTD_NUM; i++)
+		if (!oxu->qtd_used[i])
+			break;
+
+	if (i < QTD_NUM) {
+		qtd = (struct ehci_qtd *) &oxu->mem->qtd_pool[i];
+		memset(qtd, 0, sizeof *qtd);
+
+		qtd->hw_token = cpu_to_le32(QTD_STS_HALT);
+		qtd->hw_next = EHCI_LIST_END;
+		qtd->hw_alt_next = EHCI_LIST_END;
+		INIT_LIST_HEAD(&qtd->qtd_list);
+
+		qtd->qtd_dma = virt_to_phys(qtd);
+
+		oxu->qtd_used[i] = 1;
+	}
+
+	spin_unlock(&oxu->mem_lock);
+
+	return qtd;
+}
+
+static void oxu_qh_free(struct oxu_hcd *oxu, struct ehci_qh *qh)
+{
+	int index;
+
+	spin_lock(&oxu->mem_lock);
+
+	index = qh - &oxu->mem->qh_pool[0];
+	oxu->qh_used[index] = 0;
+
+	spin_unlock(&oxu->mem_lock);
+
+	return;
+}
+
+static void qh_destroy(struct kref *kref)
+{
+	struct ehci_qh *qh = container_of(kref, struct ehci_qh, kref);
+	struct oxu_hcd *oxu = qh->oxu;
+
+	/* clean qtds first, and know this is not linked */
+	if (!list_empty(&qh->qtd_list) || qh->qh_next.ptr) {
+		oxu_dbg(oxu, "unused qh not empty!\n");
+		BUG();
+	}
+	if (qh->dummy)
+		oxu_qtd_free(oxu, qh->dummy);
+	oxu_qh_free(oxu, qh);
+}
+
+static struct ehci_qh *oxu_qh_alloc(struct oxu_hcd *oxu)
+{
+	int i;
+	struct ehci_qh *qh = NULL;
+
+	spin_lock(&oxu->mem_lock);
+
+	for (i = 0; i < QHEAD_NUM; i++)
+		if (!oxu->qh_used[i])
+			break;
+
+	if (i < QHEAD_NUM) {
+		qh = (struct ehci_qh *) &oxu->mem->qh_pool[i];
+		memset(qh, 0, sizeof *qh);
+
+		kref_init(&qh->kref);
+		qh->oxu = oxu;
+		qh->qh_dma = virt_to_phys(qh);
+		INIT_LIST_HEAD(&qh->qtd_list);
+
+		/* dummy td enables safe urb queuing */
+		qh->dummy = ehci_qtd_alloc(oxu);
+		if (qh->dummy == NULL) {
+			oxu_dbg(oxu, "no dummy td\n");
+			oxu->qh_used[i] = 0;
+
+			return NULL;
+		}
+
+		oxu->qh_used[i] = 1;
+	}
+
+	spin_unlock(&oxu->mem_lock);
+
+	return qh;
+}
+
+/* to share a qh (cpu threads, or hc) */
+static inline struct ehci_qh *qh_get(struct ehci_qh *qh)
+{
+	kref_get(&qh->kref);
+	return qh;
+}
+
+static inline void qh_put(struct ehci_qh *qh)
+{
+	kref_put(&qh->kref, qh_destroy);
+}
+
+static void oxu_murb_free(struct oxu_hcd *oxu, struct oxu_murb *murb)
+{
+	int index;
+
+	spin_lock(&oxu->mem_lock);
+
+	index = murb - &oxu->murb_pool[0];
+	oxu->murb_used[index] = 0;
+
+	spin_unlock(&oxu->mem_lock);
+
+	return;
+}
+
+static struct oxu_murb *oxu_murb_alloc(struct oxu_hcd *oxu)
+
+{
+	int i;
+	struct oxu_murb *murb = NULL;
+
+	spin_lock(&oxu->mem_lock);
+
+	for (i = 0; i < MURB_NUM; i++)
+		if (!oxu->murb_used[i])
+			break;
+
+	if (i < MURB_NUM) {
+		murb = &(oxu->murb_pool)[i];
+
+		oxu->murb_used[i] = 1;
+	}
+
+	spin_unlock(&oxu->mem_lock);
+
+	return murb;
+}
+
+/* The queue heads and transfer descriptors are managed from pools tied
+ * to each of the "per device" structures.
+ * This is the initialisation and cleanup code.
+ */
+static void ehci_mem_cleanup(struct oxu_hcd *oxu)
+{
+	kfree(oxu->murb_pool);
+	oxu->murb_pool = NULL;
+
+	if (oxu->async)
+		qh_put(oxu->async);
+	oxu->async = NULL;
+
+	del_timer(&oxu->urb_timer);
+
+	oxu->periodic = NULL;
+
+	/* shadow periodic table */
+	kfree(oxu->pshadow);
+	oxu->pshadow = NULL;
+}
+
+/* Remember to add cleanup code (above) if you add anything here.
+ */
+static int ehci_mem_init(struct oxu_hcd *oxu, gfp_t flags)
+{
+	int i;
+
+	for (i = 0; i < oxu->periodic_size; i++)
+		oxu->mem->frame_list[i] = EHCI_LIST_END;
+	for (i = 0; i < QHEAD_NUM; i++)
+		oxu->qh_used[i] = 0;
+	for (i = 0; i < QTD_NUM; i++)
+		oxu->qtd_used[i] = 0;
+
+	oxu->murb_pool = kcalloc(MURB_NUM, sizeof(struct oxu_murb), flags);
+	if (!oxu->murb_pool)
+		goto fail;
+
+	for (i = 0; i < MURB_NUM; i++)
+		oxu->murb_used[i] = 0;
+
+	oxu->async = oxu_qh_alloc(oxu);
+	if (!oxu->async)
+		goto fail;
+
+	oxu->periodic = (__le32 *) &oxu->mem->frame_list;
+	oxu->periodic_dma = virt_to_phys(oxu->periodic);
+
+	for (i = 0; i < oxu->periodic_size; i++)
+		oxu->periodic[i] = EHCI_LIST_END;
+
+	/* software shadow of hardware table */
+	oxu->pshadow = kcalloc(oxu->periodic_size, sizeof(void *), flags);
+	if (oxu->pshadow != NULL)
+		return 0;
+
+fail:
+	oxu_dbg(oxu, "couldn't init memory\n");
+	ehci_mem_cleanup(oxu);
+	return -ENOMEM;
+}
+
+/* Fill a qtd, returning how much of the buffer we were able to queue up.
+ */
+static int qtd_fill(struct ehci_qtd *qtd, dma_addr_t buf, size_t len,
+				int token, int maxpacket)
+{
+	int i, count;
+	u64 addr = buf;
+
+	/* one buffer entry per 4K ... first might be short or unaligned */
+	qtd->hw_buf[0] = cpu_to_le32((u32)addr);
+	qtd->hw_buf_hi[0] = cpu_to_le32((u32)(addr >> 32));
+	count = 0x1000 - (buf & 0x0fff);	/* rest of that page */
+	if (likely(len < count))		/* ... iff needed */
+		count = len;
+	else {
+		buf +=  0x1000;
+		buf &= ~0x0fff;
+
+		/* per-qtd limit: from 16K to 20K (best alignment) */
+		for (i = 1; count < len && i < 5; i++) {
+			addr = buf;
+			qtd->hw_buf[i] = cpu_to_le32((u32)addr);
+			qtd->hw_buf_hi[i] = cpu_to_le32((u32)(addr >> 32));
+			buf += 0x1000;
+			if ((count + 0x1000) < len)
+				count += 0x1000;
+			else
+				count = len;
+		}
+
+		/* short packets may only terminate transfers */
+		if (count != len)
+			count -= (count % maxpacket);
+	}
+	qtd->hw_token = cpu_to_le32((count << 16) | token);
+	qtd->length = count;
+
+	return count;
+}
+
+static inline void qh_update(struct oxu_hcd *oxu,
+				struct ehci_qh *qh, struct ehci_qtd *qtd)
+{
+	/* writes to an active overlay are unsafe */
+	BUG_ON(qh->qh_state != QH_STATE_IDLE);
+
+	qh->hw_qtd_next = QTD_NEXT(qtd->qtd_dma);
+	qh->hw_alt_next = EHCI_LIST_END;
+
+	/* Except for control endpoints, we make hardware maintain data
+	 * toggle (like OHCI) ... here (re)initialize the toggle in the QH,
+	 * and set the pseudo-toggle in udev. Only usb_clear_halt() will
+	 * ever clear it.
+	 */
+	if (!(qh->hw_info1 & cpu_to_le32(1 << 14))) {
+		unsigned	is_out, epnum;
+
+		is_out = !(qtd->hw_token & cpu_to_le32(1 << 8));
+		epnum = (le32_to_cpup(&qh->hw_info1) >> 8) & 0x0f;
+		if (unlikely(!usb_gettoggle(qh->dev, epnum, is_out))) {
+			qh->hw_token &= ~__constant_cpu_to_le32(QTD_TOGGLE);
+			usb_settoggle(qh->dev, epnum, is_out, 1);
+		}
+	}
+
+	/* HC must see latest qtd and qh data before we clear ACTIVE+HALT */
+	wmb();
+	qh->hw_token &= __constant_cpu_to_le32(QTD_TOGGLE | QTD_STS_PING);
+}
+
+/* If it weren't for a common silicon quirk (writing the dummy into the qh
+ * overlay, so qh->hw_token wrongly becomes inactive/halted), only fault
+ * recovery (including urb dequeue) would need software changes to a QH...
+ */
+static void qh_refresh(struct oxu_hcd *oxu, struct ehci_qh *qh)
+{
+	struct ehci_qtd *qtd;
+
+	if (list_empty(&qh->qtd_list))
+		qtd = qh->dummy;
+	else {
+		qtd = list_entry(qh->qtd_list.next,
+				struct ehci_qtd, qtd_list);
+		/* first qtd may already be partially processed */
+		if (cpu_to_le32(qtd->qtd_dma) == qh->hw_current)
+			qtd = NULL;
+	}
+
+	if (qtd)
+		qh_update(oxu, qh, qtd);
+}
+
+static void qtd_copy_status(struct oxu_hcd *oxu, struct urb *urb,
+				size_t length, u32 token)
+{
+	/* count IN/OUT bytes, not SETUP (even short packets) */
+	if (likely(QTD_PID(token) != 2))
+		urb->actual_length += length - QTD_LENGTH(token);
+
+	/* don't modify error codes */
+	if (unlikely(urb->status != -EINPROGRESS))
+		return;
+
+	/* force cleanup after short read; not always an error */
+	if (unlikely(IS_SHORT_READ(token)))
+		urb->status = -EREMOTEIO;
+
+	/* serious "can't proceed" faults reported by the hardware */
+	if (token & QTD_STS_HALT) {
+		if (token & QTD_STS_BABBLE) {
+			/* FIXME "must" disable babbling device's port too */
+			urb->status = -EOVERFLOW;
+		} else if (token & QTD_STS_MMF) {
+			/* fs/ls interrupt xfer missed the complete-split */
+			urb->status = -EPROTO;
+		} else if (token & QTD_STS_DBE) {
+			urb->status = (QTD_PID(token) == 1) /* IN ? */
+				? -ENOSR  /* hc couldn't read data */
+				: -ECOMM; /* hc couldn't write data */
+		} else if (token & QTD_STS_XACT) {
+			/* timeout, bad crc, wrong PID, etc; retried */
+			if (QTD_CERR(token))
+				urb->status = -EPIPE;
+			else {
+				oxu_dbg(oxu, "devpath %s ep%d%s 3strikes\n",
+					urb->dev->devpath,
+					usb_pipeendpoint(urb->pipe),
+					usb_pipein(urb->pipe) ? "in" : "out");
+				urb->status = -EPROTO;
+			}
+		/* CERR nonzero + no errors + halt --> stall */
+		} else if (QTD_CERR(token))
+			urb->status = -EPIPE;
+		else	/* unknown */
+			urb->status = -EPROTO;
+
+		oxu_vdbg(oxu, "dev%d ep%d%s qtd token %08x --> status %d\n",
+			usb_pipedevice(urb->pipe),
+			usb_pipeendpoint(urb->pipe),
+			usb_pipein(urb->pipe) ? "in" : "out",
+			token, urb->status);
+	}
+}
+
+static void ehci_urb_done(struct oxu_hcd *oxu, struct urb *urb)
+__releases(oxu->lock)
+__acquires(oxu->lock)
+{
+	if (likely(urb->hcpriv != NULL)) {
+		struct ehci_qh	*qh = (struct ehci_qh *) urb->hcpriv;
+
+		/* S-mask in a QH means it's an interrupt urb */
+		if ((qh->hw_info2 & __constant_cpu_to_le32(QH_SMASK)) != 0) {
+
+			/* ... update hc-wide periodic stats (for usbfs) */
+			oxu_to_hcd(oxu)->self.bandwidth_int_reqs--;
+		}
+		qh_put(qh);
+	}
+
+	urb->hcpriv = NULL;
+	switch (urb->status) {
+	case -EINPROGRESS:		/* success */
+		urb->status = 0;
+	default:			/* fault */
+		break;
+	case -EREMOTEIO:		/* fault or normal */
+		if (!(urb->transfer_flags & URB_SHORT_NOT_OK))
+			urb->status = 0;
+		break;
+	case -ECONNRESET:		/* canceled */
+	case -ENOENT:
+		break;
+	}
+
+#ifdef OXU_URB_TRACE
+	oxu_dbg(oxu, "%s %s urb %p ep%d%s status %d len %d/%d\n",
+		__func__, urb->dev->devpath, urb,
+		usb_pipeendpoint(urb->pipe),
+		usb_pipein(urb->pipe) ? "in" : "out",
+		urb->status,
+		urb->actual_length, urb->transfer_buffer_length);
+#endif
+
+	/* complete() can reenter this HCD */
+	spin_unlock(&oxu->lock);
+	usb_hcd_giveback_urb(oxu_to_hcd(oxu), urb, urb->status);
+	spin_lock(&oxu->lock);
+}
+
+static void start_unlink_async(struct oxu_hcd *oxu, struct ehci_qh *qh);
+static void unlink_async(struct oxu_hcd *oxu, struct ehci_qh *qh);
+
+static void intr_deschedule(struct oxu_hcd *oxu, struct ehci_qh *qh);
+static int qh_schedule(struct oxu_hcd *oxu, struct ehci_qh *qh);
+
+#define HALT_BIT __constant_cpu_to_le32(QTD_STS_HALT)
+
+/* Process and free completed qtds for a qh, returning URBs to drivers.
+ * Chases up to qh->hw_current.  Returns number of completions called,
+ * indicating how much "real" work we did.
+ */
+static unsigned qh_completions(struct oxu_hcd *oxu, struct ehci_qh *qh)
+{
+	struct ehci_qtd *last = NULL, *end = qh->dummy;
+	struct list_head *entry, *tmp;
+	int stopped;
+	unsigned count = 0;
+	int do_status = 0;
+	u8 state;
+	struct oxu_murb *murb = NULL;
+
+	if (unlikely(list_empty(&qh->qtd_list)))
+		return count;
+
+	/* completions (or tasks on other cpus) must never clobber HALT
+	 * till we've gone through and cleaned everything up, even when
+	 * they add urbs to this qh's queue or mark them for unlinking.
+	 *
+	 * NOTE:  unlinking expects to be done in queue order.
+	 */
+	state = qh->qh_state;
+	qh->qh_state = QH_STATE_COMPLETING;
+	stopped = (state == QH_STATE_IDLE);
+
+	/* remove de-activated QTDs from front of queue.
+	 * after faults (including short reads), cleanup this urb
+	 * then let the queue advance.
+	 * if queue is stopped, handles unlinks.
+	 */
+	list_for_each_safe(entry, tmp, &qh->qtd_list) {
+		struct ehci_qtd	*qtd;
+		struct urb *urb;
+		u32 token = 0;
+
+		qtd = list_entry(entry, struct ehci_qtd, qtd_list);
+		urb = qtd->urb;
+
+		/* Clean up any state from previous QTD ...*/
+		if (last) {
+			if (likely(last->urb != urb)) {
+				if (last->urb->complete == NULL) {
+					murb = (struct oxu_murb *) last->urb;
+					last->urb = murb->main;
+					if (murb->last) {
+						ehci_urb_done(oxu, last->urb);
+						count++;
+					}
+					oxu_murb_free(oxu, murb);
+				} else {
+					ehci_urb_done(oxu, last->urb);
+					count++;
+				}
+			}
+			oxu_qtd_free(oxu, last);
+			last = NULL;
+		}
+
+		/* ignore urbs submitted during completions we reported */
+		if (qtd == end)
+			break;
+
+		/* hardware copies qtd out of qh overlay */
+		rmb();
+		token = le32_to_cpu(qtd->hw_token);
+
+		/* always clean up qtds the hc de-activated */
+		if ((token & QTD_STS_ACTIVE) == 0) {
+
+			if ((token & QTD_STS_HALT) != 0) {
+				stopped = 1;
+
+			/* magic dummy for some short reads; qh won't advance.
+			 * that silicon quirk can kick in with this dummy too.
+			 */
+			} else if (IS_SHORT_READ(token) &&
+					!(qtd->hw_alt_next & EHCI_LIST_END)) {
+				stopped = 1;
+				goto halt;
+			}
+
+		/* stop scanning when we reach qtds the hc is using */
+		} else if (likely(!stopped &&
+				HC_IS_RUNNING(oxu_to_hcd(oxu)->state))) {
+			break;
+
+		} else {
+			stopped = 1;
+
+			if (unlikely(!HC_IS_RUNNING(oxu_to_hcd(oxu)->state)))
+				urb->status = -ESHUTDOWN;
+
+			/* ignore active urbs unless some previous qtd
+			 * for the urb faulted (including short read) or
+			 * its urb was canceled.  we may patch qh or qtds.
+			 */
+			if (likely(urb->status == -EINPROGRESS))
+				continue;
+
+			/* issue status after short control reads */
+			if (unlikely(do_status != 0)
+					&& QTD_PID(token) == 0 /* OUT */) {
+				do_status = 0;
+				continue;
+			}
+
+			/* token in overlay may be most current */
+			if (state == QH_STATE_IDLE
+					&& cpu_to_le32(qtd->qtd_dma)
+						== qh->hw_current)
+				token = le32_to_cpu(qh->hw_token);
+
+			/* force halt for unlinked or blocked qh, so we'll
+			 * patch the qh later and so that completions can't
+			 * activate it while we "know" it's stopped.
+			 */
+			if ((HALT_BIT & qh->hw_token) == 0) {
+halt:
+				qh->hw_token |= HALT_BIT;
+				wmb();
+			}
+		}
+
+		/* Remove it from the queue */
+		qtd_copy_status(oxu, urb->complete ?
+					urb : ((struct oxu_murb *) urb)->main,
+				qtd->length, token);
+		if ((usb_pipein(qtd->urb->pipe)) &&
+				(NULL != qtd->transfer_buffer))
+			memcpy(qtd->transfer_buffer, qtd->buffer, qtd->length);
+		do_status = (urb->status == -EREMOTEIO)
+				&& usb_pipecontrol(urb->pipe);
+
+		if (stopped && qtd->qtd_list.prev != &qh->qtd_list) {
+			last = list_entry(qtd->qtd_list.prev,
+					struct ehci_qtd, qtd_list);
+			last->hw_next = qtd->hw_next;
+		}
+		list_del(&qtd->qtd_list);
+		last = qtd;
+	}
+
+	/* last urb's completion might still need calling */
+	if (likely(last != NULL)) {
+		if (last->urb->complete == NULL) {
+			murb = (struct oxu_murb *) last->urb;
+			last->urb = murb->main;
+			if (murb->last) {
+				ehci_urb_done(oxu, last->urb);
+				count++;
+			}
+			oxu_murb_free(oxu, murb);
+		} else {
+			ehci_urb_done(oxu, last->urb);
+			count++;
+		}
+		oxu_qtd_free(oxu, last);
+	}
+
+	/* restore original state; caller must unlink or relink */
+	qh->qh_state = state;
+
+	/* be sure the hardware's done with the qh before refreshing
+	 * it after fault cleanup, or recovering from silicon wrongly
+	 * overlaying the dummy qtd (which reduces DMA chatter).
+	 */
+	if (stopped != 0 || qh->hw_qtd_next == EHCI_LIST_END) {
+		switch (state) {
+		case QH_STATE_IDLE:
+			qh_refresh(oxu, qh);
+			break;
+		case QH_STATE_LINKED:
+			/* should be rare for periodic transfers,
+			 * except maybe high bandwidth ...
+			 */
+			if ((__constant_cpu_to_le32(QH_SMASK)
+					& qh->hw_info2) != 0) {
+				intr_deschedule(oxu, qh);
+				(void) qh_schedule(oxu, qh);
+			} else
+				unlink_async(oxu, qh);
+			break;
+		/* otherwise, unlink already started */
+		}
+	}
+
+	return count;
+}
+
+/* High bandwidth multiplier, as encoded in highspeed endpoint descriptors */
+#define hb_mult(wMaxPacketSize)		(1 + (((wMaxPacketSize) >> 11) & 0x03))
+/* ... and packet size, for any kind of endpoint descriptor */
+#define max_packet(wMaxPacketSize)	((wMaxPacketSize) & 0x07ff)
+
+/* Reverse of qh_urb_transaction: free a list of TDs.
+ * used for cleanup after errors, before HC sees an URB's TDs.
+ */
+static void qtd_list_free(struct oxu_hcd *oxu,
+				struct urb *urb, struct list_head *qtd_list)
+{
+	struct list_head *entry, *temp;
+
+	list_for_each_safe(entry, temp, qtd_list) {
+		struct ehci_qtd	*qtd;
+
+		qtd = list_entry(entry, struct ehci_qtd, qtd_list);
+		list_del(&qtd->qtd_list);
+		oxu_qtd_free(oxu, qtd);
+	}
+}
+
+/* Create a list of filled qtds for this URB; won't link into qh.
+ */
+static struct list_head *qh_urb_transaction(struct oxu_hcd *oxu,
+						struct urb *urb,
+						struct list_head *head,
+						gfp_t flags)
+{
+	struct ehci_qtd	*qtd, *qtd_prev;
+	dma_addr_t buf;
+	int len, maxpacket;
+	int is_input;
+	u32 token;
+	void *transfer_buf = NULL;
+	int ret;
+
+	/*
+	 * URBs map to sequences of QTDs: one logical transaction
+	 */
+	qtd = ehci_qtd_alloc(oxu);
+	if (unlikely(!qtd))
+		return NULL;
+	list_add_tail(&qtd->qtd_list, head);
+	qtd->urb = urb;
+
+	token = QTD_STS_ACTIVE;
+	token |= (EHCI_TUNE_CERR << 10);
+	/* for split transactions, SplitXState initialized to zero */
+
+	len = urb->transfer_buffer_length;
+	is_input = usb_pipein(urb->pipe);
+	if (!urb->transfer_buffer && urb->transfer_buffer_length && is_input)
+		urb->transfer_buffer = phys_to_virt(urb->transfer_dma);
+
+	if (usb_pipecontrol(urb->pipe)) {
+		/* SETUP pid */
+		ret = oxu_buf_alloc(oxu, qtd, sizeof(struct usb_ctrlrequest));
+		if (ret)
+			goto cleanup;
+
+		qtd_fill(qtd, qtd->buffer_dma, sizeof(struct usb_ctrlrequest),
+				token | (2 /* "setup" */ << 8), 8);
+		memcpy(qtd->buffer, qtd->urb->setup_packet,
+				sizeof(struct usb_ctrlrequest));
+
+		/* ... and always at least one more pid */
+		token ^= QTD_TOGGLE;
+		qtd_prev = qtd;
+		qtd = ehci_qtd_alloc(oxu);
+		if (unlikely(!qtd))
+			goto cleanup;
+		qtd->urb = urb;
+		qtd_prev->hw_next = QTD_NEXT(qtd->qtd_dma);
+		list_add_tail(&qtd->qtd_list, head);
+
+		/* for zero length DATA stages, STATUS is always IN */
+		if (len == 0)
+			token |= (1 /* "in" */ << 8);
+	}
+
+	/*
+	 * Data transfer stage: buffer setup
+	 */
+
+	ret = oxu_buf_alloc(oxu, qtd, len);
+	if (ret)
+		goto cleanup;
+
+	buf = qtd->buffer_dma;
+	transfer_buf = urb->transfer_buffer;
+
+	if (!is_input)
+		memcpy(qtd->buffer, qtd->urb->transfer_buffer, len);
+
+	if (is_input)
+		token |= (1 /* "in" */ << 8);
+	/* else it's already initted to "out" pid (0 << 8) */
+
+	maxpacket = max_packet(usb_maxpacket(urb->dev, urb->pipe, !is_input));
+
+	/*
+	 * buffer gets wrapped in one or more qtds;
+	 * last one may be "short" (including zero len)
+	 * and may serve as a control status ack
+	 */
+	for (;;) {
+		int this_qtd_len;
+
+		this_qtd_len = qtd_fill(qtd, buf, len, token, maxpacket);
+		qtd->transfer_buffer = transfer_buf;
+		len -= this_qtd_len;
+		buf += this_qtd_len;
+		transfer_buf += this_qtd_len;
+		if (is_input)
+			qtd->hw_alt_next = oxu->async->hw_alt_next;
+
+		/* qh makes control packets use qtd toggle; maybe switch it */
+		if ((maxpacket & (this_qtd_len + (maxpacket - 1))) == 0)
+			token ^= QTD_TOGGLE;
+
+		if (likely(len <= 0))
+			break;
+
+		qtd_prev = qtd;
+		qtd = ehci_qtd_alloc(oxu);
+		if (unlikely(!qtd))
+			goto cleanup;
+		if (likely(len > 0)) {
+			ret = oxu_buf_alloc(oxu, qtd, len);
+			if (ret)
+				goto cleanup;
+		}
+		qtd->urb = urb;
+		qtd_prev->hw_next = QTD_NEXT(qtd->qtd_dma);
+		list_add_tail(&qtd->qtd_list, head);
+	}
+
+	/* unless the bulk/interrupt caller wants a chance to clean
+	 * up after short reads, hc should advance qh past this urb
+	 */
+	if (likely((urb->transfer_flags & URB_SHORT_NOT_OK) == 0
+				|| usb_pipecontrol(urb->pipe)))
+		qtd->hw_alt_next = EHCI_LIST_END;
+
+	/*
+	 * control requests may need a terminating data "status" ack;
+	 * bulk ones may need a terminating short packet (zero length).
+	 */
+	if (likely(urb->transfer_buffer_length != 0)) {
+		int	one_more = 0;
+
+		if (usb_pipecontrol(urb->pipe)) {
+			one_more = 1;
+			token ^= 0x0100;	/* "in" <--> "out"  */
+			token |= QTD_TOGGLE;	/* force DATA1 */
+		} else if (usb_pipebulk(urb->pipe)
+				&& (urb->transfer_flags & URB_ZERO_PACKET)
+				&& !(urb->transfer_buffer_length % maxpacket)) {
+			one_more = 1;
+		}
+		if (one_more) {
+			qtd_prev = qtd;
+			qtd = ehci_qtd_alloc(oxu);
+			if (unlikely(!qtd))
+				goto cleanup;
+			qtd->urb = urb;
+			qtd_prev->hw_next = QTD_NEXT(qtd->qtd_dma);
+			list_add_tail(&qtd->qtd_list, head);
+
+			/* never any data in such packets */
+			qtd_fill(qtd, 0, 0, token, 0);
+		}
+	}
+
+	/* by default, enable interrupt on urb completion */
+		qtd->hw_token |= __constant_cpu_to_le32(QTD_IOC);
+	return head;
+
+cleanup:
+	qtd_list_free(oxu, urb, head);
+	return NULL;
+}
+
+/* Each QH holds a qtd list; a QH is used for everything except iso.
+ *
+ * For interrupt urbs, the scheduler must set the microframe scheduling
+ * mask(s) each time the QH gets scheduled.  For highspeed, that's
+ * just one microframe in the s-mask.  For split interrupt transactions
+ * there are additional complications: c-mask, maybe FSTNs.
+ */
+static struct ehci_qh *qh_make(struct oxu_hcd *oxu,
+				struct urb *urb, gfp_t flags)
+{
+	struct ehci_qh *qh = oxu_qh_alloc(oxu);
+	u32 info1 = 0, info2 = 0;
+	int is_input, type;
+	int maxp = 0;
+
+	if (!qh)
+		return qh;
+
+	/*
+	 * init endpoint/device data for this QH
+	 */
+	info1 |= usb_pipeendpoint(urb->pipe) << 8;
+	info1 |= usb_pipedevice(urb->pipe) << 0;
+
+	is_input = usb_pipein(urb->pipe);
+	type = usb_pipetype(urb->pipe);
+	maxp = usb_maxpacket(urb->dev, urb->pipe, !is_input);
+
+	/* Compute interrupt scheduling parameters just once, and save.
+	 * - allowing for high bandwidth, how many nsec/uframe are used?
+	 * - split transactions need a second CSPLIT uframe; same question
+	 * - splits also need a schedule gap (for full/low speed I/O)
+	 * - qh has a polling interval
+	 *
+	 * For control/bulk requests, the HC or TT handles these.
+	 */
+	if (type == PIPE_INTERRUPT) {
+		qh->usecs = NS_TO_US(usb_calc_bus_time(USB_SPEED_HIGH,
+								is_input, 0,
+				hb_mult(maxp) * max_packet(maxp)));
+		qh->start = NO_FRAME;
+
+		if (urb->dev->speed == USB_SPEED_HIGH) {
+			qh->c_usecs = 0;
+			qh->gap_uf = 0;
+
+			qh->period = urb->interval >> 3;
+			if (qh->period == 0 && urb->interval != 1) {
+				/* NOTE interval 2 or 4 uframes could work.
+				 * But interval 1 scheduling is simpler, and
+				 * includes high bandwidth.
+				 */
+				dbg("intr period %d uframes, NYET!",
+						urb->interval);
+				goto done;
+			}
+		} else {
+			struct usb_tt	*tt = urb->dev->tt;
+			int		think_time;
+
+			/* gap is f(FS/LS transfer times) */
+			qh->gap_uf = 1 + usb_calc_bus_time(urb->dev->speed,
+					is_input, 0, maxp) / (125 * 1000);
+
+			/* FIXME this just approximates SPLIT/CSPLIT times */
+			if (is_input) {		/* SPLIT, gap, CSPLIT+DATA */
+				qh->c_usecs = qh->usecs + HS_USECS(0);
+				qh->usecs = HS_USECS(1);
+			} else {		/* SPLIT+DATA, gap, CSPLIT */
+				qh->usecs += HS_USECS(1);
+				qh->c_usecs = HS_USECS(0);
+			}
+
+			think_time = tt ? tt->think_time : 0;
+			qh->tt_usecs = NS_TO_US(think_time +
+					usb_calc_bus_time(urb->dev->speed,
+					is_input, 0, max_packet(maxp)));
+			qh->period = urb->interval;
+		}
+	}
+
+	/* support for tt scheduling, and access to toggles */
+	qh->dev = urb->dev;
+
+	/* using TT? */
+	switch (urb->dev->speed) {
+	case USB_SPEED_LOW:
+		info1 |= (1 << 12);	/* EPS "low" */
+		/* FALL THROUGH */
+
+	case USB_SPEED_FULL:
+		/* EPS 0 means "full" */
+		if (type != PIPE_INTERRUPT)
+			info1 |= (EHCI_TUNE_RL_TT << 28);
+		if (type == PIPE_CONTROL) {
+			info1 |= (1 << 27);	/* for TT */
+			info1 |= 1 << 14;	/* toggle from qtd */
+		}
+		info1 |= maxp << 16;
+
+		info2 |= (EHCI_TUNE_MULT_TT << 30);
+		info2 |= urb->dev->ttport << 23;
+
+		/* NOTE:  if (PIPE_INTERRUPT) { scheduler sets c-mask } */
+
+		break;
+
+	case USB_SPEED_HIGH:		/* no TT involved */
+		info1 |= (2 << 12);	/* EPS "high" */
+		if (type == PIPE_CONTROL) {
+			info1 |= (EHCI_TUNE_RL_HS << 28);
+			info1 |= 64 << 16;	/* usb2 fixed maxpacket */
+			info1 |= 1 << 14;	/* toggle from qtd */
+			info2 |= (EHCI_TUNE_MULT_HS << 30);
+		} else if (type == PIPE_BULK) {
+			info1 |= (EHCI_TUNE_RL_HS << 28);
+			info1 |= 512 << 16;	/* usb2 fixed maxpacket */
+			info2 |= (EHCI_TUNE_MULT_HS << 30);
+		} else {		/* PIPE_INTERRUPT */
+			info1 |= max_packet(maxp) << 16;
+			info2 |= hb_mult(maxp) << 30;
+		}
+		break;
+	default:
+		dbg("bogus dev %p speed %d", urb->dev, urb->dev->speed);
+done:
+		qh_put(qh);
+		return NULL;
+	}
+
+	/* NOTE:  if (PIPE_INTERRUPT) { scheduler sets s-mask } */
+
+	/* init as live, toggle clear, advance to dummy */
+	qh->qh_state = QH_STATE_IDLE;
+	qh->hw_info1 = cpu_to_le32(info1);
+	qh->hw_info2 = cpu_to_le32(info2);
+	usb_settoggle(urb->dev, usb_pipeendpoint(urb->pipe), !is_input, 1);
+	qh_refresh(oxu, qh);
+	return qh;
+}
+
+/* Move qh (and its qtds) onto async queue; maybe enable queue.
+ */
+static void qh_link_async(struct oxu_hcd *oxu, struct ehci_qh *qh)
+{
+	__le32 dma = QH_NEXT(qh->qh_dma);
+	struct ehci_qh *head;
+
+	/* (re)start the async schedule? */
+	head = oxu->async;
+	timer_action_done(oxu, TIMER_ASYNC_OFF);
+	if (!head->qh_next.qh) {
+		u32	cmd = readl(&oxu->regs->command);
+
+		if (!(cmd & CMD_ASE)) {
+			/* in case a clear of CMD_ASE didn't take yet */
+			(void)handshake(oxu, &oxu->regs->status,
+					STS_ASS, 0, 150);
+			cmd |= CMD_ASE | CMD_RUN;
+			writel(cmd, &oxu->regs->command);
+			oxu_to_hcd(oxu)->state = HC_STATE_RUNNING;
+			/* posted write need not be known to HC yet ... */
+		}
+	}
+
+	/* clear halt and/or toggle; and maybe recover from silicon quirk */
+	if (qh->qh_state == QH_STATE_IDLE)
+		qh_refresh(oxu, qh);
+
+	/* splice right after start */
+	qh->qh_next = head->qh_next;
+	qh->hw_next = head->hw_next;
+	wmb();
+
+	head->qh_next.qh = qh;
+	head->hw_next = dma;
+
+	qh->qh_state = QH_STATE_LINKED;
+	/* qtd completions reported later by interrupt */
+}
+
+#define	QH_ADDR_MASK	__constant_cpu_to_le32(0x7f)
+
+/*
+ * For control/bulk/interrupt, return QH with these TDs appended.
+ * Allocates and initializes the QH if necessary.
+ * Returns null if it can't allocate a QH it needs to.
+ * If the QH has TDs (urbs) already, that's great.
+ */
+static struct ehci_qh *qh_append_tds(struct oxu_hcd *oxu,
+				struct urb *urb, struct list_head *qtd_list,
+				int epnum, void	**ptr)
+{
+	struct ehci_qh *qh = NULL;
+
+	qh = (struct ehci_qh *) *ptr;
+	if (unlikely(qh == NULL)) {
+		/* can't sleep here, we have oxu->lock... */
+		qh = qh_make(oxu, urb, GFP_ATOMIC);
+		*ptr = qh;
+	}
+	if (likely(qh != NULL)) {
+		struct ehci_qtd	*qtd;
+
+		if (unlikely(list_empty(qtd_list)))
+			qtd = NULL;
+		else
+			qtd = list_entry(qtd_list->next, struct ehci_qtd,
+					qtd_list);
+
+		/* control qh may need patching ... */
+		if (unlikely(epnum == 0)) {
+
+			/* usb_reset_device() briefly reverts to address 0 */
+			if (usb_pipedevice(urb->pipe) == 0)
+				qh->hw_info1 &= ~QH_ADDR_MASK;
+		}
+
+		/* just one way to queue requests: swap with the dummy qtd.
+		 * only hc or qh_refresh() ever modify the overlay.
+		 */
+		if (likely(qtd != NULL)) {
+			struct ehci_qtd	*dummy;
+			dma_addr_t dma;
+			__le32 token;
+
+			/* to avoid racing the HC, use the dummy td instead of
+			 * the first td of our list (becomes new dummy).  both
+			 * tds stay deactivated until we're done, when the
+			 * HC is allowed to fetch the old dummy (4.10.2).
+			 */
+			token = qtd->hw_token;
+			qtd->hw_token = HALT_BIT;
+			wmb();
+			dummy = qh->dummy;
+
+			dma = dummy->qtd_dma;
+			*dummy = *qtd;
+			dummy->qtd_dma = dma;
+
+			list_del(&qtd->qtd_list);
+			list_add(&dummy->qtd_list, qtd_list);
+			list_splice(qtd_list, qh->qtd_list.prev);
+
+			ehci_qtd_init(qtd, qtd->qtd_dma);
+			qh->dummy = qtd;
+
+			/* hc must see the new dummy at list end */
+			dma = qtd->qtd_dma;
+			qtd = list_entry(qh->qtd_list.prev,
+					struct ehci_qtd, qtd_list);
+			qtd->hw_next = QTD_NEXT(dma);
+
+			/* let the hc process these next qtds */
+			dummy->hw_token = (token & ~(0x80));
+			wmb();
+			dummy->hw_token = token;
+
+			urb->hcpriv = qh_get(qh);
+		}
+	}
+	return qh;
+}
+
+static int submit_async(struct oxu_hcd	*oxu, struct urb *urb,
+			struct list_head *qtd_list, gfp_t mem_flags)
+{
+	struct ehci_qtd	*qtd;
+	int epnum;
+	unsigned long flags;
+	struct ehci_qh *qh = NULL;
+	int rc = 0;
+
+	qtd = list_entry(qtd_list->next, struct ehci_qtd, qtd_list);
+	epnum = urb->ep->desc.bEndpointAddress;
+
+#ifdef OXU_URB_TRACE
+	oxu_dbg(oxu, "%s %s urb %p ep%d%s len %d, qtd %p [qh %p]\n",
+		__func__, urb->dev->devpath, urb,
+		epnum & 0x0f, (epnum & USB_DIR_IN) ? "in" : "out",
+		urb->transfer_buffer_length,
+		qtd, urb->ep->hcpriv);
+#endif
+
+	spin_lock_irqsave(&oxu->lock, flags);
+	if (unlikely(!test_bit(HCD_FLAG_HW_ACCESSIBLE,
+			       &oxu_to_hcd(oxu)->flags))) {
+		rc = -ESHUTDOWN;
+		goto done;
+	}
+
+	qh = qh_append_tds(oxu, urb, qtd_list, epnum, &urb->ep->hcpriv);
+	if (unlikely(qh == NULL)) {
+		rc = -ENOMEM;
+		goto done;
+	}
+
+	/* Control/bulk operations through TTs don't need scheduling,
+	 * the HC and TT handle it when the TT has a buffer ready.
+	 */
+	if (likely(qh->qh_state == QH_STATE_IDLE))
+		qh_link_async(oxu, qh_get(qh));
+done:
+	spin_unlock_irqrestore(&oxu->lock, flags);
+	if (unlikely(qh == NULL))
+		qtd_list_free(oxu, urb, qtd_list);
+	return rc;
+}
+
+/* The async qh for the qtds being reclaimed are now unlinked from the HC */
+
+static void end_unlink_async(struct oxu_hcd *oxu)
+{
+	struct ehci_qh *qh = oxu->reclaim;
+	struct ehci_qh *next;
+
+	timer_action_done(oxu, TIMER_IAA_WATCHDOG);
+
+	qh->qh_state = QH_STATE_IDLE;
+	qh->qh_next.qh = NULL;
+	qh_put(qh);			/* refcount from reclaim */
+
+	/* other unlink(s) may be pending (in QH_STATE_UNLINK_WAIT) */
+	next = qh->reclaim;
+	oxu->reclaim = next;
+	oxu->reclaim_ready = 0;
+	qh->reclaim = NULL;
+
+	qh_completions(oxu, qh);
+
+	if (!list_empty(&qh->qtd_list)
+			&& HC_IS_RUNNING(oxu_to_hcd(oxu)->state))
+		qh_link_async(oxu, qh);
+	else {
+		qh_put(qh);		/* refcount from async list */
+
+		/* it's not free to turn the async schedule on/off; leave it
+		 * active but idle for a while once it empties.
+		 */
+		if (HC_IS_RUNNING(oxu_to_hcd(oxu)->state)
+				&& oxu->async->qh_next.qh == NULL)
+			timer_action(oxu, TIMER_ASYNC_OFF);
+	}
+
+	if (next) {
+		oxu->reclaim = NULL;
+		start_unlink_async(oxu, next);
+	}
+}
+
+/* makes sure the async qh will become idle */
+/* caller must own oxu->lock */
+
+static void start_unlink_async(struct oxu_hcd *oxu, struct ehci_qh *qh)
+{
+	int cmd = readl(&oxu->regs->command);
+	struct ehci_qh *prev;
+
+#ifdef DEBUG
+	assert_spin_locked(&oxu->lock);
+	if (oxu->reclaim || (qh->qh_state != QH_STATE_LINKED
+				&& qh->qh_state != QH_STATE_UNLINK_WAIT))
+		BUG();
+#endif
+
+	/* stop async schedule right now? */
+	if (unlikely(qh == oxu->async)) {
+		/* can't get here without STS_ASS set */
+		if (oxu_to_hcd(oxu)->state != HC_STATE_HALT
+				&& !oxu->reclaim) {
+			/* ... and CMD_IAAD clear */
+			writel(cmd & ~CMD_ASE, &oxu->regs->command);
+			wmb();
+			/* handshake later, if we need to */
+			timer_action_done(oxu, TIMER_ASYNC_OFF);
+		}
+		return;
+	}
+
+	qh->qh_state = QH_STATE_UNLINK;
+	oxu->reclaim = qh = qh_get(qh);
+
+	prev = oxu->async;
+	while (prev->qh_next.qh != qh)
+		prev = prev->qh_next.qh;
+
+	prev->hw_next = qh->hw_next;
+	prev->qh_next = qh->qh_next;
+	wmb();
+
+	if (unlikely(oxu_to_hcd(oxu)->state == HC_STATE_HALT)) {
+		/* if (unlikely(qh->reclaim != 0))
+		 *	this will recurse, probably not much
+		 */
+		end_unlink_async(oxu);
+		return;
+	}
+
+	oxu->reclaim_ready = 0;
+	cmd |= CMD_IAAD;
+	writel(cmd, &oxu->regs->command);
+	(void) readl(&oxu->regs->command);
+	timer_action(oxu, TIMER_IAA_WATCHDOG);
+}
+
+static void scan_async(struct oxu_hcd *oxu)
+{
+	struct ehci_qh *qh;
+	enum ehci_timer_action action = TIMER_IO_WATCHDOG;
+
+	if (!++(oxu->stamp))
+		oxu->stamp++;
+	timer_action_done(oxu, TIMER_ASYNC_SHRINK);
+rescan:
+	qh = oxu->async->qh_next.qh;
+	if (likely(qh != NULL)) {
+		do {
+			/* clean any finished work for this qh */
+			if (!list_empty(&qh->qtd_list)
+					&& qh->stamp != oxu->stamp) {
+				int temp;
+
+				/* unlinks could happen here; completion
+				 * reporting drops the lock.  rescan using
+				 * the latest schedule, but don't rescan
+				 * qhs we already finished (no looping).
+				 */
+				qh = qh_get(qh);
+				qh->stamp = oxu->stamp;
+				temp = qh_completions(oxu, qh);
+				qh_put(qh);
+				if (temp != 0)
+					goto rescan;
+			}
+
+			/* unlink idle entries, reducing HC PCI usage as well
+			 * as HCD schedule-scanning costs.  delay for any qh
+			 * we just scanned, there's a not-unusual case that it
+			 * doesn't stay idle for long.
+			 * (plus, avoids some kind of re-activation race.)
+			 */
+			if (list_empty(&qh->qtd_list)) {
+				if (qh->stamp == oxu->stamp)
+					action = TIMER_ASYNC_SHRINK;
+				else if (!oxu->reclaim
+					    && qh->qh_state == QH_STATE_LINKED)
+					start_unlink_async(oxu, qh);
+			}
+
+			qh = qh->qh_next.qh;
+		} while (qh);
+	}
+	if (action == TIMER_ASYNC_SHRINK)
+		timer_action(oxu, TIMER_ASYNC_SHRINK);
+}
+
+/*
+ * periodic_next_shadow - return "next" pointer on shadow list
+ * @periodic: host pointer to qh/itd/sitd
+ * @tag: hardware tag for type of this record
+ */
+static union ehci_shadow *periodic_next_shadow(union ehci_shadow *periodic,
+						__le32 tag)
+{
+	switch (tag) {
+	default:
+	case Q_TYPE_QH:
+		return &periodic->qh->qh_next;
+	}
+}
+
+/* caller must hold oxu->lock */
+static void periodic_unlink(struct oxu_hcd *oxu, unsigned frame, void *ptr)
+{
+	union ehci_shadow *prev_p = &oxu->pshadow[frame];
+	__le32 *hw_p = &oxu->periodic[frame];
+	union ehci_shadow here = *prev_p;
+
+	/* find predecessor of "ptr"; hw and shadow lists are in sync */
+	while (here.ptr && here.ptr != ptr) {
+		prev_p = periodic_next_shadow(prev_p, Q_NEXT_TYPE(*hw_p));
+		hw_p = here.hw_next;
+		here = *prev_p;
+	}
+	/* an interrupt entry (at list end) could have been shared */
+	if (!here.ptr)
+		return;
+
+	/* update shadow and hardware lists ... the old "next" pointers
+	 * from ptr may still be in use, the caller updates them.
+	 */
+	*prev_p = *periodic_next_shadow(&here, Q_NEXT_TYPE(*hw_p));
+	*hw_p = *here.hw_next;
+}
+
+/* how many of the uframe's 125 usecs are allocated? */
+static unsigned short periodic_usecs(struct oxu_hcd *oxu,
+					unsigned frame, unsigned uframe)
+{
+	__le32 *hw_p = &oxu->periodic[frame];
+	union ehci_shadow *q = &oxu->pshadow[frame];
+	unsigned usecs = 0;
+
+	while (q->ptr) {
+		switch (Q_NEXT_TYPE(*hw_p)) {
+		case Q_TYPE_QH:
+		default:
+			/* is it in the S-mask? */
+			if (q->qh->hw_info2 & cpu_to_le32(1 << uframe))
+				usecs += q->qh->usecs;
+			/* ... or C-mask? */
+			if (q->qh->hw_info2 & cpu_to_le32(1 << (8 + uframe)))
+				usecs += q->qh->c_usecs;
+			hw_p = &q->qh->hw_next;
+			q = &q->qh->qh_next;
+			break;
+		}
+	}
+#ifdef DEBUG
+	if (usecs > 100)
+		oxu_err(oxu, "uframe %d sched overrun: %d usecs\n",
+						frame * 8 + uframe, usecs);
+#endif
+	return usecs;
+}
+
+static int enable_periodic(struct oxu_hcd *oxu)
+{
+	u32 cmd;
+	int status;
+
+	/* did clearing PSE did take effect yet?
+	 * takes effect only at frame boundaries...
+	 */
+	status = handshake(oxu, &oxu->regs->status, STS_PSS, 0, 9 * 125);
+	if (status != 0) {
+		oxu_to_hcd(oxu)->state = HC_STATE_HALT;
+		return status;
+	}
+
+	cmd = readl(&oxu->regs->command) | CMD_PSE;
+	writel(cmd, &oxu->regs->command);
+	/* posted write ... PSS happens later */
+	oxu_to_hcd(oxu)->state = HC_STATE_RUNNING;
+
+	/* make sure ehci_work scans these */
+	oxu->next_uframe = readl(&oxu->regs->frame_index)
+		% (oxu->periodic_size << 3);
+	return 0;
+}
+
+static int disable_periodic(struct oxu_hcd *oxu)
+{
+	u32 cmd;
+	int status;
+
+	/* did setting PSE not take effect yet?
+	 * takes effect only at frame boundaries...
+	 */
+	status = handshake(oxu, &oxu->regs->status, STS_PSS, STS_PSS, 9 * 125);
+	if (status != 0) {
+		oxu_to_hcd(oxu)->state = HC_STATE_HALT;
+		return status;
+	}
+
+	cmd = readl(&oxu->regs->command) & ~CMD_PSE;
+	writel(cmd, &oxu->regs->command);
+	/* posted write ... */
+
+	oxu->next_uframe = -1;
+	return 0;
+}
+
+/* periodic schedule slots have iso tds (normal or split) first, then a
+ * sparse tree for active interrupt transfers.
+ *
+ * this just links in a qh; caller guarantees uframe masks are set right.
+ * no FSTN support (yet; oxu 0.96+)
+ */
+static int qh_link_periodic(struct oxu_hcd *oxu, struct ehci_qh *qh)
+{
+	unsigned i;
+	unsigned period = qh->period;
+
+	dev_dbg(&qh->dev->dev,
+		"link qh%d-%04x/%p start %d [%d/%d us]\n",
+		period, le32_to_cpup(&qh->hw_info2) & (QH_CMASK | QH_SMASK),
+		qh, qh->start, qh->usecs, qh->c_usecs);
+
+	/* high bandwidth, or otherwise every microframe */
+	if (period == 0)
+		period = 1;
+
+	for (i = qh->start; i < oxu->periodic_size; i += period) {
+		union ehci_shadow	*prev = &oxu->pshadow[i];
+		__le32			*hw_p = &oxu->periodic[i];
+		union ehci_shadow	here = *prev;
+		__le32			type = 0;
+
+		/* skip the iso nodes at list head */
+		while (here.ptr) {
+			type = Q_NEXT_TYPE(*hw_p);
+			if (type == Q_TYPE_QH)
+				break;
+			prev = periodic_next_shadow(prev, type);
+			hw_p = &here.qh->hw_next;
+			here = *prev;
+		}
+
+		/* sorting each branch by period (slow-->fast)
+		 * enables sharing interior tree nodes
+		 */
+		while (here.ptr && qh != here.qh) {
+			if (qh->period > here.qh->period)
+				break;
+			prev = &here.qh->qh_next;
+			hw_p = &here.qh->hw_next;
+			here = *prev;
+		}
+		/* link in this qh, unless some earlier pass did that */
+		if (qh != here.qh) {
+			qh->qh_next = here;
+			if (here.qh)
+				qh->hw_next = *hw_p;
+			wmb();
+			prev->qh = qh;
+			*hw_p = QH_NEXT(qh->qh_dma);
+		}
+	}
+	qh->qh_state = QH_STATE_LINKED;
+	qh_get(qh);
+
+	/* update per-qh bandwidth for usbfs */
+	oxu_to_hcd(oxu)->self.bandwidth_allocated += qh->period
+		? ((qh->usecs + qh->c_usecs) / qh->period)
+		: (qh->usecs * 8);
+
+	/* maybe enable periodic schedule processing */
+	if (!oxu->periodic_sched++)
+		return enable_periodic(oxu);
+
+	return 0;
+}
+
+static void qh_unlink_periodic(struct oxu_hcd *oxu, struct ehci_qh *qh)
+{
+	unsigned i;
+	unsigned period;
+
+	/* FIXME:
+	 *   IF this isn't high speed
+	 *   and this qh is active in the current uframe
+	 *   (and overlay token SplitXstate is false?)
+	 * THEN
+	 *   qh->hw_info1 |= __constant_cpu_to_le32(1 << 7 "ignore");
+	 */
+
+	/* high bandwidth, or otherwise part of every microframe */
+	period = qh->period;
+	if (period == 0)
+		period = 1;
+
+	for (i = qh->start; i < oxu->periodic_size; i += period)
+		periodic_unlink(oxu, i, qh);
+
+	/* update per-qh bandwidth for usbfs */
+	oxu_to_hcd(oxu)->self.bandwidth_allocated -= qh->period
+		? ((qh->usecs + qh->c_usecs) / qh->period)
+		: (qh->usecs * 8);
+
+	dev_dbg(&qh->dev->dev,
+		"unlink qh%d-%04x/%p start %d [%d/%d us]\n",
+		qh->period,
+		le32_to_cpup(&qh->hw_info2) & (QH_CMASK | QH_SMASK),
+		qh, qh->start, qh->usecs, qh->c_usecs);
+
+	/* qh->qh_next still "live" to HC */
+	qh->qh_state = QH_STATE_UNLINK;
+	qh->qh_next.ptr = NULL;
+	qh_put(qh);
+
+	/* maybe turn off periodic schedule */
+	oxu->periodic_sched--;
+	if (!oxu->periodic_sched)
+		(void) disable_periodic(oxu);
+}
+
+static void intr_deschedule(struct oxu_hcd *oxu, struct ehci_qh *qh)
+{
+	unsigned wait;
+
+	qh_unlink_periodic(oxu, qh);
+
+	/* simple/paranoid:  always delay, expecting the HC needs to read
+	 * qh->hw_next or finish a writeback after SPLIT/CSPLIT ... and
+	 * expect khubd to clean up after any CSPLITs we won't issue.
+	 * active high speed queues may need bigger delays...
+	 */
+	if (list_empty(&qh->qtd_list)
+		|| (__constant_cpu_to_le32(QH_CMASK) & qh->hw_info2) != 0)
+		wait = 2;
+	else
+		wait = 55;	/* worst case: 3 * 1024 */
+
+	udelay(wait);
+	qh->qh_state = QH_STATE_IDLE;
+	qh->hw_next = EHCI_LIST_END;
+	wmb();
+}
+
+static int check_period(struct oxu_hcd *oxu,
+			unsigned frame, unsigned uframe,
+			unsigned period, unsigned usecs)
+{
+	int claimed;
+
+	/* complete split running into next frame?
+	 * given FSTN support, we could sometimes check...
+	 */
+	if (uframe >= 8)
+		return 0;
+
+	/*
+	 * 80% periodic == 100 usec/uframe available
+	 * convert "usecs we need" to "max already claimed"
+	 */
+	usecs = 100 - usecs;
+
+	/* we "know" 2 and 4 uframe intervals were rejected; so
+	 * for period 0, check _every_ microframe in the schedule.
+	 */
+	if (unlikely(period == 0)) {
+		do {
+			for (uframe = 0; uframe < 7; uframe++) {
+				claimed = periodic_usecs(oxu, frame, uframe);
+				if (claimed > usecs)
+					return 0;
+			}
+		} while ((frame += 1) < oxu->periodic_size);
+
+	/* just check the specified uframe, at that period */
+	} else {
+		do {
+			claimed = periodic_usecs(oxu, frame, uframe);
+			if (claimed > usecs)
+				return 0;
+		} while ((frame += period) < oxu->periodic_size);
+	}
+
+	return 1;
+}
+
+static int check_intr_schedule(struct oxu_hcd	*oxu,
+				unsigned frame, unsigned uframe,
+				const struct ehci_qh *qh, __le32 *c_maskp)
+{
+	int retval = -ENOSPC;
+
+	if (qh->c_usecs && uframe >= 6)		/* FSTN territory? */
+		goto done;
+
+	if (!check_period(oxu, frame, uframe, qh->period, qh->usecs))
+		goto done;
+	if (!qh->c_usecs) {
+		retval = 0;
+		*c_maskp = 0;
+		goto done;
+	}
+
+done:
+	return retval;
+}
+
+/* "first fit" scheduling policy used the first time through,
+ * or when the previous schedule slot can't be re-used.
+ */
+static int qh_schedule(struct oxu_hcd *oxu, struct ehci_qh *qh)
+{
+	int		status;
+	unsigned	uframe;
+	__le32		c_mask;
+	unsigned	frame;		/* 0..(qh->period - 1), or NO_FRAME */
+
+	qh_refresh(oxu, qh);
+	qh->hw_next = EHCI_LIST_END;
+	frame = qh->start;
+
+	/* reuse the previous schedule slots, if we can */
+	if (frame < qh->period) {
+		uframe = ffs(le32_to_cpup(&qh->hw_info2) & QH_SMASK);
+		status = check_intr_schedule(oxu, frame, --uframe,
+				qh, &c_mask);
+	} else {
+		uframe = 0;
+		c_mask = 0;
+		status = -ENOSPC;
+	}
+
+	/* else scan the schedule to find a group of slots such that all
+	 * uframes have enough periodic bandwidth available.
+	 */
+	if (status) {
+		/* "normal" case, uframing flexible except with splits */
+		if (qh->period) {
+			frame = qh->period - 1;
+			do {
+				for (uframe = 0; uframe < 8; uframe++) {
+					status = check_intr_schedule(oxu,
+							frame, uframe, qh,
+							&c_mask);
+					if (status == 0)
+						break;
+				}
+			} while (status && frame--);
+
+		/* qh->period == 0 means every uframe */
+		} else {
+			frame = 0;
+			status = check_intr_schedule(oxu, 0, 0, qh, &c_mask);
+		}
+		if (status)
+			goto done;
+		qh->start = frame;
+
+		/* reset S-frame and (maybe) C-frame masks */
+		qh->hw_info2 &= __constant_cpu_to_le32(~(QH_CMASK | QH_SMASK));
+		qh->hw_info2 |= qh->period
+			? cpu_to_le32(1 << uframe)
+			: __constant_cpu_to_le32(QH_SMASK);
+		qh->hw_info2 |= c_mask;
+	} else
+		oxu_dbg(oxu, "reused qh %p schedule\n", qh);
+
+	/* stuff into the periodic schedule */
+	status = qh_link_periodic(oxu, qh);
+done:
+	return status;
+}
+
+static int intr_submit(struct oxu_hcd *oxu, struct urb *urb,
+			struct list_head *qtd_list, gfp_t mem_flags)
+{
+	unsigned epnum;
+	unsigned long flags;
+	struct ehci_qh *qh;
+	int status = 0;
+	struct list_head	empty;
+
+	/* get endpoint and transfer/schedule data */
+	epnum = urb->ep->desc.bEndpointAddress;
+
+	spin_lock_irqsave(&oxu->lock, flags);
+
+	if (unlikely(!test_bit(HCD_FLAG_HW_ACCESSIBLE,
+			       &oxu_to_hcd(oxu)->flags))) {
+		status = -ESHUTDOWN;
+		goto done;
+	}
+
+	/* get qh and force any scheduling errors */
+	INIT_LIST_HEAD(&empty);
+	qh = qh_append_tds(oxu, urb, &empty, epnum, &urb->ep->hcpriv);
+	if (qh == NULL) {
+		status = -ENOMEM;
+		goto done;
+	}
+	if (qh->qh_state == QH_STATE_IDLE) {
+		status = qh_schedule(oxu, qh);
+		if (status != 0)
+			goto done;
+	}
+
+	/* then queue the urb's tds to the qh */
+	qh = qh_append_tds(oxu, urb, qtd_list, epnum, &urb->ep->hcpriv);
+	BUG_ON(qh == NULL);
+
+	/* ... update usbfs periodic stats */
+	oxu_to_hcd(oxu)->self.bandwidth_int_reqs++;
+
+done:
+	spin_unlock_irqrestore(&oxu->lock, flags);
+	if (status)
+		qtd_list_free(oxu, urb, qtd_list);
+
+	return status;
+}
+
+static inline int itd_submit(struct oxu_hcd *oxu, struct urb *urb,
+						gfp_t mem_flags)
+{
+	oxu_dbg(oxu, "iso support is missing!\n");
+	return -ENOSYS;
+}
+
+static inline int sitd_submit(struct oxu_hcd *oxu, struct urb *urb,
+						gfp_t mem_flags)
+{
+	oxu_dbg(oxu, "split iso support is missing!\n");
+	return -ENOSYS;
+}
+
+static void scan_periodic(struct oxu_hcd *oxu)
+{
+	unsigned frame, clock, now_uframe, mod;
+	unsigned modified;
+
+	mod = oxu->periodic_size << 3;
+
+	/*
+	 * When running, scan from last scan point up to "now"
+	 * else clean up by scanning everything that's left.
+	 * Touches as few pages as possible:  cache-friendly.
+	 */
+	now_uframe = oxu->next_uframe;
+	if (HC_IS_RUNNING(oxu_to_hcd(oxu)->state))
+		clock = readl(&oxu->regs->frame_index);
+	else
+		clock = now_uframe + mod - 1;
+	clock %= mod;
+
+	for (;;) {
+		union ehci_shadow	q, *q_p;
+		__le32			type, *hw_p;
+		unsigned		uframes;
+
+		/* don't scan past the live uframe */
+		frame = now_uframe >> 3;
+		if (frame == (clock >> 3))
+			uframes = now_uframe & 0x07;
+		else {
+			/* safe to scan the whole frame at once */
+			now_uframe |= 0x07;
+			uframes = 8;
+		}
+
+restart:
+		/* scan each element in frame's queue for completions */
+		q_p = &oxu->pshadow[frame];
+		hw_p = &oxu->periodic[frame];
+		q.ptr = q_p->ptr;
+		type = Q_NEXT_TYPE(*hw_p);
+		modified = 0;
+
+		while (q.ptr != NULL) {
+			union ehci_shadow temp;
+			int live;
+
+			live = HC_IS_RUNNING(oxu_to_hcd(oxu)->state);
+			switch (type) {
+			case Q_TYPE_QH:
+				/* handle any completions */
+				temp.qh = qh_get(q.qh);
+				type = Q_NEXT_TYPE(q.qh->hw_next);
+				q = q.qh->qh_next;
+				modified = qh_completions(oxu, temp.qh);
+				if (unlikely(list_empty(&temp.qh->qtd_list)))
+					intr_deschedule(oxu, temp.qh);
+				qh_put(temp.qh);
+				break;
+			default:
+				dbg("corrupt type %d frame %d shadow %p",
+					type, frame, q.ptr);
+				q.ptr = NULL;
+			}
+
+			/* assume completion callbacks modify the queue */
+			if (unlikely(modified))
+				goto restart;
+		}
+
+		/* Stop when we catch up to the HC */
+
+		/* FIXME:  this assumes we won't get lapped when
+		 * latencies climb; that should be rare, but...
+		 * detect it, and just go all the way around.
+		 * FLR might help detect this case, so long as latencies
+		 * don't exceed periodic_size msec (default 1.024 sec).
+		 */
+
+		/* FIXME: likewise assumes HC doesn't halt mid-scan */
+
+		if (now_uframe == clock) {
+			unsigned	now;
+
+			if (!HC_IS_RUNNING(oxu_to_hcd(oxu)->state))
+				break;
+			oxu->next_uframe = now_uframe;
+			now = readl(&oxu->regs->frame_index) % mod;
+			if (now_uframe == now)
+				break;
+
+			/* rescan the rest of this frame, then ... */
+			clock = now;
+		} else {
+			now_uframe++;
+			now_uframe %= mod;
+		}
+	}
+}
+
+/* On some systems, leaving remote wakeup enabled prevents system shutdown.
+ * The firmware seems to think that powering off is a wakeup event!
+ * This routine turns off remote wakeup and everything else, on all ports.
+ */
+static void ehci_turn_off_all_ports(struct oxu_hcd *oxu)
+{
+	int port = HCS_N_PORTS(oxu->hcs_params);
+
+	while (port--)
+		writel(PORT_RWC_BITS, &oxu->regs->port_status[port]);
+}
+
+static void ehci_port_power(struct oxu_hcd *oxu, int is_on)
+{
+	unsigned port;
+
+	if (!HCS_PPC(oxu->hcs_params))
+		return;
+
+	oxu_dbg(oxu, "...power%s ports...\n", is_on ? "up" : "down");
+	for (port = HCS_N_PORTS(oxu->hcs_params); port > 0; )
+		(void) oxu_hub_control(oxu_to_hcd(oxu),
+				is_on ? SetPortFeature : ClearPortFeature,
+				USB_PORT_FEAT_POWER,
+				port--, NULL, 0);
+	msleep(20);
+}
+
+/* Called from some interrupts, timers, and so on.
+ * It calls driver completion functions, after dropping oxu->lock.
+ */
+static void ehci_work(struct oxu_hcd *oxu)
+{
+	timer_action_done(oxu, TIMER_IO_WATCHDOG);
+	if (oxu->reclaim_ready)
+		end_unlink_async(oxu);
+
+	/* another CPU may drop oxu->lock during a schedule scan while
+	 * it reports urb completions.  this flag guards against bogus
+	 * attempts at re-entrant schedule scanning.
+	 */
+	if (oxu->scanning)
+		return;
+	oxu->scanning = 1;
+	scan_async(oxu);
+	if (oxu->next_uframe != -1)
+		scan_periodic(oxu);
+	oxu->scanning = 0;
+
+	/* the IO watchdog guards against hardware or driver bugs that
+	 * misplace IRQs, and should let us run completely without IRQs.
+	 * such lossage has been observed on both VT6202 and VT8235.
+	 */
+	if (HC_IS_RUNNING(oxu_to_hcd(oxu)->state) &&
+			(oxu->async->qh_next.ptr != NULL ||
+			 oxu->periodic_sched != 0))
+		timer_action(oxu, TIMER_IO_WATCHDOG);
+}
+
+static void unlink_async(struct oxu_hcd *oxu, struct ehci_qh *qh)
+{
+	/* if we need to use IAA and it's busy, defer */
+	if (qh->qh_state == QH_STATE_LINKED
+			&& oxu->reclaim
+			&& HC_IS_RUNNING(oxu_to_hcd(oxu)->state)) {
+		struct ehci_qh		*last;
+
+		for (last = oxu->reclaim;
+				last->reclaim;
+				last = last->reclaim)
+			continue;
+		qh->qh_state = QH_STATE_UNLINK_WAIT;
+		last->reclaim = qh;
+
+	/* bypass IAA if the hc can't care */
+	} else if (!HC_IS_RUNNING(oxu_to_hcd(oxu)->state) && oxu->reclaim)
+		end_unlink_async(oxu);
+
+	/* something else might have unlinked the qh by now */
+	if (qh->qh_state == QH_STATE_LINKED)
+		start_unlink_async(oxu, qh);
+}
+
+/*
+ * USB host controller methods
+ */
+
+static irqreturn_t oxu210_hcd_irq(struct usb_hcd *hcd)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+	u32 status, pcd_status = 0;
+	int bh;
+
+	spin_lock(&oxu->lock);
+
+	status = readl(&oxu->regs->status);
+
+	/* e.g. cardbus physical eject */
+	if (status == ~(u32) 0) {
+		oxu_dbg(oxu, "device removed\n");
+		goto dead;
+	}
+
+	status &= INTR_MASK;
+	if (!status) {			/* irq sharing? */
+		spin_unlock(&oxu->lock);
+		return IRQ_NONE;
+	}
+
+	/* clear (just) interrupts */
+	writel(status, &oxu->regs->status);
+	readl(&oxu->regs->command);	/* unblock posted write */
+	bh = 0;
+
+#ifdef OXU_VERBOSE_DEBUG
+	/* unrequested/ignored: Frame List Rollover */
+	dbg_status(oxu, "irq", status);
+#endif
+
+	/* INT, ERR, and IAA interrupt rates can be throttled */
+
+	/* normal [4.15.1.2] or error [4.15.1.1] completion */
+	if (likely((status & (STS_INT|STS_ERR)) != 0))
+		bh = 1;
+
+	/* complete the unlinking of some qh [4.15.2.3] */
+	if (status & STS_IAA) {
+		oxu->reclaim_ready = 1;
+		bh = 1;
+	}
+
+	/* remote wakeup [4.3.1] */
+	if (status & STS_PCD) {
+		unsigned i = HCS_N_PORTS(oxu->hcs_params);
+		pcd_status = status;
+
+		/* resume root hub? */
+		if (!(readl(&oxu->regs->command) & CMD_RUN))
+			usb_hcd_resume_root_hub(hcd);
+
+		while (i--) {
+			int pstatus = readl(&oxu->regs->port_status[i]);
+
+			if (pstatus & PORT_OWNER)
+				continue;
+			if (!(pstatus & PORT_RESUME)
+					|| oxu->reset_done[i] != 0)
+				continue;
+
+			/* start 20 msec resume signaling from this port,
+			 * and make khubd collect PORT_STAT_C_SUSPEND to
+			 * stop that signaling.
+			 */
+			oxu->reset_done[i] = jiffies + msecs_to_jiffies(20);
+			oxu_dbg(oxu, "port %d remote wakeup\n", i + 1);
+			mod_timer(&hcd->rh_timer, oxu->reset_done[i]);
+		}
+	}
+
+	/* PCI errors [4.15.2.4] */
+	if (unlikely((status & STS_FATAL) != 0)) {
+		/* bogus "fatal" IRQs appear on some chips... why?  */
+		status = readl(&oxu->regs->status);
+		dbg_cmd(oxu, "fatal", readl(&oxu->regs->command));
+		dbg_status(oxu, "fatal", status);
+		if (status & STS_HALT) {
+			oxu_err(oxu, "fatal error\n");
+dead:
+			ehci_reset(oxu);
+			writel(0, &oxu->regs->configured_flag);
+			/* generic layer kills/unlinks all urbs, then
+			 * uses oxu_stop to clean up the rest
+			 */
+			bh = 1;
+		}
+	}
+
+	if (bh)
+		ehci_work(oxu);
+	spin_unlock(&oxu->lock);
+	if (pcd_status & STS_PCD)
+		usb_hcd_poll_rh_status(hcd);
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t oxu_irq(struct usb_hcd *hcd)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+	int ret = IRQ_HANDLED;
+
+	u32 status = oxu_readl(hcd->regs, OXU_CHIPIRQSTATUS);
+	u32 enable = oxu_readl(hcd->regs, OXU_CHIPIRQEN_SET);
+
+	/* Disable all interrupt */
+	oxu_writel(hcd->regs, OXU_CHIPIRQEN_CLR, enable);
+
+	if ((oxu->is_otg && (status & OXU_USBOTGI)) ||
+		(!oxu->is_otg && (status & OXU_USBSPHI)))
+		oxu210_hcd_irq(hcd);
+	else
+		ret = IRQ_NONE;
+
+	/* Enable all interrupt back */
+	oxu_writel(hcd->regs, OXU_CHIPIRQEN_SET, enable);
+
+	return ret;
+}
+
+static void oxu_watchdog(unsigned long param)
+{
+	struct oxu_hcd	*oxu = (struct oxu_hcd *) param;
+	unsigned long flags;
+
+	spin_lock_irqsave(&oxu->lock, flags);
+
+	/* lost IAA irqs wedge things badly; seen with a vt8235 */
+	if (oxu->reclaim) {
+		u32 status = readl(&oxu->regs->status);
+		if (status & STS_IAA) {
+			oxu_vdbg(oxu, "lost IAA\n");
+			writel(STS_IAA, &oxu->regs->status);
+			oxu->reclaim_ready = 1;
+		}
+	}
+
+	/* stop async processing after it's idled a bit */
+	if (test_bit(TIMER_ASYNC_OFF, &oxu->actions))
+		start_unlink_async(oxu, oxu->async);
+
+	/* oxu could run by timer, without IRQs ... */
+	ehci_work(oxu);
+
+	spin_unlock_irqrestore(&oxu->lock, flags);
+}
+
+/* One-time init, only for memory state.
+ */
+static int oxu_hcd_init(struct usb_hcd *hcd)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+	u32 temp;
+	int retval;
+	u32 hcc_params;
+
+	spin_lock_init(&oxu->lock);
+
+	init_timer(&oxu->watchdog);
+	oxu->watchdog.function = oxu_watchdog;
+	oxu->watchdog.data = (unsigned long) oxu;
+
+	/*
+	 * hw default: 1K periodic list heads, one per frame.
+	 * periodic_size can shrink by USBCMD update if hcc_params allows.
+	 */
+	oxu->periodic_size = DEFAULT_I_TDPS;
+	retval = ehci_mem_init(oxu, GFP_KERNEL);
+	if (retval < 0)
+		return retval;
+
+	/* controllers may cache some of the periodic schedule ... */
+	hcc_params = readl(&oxu->caps->hcc_params);
+	if (HCC_ISOC_CACHE(hcc_params))		/* full frame cache */
+		oxu->i_thresh = 8;
+	else					/* N microframes cached */
+		oxu->i_thresh = 2 + HCC_ISOC_THRES(hcc_params);
+
+	oxu->reclaim = NULL;
+	oxu->reclaim_ready = 0;
+	oxu->next_uframe = -1;
+
+	/*
+	 * dedicate a qh for the async ring head, since we couldn't unlink
+	 * a 'real' qh without stopping the async schedule [4.8].  use it
+	 * as the 'reclamation list head' too.
+	 * its dummy is used in hw_alt_next of many tds, to prevent the qh
+	 * from automatically advancing to the next td after short reads.
+	 */
+	oxu->async->qh_next.qh = NULL;
+	oxu->async->hw_next = QH_NEXT(oxu->async->qh_dma);
+	oxu->async->hw_info1 = cpu_to_le32(QH_HEAD);
+	oxu->async->hw_token = cpu_to_le32(QTD_STS_HALT);
+	oxu->async->hw_qtd_next = EHCI_LIST_END;
+	oxu->async->qh_state = QH_STATE_LINKED;
+	oxu->async->hw_alt_next = QTD_NEXT(oxu->async->dummy->qtd_dma);
+
+	/* clear interrupt enables, set irq latency */
+	if (log2_irq_thresh < 0 || log2_irq_thresh > 6)
+		log2_irq_thresh = 0;
+	temp = 1 << (16 + log2_irq_thresh);
+	if (HCC_CANPARK(hcc_params)) {
+		/* HW default park == 3, on hardware that supports it (like
+		 * NVidia and ALI silicon), maximizes throughput on the async
+		 * schedule by avoiding QH fetches between transfers.
+		 *
+		 * With fast usb storage devices and NForce2, "park" seems to
+		 * make problems:  throughput reduction (!), data errors...
+		 */
+		if (park) {
+			park = min(park, (unsigned) 3);
+			temp |= CMD_PARK;
+			temp |= park << 8;
+		}
+		oxu_dbg(oxu, "park %d\n", park);
+	}
+	if (HCC_PGM_FRAMELISTLEN(hcc_params)) {
+		/* periodic schedule size can be smaller than default */
+		temp &= ~(3 << 2);
+		temp |= (EHCI_TUNE_FLS << 2);
+	}
+	oxu->command = temp;
+
+	return 0;
+}
+
+/* Called during probe() after chip reset completes.
+ */
+static int oxu_reset(struct usb_hcd *hcd)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+	int ret;
+
+	spin_lock_init(&oxu->mem_lock);
+	INIT_LIST_HEAD(&oxu->urb_list);
+	oxu->urb_len = 0;
+
+	/* FIMXE */
+	hcd->self.controller->dma_mask = 0UL;
+
+	if (oxu->is_otg) {
+		oxu->caps = hcd->regs + OXU_OTG_CAP_OFFSET;
+		oxu->regs = hcd->regs + OXU_OTG_CAP_OFFSET + \
+			HC_LENGTH(readl(&oxu->caps->hc_capbase));
+
+		oxu->mem = hcd->regs + OXU_SPH_MEM;
+	} else {
+		oxu->caps = hcd->regs + OXU_SPH_CAP_OFFSET;
+		oxu->regs = hcd->regs + OXU_SPH_CAP_OFFSET + \
+			HC_LENGTH(readl(&oxu->caps->hc_capbase));
+
+		oxu->mem = hcd->regs + OXU_OTG_MEM;
+	}
+
+	oxu->hcs_params = readl(&oxu->caps->hcs_params);
+	oxu->sbrn = 0x20;
+
+	ret = oxu_hcd_init(hcd);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int oxu_run(struct usb_hcd *hcd)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+	int retval;
+	u32 temp, hcc_params;
+
+	hcd->uses_new_polling = 1;
+	hcd->poll_rh = 0;
+
+	/* EHCI spec section 4.1 */
+	retval = ehci_reset(oxu);
+	if (retval != 0) {
+		ehci_mem_cleanup(oxu);
+		return retval;
+	}
+	writel(oxu->periodic_dma, &oxu->regs->frame_list);
+	writel((u32) oxu->async->qh_dma, &oxu->regs->async_next);
+
+	/* hcc_params controls whether oxu->regs->segment must (!!!)
+	 * be used; it constrains QH/ITD/SITD and QTD locations.
+	 * pci_pool consistent memory always uses segment zero.
+	 * streaming mappings for I/O buffers, like pci_map_single(),
+	 * can return segments above 4GB, if the device allows.
+	 *
+	 * NOTE:  the dma mask is visible through dma_supported(), so
+	 * drivers can pass this info along ... like NETIF_F_HIGHDMA,
+	 * Scsi_Host.highmem_io, and so forth.  It's readonly to all
+	 * host side drivers though.
+	 */
+	hcc_params = readl(&oxu->caps->hcc_params);
+	if (HCC_64BIT_ADDR(hcc_params))
+		writel(0, &oxu->regs->segment);
+
+	oxu->command &= ~(CMD_LRESET | CMD_IAAD | CMD_PSE |
+				CMD_ASE | CMD_RESET);
+	oxu->command |= CMD_RUN;
+	writel(oxu->command, &oxu->regs->command);
+	dbg_cmd(oxu, "init", oxu->command);
+
+	/*
+	 * Start, enabling full USB 2.0 functionality ... usb 1.1 devices
+	 * are explicitly handed to companion controller(s), so no TT is
+	 * involved with the root hub.  (Except where one is integrated,
+	 * and there's no companion controller unless maybe for USB OTG.)
+	 */
+	hcd->state = HC_STATE_RUNNING;
+	writel(FLAG_CF, &oxu->regs->configured_flag);
+	readl(&oxu->regs->command);	/* unblock posted writes */
+
+	temp = HC_VERSION(readl(&oxu->caps->hc_capbase));
+	oxu_info(oxu, "USB %x.%x started, quasi-EHCI %x.%02x, driver %s%s\n",
+		((oxu->sbrn & 0xf0)>>4), (oxu->sbrn & 0x0f),
+		temp >> 8, temp & 0xff, DRIVER_VERSION,
+		ignore_oc ? ", overcurrent ignored" : "");
+
+	writel(INTR_MASK, &oxu->regs->intr_enable); /* Turn On Interrupts */
+
+	return 0;
+}
+
+static void oxu_stop(struct usb_hcd *hcd)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+
+	/* Turn off port power on all root hub ports. */
+	ehci_port_power(oxu, 0);
+
+	/* no more interrupts ... */
+	del_timer_sync(&oxu->watchdog);
+
+	spin_lock_irq(&oxu->lock);
+	if (HC_IS_RUNNING(hcd->state))
+		ehci_quiesce(oxu);
+
+	ehci_reset(oxu);
+	writel(0, &oxu->regs->intr_enable);
+	spin_unlock_irq(&oxu->lock);
+
+	/* let companion controllers work when we aren't */
+	writel(0, &oxu->regs->configured_flag);
+
+	/* root hub is shut down separately (first, when possible) */
+	spin_lock_irq(&oxu->lock);
+	if (oxu->async)
+		ehci_work(oxu);
+	spin_unlock_irq(&oxu->lock);
+	ehci_mem_cleanup(oxu);
+
+	dbg_status(oxu, "oxu_stop completed", readl(&oxu->regs->status));
+}
+
+/* Kick in for silicon on any bus (not just pci, etc).
+ * This forcibly disables dma and IRQs, helping kexec and other cases
+ * where the next system software may expect clean state.
+ */
+static void oxu_shutdown(struct usb_hcd *hcd)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+
+	(void) ehci_halt(oxu);
+	ehci_turn_off_all_ports(oxu);
+
+	/* make BIOS/etc use companion controller during reboot */
+	writel(0, &oxu->regs->configured_flag);
+
+	/* unblock posted writes */
+	readl(&oxu->regs->configured_flag);
+}
+
+/* Non-error returns are a promise to giveback() the urb later
+ * we drop ownership so next owner (or urb unlink) can get it
+ *
+ * urb + dev is in hcd.self.controller.urb_list
+ * we're queueing TDs onto software and hardware lists
+ *
+ * hcd-specific init for hcpriv hasn't been done yet
+ *
+ * NOTE:  control, bulk, and interrupt share the same code to append TDs
+ * to a (possibly active) QH, and the same QH scanning code.
+ */
+static int __oxu_urb_enqueue(struct usb_hcd *hcd, struct urb *urb,
+				gfp_t mem_flags)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+	struct list_head qtd_list;
+
+	INIT_LIST_HEAD(&qtd_list);
+
+	switch (usb_pipetype(urb->pipe)) {
+	case PIPE_CONTROL:
+	case PIPE_BULK:
+	default:
+		if (!qh_urb_transaction(oxu, urb, &qtd_list, mem_flags))
+			return -ENOMEM;
+		return submit_async(oxu, urb, &qtd_list, mem_flags);
+
+	case PIPE_INTERRUPT:
+		if (!qh_urb_transaction(oxu, urb, &qtd_list, mem_flags))
+			return -ENOMEM;
+		return intr_submit(oxu, urb, &qtd_list, mem_flags);
+
+	case PIPE_ISOCHRONOUS:
+		if (urb->dev->speed == USB_SPEED_HIGH)
+			return itd_submit(oxu, urb, mem_flags);
+		else
+			return sitd_submit(oxu, urb, mem_flags);
+	}
+}
+
+/* This function is responsible for breaking URBs with big data size
+ * into smaller size and processing small urbs in sequence.
+ */
+static int oxu_urb_enqueue(struct usb_hcd *hcd, struct urb *urb,
+				gfp_t mem_flags)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+	int num, rem;
+	int transfer_buffer_length;
+	void *transfer_buffer;
+	struct urb *murb;
+	int i, ret;
+
+	/* If not bulk pipe just enqueue the URB */
+	if (!usb_pipebulk(urb->pipe))
+		return __oxu_urb_enqueue(hcd, urb, mem_flags);
+
+	/* Otherwise we should verify the USB transfer buffer size! */
+	transfer_buffer = urb->transfer_buffer;
+	transfer_buffer_length = urb->transfer_buffer_length;
+
+	num = urb->transfer_buffer_length / 4096;
+	rem = urb->transfer_buffer_length % 4096;
+	if (rem != 0)
+		num++;
+
+	/* If URB is smaller than 4096 bytes just enqueue it! */
+	if (num == 1)
+		return __oxu_urb_enqueue(hcd, urb, mem_flags);
+
+	/* Ok, we have more job to do! :) */
+
+	for (i = 0; i < num - 1; i++) {
+		/* Get free micro URB poll till a free urb is recieved */
+
+		do {
+			murb = (struct urb *) oxu_murb_alloc(oxu);
+			if (!murb)
+				schedule();
+		} while (!murb);
+
+		/* Coping the urb */
+		memcpy(murb, urb, sizeof(struct urb));
+
+		murb->transfer_buffer_length = 4096;
+		murb->transfer_buffer = transfer_buffer + i * 4096;
+
+		/* Null pointer for the encodes that this is a micro urb */
+		murb->complete = NULL;
+
+		((struct oxu_murb *) murb)->main = urb;
+		((struct oxu_murb *) murb)->last = 0;
+
+		/* This loop is to guarantee urb to be processed when there's
+		 * not enough resources at a particular time by retrying.
+		 */
+		do {
+			ret  = __oxu_urb_enqueue(hcd, murb, mem_flags);
+			if (ret)
+				schedule();
+		} while (ret);
+	}
+
+	/* Last urb requires special handling  */
+
+	/* Get free micro URB poll till a free urb is recieved */
+	do {
+		murb = (struct urb *) oxu_murb_alloc(oxu);
+		if (!murb)
+			schedule();
+	} while (!murb);
+
+	/* Coping the urb */
+	memcpy(murb, urb, sizeof(struct urb));
+
+	murb->transfer_buffer_length = rem > 0 ? rem : 4096;
+	murb->transfer_buffer = transfer_buffer + (num - 1) * 4096;
+
+	/* Null pointer for the encodes that this is a micro urb */
+	murb->complete = NULL;
+
+	((struct oxu_murb *) murb)->main = urb;
+	((struct oxu_murb *) murb)->last = 1;
+
+	do {
+		ret = __oxu_urb_enqueue(hcd, murb, mem_flags);
+		if (ret)
+			schedule();
+	} while (ret);
+
+	return ret;
+}
+
+/* Remove from hardware lists.
+ * Completions normally happen asynchronously
+ */
+static int oxu_urb_dequeue(struct usb_hcd *hcd, struct urb *urb, int status)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+	struct ehci_qh *qh;
+	unsigned long flags;
+
+	spin_lock_irqsave(&oxu->lock, flags);
+	switch (usb_pipetype(urb->pipe)) {
+	case PIPE_CONTROL:
+	case PIPE_BULK:
+	default:
+		qh = (struct ehci_qh *) urb->hcpriv;
+		if (!qh)
+			break;
+		unlink_async(oxu, qh);
+		break;
+
+	case PIPE_INTERRUPT:
+		qh = (struct ehci_qh *) urb->hcpriv;
+		if (!qh)
+			break;
+		switch (qh->qh_state) {
+		case QH_STATE_LINKED:
+			intr_deschedule(oxu, qh);
+			/* FALL THROUGH */
+		case QH_STATE_IDLE:
+			qh_completions(oxu, qh);
+			break;
+		default:
+			oxu_dbg(oxu, "bogus qh %p state %d\n",
+					qh, qh->qh_state);
+			goto done;
+		}
+
+		/* reschedule QH iff another request is queued */
+		if (!list_empty(&qh->qtd_list)
+				&& HC_IS_RUNNING(hcd->state)) {
+			int status;
+
+			status = qh_schedule(oxu, qh);
+			spin_unlock_irqrestore(&oxu->lock, flags);
+
+			if (status != 0) {
+				/* shouldn't happen often, but ...
+				 * FIXME kill those tds' urbs
+				 */
+				err("can't reschedule qh %p, err %d",
+					qh, status);
+			}
+			return status;
+		}
+		break;
+	}
+done:
+	spin_unlock_irqrestore(&oxu->lock, flags);
+	return 0;
+}
+
+/* Bulk qh holds the data toggle */
+static void oxu_endpoint_disable(struct usb_hcd *hcd,
+					struct usb_host_endpoint *ep)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+	unsigned long		flags;
+	struct ehci_qh		*qh, *tmp;
+
+	/* ASSERT:  any requests/urbs are being unlinked */
+	/* ASSERT:  nobody can be submitting urbs for this any more */
+
+rescan:
+	spin_lock_irqsave(&oxu->lock, flags);
+	qh = ep->hcpriv;
+	if (!qh)
+		goto done;
+
+	/* endpoints can be iso streams.  for now, we don't
+	 * accelerate iso completions ... so spin a while.
+	 */
+	if (qh->hw_info1 == 0) {
+		oxu_vdbg(oxu, "iso delay\n");
+		goto idle_timeout;
+	}
+
+	if (!HC_IS_RUNNING(hcd->state))
+		qh->qh_state = QH_STATE_IDLE;
+	switch (qh->qh_state) {
+	case QH_STATE_LINKED:
+		for (tmp = oxu->async->qh_next.qh;
+				tmp && tmp != qh;
+				tmp = tmp->qh_next.qh)
+			continue;
+		/* periodic qh self-unlinks on empty */
+		if (!tmp)
+			goto nogood;
+		unlink_async(oxu, qh);
+		/* FALL THROUGH */
+	case QH_STATE_UNLINK:		/* wait for hw to finish? */
+idle_timeout:
+		spin_unlock_irqrestore(&oxu->lock, flags);
+		schedule_timeout_uninterruptible(1);
+		goto rescan;
+	case QH_STATE_IDLE:		/* fully unlinked */
+		if (list_empty(&qh->qtd_list)) {
+			qh_put(qh);
+			break;
+		}
+		/* else FALL THROUGH */
+	default:
+nogood:
+		/* caller was supposed to have unlinked any requests;
+		 * that's not our job.  just leak this memory.
+		 */
+		oxu_err(oxu, "qh %p (#%02x) state %d%s\n",
+			qh, ep->desc.bEndpointAddress, qh->qh_state,
+			list_empty(&qh->qtd_list) ? "" : "(has tds)");
+		break;
+	}
+	ep->hcpriv = NULL;
+done:
+	spin_unlock_irqrestore(&oxu->lock, flags);
+	return;
+}
+
+static int oxu_get_frame(struct usb_hcd *hcd)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+
+	return (readl(&oxu->regs->frame_index) >> 3) %
+		oxu->periodic_size;
+}
+
+/* Build "status change" packet (one or two bytes) from HC registers */
+static int oxu_hub_status_data(struct usb_hcd *hcd, char *buf)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+	u32 temp, mask, status = 0;
+	int ports, i, retval = 1;
+	unsigned long flags;
+
+	/* if !USB_SUSPEND, root hub timers won't get shut down ... */
+	if (!HC_IS_RUNNING(hcd->state))
+		return 0;
+
+	/* init status to no-changes */
+	buf[0] = 0;
+	ports = HCS_N_PORTS(oxu->hcs_params);
+	if (ports > 7) {
+		buf[1] = 0;
+		retval++;
+	}
+
+	/* Some boards (mostly VIA?) report bogus overcurrent indications,
+	 * causing massive log spam unless we completely ignore them.  It
+	 * may be relevant that VIA VT8235 controlers, where PORT_POWER is
+	 * always set, seem to clear PORT_OCC and PORT_CSC when writing to
+	 * PORT_POWER; that's surprising, but maybe within-spec.
+	 */
+	if (!ignore_oc)
+		mask = PORT_CSC | PORT_PEC | PORT_OCC;
+	else
+		mask = PORT_CSC | PORT_PEC;
+
+	/* no hub change reports (bit 0) for now (power, ...) */
+
+	/* port N changes (bit N)? */
+	spin_lock_irqsave(&oxu->lock, flags);
+	for (i = 0; i < ports; i++) {
+		temp = readl(&oxu->regs->port_status[i]);
+
+		/*
+		 * Return status information even for ports with OWNER set.
+		 * Otherwise khubd wouldn't see the disconnect event when a
+		 * high-speed device is switched over to the companion
+		 * controller by the user.
+		 */
+
+		if (!(temp & PORT_CONNECT))
+			oxu->reset_done[i] = 0;
+		if ((temp & mask) != 0 || ((temp & PORT_RESUME) != 0 &&
+				time_after_eq(jiffies, oxu->reset_done[i]))) {
+			if (i < 7)
+				buf[0] |= 1 << (i + 1);
+			else
+				buf[1] |= 1 << (i - 7);
+			status = STS_PCD;
+		}
+	}
+	/* FIXME autosuspend idle root hubs */
+	spin_unlock_irqrestore(&oxu->lock, flags);
+	return status ? retval : 0;
+}
+
+/* Returns the speed of a device attached to a port on the root hub. */
+static inline unsigned int oxu_port_speed(struct oxu_hcd *oxu,
+						unsigned int portsc)
+{
+	switch ((portsc >> 26) & 3) {
+	case 0:
+		return 0;
+	case 1:
+		return 1 << USB_PORT_FEAT_LOWSPEED;
+	case 2:
+	default:
+		return 1 << USB_PORT_FEAT_HIGHSPEED;
+	}
+}
+
+#define	PORT_WAKE_BITS	(PORT_WKOC_E|PORT_WKDISC_E|PORT_WKCONN_E)
+static int oxu_hub_control(struct usb_hcd *hcd, u16 typeReq,
+				u16 wValue, u16 wIndex, char *buf, u16 wLength)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+	int ports = HCS_N_PORTS(oxu->hcs_params);
+	u32 __iomem *status_reg = &oxu->regs->port_status[wIndex - 1];
+	u32 temp, status;
+	unsigned long	flags;
+	int retval = 0;
+	unsigned selector;
+
+	/*
+	 * FIXME:  support SetPortFeatures USB_PORT_FEAT_INDICATOR.
+	 * HCS_INDICATOR may say we can change LEDs to off/amber/green.
+	 * (track current state ourselves) ... blink for diagnostics,
+	 * power, "this is the one", etc.  EHCI spec supports this.
+	 */
+
+	spin_lock_irqsave(&oxu->lock, flags);
+	switch (typeReq) {
+	case ClearHubFeature:
+		switch (wValue) {
+		case C_HUB_LOCAL_POWER:
+		case C_HUB_OVER_CURRENT:
+			/* no hub-wide feature/status flags */
+			break;
+		default:
+			goto error;
+		}
+		break;
+	case ClearPortFeature:
+		if (!wIndex || wIndex > ports)
+			goto error;
+		wIndex--;
+		temp = readl(status_reg);
+
+		/*
+		 * Even if OWNER is set, so the port is owned by the
+		 * companion controller, khubd needs to be able to clear
+		 * the port-change status bits (especially
+		 * USB_PORT_FEAT_C_CONNECTION).
+		 */
+
+		switch (wValue) {
+		case USB_PORT_FEAT_ENABLE:
+			writel(temp & ~PORT_PE, status_reg);
+			break;
+		case USB_PORT_FEAT_C_ENABLE:
+			writel((temp & ~PORT_RWC_BITS) | PORT_PEC, status_reg);
+			break;
+		case USB_PORT_FEAT_SUSPEND:
+			if (temp & PORT_RESET)
+				goto error;
+			if (temp & PORT_SUSPEND) {
+				if ((temp & PORT_PE) == 0)
+					goto error;
+				/* resume signaling for 20 msec */
+				temp &= ~(PORT_RWC_BITS | PORT_WAKE_BITS);
+				writel(temp | PORT_RESUME, status_reg);
+				oxu->reset_done[wIndex] = jiffies
+						+ msecs_to_jiffies(20);
+			}
+			break;
+		case USB_PORT_FEAT_C_SUSPEND:
+			/* we auto-clear this feature */
+			break;
+		case USB_PORT_FEAT_POWER:
+			if (HCS_PPC(oxu->hcs_params))
+				writel(temp & ~(PORT_RWC_BITS | PORT_POWER),
+					  status_reg);
+			break;
+		case USB_PORT_FEAT_C_CONNECTION:
+			writel((temp & ~PORT_RWC_BITS) | PORT_CSC, status_reg);
+			break;
+		case USB_PORT_FEAT_C_OVER_CURRENT:
+			writel((temp & ~PORT_RWC_BITS) | PORT_OCC, status_reg);
+			break;
+		case USB_PORT_FEAT_C_RESET:
+			/* GetPortStatus clears reset */
+			break;
+		default:
+			goto error;
+		}
+		readl(&oxu->regs->command);	/* unblock posted write */
+		break;
+	case GetHubDescriptor:
+		ehci_hub_descriptor(oxu, (struct usb_hub_descriptor *)
+			buf);
+		break;
+	case GetHubStatus:
+		/* no hub-wide feature/status flags */
+		memset(buf, 0, 4);
+		break;
+	case GetPortStatus:
+		if (!wIndex || wIndex > ports)
+			goto error;
+		wIndex--;
+		status = 0;
+		temp = readl(status_reg);
+
+		/* wPortChange bits */
+		if (temp & PORT_CSC)
+			status |= 1 << USB_PORT_FEAT_C_CONNECTION;
+		if (temp & PORT_PEC)
+			status |= 1 << USB_PORT_FEAT_C_ENABLE;
+		if ((temp & PORT_OCC) && !ignore_oc)
+			status |= 1 << USB_PORT_FEAT_C_OVER_CURRENT;
+
+		/* whoever resumes must GetPortStatus to complete it!! */
+		if (temp & PORT_RESUME) {
+
+			/* Remote Wakeup received? */
+			if (!oxu->reset_done[wIndex]) {
+				/* resume signaling for 20 msec */
+				oxu->reset_done[wIndex] = jiffies
+						+ msecs_to_jiffies(20);
+				/* check the port again */
+				mod_timer(&oxu_to_hcd(oxu)->rh_timer,
+						oxu->reset_done[wIndex]);
+			}
+
+			/* resume completed? */
+			else if (time_after_eq(jiffies,
+					oxu->reset_done[wIndex])) {
+				status |= 1 << USB_PORT_FEAT_C_SUSPEND;
+				oxu->reset_done[wIndex] = 0;
+
+				/* stop resume signaling */
+				temp = readl(status_reg);
+				writel(temp & ~(PORT_RWC_BITS | PORT_RESUME),
+					status_reg);
+				retval = handshake(oxu, status_reg,
+					   PORT_RESUME, 0, 2000 /* 2msec */);
+				if (retval != 0) {
+					oxu_err(oxu,
+						"port %d resume error %d\n",
+						wIndex + 1, retval);
+					goto error;
+				}
+				temp &= ~(PORT_SUSPEND|PORT_RESUME|(3<<10));
+			}
+		}
+
+		/* whoever resets must GetPortStatus to complete it!! */
+		if ((temp & PORT_RESET)
+				&& time_after_eq(jiffies,
+					oxu->reset_done[wIndex])) {
+			status |= 1 << USB_PORT_FEAT_C_RESET;
+			oxu->reset_done[wIndex] = 0;
+
+			/* force reset to complete */
+			writel(temp & ~(PORT_RWC_BITS | PORT_RESET),
+					status_reg);
+			/* REVISIT:  some hardware needs 550+ usec to clear
+			 * this bit; seems too long to spin routinely...
+			 */
+			retval = handshake(oxu, status_reg,
+					PORT_RESET, 0, 750);
+			if (retval != 0) {
+				oxu_err(oxu, "port %d reset error %d\n",
+					wIndex + 1, retval);
+				goto error;
+			}
+
+			/* see what we found out */
+			temp = check_reset_complete(oxu, wIndex, status_reg,
+					readl(status_reg));
+		}
+
+		/* transfer dedicated ports to the companion hc */
+		if ((temp & PORT_CONNECT) &&
+				test_bit(wIndex, &oxu->companion_ports)) {
+			temp &= ~PORT_RWC_BITS;
+			temp |= PORT_OWNER;
+			writel(temp, status_reg);
+			oxu_dbg(oxu, "port %d --> companion\n", wIndex + 1);
+			temp = readl(status_reg);
+		}
+
+		/*
+		 * Even if OWNER is set, there's no harm letting khubd
+		 * see the wPortStatus values (they should all be 0 except
+		 * for PORT_POWER anyway).
+		 */
+
+		if (temp & PORT_CONNECT) {
+			status |= 1 << USB_PORT_FEAT_CONNECTION;
+			/* status may be from integrated TT */
+			status |= oxu_port_speed(oxu, temp);
+		}
+		if (temp & PORT_PE)
+			status |= 1 << USB_PORT_FEAT_ENABLE;
+		if (temp & (PORT_SUSPEND|PORT_RESUME))
+			status |= 1 << USB_PORT_FEAT_SUSPEND;
+		if (temp & PORT_OC)
+			status |= 1 << USB_PORT_FEAT_OVER_CURRENT;
+		if (temp & PORT_RESET)
+			status |= 1 << USB_PORT_FEAT_RESET;
+		if (temp & PORT_POWER)
+			status |= 1 << USB_PORT_FEAT_POWER;
+
+#ifndef	OXU_VERBOSE_DEBUG
+	if (status & ~0xffff)	/* only if wPortChange is interesting */
+#endif
+		dbg_port(oxu, "GetStatus", wIndex + 1, temp);
+		put_unaligned(cpu_to_le32(status), (__le32 *) buf);
+		break;
+	case SetHubFeature:
+		switch (wValue) {
+		case C_HUB_LOCAL_POWER:
+		case C_HUB_OVER_CURRENT:
+			/* no hub-wide feature/status flags */
+			break;
+		default:
+			goto error;
+		}
+		break;
+	case SetPortFeature:
+		selector = wIndex >> 8;
+		wIndex &= 0xff;
+		if (!wIndex || wIndex > ports)
+			goto error;
+		wIndex--;
+		temp = readl(status_reg);
+		if (temp & PORT_OWNER)
+			break;
+
+		temp &= ~PORT_RWC_BITS;
+		switch (wValue) {
+		case USB_PORT_FEAT_SUSPEND:
+			if ((temp & PORT_PE) == 0
+					|| (temp & PORT_RESET) != 0)
+				goto error;
+			if (device_may_wakeup(&hcd->self.root_hub->dev))
+				temp |= PORT_WAKE_BITS;
+			writel(temp | PORT_SUSPEND, status_reg);
+			break;
+		case USB_PORT_FEAT_POWER:
+			if (HCS_PPC(oxu->hcs_params))
+				writel(temp | PORT_POWER, status_reg);
+			break;
+		case USB_PORT_FEAT_RESET:
+			if (temp & PORT_RESUME)
+				goto error;
+			/* line status bits may report this as low speed,
+			 * which can be fine if this root hub has a
+			 * transaction translator built in.
+			 */
+			oxu_vdbg(oxu, "port %d reset\n", wIndex + 1);
+			temp |= PORT_RESET;
+			temp &= ~PORT_PE;
+
+			/*
+			 * caller must wait, then call GetPortStatus
+			 * usb 2.0 spec says 50 ms resets on root
+			 */
+			oxu->reset_done[wIndex] = jiffies
+					+ msecs_to_jiffies(50);
+			writel(temp, status_reg);
+			break;
+
+		/* For downstream facing ports (these):  one hub port is put
+		 * into test mode according to USB2 11.24.2.13, then the hub
+		 * must be reset (which for root hub now means rmmod+modprobe,
+		 * or else system reboot).  See EHCI 2.3.9 and 4.14 for info
+		 * about the EHCI-specific stuff.
+		 */
+		case USB_PORT_FEAT_TEST:
+			if (!selector || selector > 5)
+				goto error;
+			ehci_quiesce(oxu);
+			ehci_halt(oxu);
+			temp |= selector << 16;
+			writel(temp, status_reg);
+			break;
+
+		default:
+			goto error;
+		}
+		readl(&oxu->regs->command);	/* unblock posted writes */
+		break;
+
+	default:
+error:
+		/* "stall" on error */
+		retval = -EPIPE;
+	}
+	spin_unlock_irqrestore(&oxu->lock, flags);
+	return retval;
+}
+
+#ifdef CONFIG_PM
+
+static int oxu_bus_suspend(struct usb_hcd *hcd)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+	int port;
+	int mask;
+
+	oxu_dbg(oxu, "suspend root hub\n");
+
+	if (time_before(jiffies, oxu->next_statechange))
+		msleep(5);
+
+	port = HCS_N_PORTS(oxu->hcs_params);
+	spin_lock_irq(&oxu->lock);
+
+	/* stop schedules, clean any completed work */
+	if (HC_IS_RUNNING(hcd->state)) {
+		ehci_quiesce(oxu);
+		hcd->state = HC_STATE_QUIESCING;
+	}
+	oxu->command = readl(&oxu->regs->command);
+	if (oxu->reclaim)
+		oxu->reclaim_ready = 1;
+	ehci_work(oxu);
+
+	/* Unlike other USB host controller types, EHCI doesn't have
+	 * any notion of "global" or bus-wide suspend.  The driver has
+	 * to manually suspend all the active unsuspended ports, and
+	 * then manually resume them in the bus_resume() routine.
+	 */
+	oxu->bus_suspended = 0;
+	while (port--) {
+		u32 __iomem *reg = &oxu->regs->port_status[port];
+		u32 t1 = readl(reg) & ~PORT_RWC_BITS;
+		u32 t2 = t1;
+
+		/* keep track of which ports we suspend */
+		if ((t1 & PORT_PE) && !(t1 & PORT_OWNER) &&
+				!(t1 & PORT_SUSPEND)) {
+			t2 |= PORT_SUSPEND;
+			set_bit(port, &oxu->bus_suspended);
+		}
+
+		/* enable remote wakeup on all ports */
+		if (device_may_wakeup(&hcd->self.root_hub->dev))
+			t2 |= PORT_WKOC_E|PORT_WKDISC_E|PORT_WKCONN_E;
+		else
+			t2 &= ~(PORT_WKOC_E|PORT_WKDISC_E|PORT_WKCONN_E);
+
+		if (t1 != t2) {
+			oxu_vdbg(oxu, "port %d, %08x -> %08x\n",
+				port + 1, t1, t2);
+			writel(t2, reg);
+		}
+	}
+
+	/* turn off now-idle HC */
+	del_timer_sync(&oxu->watchdog);
+	ehci_halt(oxu);
+	hcd->state = HC_STATE_SUSPENDED;
+
+	/* allow remote wakeup */
+	mask = INTR_MASK;
+	if (!device_may_wakeup(&hcd->self.root_hub->dev))
+		mask &= ~STS_PCD;
+	writel(mask, &oxu->regs->intr_enable);
+	readl(&oxu->regs->intr_enable);
+
+	oxu->next_statechange = jiffies + msecs_to_jiffies(10);
+	spin_unlock_irq(&oxu->lock);
+	return 0;
+}
+
+/* Caller has locked the root hub, and should reset/reinit on error */
+static int oxu_bus_resume(struct usb_hcd *hcd)
+{
+	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
+	u32 temp;
+	int i;
+
+	if (time_before(jiffies, oxu->next_statechange))
+		msleep(5);
+	spin_lock_irq(&oxu->lock);
+
+	/* Ideally and we've got a real resume here, and no port's power
+	 * was lost.  (For PCI, that means Vaux was maintained.)  But we
+	 * could instead be restoring a swsusp snapshot -- so that BIOS was
+	 * the last user of the controller, not reset/pm hardware keeping
+	 * state we gave to it.
+	 */
+	temp = readl(&oxu->regs->intr_enable);
+	oxu_dbg(oxu, "resume root hub%s\n", temp ? "" : " after power loss");
+
+	/* at least some APM implementations will try to deliver
+	 * IRQs right away, so delay them until we're ready.
+	 */
+	writel(0, &oxu->regs->intr_enable);
+
+	/* re-init operational registers */
+	writel(0, &oxu->regs->segment);
+	writel(oxu->periodic_dma, &oxu->regs->frame_list);
+	writel((u32) oxu->async->qh_dma, &oxu->regs->async_next);
+
+	/* restore CMD_RUN, framelist size, and irq threshold */
+	writel(oxu->command, &oxu->regs->command);
+
+	/* Some controller/firmware combinations need a delay during which
+	 * they set up the port statuses.  See Bugzilla #8190. */
+	mdelay(8);
+
+	/* manually resume the ports we suspended during bus_suspend() */
+	i = HCS_N_PORTS(oxu->hcs_params);
+	while (i--) {
+		temp = readl(&oxu->regs->port_status[i]);
+		temp &= ~(PORT_RWC_BITS
+			| PORT_WKOC_E | PORT_WKDISC_E | PORT_WKCONN_E);
+		if (test_bit(i, &oxu->bus_suspended) && (temp & PORT_SUSPEND)) {
+			oxu->reset_done[i] = jiffies + msecs_to_jiffies(20);
+			temp |= PORT_RESUME;
+		}
+		writel(temp, &oxu->regs->port_status[i]);
+	}
+	i = HCS_N_PORTS(oxu->hcs_params);
+	mdelay(20);
+	while (i--) {
+		temp = readl(&oxu->regs->port_status[i]);
+		if (test_bit(i, &oxu->bus_suspended) && (temp & PORT_SUSPEND)) {
+			temp &= ~(PORT_RWC_BITS | PORT_RESUME);
+			writel(temp, &oxu->regs->port_status[i]);
+			oxu_vdbg(oxu, "resumed port %d\n", i + 1);
+		}
+	}
+	(void) readl(&oxu->regs->command);
+
+	/* maybe re-activate the schedule(s) */
+	temp = 0;
+	if (oxu->async->qh_next.qh)
+		temp |= CMD_ASE;
+	if (oxu->periodic_sched)
+		temp |= CMD_PSE;
+	if (temp) {
+		oxu->command |= temp;
+		writel(oxu->command, &oxu->regs->command);
+	}
+
+	oxu->next_statechange = jiffies + msecs_to_jiffies(5);
+	hcd->state = HC_STATE_RUNNING;
+
+	/* Now we can safely re-enable irqs */
+	writel(INTR_MASK, &oxu->regs->intr_enable);
+
+	spin_unlock_irq(&oxu->lock);
+	return 0;
+}
+
+#else
+
+static int oxu_bus_suspend(struct usb_hcd *hcd)
+{
+	return 0;
+}
+
+static int oxu_bus_resume(struct usb_hcd *hcd)
+{
+	return 0;
+}
+
+#endif	/* CONFIG_PM */
+
+static const struct hc_driver oxu_hc_driver = {
+	.description =		"oxu210hp_hcd",
+	.product_desc =		"oxu210hp HCD",
+	.hcd_priv_size =	sizeof(struct oxu_hcd),
+
+	/*
+	 * Generic hardware linkage
+	 */
+	.irq =			oxu_irq,
+	.flags =		HCD_MEMORY | HCD_USB2,
+
+	/*
+	 * Basic lifecycle operations
+	 */
+	.reset =		oxu_reset,
+	.start =		oxu_run,
+	.stop =			oxu_stop,
+	.shutdown =		oxu_shutdown,
+
+	/*
+	 * Managing i/o requests and associated device resources
+	 */
+	.urb_enqueue =		oxu_urb_enqueue,
+	.urb_dequeue =		oxu_urb_dequeue,
+	.endpoint_disable =	oxu_endpoint_disable,
+
+	/*
+	 * Scheduling support
+	 */
+	.get_frame_number =	oxu_get_frame,
+
+	/*
+	 * Root hub support
+	 */
+	.hub_status_data =	oxu_hub_status_data,
+	.hub_control =		oxu_hub_control,
+	.bus_suspend =		oxu_bus_suspend,
+	.bus_resume =		oxu_bus_resume,
+};
+
+/*
+ * Module stuff
+ */
+
+static void oxu_configuration(struct platform_device *pdev, void *base)
+{
+	u32 tmp;
+
+	/* Initialize top level registers.
+	 * First write ever
+	 */
+	oxu_writel(base, OXU_HOSTIFCONFIG, 0x0000037D);
+	oxu_writel(base, OXU_SOFTRESET, OXU_SRESET);
+	oxu_writel(base, OXU_HOSTIFCONFIG, 0x0000037D);
+
+	tmp = oxu_readl(base, OXU_PIOBURSTREADCTRL);
+	oxu_writel(base, OXU_PIOBURSTREADCTRL, tmp | 0x0040);
+
+	oxu_writel(base, OXU_ASO, OXU_SPHPOEN | OXU_OVRCCURPUPDEN |
+					OXU_COMPARATOR | OXU_ASO_OP);
+
+	tmp = oxu_readl(base, OXU_CLKCTRL_SET);
+	oxu_writel(base, OXU_CLKCTRL_SET, tmp | OXU_SYSCLKEN | OXU_USBOTGCLKEN);
+
+	/* Clear all top interrupt enable */
+	oxu_writel(base, OXU_CHIPIRQEN_CLR, 0xff);
+
+	/* Clear all top interrupt status */
+	oxu_writel(base, OXU_CHIPIRQSTATUS, 0xff);
+
+	/* Enable all needed top interrupt except OTG SPH core */
+	oxu_writel(base, OXU_CHIPIRQEN_SET, OXU_USBSPHLPWUI | OXU_USBOTGLPWUI);
+}
+
+static int oxu_verify_id(struct platform_device *pdev, void *base)
+{
+	u32 id;
+	char *bo[] = {
+		"reserved",
+		"128-pin LQFP",
+		"84-pin TFBGA",
+		"reserved",
+	};
+
+	/* Read controller signature register to find a match */
+	id = oxu_readl(base, OXU_DEVICEID);
+	dev_info(&pdev->dev, "device ID %x\n", id);
+	if ((id & OXU_REV_MASK) != (OXU_REV_2100 << OXU_REV_SHIFT))
+		return -1;
+
+	dev_info(&pdev->dev, "found device %x %s (%04x:%04x)\n",
+		id >> OXU_REV_SHIFT,
+		bo[(id & OXU_BO_MASK) >> OXU_BO_SHIFT],
+		(id & OXU_MAJ_REV_MASK) >> OXU_MAJ_REV_SHIFT,
+		(id & OXU_MIN_REV_MASK) >> OXU_MIN_REV_SHIFT);
+
+	return 0;
+}
+
+static const struct hc_driver oxu_hc_driver;
+static struct usb_hcd *oxu_create(struct platform_device *pdev,
+				unsigned long memstart, unsigned long memlen,
+				void *base, int irq, int otg)
+{
+	struct device *dev = &pdev->dev;
+
+	struct usb_hcd *hcd;
+	struct oxu_hcd *oxu;
+	int ret;
+
+	/* Set endian mode and host mode */
+	oxu_writel(base + (otg ? OXU_OTG_CORE_OFFSET : OXU_SPH_CORE_OFFSET),
+				OXU_USBMODE,
+				OXU_CM_HOST_ONLY | OXU_ES_LITTLE | OXU_VBPS);
+
+	hcd = usb_create_hcd(&oxu_hc_driver, dev,
+				otg ? "oxu210hp_otg" : "oxu210hp_sph");
+	if (!hcd)
+		return ERR_PTR(-ENOMEM);
+
+	hcd->rsrc_start = memstart;
+	hcd->rsrc_len = memlen;
+	hcd->regs = base;
+	hcd->irq = irq;
+	hcd->state = HC_STATE_HALT;
+
+	oxu = hcd_to_oxu(hcd);
+	oxu->is_otg = otg;
+
+	ret = usb_add_hcd(hcd, irq, IRQF_SHARED);
+	if (ret < 0)
+		return ERR_PTR(ret);
+
+	return hcd;
+}
+
+static int oxu_init(struct platform_device *pdev,
+				unsigned long memstart, unsigned long memlen,
+				void *base, int irq)
+{
+	struct oxu_info *info = platform_get_drvdata(pdev);
+	struct usb_hcd *hcd;
+	int ret;
+
+	/* First time configuration at start up */
+	oxu_configuration(pdev, base);
+
+	ret = oxu_verify_id(pdev, base);
+	if (ret) {
+		dev_err(&pdev->dev, "no devices found!\n");
+		return -ENODEV;
+	}
+
+	/* Create the OTG controller */
+	hcd = oxu_create(pdev, memstart, memlen, base, irq, 1);
+	if (IS_ERR(hcd)) {
+		dev_err(&pdev->dev, "cannot create OTG controller!\n");
+		ret = PTR_ERR(hcd);
+		goto error_create_otg;
+	}
+	info->hcd[0] = hcd;
+
+	/* Create the SPH host controller */
+	hcd = oxu_create(pdev, memstart, memlen, base, irq, 0);
+	if (IS_ERR(hcd)) {
+		dev_err(&pdev->dev, "cannot create SPH controller!\n");
+		ret = PTR_ERR(hcd);
+		goto error_create_sph;
+	}
+	info->hcd[1] = hcd;
+
+	oxu_writel(base, OXU_CHIPIRQEN_SET,
+		oxu_readl(base, OXU_CHIPIRQEN_SET) | 3);
+
+	return 0;
+
+error_create_sph:
+	usb_remove_hcd(info->hcd[0]);
+	usb_put_hcd(info->hcd[0]);
+
+error_create_otg:
+	return ret;
+}
+
+static int oxu_drv_probe(struct platform_device *pdev)
+{
+	struct resource *res;
+	void *base;
+	unsigned long memstart, memlen;
+	int irq, ret;
+	struct oxu_info *info;
+
+	if (usb_disabled())
+		return -ENODEV;
+
+	/*
+	 * Get the platform resources
+	 */
+	res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	if (!res) {
+		dev_err(&pdev->dev,
+			"no IRQ! Check %s setup!\n", pdev->dev.bus_id);
+		return -ENODEV;
+	}
+	irq = res->start;
+	dev_dbg(&pdev->dev, "IRQ resource %d\n", irq);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res) {
+		dev_err(&pdev->dev, "no registers address! Check %s setup!\n",
+			pdev->dev.bus_id);
+		return -ENODEV;
+	}
+	memstart = res->start;
+	memlen = res->end - res->start + 1;
+	dev_dbg(&pdev->dev, "MEM resource %lx-%lx\n", memstart, memlen);
+	if (!request_mem_region(memstart, memlen,
+				oxu_hc_driver.description)) {
+		dev_dbg(&pdev->dev, "memory area already in use\n");
+		return -EBUSY;
+	}
+
+	ret = set_irq_type(irq, IRQF_TRIGGER_FALLING);
+	if (ret) {
+		dev_err(&pdev->dev, "error setting irq type\n");
+		ret = -EFAULT;
+		goto error_set_irq_type;
+	}
+
+	base = ioremap(memstart, memlen);
+	if (!base) {
+		dev_dbg(&pdev->dev, "error mapping memory\n");
+		ret = -EFAULT;
+		goto error_ioremap;
+	}
+
+	/* Allocate a driver data struct to hold useful info for both
+	 * SPH & OTG devices
+	 */
+	info = kzalloc(sizeof(struct oxu_info), GFP_KERNEL);
+	if (!info) {
+		dev_dbg(&pdev->dev, "error allocating memory\n");
+		ret = -EFAULT;
+		goto error_alloc;
+	}
+	platform_set_drvdata(pdev, info);
+
+	ret = oxu_init(pdev, memstart, memlen, base, irq);
+	if (ret < 0) {
+		dev_dbg(&pdev->dev, "cannot init USB devices\n");
+		goto error_init;
+	}
+
+	dev_info(&pdev->dev, "devices enabled and running\n");
+	platform_set_drvdata(pdev, info);
+
+	return 0;
+
+error_init:
+	kfree(info);
+	platform_set_drvdata(pdev, NULL);
+
+error_alloc:
+	iounmap(base);
+
+error_set_irq_type:
+error_ioremap:
+	release_mem_region(memstart, memlen);
+
+	dev_err(&pdev->dev, "init %s fail, %d\n", pdev->dev.bus_id, ret);
+	return ret;
+}
+
+static void oxu_remove(struct platform_device *pdev, struct usb_hcd *hcd)
+{
+	usb_remove_hcd(hcd);
+	usb_put_hcd(hcd);
+}
+
+static int oxu_drv_remove(struct platform_device *pdev)
+{
+	struct oxu_info *info = platform_get_drvdata(pdev);
+	unsigned long memstart = info->hcd[0]->rsrc_start,
+			memlen = info->hcd[0]->rsrc_len;
+	void *base = info->hcd[0]->regs;
+
+	oxu_remove(pdev, info->hcd[0]);
+	oxu_remove(pdev, info->hcd[1]);
+
+	iounmap(base);
+	release_mem_region(memstart, memlen);
+
+	kfree(info);
+	platform_set_drvdata(pdev, NULL);
+
+	return 0;
+}
+
+static void oxu_drv_shutdown(struct platform_device *pdev)
+{
+	oxu_drv_remove(pdev);
+}
+
+#if 0
+/* FIXME: TODO */
+static int oxu_drv_suspend(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct usb_hcd *hcd = dev_get_drvdata(dev);
+
+	return 0;
+}
+
+static int oxu_drv_resume(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct usb_hcd *hcd = dev_get_drvdata(dev);
+
+	return 0;
+}
+#else
+#define oxu_drv_suspend	NULL
+#define oxu_drv_resume	NULL
+#endif
+
+static struct platform_driver oxu_driver = {
+	.probe		= oxu_drv_probe,
+	.remove		= oxu_drv_remove,
+	.shutdown	= oxu_drv_shutdown,
+	.suspend	= oxu_drv_suspend,
+	.resume		= oxu_drv_resume,
+	.driver = {
+		.name = "oxu210hp-hcd",
+		.bus = &platform_bus_type
+	}
+};
+
+static int __init oxu_module_init(void)
+{
+	int retval = 0;
+
+	retval = platform_driver_register(&oxu_driver);
+	if (retval < 0)
+		return retval;
+
+	return retval;
+}
+
+static void __exit oxu_module_cleanup(void)
+{
+	platform_driver_unregister(&oxu_driver);
+}
+
+module_init(oxu_module_init);
+module_exit(oxu_module_cleanup);
+
+MODULE_DESCRIPTION("Oxford OXU210HP HCD driver - ver. " DRIVER_VERSION);
+MODULE_AUTHOR("Rodolfo Giometti <giometti@linux.it>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/usb/host/oxu210hp.h b/drivers/usb/host/oxu210hp.h
new file mode 100644
index 000000000000..8910e271cc7d
--- /dev/null
+++ b/drivers/usb/host/oxu210hp.h
@@ -0,0 +1,447 @@
+/*
+ * Host interface registers
+ */
+
+#define OXU_DEVICEID			0x00
+	#define OXU_REV_MASK		0xffff0000
+	#define OXU_REV_SHIFT		16
+	#define OXU_REV_2100		0x2100
+	#define OXU_BO_SHIFT		8
+	#define OXU_BO_MASK		(0x3 << OXU_BO_SHIFT)
+	#define OXU_MAJ_REV_SHIFT	4
+	#define OXU_MAJ_REV_MASK	(0xf << OXU_MAJ_REV_SHIFT)
+	#define OXU_MIN_REV_SHIFT	0
+	#define OXU_MIN_REV_MASK	(0xf << OXU_MIN_REV_SHIFT)
+#define OXU_HOSTIFCONFIG		0x04
+#define OXU_SOFTRESET			0x08
+	#define OXU_SRESET		(1 << 0)
+
+#define OXU_PIOBURSTREADCTRL		0x0C
+
+#define OXU_CHIPIRQSTATUS		0x10
+#define OXU_CHIPIRQEN_SET		0x14
+#define OXU_CHIPIRQEN_CLR		0x18
+	#define OXU_USBSPHLPWUI		0x00000080
+	#define OXU_USBOTGLPWUI		0x00000040
+	#define OXU_USBSPHI		0x00000002
+	#define OXU_USBOTGI		0x00000001
+
+#define OXU_CLKCTRL_SET			0x1C
+	#define OXU_SYSCLKEN		0x00000008
+	#define OXU_USBSPHCLKEN		0x00000002
+	#define OXU_USBOTGCLKEN		0x00000001
+
+#define OXU_ASO				0x68
+	#define OXU_SPHPOEN		0x00000100
+	#define OXU_OVRCCURPUPDEN	0x00000800
+	#define OXU_ASO_OP		(1 << 10)
+	#define OXU_COMPARATOR		0x000004000
+
+#define OXU_USBMODE			0x1A8
+	#define OXU_VBPS		0x00000020
+	#define OXU_ES_LITTLE		0x00000000
+	#define OXU_CM_HOST_ONLY	0x00000003
+
+/*
+ * Proper EHCI structs & defines
+ */
+
+/* Magic numbers that can affect system performance */
+#define EHCI_TUNE_CERR		3	/* 0-3 qtd retries; 0 == don't stop */
+#define EHCI_TUNE_RL_HS		4	/* nak throttle; see 4.9 */
+#define EHCI_TUNE_RL_TT		0
+#define EHCI_TUNE_MULT_HS	1	/* 1-3 transactions/uframe; 4.10.3 */
+#define EHCI_TUNE_MULT_TT	1
+#define EHCI_TUNE_FLS		2	/* (small) 256 frame schedule */
+
+struct oxu_hcd;
+
+/* EHCI register interface, corresponds to EHCI Revision 0.95 specification */
+
+/* Section 2.2 Host Controller Capability Registers */
+struct ehci_caps {
+	/* these fields are specified as 8 and 16 bit registers,
+	 * but some hosts can't perform 8 or 16 bit PCI accesses.
+	 */
+	u32		hc_capbase;
+#define HC_LENGTH(p)		(((p)>>00)&0x00ff)	/* bits 7:0 */
+#define HC_VERSION(p)		(((p)>>16)&0xffff)	/* bits 31:16 */
+	u32		hcs_params;     /* HCSPARAMS - offset 0x4 */
+#define HCS_DEBUG_PORT(p)	(((p)>>20)&0xf)	/* bits 23:20, debug port? */
+#define HCS_INDICATOR(p)	((p)&(1 << 16))	/* true: has port indicators */
+#define HCS_N_CC(p)		(((p)>>12)&0xf)	/* bits 15:12, #companion HCs */
+#define HCS_N_PCC(p)		(((p)>>8)&0xf)	/* bits 11:8, ports per CC */
+#define HCS_PORTROUTED(p)	((p)&(1 << 7))	/* true: port routing */
+#define HCS_PPC(p)		((p)&(1 << 4))	/* true: port power control */
+#define HCS_N_PORTS(p)		(((p)>>0)&0xf)	/* bits 3:0, ports on HC */
+
+	u32		hcc_params;      /* HCCPARAMS - offset 0x8 */
+#define HCC_EXT_CAPS(p)		(((p)>>8)&0xff)	/* for pci extended caps */
+#define HCC_ISOC_CACHE(p)       ((p)&(1 << 7))  /* true: can cache isoc frame */
+#define HCC_ISOC_THRES(p)       (((p)>>4)&0x7)  /* bits 6:4, uframes cached */
+#define HCC_CANPARK(p)		((p)&(1 << 2))  /* true: can park on async qh */
+#define HCC_PGM_FRAMELISTLEN(p) ((p)&(1 << 1))  /* true: periodic_size changes*/
+#define HCC_64BIT_ADDR(p)       ((p)&(1))       /* true: can use 64-bit addr */
+	u8		portroute[8];	 /* nibbles for routing - offset 0xC */
+} __attribute__ ((packed));
+
+
+/* Section 2.3 Host Controller Operational Registers */
+struct ehci_regs {
+	/* USBCMD: offset 0x00 */
+	u32		command;
+/* 23:16 is r/w intr rate, in microframes; default "8" == 1/msec */
+#define CMD_PARK	(1<<11)		/* enable "park" on async qh */
+#define CMD_PARK_CNT(c)	(((c)>>8)&3)	/* how many transfers to park for */
+#define CMD_LRESET	(1<<7)		/* partial reset (no ports, etc) */
+#define CMD_IAAD	(1<<6)		/* "doorbell" interrupt async advance */
+#define CMD_ASE		(1<<5)		/* async schedule enable */
+#define CMD_PSE		(1<<4)		/* periodic schedule enable */
+/* 3:2 is periodic frame list size */
+#define CMD_RESET	(1<<1)		/* reset HC not bus */
+#define CMD_RUN		(1<<0)		/* start/stop HC */
+
+	/* USBSTS: offset 0x04 */
+	u32		status;
+#define STS_ASS		(1<<15)		/* Async Schedule Status */
+#define STS_PSS		(1<<14)		/* Periodic Schedule Status */
+#define STS_RECL	(1<<13)		/* Reclamation */
+#define STS_HALT	(1<<12)		/* Not running (any reason) */
+/* some bits reserved */
+	/* these STS_* flags are also intr_enable bits (USBINTR) */
+#define STS_IAA		(1<<5)		/* Interrupted on async advance */
+#define STS_FATAL	(1<<4)		/* such as some PCI access errors */
+#define STS_FLR		(1<<3)		/* frame list rolled over */
+#define STS_PCD		(1<<2)		/* port change detect */
+#define STS_ERR		(1<<1)		/* "error" completion (overflow, ...) */
+#define STS_INT		(1<<0)		/* "normal" completion (short, ...) */
+
+#define INTR_MASK (STS_IAA | STS_FATAL | STS_PCD | STS_ERR | STS_INT)
+
+	/* USBINTR: offset 0x08 */
+	u32		intr_enable;
+
+	/* FRINDEX: offset 0x0C */
+	u32		frame_index;	/* current microframe number */
+	/* CTRLDSSEGMENT: offset 0x10 */
+	u32		segment;	/* address bits 63:32 if needed */
+	/* PERIODICLISTBASE: offset 0x14 */
+	u32		frame_list;	/* points to periodic list */
+	/* ASYNCLISTADDR: offset 0x18 */
+	u32		async_next;	/* address of next async queue head */
+
+	u32		reserved[9];
+
+	/* CONFIGFLAG: offset 0x40 */
+	u32		configured_flag;
+#define FLAG_CF		(1<<0)		/* true: we'll support "high speed" */
+
+	/* PORTSC: offset 0x44 */
+	u32		port_status[0];	/* up to N_PORTS */
+/* 31:23 reserved */
+#define PORT_WKOC_E	(1<<22)		/* wake on overcurrent (enable) */
+#define PORT_WKDISC_E	(1<<21)		/* wake on disconnect (enable) */
+#define PORT_WKCONN_E	(1<<20)		/* wake on connect (enable) */
+/* 19:16 for port testing */
+#define PORT_LED_OFF	(0<<14)
+#define PORT_LED_AMBER	(1<<14)
+#define PORT_LED_GREEN	(2<<14)
+#define PORT_LED_MASK	(3<<14)
+#define PORT_OWNER	(1<<13)		/* true: companion hc owns this port */
+#define PORT_POWER	(1<<12)		/* true: has power (see PPC) */
+#define PORT_USB11(x) (((x)&(3<<10)) == (1<<10))	/* USB 1.1 device */
+/* 11:10 for detecting lowspeed devices (reset vs release ownership) */
+/* 9 reserved */
+#define PORT_RESET	(1<<8)		/* reset port */
+#define PORT_SUSPEND	(1<<7)		/* suspend port */
+#define PORT_RESUME	(1<<6)		/* resume it */
+#define PORT_OCC	(1<<5)		/* over current change */
+#define PORT_OC		(1<<4)		/* over current active */
+#define PORT_PEC	(1<<3)		/* port enable change */
+#define PORT_PE		(1<<2)		/* port enable */
+#define PORT_CSC	(1<<1)		/* connect status change */
+#define PORT_CONNECT	(1<<0)		/* device connected */
+#define PORT_RWC_BITS   (PORT_CSC | PORT_PEC | PORT_OCC)
+} __attribute__ ((packed));
+
+/* Appendix C, Debug port ... intended for use with special "debug devices"
+ * that can help if there's no serial console.  (nonstandard enumeration.)
+ */
+struct ehci_dbg_port {
+	u32	control;
+#define DBGP_OWNER	(1<<30)
+#define DBGP_ENABLED	(1<<28)
+#define DBGP_DONE	(1<<16)
+#define DBGP_INUSE	(1<<10)
+#define DBGP_ERRCODE(x)	(((x)>>7)&0x07)
+#	define DBGP_ERR_BAD	1
+#	define DBGP_ERR_SIGNAL	2
+#define DBGP_ERROR	(1<<6)
+#define DBGP_GO		(1<<5)
+#define DBGP_OUT	(1<<4)
+#define DBGP_LEN(x)	(((x)>>0)&0x0f)
+	u32	pids;
+#define DBGP_PID_GET(x)		(((x)>>16)&0xff)
+#define DBGP_PID_SET(data, tok)	(((data)<<8)|(tok))
+	u32	data03;
+	u32	data47;
+	u32	address;
+#define DBGP_EPADDR(dev, ep)	(((dev)<<8)|(ep))
+} __attribute__ ((packed));
+
+
+#define	QTD_NEXT(dma)	cpu_to_le32((u32)dma)
+
+/*
+ * EHCI Specification 0.95 Section 3.5
+ * QTD: describe data transfer components (buffer, direction, ...)
+ * See Fig 3-6 "Queue Element Transfer Descriptor Block Diagram".
+ *
+ * These are associated only with "QH" (Queue Head) structures,
+ * used with control, bulk, and interrupt transfers.
+ */
+struct ehci_qtd {
+	/* first part defined by EHCI spec */
+	__le32			hw_next;		/* see EHCI 3.5.1 */
+	__le32			hw_alt_next;		/* see EHCI 3.5.2 */
+	__le32			hw_token;		/* see EHCI 3.5.3 */
+#define	QTD_TOGGLE	(1 << 31)	/* data toggle */
+#define	QTD_LENGTH(tok)	(((tok)>>16) & 0x7fff)
+#define	QTD_IOC		(1 << 15)	/* interrupt on complete */
+#define	QTD_CERR(tok)	(((tok)>>10) & 0x3)
+#define	QTD_PID(tok)	(((tok)>>8) & 0x3)
+#define	QTD_STS_ACTIVE	(1 << 7)	/* HC may execute this */
+#define	QTD_STS_HALT	(1 << 6)	/* halted on error */
+#define	QTD_STS_DBE	(1 << 5)	/* data buffer error (in HC) */
+#define	QTD_STS_BABBLE	(1 << 4)	/* device was babbling (qtd halted) */
+#define	QTD_STS_XACT	(1 << 3)	/* device gave illegal response */
+#define	QTD_STS_MMF	(1 << 2)	/* incomplete split transaction */
+#define	QTD_STS_STS	(1 << 1)	/* split transaction state */
+#define	QTD_STS_PING	(1 << 0)	/* issue PING? */
+	__le32			hw_buf[5];		/* see EHCI 3.5.4 */
+	__le32			hw_buf_hi[5];		/* Appendix B */
+
+	/* the rest is HCD-private */
+	dma_addr_t		qtd_dma;		/* qtd address */
+	struct list_head	qtd_list;		/* sw qtd list */
+	struct urb		*urb;			/* qtd's urb */
+	size_t			length;			/* length of buffer */
+
+	u32			qtd_buffer_len;
+	void			*buffer;
+	dma_addr_t		buffer_dma;
+	void			*transfer_buffer;
+	void			*transfer_dma;
+} __attribute__ ((aligned(32)));
+
+/* mask NakCnt+T in qh->hw_alt_next */
+#define QTD_MASK __constant_cpu_to_le32 (~0x1f)
+
+#define IS_SHORT_READ(token) (QTD_LENGTH(token) != 0 && QTD_PID(token) == 1)
+
+/* Type tag from {qh, itd, sitd, fstn}->hw_next */
+#define Q_NEXT_TYPE(dma) ((dma) & __constant_cpu_to_le32 (3 << 1))
+
+/* values for that type tag */
+#define Q_TYPE_QH	__constant_cpu_to_le32 (1 << 1)
+
+/* next async queue entry, or pointer to interrupt/periodic QH */
+#define	QH_NEXT(dma)	(cpu_to_le32(((u32)dma)&~0x01f)|Q_TYPE_QH)
+
+/* for periodic/async schedules and qtd lists, mark end of list */
+#define	EHCI_LIST_END	__constant_cpu_to_le32(1) /* "null pointer" to hw */
+
+/*
+ * Entries in periodic shadow table are pointers to one of four kinds
+ * of data structure.  That's dictated by the hardware; a type tag is
+ * encoded in the low bits of the hardware's periodic schedule.  Use
+ * Q_NEXT_TYPE to get the tag.
+ *
+ * For entries in the async schedule, the type tag always says "qh".
+ */
+union ehci_shadow {
+	struct ehci_qh		*qh;		/* Q_TYPE_QH */
+	__le32			*hw_next;	/* (all types) */
+	void			*ptr;
+};
+
+/*
+ * EHCI Specification 0.95 Section 3.6
+ * QH: describes control/bulk/interrupt endpoints
+ * See Fig 3-7 "Queue Head Structure Layout".
+ *
+ * These appear in both the async and (for interrupt) periodic schedules.
+ */
+
+struct ehci_qh {
+	/* first part defined by EHCI spec */
+	__le32			hw_next;	 /* see EHCI 3.6.1 */
+	__le32			hw_info1;	/* see EHCI 3.6.2 */
+#define	QH_HEAD		0x00008000
+	__le32			hw_info2;	/* see EHCI 3.6.2 */
+#define	QH_SMASK	0x000000ff
+#define	QH_CMASK	0x0000ff00
+#define	QH_HUBADDR	0x007f0000
+#define	QH_HUBPORT	0x3f800000
+#define	QH_MULT		0xc0000000
+	__le32			hw_current;	 /* qtd list - see EHCI 3.6.4 */
+
+	/* qtd overlay (hardware parts of a struct ehci_qtd) */
+	__le32			hw_qtd_next;
+	__le32			hw_alt_next;
+	__le32			hw_token;
+	__le32			hw_buf[5];
+	__le32			hw_buf_hi[5];
+
+	/* the rest is HCD-private */
+	dma_addr_t		qh_dma;		/* address of qh */
+	union ehci_shadow	qh_next;	/* ptr to qh; or periodic */
+	struct list_head	qtd_list;	/* sw qtd list */
+	struct ehci_qtd		*dummy;
+	struct ehci_qh		*reclaim;	/* next to reclaim */
+
+	struct oxu_hcd		*oxu;
+	struct kref		kref;
+	unsigned		stamp;
+
+	u8			qh_state;
+#define	QH_STATE_LINKED		1		/* HC sees this */
+#define	QH_STATE_UNLINK		2		/* HC may still see this */
+#define	QH_STATE_IDLE		3		/* HC doesn't see this */
+#define	QH_STATE_UNLINK_WAIT	4		/* LINKED and on reclaim q */
+#define	QH_STATE_COMPLETING	5		/* don't touch token.HALT */
+
+	/* periodic schedule info */
+	u8			usecs;		/* intr bandwidth */
+	u8			gap_uf;		/* uframes split/csplit gap */
+	u8			c_usecs;	/* ... split completion bw */
+	u16			tt_usecs;	/* tt downstream bandwidth */
+	unsigned short		period;		/* polling interval */
+	unsigned short		start;		/* where polling starts */
+#define NO_FRAME ((unsigned short)~0)			/* pick new start */
+	struct usb_device	*dev;		/* access to TT */
+} __attribute__ ((aligned(32)));
+
+/*
+ * Proper OXU210HP structs
+ */
+
+#define OXU_OTG_CORE_OFFSET	0x00400
+#define OXU_OTG_CAP_OFFSET	(OXU_OTG_CORE_OFFSET + 0x100)
+#define OXU_SPH_CORE_OFFSET	0x00800
+#define OXU_SPH_CAP_OFFSET	(OXU_SPH_CORE_OFFSET + 0x100)
+
+#define OXU_OTG_MEM		0xE000
+#define OXU_SPH_MEM		0x16000
+
+/* Only how many elements & element structure are specifies here. */
+/* 2 host controllers are enabled - total size <= 28 kbytes */
+#define	DEFAULT_I_TDPS		1024
+#define QHEAD_NUM		16
+#define QTD_NUM			32
+#define SITD_NUM		8
+#define MURB_NUM		8
+
+#define BUFFER_NUM		8
+#define BUFFER_SIZE		512
+
+struct oxu_info {
+	struct usb_hcd *hcd[2];
+};
+
+struct oxu_buf {
+	u8			buffer[BUFFER_SIZE];
+} __attribute__ ((aligned(BUFFER_SIZE)));
+
+struct oxu_onchip_mem {
+	struct oxu_buf		db_pool[BUFFER_NUM];
+
+	u32			frame_list[DEFAULT_I_TDPS];
+	struct ehci_qh		qh_pool[QHEAD_NUM];
+	struct ehci_qtd		qtd_pool[QTD_NUM];
+} __attribute__ ((aligned(4 << 10)));
+
+#define	EHCI_MAX_ROOT_PORTS	15		/* see HCS_N_PORTS */
+
+struct oxu_murb {
+	struct urb		urb;
+	struct urb		*main;
+	u8			last;
+};
+
+struct oxu_hcd {				/* one per controller */
+	unsigned int		is_otg:1;
+
+	u8			qh_used[QHEAD_NUM];
+	u8			qtd_used[QTD_NUM];
+	u8			db_used[BUFFER_NUM];
+	u8			murb_used[MURB_NUM];
+
+	struct oxu_onchip_mem	__iomem *mem;
+	spinlock_t		mem_lock;
+
+	struct timer_list	urb_timer;
+
+	struct ehci_caps __iomem *caps;
+	struct ehci_regs __iomem *regs;
+
+	__u32			hcs_params;	/* cached register copy */
+	spinlock_t		lock;
+
+	/* async schedule support */
+	struct ehci_qh		*async;
+	struct ehci_qh		*reclaim;
+	unsigned		reclaim_ready:1;
+	unsigned		scanning:1;
+
+	/* periodic schedule support */
+	unsigned		periodic_size;
+	__le32			*periodic;	/* hw periodic table */
+	dma_addr_t		periodic_dma;
+	unsigned		i_thresh;	/* uframes HC might cache */
+
+	union ehci_shadow	*pshadow;	/* mirror hw periodic table */
+	int			next_uframe;	/* scan periodic, start here */
+	unsigned		periodic_sched;	/* periodic activity count */
+
+	/* per root hub port */
+	unsigned long		reset_done[EHCI_MAX_ROOT_PORTS];
+	/* bit vectors (one bit per port) */
+	unsigned long		bus_suspended;	/* which ports were
+						 * already suspended at the
+						 * start of a bus suspend
+						 */
+	unsigned long		companion_ports;/* which ports are dedicated
+						 * to the companion controller
+						 */
+
+	struct timer_list	watchdog;
+	unsigned long		actions;
+	unsigned		stamp;
+	unsigned long		next_statechange;
+	u32			command;
+
+	/* SILICON QUIRKS */
+	struct list_head	urb_list;	/* this is the head to urb
+						 * queue that didn't get enough
+						 * resources
+						 */
+	struct oxu_murb		*murb_pool;	/* murb per split big urb */
+	unsigned urb_len;
+
+	u8			sbrn;		/* packed release number */
+};
+
+#define EHCI_IAA_JIFFIES	(HZ/100)	/* arbitrary; ~10 msec */
+#define EHCI_IO_JIFFIES	 	(HZ/10)		/* io watchdog > irq_thresh */
+#define EHCI_ASYNC_JIFFIES      (HZ/20)		/* async idle timeout */
+#define EHCI_SHRINK_JIFFIES     (HZ/200)	/* async qh unlink delay */
+
+enum ehci_timer_action {
+	TIMER_IO_WATCHDOG,
+	TIMER_IAA_WATCHDOG,
+	TIMER_ASYNC_SHRINK,
+	TIMER_ASYNC_OFF,
+};
+
+#include <linux/oxu210hp.h>
diff --git a/include/linux/oxu210hp.h b/include/linux/oxu210hp.h
new file mode 100644
index 000000000000..0bf96eae5389
--- /dev/null
+++ b/include/linux/oxu210hp.h
@@ -0,0 +1,7 @@
+/* platform data for the OXU210HP HCD */
+
+struct oxu210hp_platform_data {
+	unsigned int bus16:1;
+	unsigned int use_hcd_otg:1;
+	unsigned int use_hcd_sph:1;
+};
-- 
cgit v1.2.3


From d767d888750a8e15656b7ee15d68f90a151b8936 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Thu, 6 Nov 2008 22:32:15 -0800
Subject: USB: wusb: annotate association types withe proper endianness

Also a trivial annotation in rh.c for:
drivers/usb/wusbcore/rh.c:366:9: warning: incorrect type in assignment (different base types)
drivers/usb/wusbcore/rh.c:366:9:    expected unsigned short [unsigned] [short] [usertype] <noident>
drivers/usb/wusbcore/rh.c:366:9:    got restricted __le16 [usertype] <noident>
drivers/usb/wusbcore/rh.c:367:9: warning: incorrect type in assignment (different base types)
drivers/usb/wusbcore/rh.c:367:9:    expected unsigned short [unsigned] [short] [usertype] <noident>
drivers/usb/wusbcore/rh.c:367:9:    got restricted __le16 [usertype] <noident>

Association types annotation fixes piles of warnings similar to:
drivers/usb/wusbcore/cbaf.c:238:30: warning: incorrect type in initializer (different base types)
drivers/usb/wusbcore/cbaf.c:238:30:    expected restricted __le16 [usertype] id
drivers/usb/wusbcore/cbaf.c:238:30:    got int
drivers/usb/wusbcore/cbaf.c:238:30: warning: incorrect type in initializer (different base types)
drivers/usb/wusbcore/cbaf.c:238:30:    expected restricted __le16 [usertype] len
drivers/usb/wusbcore/cbaf.c:238:30:    got int

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: David Vrabel <david.vrabel@csr.com>
Cc: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/wusbcore/rh.c       |  2 +-
 include/linux/usb/association.h | 22 +++++++++++-----------
 2 files changed, 12 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/wusbcore/rh.c b/drivers/usb/wusbcore/rh.c
index 95c6fa3bf6b2..3937bf6f8cef 100644
--- a/drivers/usb/wusbcore/rh.c
+++ b/drivers/usb/wusbcore/rh.c
@@ -326,7 +326,7 @@ static int wusbhc_rh_clear_port_feat(struct wusbhc *wusbhc, u16 feature,
 static int wusbhc_rh_get_port_status(struct wusbhc *wusbhc, u16 port_idx,
 				     u32 *_buf, u16 wLength)
 {
-	u16 *buf = (u16 *) _buf;
+	__le16 *buf = (__le16 *)_buf;
 
 	if (port_idx > wusbhc->ports_max)
 		return -EINVAL;
diff --git a/include/linux/usb/association.h b/include/linux/usb/association.h
index 07c5e3cf5898..0a4a18b3c1bb 100644
--- a/include/linux/usb/association.h
+++ b/include/linux/usb/association.h
@@ -28,17 +28,17 @@ struct wusb_am_attr {
 };
 
 /* Different fields defined by the spec */
-#define WUSB_AR_AssociationTypeId	{ .id = 0x0000, .len =  2 }
-#define WUSB_AR_AssociationSubTypeId	{ .id = 0x0001, .len =  2 }
-#define WUSB_AR_Length			{ .id = 0x0002, .len =  4 }
-#define WUSB_AR_AssociationStatus	{ .id = 0x0004, .len =  4 }
-#define WUSB_AR_LangID			{ .id = 0x0008, .len =  2 }
-#define WUSB_AR_DeviceFriendlyName	{ .id = 0x000b, .len = 64 } /* max */
-#define WUSB_AR_HostFriendlyName	{ .id = 0x000c, .len = 64 } /* max */
-#define WUSB_AR_CHID			{ .id = 0x1000, .len = 16 }
-#define WUSB_AR_CDID			{ .id = 0x1001, .len = 16 }
-#define WUSB_AR_ConnectionContext	{ .id = 0x1002, .len = 48 }
-#define WUSB_AR_BandGroups		{ .id = 0x1004, .len =  2 }
+#define WUSB_AR_AssociationTypeId	{ .id = cpu_to_le16(0x0000), .len = cpu_to_le16(2) }
+#define WUSB_AR_AssociationSubTypeId	{ .id = cpu_to_le16(0x0001), .len = cpu_to_le16(2) }
+#define WUSB_AR_Length			{ .id = cpu_to_le16(0x0002), .len = cpu_to_le16(4) }
+#define WUSB_AR_AssociationStatus	{ .id = cpu_to_le16(0x0004), .len = cpu_to_le16(4) }
+#define WUSB_AR_LangID			{ .id = cpu_to_le16(0x0008), .len = cpu_to_le16(2) }
+#define WUSB_AR_DeviceFriendlyName	{ .id = cpu_to_le16(0x000b), .len = cpu_to_le16(64) } /* max */
+#define WUSB_AR_HostFriendlyName	{ .id = cpu_to_le16(0x000c), .len = cpu_to_le16(64) } /* max */
+#define WUSB_AR_CHID			{ .id = cpu_to_le16(0x1000), .len = cpu_to_le16(16) }
+#define WUSB_AR_CDID			{ .id = cpu_to_le16(0x1001), .len = cpu_to_le16(16) }
+#define WUSB_AR_ConnectionContext	{ .id = cpu_to_le16(0x1002), .len = cpu_to_le16(48) }
+#define WUSB_AR_BandGroups		{ .id = cpu_to_le16(0x1004), .len = cpu_to_le16(2) }
 
 /* CBAF Control Requests (AMS1.0[T4-1] */
 enum {
-- 
cgit v1.2.3


From 9ac39f28b5237a629e41ccfc1f73d3a55723045c Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 12 Nov 2008 16:19:49 -0500
Subject: USB: add asynchronous autosuspend/autoresume support

This patch (as1160b) adds support routines for asynchronous autosuspend
and autoresume, with accompanying documentation updates.  There
already are several potential users of this interface, and others are
likely to arise as autosuspend support becomes more widespread.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/usb/power-management.txt | 22 +++++++--
 drivers/usb/core/driver.c              | 86 ++++++++++++++++++++++++++++++++++
 drivers/usb/core/hub.c                 |  5 +-
 drivers/usb/core/usb.c                 |  1 +
 drivers/usb/core/usb.h                 |  1 +
 include/linux/usb.h                    |  9 ++++
 6 files changed, 117 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/usb/power-management.txt b/Documentation/usb/power-management.txt
index e48ea1d51010..ad642615ad4c 100644
--- a/Documentation/usb/power-management.txt
+++ b/Documentation/usb/power-management.txt
@@ -313,11 +313,13 @@ three of the methods listed above.  In addition, a driver indicates
 that it supports autosuspend by setting the .supports_autosuspend flag
 in its usb_driver structure.  It is then responsible for informing the
 USB core whenever one of its interfaces becomes busy or idle.  The
-driver does so by calling these three functions:
+driver does so by calling these five functions:
 
 	int  usb_autopm_get_interface(struct usb_interface *intf);
 	void usb_autopm_put_interface(struct usb_interface *intf);
 	int  usb_autopm_set_interface(struct usb_interface *intf);
+	int  usb_autopm_get_interface_async(struct usb_interface *intf);
+	void usb_autopm_put_interface_async(struct usb_interface *intf);
 
 The functions work by maintaining a counter in the usb_interface
 structure.  When intf->pm_usage_count is > 0 then the interface is
@@ -330,10 +332,12 @@ associated with the device itself rather than any of its interfaces.
 This field is used only by the USB core.)
 
 The driver owns intf->pm_usage_count; it can modify the value however
-and whenever it likes.  A nice aspect of the usb_autopm_* routines is
-that the changes they make are protected by the usb_device structure's
-PM mutex (udev->pm_mutex); however drivers may change pm_usage_count
-without holding the mutex.
+and whenever it likes.  A nice aspect of the non-async usb_autopm_*
+routines is that the changes they make are protected by the usb_device
+structure's PM mutex (udev->pm_mutex); however drivers may change
+pm_usage_count without holding the mutex.  Drivers using the async
+routines are responsible for their own synchronization and mutual
+exclusion.
 
 	usb_autopm_get_interface() increments pm_usage_count and
 	attempts an autoresume if the new value is > 0 and the
@@ -348,6 +352,14 @@ without holding the mutex.
 	is suspended, and it attempts an autosuspend if the value is
 	<= 0 and the device isn't suspended.
 
+	usb_autopm_get_interface_async() and
+	usb_autopm_put_interface_async() do almost the same things as
+	their non-async counterparts.  The differences are: they do
+	not acquire the PM mutex, and they use a workqueue to do their
+	jobs.  As a result they can be called in an atomic context,
+	such as an URB's completion handler, but when they return the
+	device will not generally not yet be in the desired state.
+
 There also are a couple of utility routines drivers can use:
 
 	usb_autopm_enable() sets pm_usage_cnt to 0 and then calls
diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index 8c081308b0e2..23b3c7e79d4b 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -1341,6 +1341,19 @@ void usb_autosuspend_work(struct work_struct *work)
 	usb_autopm_do_device(udev, 0);
 }
 
+/* usb_autoresume_work - callback routine to autoresume a USB device */
+void usb_autoresume_work(struct work_struct *work)
+{
+	struct usb_device *udev =
+		container_of(work, struct usb_device, autoresume);
+
+	/* Wake it up, let the drivers do their thing, and then put it
+	 * back to sleep.
+	 */
+	if (usb_autopm_do_device(udev, 1) == 0)
+		usb_autopm_do_device(udev, -1);
+}
+
 /**
  * usb_autosuspend_device - delayed autosuspend of a USB device and its interfaces
  * @udev: the usb_device to autosuspend
@@ -1491,6 +1504,45 @@ void usb_autopm_put_interface(struct usb_interface *intf)
 }
 EXPORT_SYMBOL_GPL(usb_autopm_put_interface);
 
+/**
+ * usb_autopm_put_interface_async - decrement a USB interface's PM-usage counter
+ * @intf: the usb_interface whose counter should be decremented
+ *
+ * This routine does essentially the same thing as
+ * usb_autopm_put_interface(): it decrements @intf's usage counter and
+ * queues a delayed autosuspend request if the counter is <= 0.  The
+ * difference is that it does not acquire the device's pm_mutex;
+ * callers must handle all synchronization issues themselves.
+ *
+ * Typically a driver would call this routine during an URB's completion
+ * handler, if no more URBs were pending.
+ *
+ * This routine can run in atomic context.
+ */
+void usb_autopm_put_interface_async(struct usb_interface *intf)
+{
+	struct usb_device	*udev = interface_to_usbdev(intf);
+	int			status = 0;
+
+	if (intf->condition == USB_INTERFACE_UNBOUND) {
+		status = -ENODEV;
+	} else {
+		udev->last_busy = jiffies;
+		--intf->pm_usage_cnt;
+		if (udev->autosuspend_disabled || udev->autosuspend_delay < 0)
+			status = -EPERM;
+		else if (intf->pm_usage_cnt <= 0 &&
+				!timer_pending(&udev->autosuspend.timer)) {
+			queue_delayed_work(ksuspend_usb_wq, &udev->autosuspend,
+					round_jiffies_relative(
+						udev->autosuspend_delay));
+		}
+	}
+	dev_vdbg(&intf->dev, "%s: status %d cnt %d\n",
+			__func__, status, intf->pm_usage_cnt);
+}
+EXPORT_SYMBOL_GPL(usb_autopm_put_interface_async);
+
 /**
  * usb_autopm_get_interface - increment a USB interface's PM-usage counter
  * @intf: the usb_interface whose counter should be incremented
@@ -1536,6 +1588,37 @@ int usb_autopm_get_interface(struct usb_interface *intf)
 }
 EXPORT_SYMBOL_GPL(usb_autopm_get_interface);
 
+/**
+ * usb_autopm_get_interface_async - increment a USB interface's PM-usage counter
+ * @intf: the usb_interface whose counter should be incremented
+ *
+ * This routine does much the same thing as
+ * usb_autopm_get_interface(): it increments @intf's usage counter and
+ * queues an autoresume request if the result is > 0.  The differences
+ * are that it does not acquire the device's pm_mutex (callers must
+ * handle all synchronization issues themselves), and it does not
+ * autoresume the device directly (it only queues a request).  After a
+ * successful call, the device will generally not yet be resumed.
+ *
+ * This routine can run in atomic context.
+ */
+int usb_autopm_get_interface_async(struct usb_interface *intf)
+{
+	struct usb_device	*udev = interface_to_usbdev(intf);
+	int			status = 0;
+
+	if (intf->condition == USB_INTERFACE_UNBOUND)
+		status = -ENODEV;
+	else if (udev->autoresume_disabled)
+		status = -EPERM;
+	else if (++intf->pm_usage_cnt > 0 && udev->state == USB_STATE_SUSPENDED)
+		queue_work(ksuspend_usb_wq, &udev->autoresume);
+	dev_vdbg(&intf->dev, "%s: status %d cnt %d\n",
+			__func__, status, intf->pm_usage_cnt);
+	return status;
+}
+EXPORT_SYMBOL_GPL(usb_autopm_get_interface_async);
+
 /**
  * usb_autopm_set_interface - set a USB interface's autosuspend state
  * @intf: the usb_interface whose state should be set
@@ -1563,6 +1646,9 @@ EXPORT_SYMBOL_GPL(usb_autopm_set_interface);
 void usb_autosuspend_work(struct work_struct *work)
 {}
 
+void usb_autoresume_work(struct work_struct *work)
+{}
+
 #endif /* CONFIG_USB_SUSPEND */
 
 /**
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index b19cbfcd51da..95fb3104ba4f 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1374,8 +1374,9 @@ static void usb_stop_pm(struct usb_device *udev)
 		usb_autosuspend_device(udev->parent);
 	usb_pm_unlock(udev);
 
-	/* Stop any autosuspend requests already submitted */
-	cancel_rearming_delayed_work(&udev->autosuspend);
+	/* Stop any autosuspend or autoresume requests already submitted */
+	cancel_delayed_work_sync(&udev->autosuspend);
+	cancel_work_sync(&udev->autoresume);
 }
 
 #else
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 400fa4cc9a34..44f2fc750b6d 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -402,6 +402,7 @@ struct usb_device *usb_alloc_dev(struct usb_device *parent,
 #ifdef	CONFIG_PM
 	mutex_init(&dev->pm_mutex);
 	INIT_DELAYED_WORK(&dev->autosuspend, usb_autosuspend_work);
+	INIT_WORK(&dev->autoresume, usb_autoresume_work);
 	dev->autosuspend_delay = usb_autosuspend_delay * HZ;
 	dev->connect_time = jiffies;
 	dev->active_duration = -jiffies;
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 9a1a45ac3add..b60ebb4de1a8 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -45,6 +45,7 @@ extern int usb_suspend(struct device *dev, pm_message_t msg);
 extern int usb_resume(struct device *dev);
 
 extern void usb_autosuspend_work(struct work_struct *work);
+extern void usb_autoresume_work(struct work_struct *work);
 extern int usb_port_suspend(struct usb_device *dev);
 extern int usb_port_resume(struct usb_device *dev);
 extern int usb_external_suspend_device(struct usb_device *udev,
diff --git a/include/linux/usb.h b/include/linux/usb.h
index f72aa51f7bcd..859a88e6ce9c 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -398,6 +398,7 @@ struct usb_tt;
  * @urbnum: number of URBs submitted for the whole device
  * @active_duration: total time device is not suspended
  * @autosuspend: for delayed autosuspends
+ * @autoresume: for autoresumes requested while in_interrupt
  * @pm_mutex: protects PM operations
  * @last_busy: time of last use
  * @autosuspend_delay: in jiffies
@@ -476,6 +477,7 @@ struct usb_device {
 
 #ifdef CONFIG_PM
 	struct delayed_work autosuspend;
+	struct work_struct autoresume;
 	struct mutex pm_mutex;
 
 	unsigned long last_busy;
@@ -513,6 +515,8 @@ extern struct usb_device *usb_find_device(u16 vendor_id, u16 product_id);
 extern int usb_autopm_set_interface(struct usb_interface *intf);
 extern int usb_autopm_get_interface(struct usb_interface *intf);
 extern void usb_autopm_put_interface(struct usb_interface *intf);
+extern int usb_autopm_get_interface_async(struct usb_interface *intf);
+extern void usb_autopm_put_interface_async(struct usb_interface *intf);
 
 static inline void usb_autopm_enable(struct usb_interface *intf)
 {
@@ -539,8 +543,13 @@ static inline int usb_autopm_set_interface(struct usb_interface *intf)
 static inline int usb_autopm_get_interface(struct usb_interface *intf)
 { return 0; }
 
+static inline int usb_autopm_get_interface_async(struct usb_interface *intf)
+{ return 0; }
+
 static inline void usb_autopm_put_interface(struct usb_interface *intf)
 { }
+static inline void usb_autopm_put_interface_async(struct usb_interface *intf)
+{ }
 static inline void usb_autopm_enable(struct usb_interface *intf)
 { }
 static inline void usb_autopm_disable(struct usb_interface *intf)
-- 
cgit v1.2.3


From dc023dceec861c60bc1d1a17a2c6496ddac26ee7 Mon Sep 17 00:00:00 2001
From: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Date: Thu, 13 Nov 2008 10:31:35 -0800
Subject: USB: Introduce usb_queue_reset() to do resets from atomic contexts

This patch introduces a new call to be able to do a USB reset from an
atomic contect. This is quite helpful in USB callbacks to handle
errors (when the only thing that can be done is to do a device
reset).

It is done queuing a work struct that will do the actual reset. The
struct is "attached" to an interface so pending requests from an
interface are removed when said interface is unbound from the driver.

The call flow then becomes:

usb_queue_reset_device()
  __usb_queue_reset_device() [workqueue]
    usb_reset_device()

usb_probe_interface()
  usb_cancel_queue_reset()      [error path]

usb_unbind_interface()
  usb_cancel_queue_reset()

usb_driver_release_interface()
  usb_cancel_queue_reset()

Note usb_cancel_queue_reset() needs smarts to try not to unqueue when
it is actually being executed. This happens when we run the reset from
the workqueue: usb_reset_device() is called and on interface unbind
time, usb_cancel_queue_reset() would be called. That would deadlock on
cancel_work_sync(). To avoid that, we set (before running
usb_reset_device()) usb_intf->reset_running and clear it inmediately
after returning.

Patch is against 2.6.28-rc2 and depends on
http://marc.info/?l=linux-usb&m=122581634925308&w=2 (as submitted by
Alan Stern).

Signed-off-by: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c  | 23 +++++++++++++++++++++--
 drivers/usb/core/hub.c     | 43 +++++++++++++++++++++++++++++++++++++++++++
 drivers/usb/core/message.c | 41 +++++++++++++++++++++++++++++++++++++++++
 include/linux/usb.h        |  8 ++++++++
 4 files changed, 113 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index 23b3c7e79d4b..7e26fb3c2759 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -184,6 +184,20 @@ static int usb_unbind_device(struct device *dev)
 	return 0;
 }
 
+/*
+ * Cancel any pending scheduled resets
+ *
+ * [see usb_queue_reset_device()]
+ *
+ * Called after unconfiguring / when releasing interfaces. See
+ * comments in __usb_queue_reset_device() regarding
+ * udev->reset_running.
+ */
+static void usb_cancel_queued_reset(struct usb_interface *iface)
+{
+	if (iface->reset_running == 0)
+		cancel_work_sync(&iface->reset_ws);
+}
 
 /* called from driver core with dev locked */
 static int usb_probe_interface(struct device *dev)
@@ -242,6 +256,7 @@ static int usb_probe_interface(struct device *dev)
 			mark_quiesced(intf);
 			intf->needs_remote_wakeup = 0;
 			intf->condition = USB_INTERFACE_UNBOUND;
+			usb_cancel_queued_reset(intf);
 		} else
 			intf->condition = USB_INTERFACE_BOUND;
 
@@ -272,6 +287,7 @@ static int usb_unbind_interface(struct device *dev)
 		usb_disable_interface(udev, intf);
 
 	driver->disconnect(intf);
+	usb_cancel_queued_reset(intf);
 
 	/* Reset other interface state.
 	 * We cannot do a Set-Interface if the device is suspended or
@@ -380,8 +396,10 @@ void usb_driver_release_interface(struct usb_driver *driver,
 	if (device_is_registered(dev)) {
 		iface->condition = USB_INTERFACE_UNBINDING;
 		device_release_driver(dev);
+	} else {
+		iface->condition = USB_INTERFACE_UNBOUND;
+		usb_cancel_queued_reset(iface);
 	}
-
 	dev->driver = NULL;
 	usb_set_intfdata(iface, NULL);
 
@@ -942,7 +960,8 @@ static int usb_suspend_interface(struct usb_device *udev,
 	if (udev->state == USB_STATE_NOTATTACHED || !is_active(intf))
 		goto done;
 
-	if (intf->condition == USB_INTERFACE_UNBOUND)	/* This can't happen */
+	/* This can happen; see usb_driver_release_interface() */
+	if (intf->condition == USB_INTERFACE_UNBOUND)
 		goto done;
 	driver = to_usb_driver(intf->dev.driver);
 
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 95fb3104ba4f..e65881899c8f 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -3518,3 +3518,46 @@ int usb_reset_device(struct usb_device *udev)
 	return ret;
 }
 EXPORT_SYMBOL_GPL(usb_reset_device);
+
+
+/**
+ * usb_queue_reset_device - Reset a USB device from an atomic context
+ * @iface: USB interface belonging to the device to reset
+ *
+ * This function can be used to reset a USB device from an atomic
+ * context, where usb_reset_device() won't work (as it blocks).
+ *
+ * Doing a reset via this method is functionally equivalent to calling
+ * usb_reset_device(), except for the fact that it is delayed to a
+ * workqueue. This means that any drivers bound to other interfaces
+ * might be unbound, as well as users from usbfs in user space.
+ *
+ * Corner cases:
+ *
+ * - Scheduling two resets at the same time from two different drivers
+ *   attached to two different interfaces of the same device is
+ *   possible; depending on how the driver attached to each interface
+ *   handles ->pre_reset(), the second reset might happen or not.
+ *
+ * - If a driver is unbound and it had a pending reset, the reset will
+ *   be cancelled.
+ *
+ * - This function can be called during .probe() or .disconnect()
+ *   times. On return from .disconnect(), any pending resets will be
+ *   cancelled.
+ *
+ * There is no no need to lock/unlock the @reset_ws as schedule_work()
+ * does its own.
+ *
+ * NOTE: We don't do any reference count tracking because it is not
+ *     needed. The lifecycle of the work_struct is tied to the
+ *     usb_interface. Before destroying the interface we cancel the
+ *     work_struct, so the fact that work_struct is queued and or
+ *     running means the interface (and thus, the device) exist and
+ *     are referenced.
+ */
+void usb_queue_reset_device(struct usb_interface *iface)
+{
+	schedule_work(&iface->reset_ws);
+}
+EXPORT_SYMBOL_GPL(usb_queue_reset_device);
diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c
index cc47d36798b1..aadf29f09c45 100644
--- a/drivers/usb/core/message.c
+++ b/drivers/usb/core/message.c
@@ -1441,6 +1441,46 @@ static struct usb_interface_assoc_descriptor *find_iad(struct usb_device *dev,
 	return retval;
 }
 
+
+/*
+ * Internal function to queue a device reset
+ *
+ * This is initialized into the workstruct in 'struct
+ * usb_device->reset_ws' that is launched by
+ * message.c:usb_set_configuration() when initializing each 'struct
+ * usb_interface'.
+ *
+ * It is safe to get the USB device without reference counts because
+ * the life cycle of @iface is bound to the life cycle of @udev. Then,
+ * this function will be ran only if @iface is alive (and before
+ * freeing it any scheduled instances of it will have been cancelled).
+ *
+ * We need to set a flag (usb_dev->reset_running) because when we call
+ * the reset, the interfaces might be unbound. The current interface
+ * cannot try to remove the queued work as it would cause a deadlock
+ * (you cannot remove your work from within your executing
+ * workqueue). This flag lets it know, so that
+ * usb_cancel_queued_reset() doesn't try to do it.
+ *
+ * See usb_queue_reset_device() for more details
+ */
+void __usb_queue_reset_device(struct work_struct *ws)
+{
+	int rc;
+	struct usb_interface *iface =
+		container_of(ws, struct usb_interface, reset_ws);
+	struct usb_device *udev = interface_to_usbdev(iface);
+
+	rc = usb_lock_device_for_reset(udev, iface);
+	if (rc >= 0) {
+		iface->reset_running = 1;
+		usb_reset_device(udev);
+		iface->reset_running = 0;
+		usb_unlock_device(udev);
+	}
+}
+
+
 /*
  * usb_set_configuration - Makes a particular device setting be current
  * @dev: the device whose configuration is being updated
@@ -1611,6 +1651,7 @@ free_interfaces:
 		intf->dev.type = &usb_if_device_type;
 		intf->dev.groups = usb_interface_groups;
 		intf->dev.dma_mask = dev->dev.dma_mask;
+		INIT_WORK(&intf->reset_ws, __usb_queue_reset_device);
 		device_initialize(&intf->dev);
 		mark_quiesced(intf);
 		dev_set_name(&intf->dev, "%d-%s:%d.%d",
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 859a88e6ce9c..c8e55aa979de 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -120,6 +120,11 @@ enum usb_interface_condition {
  *	to the sysfs representation for that device.
  * @pm_usage_cnt: PM usage counter for this interface; autosuspend is not
  *	allowed unless the counter is 0.
+ * @reset_ws: Used for scheduling resets from atomic context.
+ * @reset_running: set to 1 if the interface is currently running a
+ *      queued reset so that usb_cancel_queued_reset() doesn't try to
+ *      remove from the workqueue when running inside the worker
+ *      thread. See __usb_queue_reset_device().
  *
  * USB device drivers attach to interfaces on a physical device.  Each
  * interface encapsulates a single high level function, such as feeding
@@ -168,10 +173,12 @@ struct usb_interface {
 	unsigned needs_remote_wakeup:1;	/* driver requires remote wakeup */
 	unsigned needs_altsetting0:1;	/* switch to altsetting 0 is pending */
 	unsigned needs_binding:1;	/* needs delayed unbind/rebind */
+	unsigned reset_running:1;
 
 	struct device dev;		/* interface specific device info */
 	struct device *usb_dev;
 	int pm_usage_cnt;		/* usage counter for autosuspend */
+	struct work_struct reset_ws;	/* for resets in atomic context */
 };
 #define	to_usb_interface(d) container_of(d, struct usb_interface, dev)
 #define	interface_to_usbdev(intf) \
@@ -507,6 +514,7 @@ extern int usb_lock_device_for_reset(struct usb_device *udev,
 
 /* USB port reset for device reinitialization */
 extern int usb_reset_device(struct usb_device *dev);
+extern void usb_queue_reset_device(struct usb_interface *dev);
 
 extern struct usb_device *usb_find_device(u16 vendor_id, u16 product_id);
 
-- 
cgit v1.2.3


From f150fa1afbf69a87f54752579ff2bb769aad88b3 Mon Sep 17 00:00:00 2001
From: Pete Zaitcev <zaitcev@redhat.com>
Date: Thu, 13 Nov 2008 21:31:21 -0700
Subject: USB: Allow usbmon as a module even if usbcore is builtin

usbmon can only be built as a module if usbcore is a module too. Trivial
changes to the relevant Kconfig and Makefile (and a few trivial changes
elsewhere) allow usbmon to be built as a module even if usbcore is
builtin.

This is verified to work in all 9 permutations (3 correctly prohibited
by Kconfig, 6 build a suitable result).

Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Pete Zaitcev <zaitcev@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd.c   |  4 ++--
 drivers/usb/core/hcd.h   |  4 ++--
 drivers/usb/mon/Kconfig  | 13 ++++++-------
 drivers/usb/mon/Makefile |  3 +--
 include/linux/usb.h      |  2 +-
 5 files changed, 12 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index e1b42626d04d..7403ed871abd 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -2028,7 +2028,7 @@ EXPORT_SYMBOL_GPL(usb_hcd_platform_shutdown);
 
 /*-------------------------------------------------------------------------*/
 
-#if defined(CONFIG_USB_MON)
+#if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE)
 
 struct usb_mon_operations *mon_ops;
 
@@ -2064,4 +2064,4 @@ void usb_mon_deregister (void)
 }
 EXPORT_SYMBOL_GPL (usb_mon_deregister);
 
-#endif /* CONFIG_USB_MON */
+#endif /* CONFIG_USB_MON || CONFIG_USB_MON_MODULE */
diff --git a/drivers/usb/core/hcd.h b/drivers/usb/core/hcd.h
index d202342b36b6..0aaa9cea6b38 100644
--- a/drivers/usb/core/hcd.h
+++ b/drivers/usb/core/hcd.h
@@ -421,7 +421,7 @@ static inline void usbfs_cleanup(void) { }
 
 /*-------------------------------------------------------------------------*/
 
-#if defined(CONFIG_USB_MON)
+#if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE)
 
 struct usb_mon_operations {
 	void (*urb_submit)(struct usb_bus *bus, struct urb *urb);
@@ -463,7 +463,7 @@ static inline void usbmon_urb_submit_error(struct usb_bus *bus, struct urb *urb,
 static inline void usbmon_urb_complete(struct usb_bus *bus, struct urb *urb,
 		int status) {}
 
-#endif /* CONFIG_USB_MON */
+#endif /* CONFIG_USB_MON || CONFIG_USB_MON_MODULE */
 
 /*-------------------------------------------------------------------------*/
 
diff --git a/drivers/usb/mon/Kconfig b/drivers/usb/mon/Kconfig
index deb9ddffa402..f28f350cd96a 100644
--- a/drivers/usb/mon/Kconfig
+++ b/drivers/usb/mon/Kconfig
@@ -3,14 +3,13 @@
 #
 
 config USB_MON
-	bool "USB Monitor"
-	depends on USB!=n
-	default y
+	tristate "USB Monitor"
+	depends on USB
+	default y if USB=y
+	default m if USB=m
 	help
-	  If you say Y here, a component which captures the USB traffic
+	  If you select this option, a component which captures the USB traffic
 	  between peripheral-specific drivers and HC drivers will be built.
 	  For more information, see <file:Documentation/usb/usbmon.txt>.
 
-	  This is somewhat experimental at this time, but it should be safe.
-
-	  If unsure, say Y.
+	  If unsure, say Y (if allowed), otherwise M.
diff --git a/drivers/usb/mon/Makefile b/drivers/usb/mon/Makefile
index 0f76ed5e1617..c6516b566731 100644
--- a/drivers/usb/mon/Makefile
+++ b/drivers/usb/mon/Makefile
@@ -4,5 +4,4 @@
 
 usbmon-objs	:= mon_main.o mon_stat.o mon_text.o mon_bin.o mon_dma.o
 
-# This does not use CONFIG_USB_MON because we want this to use a tristate.
-obj-$(CONFIG_USB)	+= usbmon.o
+obj-$(CONFIG_USB_MON)	+= usbmon.o
diff --git a/include/linux/usb.h b/include/linux/usb.h
index c8e55aa979de..8bc81bffc195 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -336,7 +336,7 @@ struct usb_bus {
 #endif
 	struct device *dev;		/* device for this bus */
 
-#if defined(CONFIG_USB_MON)
+#if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE)
 	struct mon_bus *mon_bus;	/* non-null when associated */
 	int monitored;			/* non-zero when monitored */
 #endif
-- 
cgit v1.2.3


From 1537e0ad944acf3a4c2b311a646d7993b89499f7 Mon Sep 17 00:00:00 2001
From: Ben Efros <ben@pc-doctor.com>
Date: Tue, 18 Nov 2008 13:31:13 -0800
Subject: USB: storage devices and SAT

Add the SANE SENSE flag to indicate that a device is capable of handling
more than 18-bytes of sense data.  This functionality is required for
USB-ATA bridges implementing SAT.  A future patch will actually enable this
function for several devices.

The logic behind this is that we can detect support for SANE_SENSE in a few ways:
 1) ATA PASS THROUGH (12) or (16) execute successfully
 2) SPC-3 or higher is in use
 3) A previous CHECK CONDITION occurred with sense format 70-73 and had
    a length greater than 18-bytes total

Signed-off-by: Ben Efros <ben@pc-doctor.com>
Signed-off-by: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/scsiglue.c  |  4 ++++
 drivers/usb/storage/transport.c | 40 +++++++++++++++++++++++++++++++++++++++-
 include/linux/usb_usual.h       |  5 +++--
 3 files changed, 46 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/storage/scsiglue.c b/drivers/usb/storage/scsiglue.c
index 09779f6a8179..1b35e011a34f 100644
--- a/drivers/usb/storage/scsiglue.c
+++ b/drivers/usb/storage/scsiglue.c
@@ -170,6 +170,10 @@ static int slave_configure(struct scsi_device *sdev)
 		if (us->fflags & US_FL_CAPACITY_HEURISTICS)
 			sdev->guess_capacity = 1;
 
+		/* assume SPC3 or latter devices support sense size > 18 */
+		if (sdev->scsi_level > SCSI_SPC_2)
+			us->fflags |= US_FL_SANE_SENSE;
+
 		/* Some devices report a SCSI revision level above 2 but are
 		 * unable to handle the REPORT LUNS command (for which
 		 * support is mandatory at level 3).  Since we already have
diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
index 2058db41618c..f584e72cc689 100644
--- a/drivers/usb/storage/transport.c
+++ b/drivers/usb/storage/transport.c
@@ -578,6 +578,20 @@ void usb_stor_invoke_transport(struct scsi_cmnd *srb, struct us_data *us)
 		need_auto_sense = 1;
 	}
 
+	/*
+	 * Determine if this device is SAT by seeing if the
+	 * command executed successfully.  Otherwise we'll have
+	 * to wait for at least one CHECK_CONDITION to determine
+	 * SANE_SENSE support
+	 */
+	if ((srb->cmnd[0] == ATA_16 || srb->cmnd[0] == ATA_12) &&
+	    result == USB_STOR_TRANSPORT_GOOD &&
+	    !(us->fflags & US_FL_SANE_SENSE) &&
+	    !(srb->cmnd[2] & 0x20)) {
+		US_DEBUGP("-- SAT supported, increasing auto-sense\n");
+		us->fflags |= US_FL_SANE_SENSE;
+	}
+
 	/*
 	 * A short transfer on a command where we don't expect it
 	 * is unusual, but it doesn't mean we need to auto-sense.
@@ -595,10 +609,15 @@ void usb_stor_invoke_transport(struct scsi_cmnd *srb, struct us_data *us)
 	if (need_auto_sense) {
 		int temp_result;
 		struct scsi_eh_save ses;
+		int sense_size = US_SENSE_SIZE;
+
+		/* device supports and needs bigger sense buffer */
+		if (us->fflags & US_FL_SANE_SENSE)
+			sense_size = ~0;
 
 		US_DEBUGP("Issuing auto-REQUEST_SENSE\n");
 
-		scsi_eh_prep_cmnd(srb, &ses, NULL, 0, US_SENSE_SIZE);
+		scsi_eh_prep_cmnd(srb, &ses, NULL, 0, sense_size);
 
 		/* FIXME: we must do the protocol translation here */
 		if (us->subclass == US_SC_RBC || us->subclass == US_SC_SCSI ||
@@ -632,6 +651,25 @@ void usb_stor_invoke_transport(struct scsi_cmnd *srb, struct us_data *us)
 			return;
 		}
 
+		/* If the sense data returned is larger than 18-bytes then we
+		 * assume this device supports requesting more in the future.
+		 * The response code must be 70h through 73h inclusive.
+		 */
+		if (srb->sense_buffer[7] > (US_SENSE_SIZE - 8) &&
+		    !(us->fflags & US_FL_SANE_SENSE) &&
+		    (srb->sense_buffer[0] & 0x7C) == 0x70) {
+			US_DEBUGP("-- SANE_SENSE support enabled\n");
+			us->fflags |= US_FL_SANE_SENSE;
+
+			/* Indicate to the user that we truncated their sense
+			 * because we didn't know it supported larger sense.
+			 */
+			US_DEBUGP("-- Sense data truncated to %i from %i\n",
+			          US_SENSE_SIZE,
+			          srb->sense_buffer[7] + 8);
+			srb->sense_buffer[7] = (US_SENSE_SIZE - 8);
+		}
+
 		US_DEBUGP("-- Result from auto-sense is %d\n", temp_result);
 		US_DEBUGP("-- code: 0x%x, key: 0x%x, ASC: 0x%x, ASCQ: 0x%x\n",
 			  srb->sense_buffer[0],
diff --git a/include/linux/usb_usual.h b/include/linux/usb_usual.h
index d9a3bbe38e6b..998e5cbbf29e 100644
--- a/include/linux/usb_usual.h
+++ b/include/linux/usb_usual.h
@@ -52,8 +52,9 @@
 	US_FLAG(MAX_SECTORS_MIN,0x00002000)			\
 		/* Sets max_sectors to arch min */		\
 	US_FLAG(BULK_IGNORE_TAG,0x00004000)			\
-		/* Ignore tag mismatch in bulk operations */
-
+		/* Ignore tag mismatch in bulk operations */    \
+	US_FLAG(SANE_SENSE,     0x00008000)
+		/* Sane Sense (> 18 bytes) */
 
 #define US_FLAG(name, value)	US_FL_##name = value ,
 enum { US_DO_ALL_FLAGS };
-- 
cgit v1.2.3


From 6084f1bf0c51a99cbba612ee90a4607cffb8b042 Mon Sep 17 00:00:00 2001
From: Philipp Zabel <philipp.zabel@gmail.com>
Date: Mon, 24 Nov 2008 12:00:01 -0800
Subject: USB: otg: gpio_vbus transceiver stub

gpio_vbus provides simple GPIO VBUS sensing for peripheral
controllers with an internal transceiver.
Optionally, a second GPIO can be used to control D+ pullup.

It also interfaces with the regulator framework to limit charging
currents when powered via USB. gpio_vbus requests the regulator
supplying "vbus_draw" and can enable/disable it or limit its
current depending on USB state.

[dbrownell@users.sourceforge.net: use drivers/otg, cleanups ]

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/otg/Kconfig       |  13 ++
 drivers/usb/otg/Makefile      |   5 +
 drivers/usb/otg/gpio_vbus.c   | 335 ++++++++++++++++++++++++++++++++++++++++++
 include/linux/usb/gpio_vbus.h |  30 ++++
 4 files changed, 383 insertions(+)
 create mode 100644 drivers/usb/otg/gpio_vbus.c
 create mode 100644 include/linux/usb/gpio_vbus.h

(limited to 'include/linux')

diff --git a/drivers/usb/otg/Kconfig b/drivers/usb/otg/Kconfig
index 860342902bbb..afe91bfea7f2 100644
--- a/drivers/usb/otg/Kconfig
+++ b/drivers/usb/otg/Kconfig
@@ -14,6 +14,19 @@ config USB_OTG_UTILS
 	  Select this to make sure the build includes objects from
 	  the OTG infrastructure directory.
 
+#
+# USB Transceiver Drivers
+#
+config USB_GPIO_VBUS
+	tristate "GPIO based peripheral-only VBUS sensing 'transceiver'"
+	depends on GENERIC_GPIO
+	select USB_OTG_UTILS
+	help
+	  Provides simple GPIO VBUS sensing for controllers with an
+	  internal transceiver via the otg_transceiver interface, and
+	  optionally control of a D+ pullup GPIO as well as a VBUS
+	  current limit regulator.
+
 config ISP1301_OMAP
 	tristate "Philips ISP1301 with OMAP OTG"
 	depends on I2C && ARCH_OMAP_OTG
diff --git a/drivers/usb/otg/Makefile b/drivers/usb/otg/Makefile
index 483816685d00..6c58b36ca7cf 100644
--- a/drivers/usb/otg/Makefile
+++ b/drivers/usb/otg/Makefile
@@ -1,3 +1,8 @@
+#
+# OTG infrastructure and transceiver drivers
+#
+
+obj-$(CONFIG_USB_GPIO_VBUS)	+= gpio_vbus.o
 obj-$(CONFIG_ISP1301_OMAP)	+= isp1301_omap.o
 
 ifeq ($(CONFIG_USB_DEBUG),y)
diff --git a/drivers/usb/otg/gpio_vbus.c b/drivers/usb/otg/gpio_vbus.c
new file mode 100644
index 000000000000..63a6036f04be
--- /dev/null
+++ b/drivers/usb/otg/gpio_vbus.c
@@ -0,0 +1,335 @@
+/*
+ * gpio-vbus.c - simple GPIO VBUS sensing driver for B peripheral devices
+ *
+ * Copyright (c) 2008 Philipp Zabel <philipp.zabel@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/gpio.h>
+#include <linux/interrupt.h>
+#include <linux/usb.h>
+
+#include <linux/regulator/consumer.h>
+
+#include <linux/usb/gadget.h>
+#include <linux/usb/gpio_vbus.h>
+#include <linux/usb/otg.h>
+
+
+/*
+ * A simple GPIO VBUS sensing driver for B peripheral only devices
+ * with internal transceivers. It can control a D+ pullup GPIO and
+ * a regulator to limit the current drawn from VBUS.
+ *
+ * Needs to be loaded before the UDC driver that will use it.
+ */
+struct gpio_vbus_data {
+	struct otg_transceiver otg;
+	struct device          *dev;
+	struct regulator       *vbus_draw;
+	int			vbus_draw_enabled;
+	unsigned		mA;
+};
+
+
+/*
+ * This driver relies on "both edges" triggering.  VBUS has 100 msec to
+ * stabilize, so the peripheral controller driver may need to cope with
+ * some bouncing due to current surges (e.g. charging local capacitance)
+ * and contact chatter.
+ *
+ * REVISIT in desperate straits, toggling between rising and falling
+ * edges might be workable.
+ */
+#define VBUS_IRQ_FLAGS \
+	( IRQF_SAMPLE_RANDOM | IRQF_SHARED \
+	| IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING )
+
+
+/* interface to regulator framework */
+static void set_vbus_draw(struct gpio_vbus_data *gpio_vbus, unsigned mA)
+{
+	struct regulator *vbus_draw = gpio_vbus->vbus_draw;
+	int enabled;
+
+	if (!vbus_draw)
+		return;
+
+	enabled = gpio_vbus->vbus_draw_enabled;
+	if (mA) {
+		regulator_set_current_limit(vbus_draw, 0, 1000 * mA);
+		if (!enabled) {
+			regulator_enable(vbus_draw);
+			gpio_vbus->vbus_draw_enabled = 1;
+		}
+	} else {
+		if (enabled) {
+			regulator_disable(vbus_draw);
+			gpio_vbus->vbus_draw_enabled = 0;
+		}
+	}
+	gpio_vbus->mA = mA;
+}
+
+/* VBUS change IRQ handler */
+static irqreturn_t gpio_vbus_irq(int irq, void *data)
+{
+	struct platform_device *pdev = data;
+	struct gpio_vbus_mach_info *pdata = pdev->dev.platform_data;
+	struct gpio_vbus_data *gpio_vbus = platform_get_drvdata(pdev);
+	int gpio, vbus;
+
+	vbus = gpio_get_value(pdata->gpio_vbus);
+	if (pdata->gpio_vbus_inverted)
+		vbus = !vbus;
+
+	dev_dbg(&pdev->dev, "VBUS %s (gadget: %s)\n",
+		vbus ? "supplied" : "inactive",
+		gpio_vbus->otg.gadget ? gpio_vbus->otg.gadget->name : "none");
+
+	if (!gpio_vbus->otg.gadget)
+		return IRQ_HANDLED;
+
+	/* Peripheral controllers which manage the pullup themselves won't have
+	 * gpio_pullup configured here.  If it's configured here, we'll do what
+	 * isp1301_omap::b_peripheral() does and enable the pullup here... although
+	 * that may complicate usb_gadget_{,dis}connect() support.
+	 */
+	gpio = pdata->gpio_pullup;
+	if (vbus) {
+		gpio_vbus->otg.state = OTG_STATE_B_PERIPHERAL;
+		usb_gadget_vbus_connect(gpio_vbus->otg.gadget);
+
+		/* drawing a "unit load" is *always* OK, except for OTG */
+		set_vbus_draw(gpio_vbus, 100);
+
+		/* optionally enable D+ pullup */
+		if (gpio_is_valid(gpio))
+			gpio_set_value(gpio, !pdata->gpio_pullup_inverted);
+	} else {
+		/* optionally disable D+ pullup */
+		if (gpio_is_valid(gpio))
+			gpio_set_value(gpio, pdata->gpio_pullup_inverted);
+
+		set_vbus_draw(gpio_vbus, 0);
+
+		usb_gadget_vbus_disconnect(gpio_vbus->otg.gadget);
+		gpio_vbus->otg.state = OTG_STATE_B_IDLE;
+	}
+
+	return IRQ_HANDLED;
+}
+
+/* OTG transceiver interface */
+
+/* bind/unbind the peripheral controller */
+static int gpio_vbus_set_peripheral(struct otg_transceiver *otg,
+				struct usb_gadget *gadget)
+{
+	struct gpio_vbus_data *gpio_vbus;
+	struct gpio_vbus_mach_info *pdata;
+	struct platform_device *pdev;
+	int gpio, irq;
+
+	gpio_vbus = container_of(otg, struct gpio_vbus_data, otg);
+	pdev = to_platform_device(gpio_vbus->dev);
+	pdata = gpio_vbus->dev->platform_data;
+	irq = gpio_to_irq(pdata->gpio_vbus);
+	gpio = pdata->gpio_pullup;
+
+	if (!gadget) {
+		dev_dbg(&pdev->dev, "unregistering gadget '%s'\n",
+			otg->gadget->name);
+
+		/* optionally disable D+ pullup */
+		if (gpio_is_valid(gpio))
+			gpio_set_value(gpio, pdata->gpio_pullup_inverted);
+
+		set_vbus_draw(gpio_vbus, 0);
+
+		usb_gadget_vbus_disconnect(otg->gadget);
+		otg->state = OTG_STATE_UNDEFINED;
+
+		otg->gadget = NULL;
+		return 0;
+	}
+
+	otg->gadget = gadget;
+	dev_dbg(&pdev->dev, "registered gadget '%s'\n", gadget->name);
+
+	/* initialize connection state */
+	gpio_vbus_irq(irq, pdev);
+	return 0;
+}
+
+/* effective for B devices, ignored for A-peripheral */
+static int gpio_vbus_set_power(struct otg_transceiver *otg, unsigned mA)
+{
+	struct gpio_vbus_data *gpio_vbus;
+
+	gpio_vbus = container_of(otg, struct gpio_vbus_data, otg);
+
+	if (otg->state == OTG_STATE_B_PERIPHERAL)
+		set_vbus_draw(gpio_vbus, mA);
+	return 0;
+}
+
+/* for non-OTG B devices: set/clear transceiver suspend mode */
+static int gpio_vbus_set_suspend(struct otg_transceiver *otg, int suspend)
+{
+	struct gpio_vbus_data *gpio_vbus;
+
+	gpio_vbus = container_of(otg, struct gpio_vbus_data, otg);
+
+	/* draw max 0 mA from vbus in suspend mode; or the previously
+	 * recorded amount of current if not suspended
+	 *
+	 * NOTE: high powered configs (mA > 100) may draw up to 2.5 mA
+	 * if they're wake-enabled ... we don't handle that yet.
+	 */
+	return gpio_vbus_set_power(otg, suspend ? 0 : gpio_vbus->mA);
+}
+
+/* platform driver interface */
+
+static int __init gpio_vbus_probe(struct platform_device *pdev)
+{
+	struct gpio_vbus_mach_info *pdata = pdev->dev.platform_data;
+	struct gpio_vbus_data *gpio_vbus;
+	struct resource *res;
+	int err, gpio, irq;
+
+	if (!pdata || !gpio_is_valid(pdata->gpio_vbus))
+		return -EINVAL;
+	gpio = pdata->gpio_vbus;
+
+	gpio_vbus = kzalloc(sizeof(struct gpio_vbus_data), GFP_KERNEL);
+	if (!gpio_vbus)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, gpio_vbus);
+	gpio_vbus->dev = &pdev->dev;
+	gpio_vbus->otg.label = "gpio-vbus";
+	gpio_vbus->otg.state = OTG_STATE_UNDEFINED;
+	gpio_vbus->otg.set_peripheral = gpio_vbus_set_peripheral;
+	gpio_vbus->otg.set_power = gpio_vbus_set_power;
+	gpio_vbus->otg.set_suspend = gpio_vbus_set_suspend;
+
+	err = gpio_request(gpio, "vbus_detect");
+	if (err) {
+		dev_err(&pdev->dev, "can't request vbus gpio %d, err: %d\n",
+			gpio, err);
+		goto err_gpio;
+	}
+	gpio_direction_input(gpio);
+
+	res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	if (res) {
+		irq = res->start;
+		res->flags &= IRQF_TRIGGER_MASK;
+		res->flags |= IRQF_SAMPLE_RANDOM | IRQF_SHARED;
+	} else
+		irq = gpio_to_irq(gpio);
+
+	/* if data line pullup is in use, initialize it to "not pulling up" */
+	gpio = pdata->gpio_pullup;
+	if (gpio_is_valid(gpio)) {
+		err = gpio_request(gpio, "udc_pullup");
+		if (err) {
+			dev_err(&pdev->dev,
+				"can't request pullup gpio %d, err: %d\n",
+				gpio, err);
+			gpio_free(pdata->gpio_vbus);
+			goto err_gpio;
+		}
+		gpio_direction_output(gpio, pdata->gpio_pullup_inverted);
+	}
+
+	err = request_irq(irq, gpio_vbus_irq, VBUS_IRQ_FLAGS,
+		"vbus_detect", pdev);
+	if (err) {
+		dev_err(&pdev->dev, "can't request irq %i, err: %d\n",
+			irq, err);
+		goto err_irq;
+	}
+
+	/* only active when a gadget is registered */
+	err = otg_set_transceiver(&gpio_vbus->otg);
+	if (err) {
+		dev_err(&pdev->dev, "can't register transceiver, err: %d\n",
+			err);
+		goto err_otg;
+	}
+
+	gpio_vbus->vbus_draw = regulator_get(&pdev->dev, "vbus_draw");
+	if (IS_ERR(gpio_vbus->vbus_draw)) {
+		dev_dbg(&pdev->dev, "can't get vbus_draw regulator, err: %ld\n",
+			PTR_ERR(gpio_vbus->vbus_draw));
+		gpio_vbus->vbus_draw = NULL;
+	}
+
+	return 0;
+err_otg:
+	free_irq(irq, &pdev->dev);
+err_irq:
+	if (gpio_is_valid(pdata->gpio_pullup))
+		gpio_free(pdata->gpio_pullup);
+	gpio_free(pdata->gpio_vbus);
+err_gpio:
+	platform_set_drvdata(pdev, NULL);
+	kfree(gpio_vbus);
+	return err;
+}
+
+static int __exit gpio_vbus_remove(struct platform_device *pdev)
+{
+	struct gpio_vbus_data *gpio_vbus = platform_get_drvdata(pdev);
+	struct gpio_vbus_mach_info *pdata = pdev->dev.platform_data;
+	int gpio = pdata->gpio_vbus;
+
+	regulator_put(gpio_vbus->vbus_draw);
+
+	otg_set_transceiver(NULL);
+
+	free_irq(gpio_to_irq(gpio), &pdev->dev);
+	if (gpio_is_valid(pdata->gpio_pullup))
+		gpio_free(pdata->gpio_pullup);
+	gpio_free(gpio);
+	platform_set_drvdata(pdev, NULL);
+	kfree(gpio_vbus);
+
+	return 0;
+}
+
+/* NOTE:  the gpio-vbus device may *NOT* be hotplugged */
+
+MODULE_ALIAS("platform:gpio-vbus");
+
+static struct platform_driver gpio_vbus_driver = {
+	.driver = {
+		.name  = "gpio-vbus",
+		.owner = THIS_MODULE,
+	},
+	.remove  = __exit_p(gpio_vbus_remove),
+};
+
+static int __init gpio_vbus_init(void)
+{
+	return platform_driver_probe(&gpio_vbus_driver, gpio_vbus_probe);
+}
+module_init(gpio_vbus_init);
+
+static void __exit gpio_vbus_exit(void)
+{
+	platform_driver_unregister(&gpio_vbus_driver);
+}
+module_exit(gpio_vbus_exit);
+
+MODULE_DESCRIPTION("simple GPIO controlled OTG transceiver driver");
+MODULE_AUTHOR("Philipp Zabel");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/usb/gpio_vbus.h b/include/linux/usb/gpio_vbus.h
new file mode 100644
index 000000000000..d9f03ccc2d60
--- /dev/null
+++ b/include/linux/usb/gpio_vbus.h
@@ -0,0 +1,30 @@
+/*
+ * A simple GPIO VBUS sensing driver for B peripheral only devices
+ * with internal transceivers.
+ * Optionally D+ pullup can be controlled by a second GPIO.
+ *
+ * Copyright (c) 2008 Philipp Zabel <philipp.zabel@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+/**
+ * struct gpio_vbus_mach_info - configuration for gpio_vbus
+ * @gpio_vbus: VBUS sensing GPIO
+ * @gpio_pullup: optional D+ or D- pullup GPIO (else negative/invalid)
+ * @gpio_vbus_inverted: true if gpio_vbus is active low
+ * @gpio_pullup_inverted: true if gpio_pullup is active low
+ *
+ * The VBUS sensing GPIO should have a pulldown, which will normally be
+ * part of a resistor ladder turning a 4.0V-5.25V level on VBUS into a
+ * value the GPIO detects as active.  Some systems will use comparators.
+ */
+struct gpio_vbus_mach_info {
+	int gpio_vbus;
+	int gpio_pullup;
+	bool gpio_vbus_inverted;
+	bool gpio_pullup_inverted;
+};
-- 
cgit v1.2.3


From 68144e0cc92125f41157ede7b060f83367bc4fe7 Mon Sep 17 00:00:00 2001
From: Philipp Zabel <philipp.zabel@gmail.com>
Date: Mon, 24 Nov 2008 12:01:17 -0800
Subject: USB: otg: add otg_put_transceiver()

As Russell King points out, calling put_device(otg_transceiver->dev)
directly in driver cleanup paths makes assumptions about otg_transceiver
internals.

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/arm/plat-omap/usb.c      | 6 ++++++
 drivers/usb/gadget/omap_udc.c | 4 ++--
 include/linux/usb/otg.h       | 1 +
 3 files changed, 9 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/plat-omap/usb.c b/arch/arm/plat-omap/usb.c
index 67ca1e216df7..2f88ca8b8f38 100644
--- a/arch/arm/plat-omap/usb.c
+++ b/arch/arm/plat-omap/usb.c
@@ -96,6 +96,12 @@ struct otg_transceiver *otg_get_transceiver(void)
 }
 EXPORT_SYMBOL(otg_get_transceiver);
 
+void otg_put_transceiver(struct otg_transceiver *x)
+{
+	put_device(x->dev);
+}
+EXPORT_SYMBOL(otg_put_transceiver);
+
 int otg_set_transceiver(struct otg_transceiver *x)
 {
 	if (xceiv && x)
diff --git a/drivers/usb/gadget/omap_udc.c b/drivers/usb/gadget/omap_udc.c
index 34e9e393f929..57d9641c6bf8 100644
--- a/drivers/usb/gadget/omap_udc.c
+++ b/drivers/usb/gadget/omap_udc.c
@@ -3006,7 +3006,7 @@ cleanup1:
 
 cleanup0:
 	if (xceiv)
-		put_device(xceiv->dev);
+		otg_put_transceiver(xceiv);
 
 	if (cpu_is_omap16xx() || cpu_is_omap24xx()) {
 		clk_disable(hhc_clk);
@@ -3034,7 +3034,7 @@ static int __exit omap_udc_remove(struct platform_device *pdev)
 
 	pullup_disable(udc);
 	if (udc->transceiver) {
-		put_device(udc->transceiver->dev);
+		otg_put_transceiver(udc->transceiver);
 		udc->transceiver = NULL;
 	}
 	omap_writew(0, UDC_SYSCON1);
diff --git a/include/linux/usb/otg.h b/include/linux/usb/otg.h
index 1db25d152ad8..94df4fe6c6c0 100644
--- a/include/linux/usb/otg.h
+++ b/include/linux/usb/otg.h
@@ -84,6 +84,7 @@ extern int otg_set_transceiver(struct otg_transceiver *);
 
 /* for usb host and peripheral controller drivers */
 extern struct otg_transceiver *otg_get_transceiver(void);
+extern void otg_put_transceiver(struct otg_transceiver *);
 
 static inline int
 otg_start_hnp(struct otg_transceiver *otg)
-- 
cgit v1.2.3


From 65bfd2967c906ca322a4bb69a285fe0de8916ac6 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Tue, 25 Nov 2008 16:39:18 -0500
Subject: USB: Enhance usage of pm_message_t

This patch (as1177) modifies the USB core suspend and resume
routines.  The resume functions now will take a pm_message_t argument,
so they will know what sort of resume is occurring.  The new argument
is also passed to the port suspend/resume and bus suspend/resume
routines (although they don't use it for anything but debugging).

In addition, special pm_message_t values are used for user-initiated,
device-initiated (i.e., remote wakeup), and automatic suspend/resume.
By testing these values, drivers can tell whether or not a particular
suspend was an autosuspend.  Unfortunately, they can't do the same for
resumes -- not until the pm_message_t argument is also passed to the
drivers' resume methods.  That will require a bigger change.

IMO, the whole Power Management framework should have been set up this
way in the first place.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/class/cdc-acm.c |  2 +-
 drivers/usb/class/cdc-wdm.c |  3 ++-
 drivers/usb/core/driver.c   | 61 +++++++++++++++++++++++++--------------------
 drivers/usb/core/generic.c  | 10 ++++----
 drivers/usb/core/hcd.c      | 10 ++++----
 drivers/usb/core/hcd.h      |  4 +--
 drivers/usb/core/hub.c      | 16 ++++++------
 drivers/usb/core/sysfs.c    |  6 ++---
 drivers/usb/core/usb.c      |  8 +++---
 drivers/usb/core/usb.h      | 15 ++++++-----
 include/linux/usb.h         |  2 +-
 11 files changed, 74 insertions(+), 63 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c
index d50a99f70aee..00b47ea24f86 100644
--- a/drivers/usb/class/cdc-acm.c
+++ b/drivers/usb/class/cdc-acm.c
@@ -1275,7 +1275,7 @@ static int acm_suspend(struct usb_interface *intf, pm_message_t message)
 	struct acm *acm = usb_get_intfdata(intf);
 	int cnt;
 
-	if (acm->dev->auto_pm) {
+	if (message.event & PM_EVENT_AUTO) {
 		int b;
 
 		spin_lock_irq(&acm->read_lock);
diff --git a/drivers/usb/class/cdc-wdm.c b/drivers/usb/class/cdc-wdm.c
index 5a8ecc045e3f..3771d6e6d0cc 100644
--- a/drivers/usb/class/cdc-wdm.c
+++ b/drivers/usb/class/cdc-wdm.c
@@ -764,7 +764,8 @@ static int wdm_suspend(struct usb_interface *intf, pm_message_t message)
 
 	mutex_lock(&desc->plock);
 #ifdef CONFIG_PM
-	if (interface_to_usbdev(desc->intf)->auto_pm && test_bit(WDM_IN_USE, &desc->flags)) {
+	if ((message.event & PM_EVENT_AUTO) &&
+			test_bit(WDM_IN_USE, &desc->flags)) {
 		rv = -EBUSY;
 	} else {
 #endif
diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index 0226e019326a..41c06025506e 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -922,7 +922,7 @@ static int usb_suspend_device(struct usb_device *udev, pm_message_t msg)
 }
 
 /* Caller has locked udev's pm_mutex */
-static int usb_resume_device(struct usb_device *udev)
+static int usb_resume_device(struct usb_device *udev, pm_message_t msg)
 {
 	struct usb_device_driver	*udriver;
 	int				status = 0;
@@ -940,7 +940,7 @@ static int usb_resume_device(struct usb_device *udev)
 		udev->reset_resume = 1;
 
 	udriver = to_usb_device_driver(udev->dev.driver);
-	status = udriver->resume(udev);
+	status = udriver->resume(udev, msg);
 
  done:
 	dev_vdbg(&udev->dev, "%s: status %d\n", __func__, status);
@@ -969,7 +969,7 @@ static int usb_suspend_interface(struct usb_device *udev,
 		status = driver->suspend(intf, msg);
 		if (status == 0)
 			mark_quiesced(intf);
-		else if (!udev->auto_pm)
+		else if (!(msg.event & PM_EVENT_AUTO))
 			dev_err(&intf->dev, "%s error %d\n",
 					"suspend", status);
 	} else {
@@ -987,7 +987,7 @@ static int usb_suspend_interface(struct usb_device *udev,
 
 /* Caller has locked intf's usb_device's pm_mutex */
 static int usb_resume_interface(struct usb_device *udev,
-		struct usb_interface *intf, int reset_resume)
+		struct usb_interface *intf, pm_message_t msg, int reset_resume)
 {
 	struct usb_driver	*driver;
 	int			status = 0;
@@ -1138,10 +1138,9 @@ static inline int autosuspend_check(struct usb_device *udev, int reschedule)
  * all the interfaces which were suspended are resumed so that they remain
  * in the same state as the device.
  *
- * If an autosuspend is in progress (@udev->auto_pm is set), the routine
- * checks first to make sure that neither the device itself or any of its
- * active interfaces is in use (pm_usage_cnt is greater than 0).  If they
- * are, the autosuspend fails.
+ * If an autosuspend is in progress the routine checks first to make sure
+ * that neither the device itself or any of its active interfaces is in use
+ * (pm_usage_cnt is greater than 0).  If they are, the autosuspend fails.
  *
  * If the suspend succeeds, the routine recursively queues an autosuspend
  * request for @udev's parent device, thereby propagating the change up
@@ -1176,7 +1175,7 @@ static int usb_suspend_both(struct usb_device *udev, pm_message_t msg)
 
 	udev->do_remote_wakeup = device_may_wakeup(&udev->dev);
 
-	if (udev->auto_pm) {
+	if (msg.event & PM_EVENT_AUTO) {
 		status = autosuspend_check(udev, 0);
 		if (status < 0)
 			goto done;
@@ -1196,13 +1195,16 @@ static int usb_suspend_both(struct usb_device *udev, pm_message_t msg)
 
 	/* If the suspend failed, resume interfaces that did get suspended */
 	if (status != 0) {
+		pm_message_t msg2;
+
+		msg2.event = msg.event ^ (PM_EVENT_SUSPEND | PM_EVENT_RESUME);
 		while (--i >= 0) {
 			intf = udev->actconfig->interface[i];
-			usb_resume_interface(udev, intf, 0);
+			usb_resume_interface(udev, intf, msg2, 0);
 		}
 
 		/* Try another autosuspend when the interfaces aren't busy */
-		if (udev->auto_pm)
+		if (msg.event & PM_EVENT_AUTO)
 			autosuspend_check(udev, status == -EBUSY);
 
 	/* If the suspend succeeded then prevent any more URB submissions,
@@ -1232,6 +1234,7 @@ static int usb_suspend_both(struct usb_device *udev, pm_message_t msg)
 /**
  * usb_resume_both - resume a USB device and its interfaces
  * @udev: the usb_device to resume
+ * @msg: Power Management message describing this state transition
  *
  * This is the central routine for resuming USB devices.  It calls the
  * the resume method for @udev and then calls the resume methods for all
@@ -1257,7 +1260,7 @@ static int usb_suspend_both(struct usb_device *udev, pm_message_t msg)
  *
  * This routine can run only in process context.
  */
-static int usb_resume_both(struct usb_device *udev)
+static int usb_resume_both(struct usb_device *udev, pm_message_t msg)
 {
 	int			status = 0;
 	int			i;
@@ -1273,14 +1276,15 @@ static int usb_resume_both(struct usb_device *udev)
 
 	/* Propagate the resume up the tree, if necessary */
 	if (udev->state == USB_STATE_SUSPENDED) {
-		if (udev->auto_pm && udev->autoresume_disabled) {
+		if ((msg.event & PM_EVENT_AUTO) &&
+				udev->autoresume_disabled) {
 			status = -EPERM;
 			goto done;
 		}
 		if (parent) {
 			status = usb_autoresume_device(parent);
 			if (status == 0) {
-				status = usb_resume_device(udev);
+				status = usb_resume_device(udev, msg);
 				if (status || udev->state ==
 						USB_STATE_NOTATTACHED) {
 					usb_autosuspend_device(parent);
@@ -1303,15 +1307,16 @@ static int usb_resume_both(struct usb_device *udev)
 			/* We can't progagate beyond the USB subsystem,
 			 * so if a root hub's controller is suspended
 			 * then we're stuck. */
-			status = usb_resume_device(udev);
+			status = usb_resume_device(udev, msg);
 		}
 	} else if (udev->reset_resume)
-		status = usb_resume_device(udev);
+		status = usb_resume_device(udev, msg);
 
 	if (status == 0 && udev->actconfig) {
 		for (i = 0; i < udev->actconfig->desc.bNumInterfaces; i++) {
 			intf = udev->actconfig->interface[i];
-			usb_resume_interface(udev, intf, udev->reset_resume);
+			usb_resume_interface(udev, intf, msg,
+					udev->reset_resume);
 		}
 	}
 
@@ -1339,13 +1344,13 @@ static int usb_autopm_do_device(struct usb_device *udev, int inc_usage_cnt)
 		udev->last_busy = jiffies;
 	if (inc_usage_cnt >= 0 && udev->pm_usage_cnt > 0) {
 		if (udev->state == USB_STATE_SUSPENDED)
-			status = usb_resume_both(udev);
+			status = usb_resume_both(udev, PMSG_AUTO_RESUME);
 		if (status != 0)
 			udev->pm_usage_cnt -= inc_usage_cnt;
 		else if (inc_usage_cnt)
 			udev->last_busy = jiffies;
 	} else if (inc_usage_cnt <= 0 && udev->pm_usage_cnt <= 0) {
-		status = usb_suspend_both(udev, PMSG_SUSPEND);
+		status = usb_suspend_both(udev, PMSG_AUTO_SUSPEND);
 	}
 	usb_pm_unlock(udev);
 	return status;
@@ -1469,13 +1474,14 @@ static int usb_autopm_do_interface(struct usb_interface *intf,
 		udev->last_busy = jiffies;
 		if (inc_usage_cnt >= 0 && intf->pm_usage_cnt > 0) {
 			if (udev->state == USB_STATE_SUSPENDED)
-				status = usb_resume_both(udev);
+				status = usb_resume_both(udev,
+						PMSG_AUTO_RESUME);
 			if (status != 0)
 				intf->pm_usage_cnt -= inc_usage_cnt;
 			else
 				udev->last_busy = jiffies;
 		} else if (inc_usage_cnt <= 0 && intf->pm_usage_cnt <= 0) {
-			status = usb_suspend_both(udev, PMSG_SUSPEND);
+			status = usb_suspend_both(udev, PMSG_AUTO_SUSPEND);
 		}
 	}
 	usb_pm_unlock(udev);
@@ -1700,6 +1706,7 @@ int usb_external_suspend_device(struct usb_device *udev, pm_message_t msg)
 /**
  * usb_external_resume_device - external resume of a USB device and its interfaces
  * @udev: the usb_device to resume
+ * @msg: Power Management message describing this state transition
  *
  * This routine handles external resume requests: ones not generated
  * internally by a USB driver (autoresume) but rather coming from the user
@@ -1708,13 +1715,13 @@ int usb_external_suspend_device(struct usb_device *udev, pm_message_t msg)
  *
  * The caller must hold @udev's device lock.
  */
-int usb_external_resume_device(struct usb_device *udev)
+int usb_external_resume_device(struct usb_device *udev, pm_message_t msg)
 {
 	int	status;
 
 	usb_pm_lock(udev);
 	udev->auto_pm = 0;
-	status = usb_resume_both(udev);
+	status = usb_resume_both(udev, msg);
 	udev->last_busy = jiffies;
 	usb_pm_unlock(udev);
 	if (status == 0)
@@ -1727,7 +1734,7 @@ int usb_external_resume_device(struct usb_device *udev)
 	return status;
 }
 
-int usb_suspend(struct device *dev, pm_message_t message)
+int usb_suspend(struct device *dev, pm_message_t msg)
 {
 	struct usb_device	*udev;
 
@@ -1746,10 +1753,10 @@ int usb_suspend(struct device *dev, pm_message_t message)
 	}
 
 	udev->skip_sys_resume = 0;
-	return usb_external_suspend_device(udev, message);
+	return usb_external_suspend_device(udev, msg);
 }
 
-int usb_resume(struct device *dev)
+int usb_resume(struct device *dev, pm_message_t msg)
 {
 	struct usb_device	*udev;
 
@@ -1761,7 +1768,7 @@ int usb_resume(struct device *dev)
 	 */
 	if (udev->skip_sys_resume)
 		return 0;
-	return usb_external_resume_device(udev);
+	return usb_external_resume_device(udev, msg);
 }
 
 #endif /* CONFIG_PM */
diff --git a/drivers/usb/core/generic.c b/drivers/usb/core/generic.c
index 7e912f21fd36..30ecac3af15a 100644
--- a/drivers/usb/core/generic.c
+++ b/drivers/usb/core/generic.c
@@ -200,18 +200,18 @@ static int generic_suspend(struct usb_device *udev, pm_message_t msg)
 	 * interfaces manually by doing a bus (or "global") suspend.
 	 */
 	if (!udev->parent)
-		rc = hcd_bus_suspend(udev);
+		rc = hcd_bus_suspend(udev, msg);
 
 	/* Non-root devices don't need to do anything for FREEZE or PRETHAW */
 	else if (msg.event == PM_EVENT_FREEZE || msg.event == PM_EVENT_PRETHAW)
 		rc = 0;
 	else
-		rc = usb_port_suspend(udev);
+		rc = usb_port_suspend(udev, msg);
 
 	return rc;
 }
 
-static int generic_resume(struct usb_device *udev)
+static int generic_resume(struct usb_device *udev, pm_message_t msg)
 {
 	int rc;
 
@@ -221,9 +221,9 @@ static int generic_resume(struct usb_device *udev)
 	 * interfaces manually by doing a bus (or "global") resume.
 	 */
 	if (!udev->parent)
-		rc = hcd_bus_resume(udev);
+		rc = hcd_bus_resume(udev, msg);
 	else
-		rc = usb_port_resume(udev);
+		rc = usb_port_resume(udev, msg);
 	return rc;
 }
 
diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index 7403ed871abd..a0079876d74e 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -1573,14 +1573,14 @@ int usb_hcd_get_frame_number (struct usb_device *udev)
 
 #ifdef	CONFIG_PM
 
-int hcd_bus_suspend(struct usb_device *rhdev)
+int hcd_bus_suspend(struct usb_device *rhdev, pm_message_t msg)
 {
 	struct usb_hcd	*hcd = container_of(rhdev->bus, struct usb_hcd, self);
 	int		status;
 	int		old_state = hcd->state;
 
 	dev_dbg(&rhdev->dev, "bus %s%s\n",
-			rhdev->auto_pm ? "auto-" : "", "suspend");
+			(msg.event & PM_EVENT_AUTO ? "auto-" : ""), "suspend");
 	if (!hcd->driver->bus_suspend) {
 		status = -ENOENT;
 	} else {
@@ -1598,14 +1598,14 @@ int hcd_bus_suspend(struct usb_device *rhdev)
 	return status;
 }
 
-int hcd_bus_resume(struct usb_device *rhdev)
+int hcd_bus_resume(struct usb_device *rhdev, pm_message_t msg)
 {
 	struct usb_hcd	*hcd = container_of(rhdev->bus, struct usb_hcd, self);
 	int		status;
 	int		old_state = hcd->state;
 
 	dev_dbg(&rhdev->dev, "usb %s%s\n",
-			rhdev->auto_pm ? "auto-" : "", "resume");
+			(msg.event & PM_EVENT_AUTO ? "auto-" : ""), "resume");
 	if (!hcd->driver->bus_resume)
 		return -ENOENT;
 	if (hcd->state == HC_STATE_RUNNING)
@@ -1638,7 +1638,7 @@ static void hcd_resume_work(struct work_struct *work)
 
 	usb_lock_device(udev);
 	usb_mark_last_busy(udev);
-	usb_external_resume_device(udev);
+	usb_external_resume_device(udev, PMSG_REMOTE_RESUME);
 	usb_unlock_device(udev);
 }
 
diff --git a/drivers/usb/core/hcd.h b/drivers/usb/core/hcd.h
index 0aaa9cea6b38..aa5da82d9071 100644
--- a/drivers/usb/core/hcd.h
+++ b/drivers/usb/core/hcd.h
@@ -388,8 +388,8 @@ extern int usb_find_interface_driver(struct usb_device *dev,
 #ifdef CONFIG_PM
 extern void usb_hcd_resume_root_hub(struct usb_hcd *hcd);
 extern void usb_root_hub_lost_power(struct usb_device *rhdev);
-extern int hcd_bus_suspend(struct usb_device *rhdev);
-extern int hcd_bus_resume(struct usb_device *rhdev);
+extern int hcd_bus_suspend(struct usb_device *rhdev, pm_message_t msg);
+extern int hcd_bus_resume(struct usb_device *rhdev, pm_message_t msg);
 #else
 static inline void usb_hcd_resume_root_hub(struct usb_hcd *hcd)
 {
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index ff066edf4dca..fc99ef67761d 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1984,7 +1984,7 @@ static int check_port_resume_type(struct usb_device *udev,
  *
  * Returns 0 on success, else negative errno.
  */
-int usb_port_suspend(struct usb_device *udev)
+int usb_port_suspend(struct usb_device *udev, pm_message_t msg)
 {
 	struct usb_hub	*hub = hdev_to_hub(udev->parent);
 	int		port1 = udev->portnum;
@@ -2023,7 +2023,7 @@ int usb_port_suspend(struct usb_device *udev)
 	} else {
 		/* device has up to 10 msec to fully suspend */
 		dev_dbg(&udev->dev, "usb %ssuspend\n",
-				udev->auto_pm ? "auto-" : "");
+				(msg.event & PM_EVENT_AUTO ? "auto-" : ""));
 		usb_set_device_state(udev, USB_STATE_SUSPENDED);
 		msleep(10);
 	}
@@ -2142,7 +2142,7 @@ static int finish_port_resume(struct usb_device *udev)
  *
  * Returns 0 on success, else negative errno.
  */
-int usb_port_resume(struct usb_device *udev)
+int usb_port_resume(struct usb_device *udev, pm_message_t msg)
 {
 	struct usb_hub	*hub = hdev_to_hub(udev->parent);
 	int		port1 = udev->portnum;
@@ -2167,7 +2167,7 @@ int usb_port_resume(struct usb_device *udev)
 	} else {
 		/* drive resume for at least 20 msec */
 		dev_dbg(&udev->dev, "usb %sresume\n",
-				udev->auto_pm ? "auto-" : "");
+				(msg.event & PM_EVENT_AUTO ? "auto-" : ""));
 		msleep(25);
 
 		/* Virtual root hubs can trigger on GET_PORT_STATUS to
@@ -2208,7 +2208,7 @@ static int remote_wakeup(struct usb_device *udev)
 	if (udev->state == USB_STATE_SUSPENDED) {
 		dev_dbg(&udev->dev, "usb %sresume\n", "wakeup-");
 		usb_mark_last_busy(udev);
-		status = usb_external_resume_device(udev);
+		status = usb_external_resume_device(udev, PMSG_REMOTE_RESUME);
 	}
 	return status;
 }
@@ -2217,14 +2217,14 @@ static int remote_wakeup(struct usb_device *udev)
 
 /* When CONFIG_USB_SUSPEND isn't set, we never suspend or resume any ports. */
 
-int usb_port_suspend(struct usb_device *udev)
+int usb_port_suspend(struct usb_device *udev, pm_message_t msg)
 {
 	return 0;
 }
 
 /* However we may need to do a reset-resume */
 
-int usb_port_resume(struct usb_device *udev)
+int usb_port_resume(struct usb_device *udev, pm_message_t msg)
 {
 	struct usb_hub	*hub = hdev_to_hub(udev->parent);
 	int		port1 = udev->portnum;
@@ -2264,7 +2264,7 @@ static int hub_suspend(struct usb_interface *intf, pm_message_t msg)
 
 		udev = hdev->children [port1-1];
 		if (udev && udev->can_submit) {
-			if (!hdev->auto_pm)
+			if (!(msg.event & PM_EVENT_AUTO))
 				dev_dbg(&intf->dev, "port %d nyet suspended\n",
 						port1);
 			return -EBUSY;
diff --git a/drivers/usb/core/sysfs.c b/drivers/usb/core/sysfs.c
index 8c65fa75b5c2..0f0ccf640114 100644
--- a/drivers/usb/core/sysfs.c
+++ b/drivers/usb/core/sysfs.c
@@ -359,19 +359,19 @@ set_level(struct device *dev, struct device_attribute *attr,
 			strncmp(buf, on_string, len) == 0) {
 		udev->autosuspend_disabled = 1;
 		udev->autoresume_disabled = 0;
-		rc = usb_external_resume_device(udev);
+		rc = usb_external_resume_device(udev, PMSG_USER_RESUME);
 
 	} else if (len == sizeof auto_string - 1 &&
 			strncmp(buf, auto_string, len) == 0) {
 		udev->autosuspend_disabled = 0;
 		udev->autoresume_disabled = 0;
-		rc = usb_external_resume_device(udev);
+		rc = usb_external_resume_device(udev, PMSG_USER_RESUME);
 
 	} else if (len == sizeof suspend_string - 1 &&
 			strncmp(buf, suspend_string, len) == 0) {
 		udev->autosuspend_disabled = 0;
 		udev->autoresume_disabled = 1;
-		rc = usb_external_suspend_device(udev, PMSG_SUSPEND);
+		rc = usb_external_suspend_device(udev, PMSG_USER_SUSPEND);
 
 	} else
 		rc = -EINVAL;
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 51854c2bc912..4c98f3975afe 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -253,7 +253,7 @@ static int usb_dev_prepare(struct device *dev)
 static void usb_dev_complete(struct device *dev)
 {
 	/* Currently used only for rebinding interfaces */
-	usb_resume(dev);	/* Implement eventually? */
+	usb_resume(dev, PMSG_RESUME);	/* Message event is meaningless */
 }
 
 static int usb_dev_suspend(struct device *dev)
@@ -263,7 +263,7 @@ static int usb_dev_suspend(struct device *dev)
 
 static int usb_dev_resume(struct device *dev)
 {
-	return usb_resume(dev);
+	return usb_resume(dev, PMSG_RESUME);
 }
 
 static int usb_dev_freeze(struct device *dev)
@@ -273,7 +273,7 @@ static int usb_dev_freeze(struct device *dev)
 
 static int usb_dev_thaw(struct device *dev)
 {
-	return usb_resume(dev);
+	return usb_resume(dev, PMSG_THAW);
 }
 
 static int usb_dev_poweroff(struct device *dev)
@@ -283,7 +283,7 @@ static int usb_dev_poweroff(struct device *dev)
 
 static int usb_dev_restore(struct device *dev)
 {
-	return usb_resume(dev);
+	return usb_resume(dev, PMSG_RESTORE);
 }
 
 static struct dev_pm_ops usb_device_pm_ops = {
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index b60ebb4de1a8..9fb195665fa8 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -1,3 +1,5 @@
+#include <linux/pm.h>
+
 /* Functions local to drivers/usb/core/ */
 
 extern int usb_create_sysfs_dev_files(struct usb_device *dev);
@@ -42,15 +44,16 @@ extern void usb_host_cleanup(void);
 #ifdef	CONFIG_PM
 
 extern int usb_suspend(struct device *dev, pm_message_t msg);
-extern int usb_resume(struct device *dev);
+extern int usb_resume(struct device *dev, pm_message_t msg);
 
 extern void usb_autosuspend_work(struct work_struct *work);
 extern void usb_autoresume_work(struct work_struct *work);
-extern int usb_port_suspend(struct usb_device *dev);
-extern int usb_port_resume(struct usb_device *dev);
+extern int usb_port_suspend(struct usb_device *dev, pm_message_t msg);
+extern int usb_port_resume(struct usb_device *dev, pm_message_t msg);
 extern int usb_external_suspend_device(struct usb_device *udev,
 		pm_message_t msg);
-extern int usb_external_resume_device(struct usb_device *udev);
+extern int usb_external_resume_device(struct usb_device *udev,
+		pm_message_t msg);
 
 static inline void usb_pm_lock(struct usb_device *udev)
 {
@@ -64,12 +67,12 @@ static inline void usb_pm_unlock(struct usb_device *udev)
 
 #else
 
-static inline int usb_port_suspend(struct usb_device *udev)
+static inline int usb_port_suspend(struct usb_device *udev, pm_message_t msg)
 {
 	return 0;
 }
 
-static inline int usb_port_resume(struct usb_device *udev)
+static inline int usb_port_resume(struct usb_device *udev, pm_message_t msg)
 {
 	return 0;
 }
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 8bc81bffc195..74d0b9990c73 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -1067,7 +1067,7 @@ struct usb_device_driver {
 	void (*disconnect) (struct usb_device *udev);
 
 	int (*suspend) (struct usb_device *udev, pm_message_t message);
-	int (*resume) (struct usb_device *udev);
+	int (*resume) (struct usb_device *udev, pm_message_t message);
 	struct usbdrv_wrap drvwrap;
 	unsigned int supports_autosuspend:1;
 };
-- 
cgit v1.2.3


From 2ffcdb3bdadaf8260986e96384df26c94a6ad42c Mon Sep 17 00:00:00 2001
From: Bryan Wu <cooloney@kernel.org>
Date: Tue, 2 Dec 2008 21:33:43 +0200
Subject: USB: musb: use new platform data interface of musb to replace old one

Signed-off-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: Felipe Balbi <felipe.balbi@nokia.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb/musb.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/usb/musb.h b/include/linux/usb/musb.h
index 630962c04ca4..d6aad0ea6033 100644
--- a/include/linux/usb/musb.h
+++ b/include/linux/usb/musb.h
@@ -47,6 +47,11 @@ struct musb_hdrc_config {
 	u8		ram_bits;	/* ram address size */
 
 	struct musb_hdrc_eps_bits *eps_bits;
+#ifdef CONFIG_BLACKFIN
+        /* A GPIO controlling VRSEL in Blackfin */
+        unsigned int    gpio_vrsel;
+#endif
+
 };
 
 struct musb_hdrc_platform_data {
-- 
cgit v1.2.3


From 3b23dd6f8a718e5339de4f7d86ce76a078b5f771 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Fri, 5 Dec 2008 14:10:34 -0500
Subject: USB: utilize the bus notifiers

This patch (as1185) makes usbcore take advantage of the bus
notifications sent out by the driver core.  Now we can create all our
device and interface attribute files before the device or interface
uevent is broadcast.

A side effect is that we no longer create the endpoint "pseudo"
devices at the same time as a device or interface is registered -- it
seems like a bad idea to try registering an endpoint before the
registration of its parent is complete.  So the routines for creating
and removing endpoint devices have been split out and renamed, and
they are called explicitly when needed.  A new bitflag is used for
keeping track of whether or not the interface's endpoint devices have
been created, since (just as with the interface attributes) they vary
with the altsetting and hence can be changed at random times.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/endpoint.c |  4 ++--
 drivers/usb/core/hub.c      | 18 ++++++----------
 drivers/usb/core/message.c  | 50 ++++++++++++++++++++++++++++++++++++++-------
 drivers/usb/core/sysfs.c    | 22 +-------------------
 drivers/usb/core/usb.c      | 37 +++++++++++++++++++++++++++++++++
 drivers/usb/core/usb.h      |  4 ++--
 include/linux/usb.h         |  2 ++
 7 files changed, 93 insertions(+), 44 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/core/endpoint.c b/drivers/usb/core/endpoint.c
index 946fae43d622..e1710f260b4f 100644
--- a/drivers/usb/core/endpoint.c
+++ b/drivers/usb/core/endpoint.c
@@ -276,7 +276,7 @@ static void ep_device_release(struct device *dev)
 	kfree(ep_dev);
 }
 
-int usb_create_ep_files(struct device *parent,
+int usb_create_ep_devs(struct device *parent,
 			struct usb_host_endpoint *endpoint,
 			struct usb_device *udev)
 {
@@ -340,7 +340,7 @@ exit:
 	return retval;
 }
 
-void usb_remove_ep_files(struct usb_host_endpoint *endpoint)
+void usb_remove_ep_devs(struct usb_host_endpoint *endpoint)
 {
 	struct ep_device *ep_dev = endpoint->ep_dev;
 
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 5abdc11be1e5..756b8d9993fc 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1437,17 +1437,12 @@ void usb_disconnect(struct usb_device **pdev)
 	usb_disable_device(udev, 0);
 	usb_hcd_synchronize_unlinks(udev);
 
+	usb_remove_ep_devs(&udev->ep0);
 	usb_unlock_device(udev);
 
-	/* Remove the device-specific files from sysfs.  This must be
-	 * done with udev unlocked, because some of the attribute
-	 * routines try to acquire the device lock.
-	 */
-	usb_remove_sysfs_dev_files(udev);
-
 	/* Unregister the device.  The device driver is responsible
-	 * for removing the device files from usbfs and sysfs and for
-	 * de-configuring the device.
+	 * for de-configuring the device and invoking the remove-device
+	 * notifier chain (used by usbfs and possibly others).
 	 */
 	device_del(&udev->dev);
 
@@ -1654,8 +1649,8 @@ int usb_new_device(struct usb_device *udev)
 	announce_device(udev);
 
 	/* Register the device.  The device driver is responsible
-	 * for adding the device files to sysfs and for configuring
-	 * the device.
+	 * for configuring the device and invoking the add-device
+	 * notifier chain (used by usbfs and possibly others).
 	 */
 	err = device_add(&udev->dev);
 	if (err) {
@@ -1663,8 +1658,7 @@ int usb_new_device(struct usb_device *udev)
 		goto fail;
 	}
 
-	/* put device-specific files into sysfs */
-	usb_create_sysfs_dev_files(udev);
+	(void) usb_create_ep_devs(&udev->dev, &udev->ep0, udev);
 	return err;
 
 fail:
diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c
index aadf29f09c45..7943901c641c 100644
--- a/drivers/usb/core/message.c
+++ b/drivers/usb/core/message.c
@@ -1004,6 +1004,34 @@ int usb_clear_halt(struct usb_device *dev, int pipe)
 }
 EXPORT_SYMBOL_GPL(usb_clear_halt);
 
+static int create_intf_ep_devs(struct usb_interface *intf)
+{
+	struct usb_device *udev = interface_to_usbdev(intf);
+	struct usb_host_interface *alt = intf->cur_altsetting;
+	int i;
+
+	if (intf->ep_devs_created || intf->unregistering)
+		return 0;
+
+	for (i = 0; i < alt->desc.bNumEndpoints; ++i)
+		(void) usb_create_ep_devs(&intf->dev, &alt->endpoint[i], udev);
+	intf->ep_devs_created = 1;
+	return 0;
+}
+
+static void remove_intf_ep_devs(struct usb_interface *intf)
+{
+	struct usb_host_interface *alt = intf->cur_altsetting;
+	int i;
+
+	if (!intf->ep_devs_created)
+		return;
+
+	for (i = 0; i < alt->desc.bNumEndpoints; ++i)
+		usb_remove_ep_devs(&alt->endpoint[i]);
+	intf->ep_devs_created = 0;
+}
+
 /**
  * usb_disable_endpoint -- Disable an endpoint by address
  * @dev: the device whose endpoint is being disabled
@@ -1092,7 +1120,7 @@ void usb_disable_device(struct usb_device *dev, int skip_ep0)
 			dev_dbg(&dev->dev, "unregistering interface %s\n",
 				dev_name(&interface->dev));
 			interface->unregistering = 1;
-			usb_remove_sysfs_intf_files(interface);
+			remove_intf_ep_devs(interface);
 			device_del(&interface->dev);
 		}
 
@@ -1235,8 +1263,10 @@ int usb_set_interface(struct usb_device *dev, int interface, int alternate)
 	 */
 
 	/* prevent submissions using previous endpoint settings */
-	if (iface->cur_altsetting != alt)
+	if (iface->cur_altsetting != alt) {
+		remove_intf_ep_devs(iface);
 		usb_remove_sysfs_intf_files(iface);
+	}
 	usb_disable_interface(dev, iface);
 
 	iface->cur_altsetting = alt;
@@ -1272,9 +1302,10 @@ int usb_set_interface(struct usb_device *dev, int interface, int alternate)
 	 * (Likewise, EP0 never "halts" on well designed devices.)
 	 */
 	usb_enable_interface(dev, iface);
-	if (device_is_registered(&iface->dev))
+	if (device_is_registered(&iface->dev)) {
 		usb_create_sysfs_intf_files(iface);
-
+		create_intf_ep_devs(iface);
+	}
 	return 0;
 }
 EXPORT_SYMBOL_GPL(usb_set_interface);
@@ -1334,7 +1365,6 @@ int usb_reset_configuration(struct usb_device *dev)
 		struct usb_interface *intf = config->interface[i];
 		struct usb_host_interface *alt;
 
-		usb_remove_sysfs_intf_files(intf);
 		alt = usb_altnum_to_altsetting(intf, 0);
 
 		/* No altsetting 0?  We'll assume the first altsetting.
@@ -1345,10 +1375,16 @@ int usb_reset_configuration(struct usb_device *dev)
 		if (!alt)
 			alt = &intf->altsetting[0];
 
+		if (alt != intf->cur_altsetting) {
+			remove_intf_ep_devs(intf);
+			usb_remove_sysfs_intf_files(intf);
+		}
 		intf->cur_altsetting = alt;
 		usb_enable_interface(dev, intf);
-		if (device_is_registered(&intf->dev))
+		if (device_is_registered(&intf->dev)) {
 			usb_create_sysfs_intf_files(intf);
+			create_intf_ep_devs(intf);
+		}
 	}
 	return 0;
 }
@@ -1682,7 +1718,7 @@ free_interfaces:
 				dev_name(&intf->dev), ret);
 			continue;
 		}
-		usb_create_sysfs_intf_files(intf);
+		create_intf_ep_devs(intf);
 	}
 
 	usb_autosuspend_device(dev);
diff --git a/drivers/usb/core/sysfs.c b/drivers/usb/core/sysfs.c
index 0f0ccf640114..4cc2456ef3be 100644
--- a/drivers/usb/core/sysfs.c
+++ b/drivers/usb/core/sysfs.c
@@ -629,9 +629,6 @@ int usb_create_sysfs_dev_files(struct usb_device *udev)
 	struct device *dev = &udev->dev;
 	int retval;
 
-	/* Unforunately these attributes cannot be created before
-	 * the uevent is broadcast.
-	 */
 	retval = device_create_bin_file(dev, &dev_bin_attr_descriptors);
 	if (retval)
 		goto error;
@@ -643,11 +640,7 @@ int usb_create_sysfs_dev_files(struct usb_device *udev)
 	retval = add_power_attributes(dev);
 	if (retval)
 		goto error;
-
-	retval = usb_create_ep_files(dev, &udev->ep0, udev);
-	if (retval)
-		goto error;
-	return 0;
+	return retval;
 error:
 	usb_remove_sysfs_dev_files(udev);
 	return retval;
@@ -657,7 +650,6 @@ void usb_remove_sysfs_dev_files(struct usb_device *udev)
 {
 	struct device *dev = &udev->dev;
 
-	usb_remove_ep_files(&udev->ep0);
 	remove_power_attributes(dev);
 	remove_persist_attributes(dev);
 	device_remove_bin_file(dev, &dev_bin_attr_descriptors);
@@ -816,36 +808,24 @@ int usb_create_sysfs_intf_files(struct usb_interface *intf)
 {
 	struct usb_device *udev = interface_to_usbdev(intf);
 	struct usb_host_interface *alt = intf->cur_altsetting;
-	int i;
 	int retval;
 
 	if (intf->sysfs_files_created || intf->unregistering)
 		return 0;
 
-	/* The interface string may be present in some altsettings
-	 * and missing in others.  Hence its attribute cannot be created
-	 * before the uevent is broadcast.
-	 */
 	if (alt->string == NULL)
 		alt->string = usb_cache_string(udev, alt->desc.iInterface);
 	if (alt->string)
 		retval = device_create_file(&intf->dev, &dev_attr_interface);
-	for (i = 0; i < alt->desc.bNumEndpoints; ++i)
-		usb_create_ep_files(&intf->dev, &alt->endpoint[i], udev);
 	intf->sysfs_files_created = 1;
 	return 0;
 }
 
 void usb_remove_sysfs_intf_files(struct usb_interface *intf)
 {
-	struct usb_host_interface *alt = intf->cur_altsetting;
-	int i;
-
 	if (!intf->sysfs_files_created)
 		return;
 
-	for (i = 0; i < alt->desc.bNumEndpoints; ++i)
-		usb_remove_ep_files(&alt->endpoint[i]);
 	device_remove_file(&intf->dev, &dev_attr_interface);
 	intf->sysfs_files_created = 0;
 }
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 4c98f3975afe..c0821564a3fe 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -970,6 +970,37 @@ int usb_disabled(void)
 }
 EXPORT_SYMBOL_GPL(usb_disabled);
 
+/*
+ * Notifications of device and interface registration
+ */
+static int usb_bus_notify(struct notifier_block *nb, unsigned long action,
+		void *data)
+{
+	struct device *dev = data;
+
+	switch (action) {
+	case BUS_NOTIFY_ADD_DEVICE:
+		if (dev->type == &usb_device_type)
+			(void) usb_create_sysfs_dev_files(to_usb_device(dev));
+		else if (dev->type == &usb_if_device_type)
+			(void) usb_create_sysfs_intf_files(
+					to_usb_interface(dev));
+		break;
+
+	case BUS_NOTIFY_DEL_DEVICE:
+		if (dev->type == &usb_device_type)
+			usb_remove_sysfs_dev_files(to_usb_device(dev));
+		else if (dev->type == &usb_if_device_type)
+			usb_remove_sysfs_intf_files(to_usb_interface(dev));
+		break;
+	}
+	return 0;
+}
+
+static struct notifier_block usb_bus_nb = {
+	.notifier_call = usb_bus_notify,
+};
+
 /*
  * Init
  */
@@ -987,6 +1018,9 @@ static int __init usb_init(void)
 	retval = bus_register(&usb_bus_type);
 	if (retval)
 		goto bus_register_failed;
+	retval = bus_register_notifier(&usb_bus_type, &usb_bus_nb);
+	if (retval)
+		goto bus_notifier_failed;
 	retval = usb_host_init();
 	if (retval)
 		goto host_init_failed;
@@ -1021,6 +1055,8 @@ driver_register_failed:
 major_init_failed:
 	usb_host_cleanup();
 host_init_failed:
+	bus_unregister_notifier(&usb_bus_type, &usb_bus_nb);
+bus_notifier_failed:
 	bus_unregister(&usb_bus_type);
 bus_register_failed:
 	ksuspend_usb_cleanup();
@@ -1044,6 +1080,7 @@ static void __exit usb_exit(void)
 	usb_devio_cleanup();
 	usb_hub_cleanup();
 	usb_host_cleanup();
+	bus_unregister_notifier(&usb_bus_type, &usb_bus_nb);
 	bus_unregister(&usb_bus_type);
 	ksuspend_usb_cleanup();
 }
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 9fb195665fa8..381eae90c3b7 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -6,10 +6,10 @@ extern int usb_create_sysfs_dev_files(struct usb_device *dev);
 extern void usb_remove_sysfs_dev_files(struct usb_device *dev);
 extern int usb_create_sysfs_intf_files(struct usb_interface *intf);
 extern void usb_remove_sysfs_intf_files(struct usb_interface *intf);
-extern int usb_create_ep_files(struct device *parent,
+extern int usb_create_ep_devs(struct device *parent,
 				struct usb_host_endpoint *endpoint,
 				struct usb_device *udev);
-extern void usb_remove_ep_files(struct usb_host_endpoint *endpoint);
+extern void usb_remove_ep_devs(struct usb_host_endpoint *endpoint);
 
 extern void usb_enable_endpoint(struct usb_device *dev,
 		struct usb_host_endpoint *ep);
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 74d0b9990c73..e9d63562325a 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -108,6 +108,7 @@ enum usb_interface_condition {
  *	(in probe()), bound to a driver, or unbinding (in disconnect())
  * @is_active: flag set when the interface is bound and not suspended.
  * @sysfs_files_created: sysfs attributes exist
+ * @ep_devs_created: endpoint child pseudo-devices exist
  * @unregistering: flag set when the interface is being unregistered
  * @needs_remote_wakeup: flag set when the driver requires remote-wakeup
  *	capability during autosuspend.
@@ -169,6 +170,7 @@ struct usb_interface {
 	enum usb_interface_condition condition;		/* state of binding */
 	unsigned is_active:1;		/* the interface is not suspended */
 	unsigned sysfs_files_created:1;	/* the sysfs attributes exist */
+	unsigned ep_devs_created:1;	/* endpoint "devices" exist */
 	unsigned unregistering:1;	/* unregistration is in progress */
 	unsigned needs_remote_wakeup:1;	/* driver requires remote wakeup */
 	unsigned needs_altsetting0:1;	/* switch to altsetting 0 is pending */
-- 
cgit v1.2.3


From 49367d8f1d9f26482cf7089489e90f0afd0a942c Mon Sep 17 00:00:00 2001
From: Ming Lei <tom.leiming@gmail.com>
Date: Fri, 12 Dec 2008 21:38:45 +0800
Subject: USB: mark "reject" field of struct urb as atomic_t

It is enough to protect accesses to reject field of urb
by marking it as atomic_t,also it is the only reason of
existence of usb_reject_lock,so remove the lock to make
code more clean.

Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Acked-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd.c |  6 +++---
 drivers/usb/core/urb.c | 23 +++++------------------
 include/linux/usb.h    |  2 +-
 3 files changed, 9 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index a0079876d74e..3c711db55d86 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -1010,7 +1010,7 @@ int usb_hcd_link_urb_to_ep(struct usb_hcd *hcd, struct urb *urb)
 	spin_lock(&hcd_urb_list_lock);
 
 	/* Check that the URB isn't being killed */
-	if (unlikely(urb->reject)) {
+	if (unlikely(atomic_read(&urb->reject))) {
 		rc = -EPERM;
 		goto done;
 	}
@@ -1340,7 +1340,7 @@ int usb_hcd_submit_urb (struct urb *urb, gfp_t mem_flags)
 		INIT_LIST_HEAD(&urb->urb_list);
 		atomic_dec(&urb->use_count);
 		atomic_dec(&urb->dev->urbnum);
-		if (urb->reject)
+		if (atomic_read(&urb->reject))
 			wake_up(&usb_kill_urb_queue);
 		usb_put_urb(urb);
 	}
@@ -1444,7 +1444,7 @@ void usb_hcd_giveback_urb(struct usb_hcd *hcd, struct urb *urb, int status)
 	urb->status = status;
 	urb->complete (urb);
 	atomic_dec (&urb->use_count);
-	if (unlikely (urb->reject))
+	if (unlikely(atomic_read(&urb->reject)))
 		wake_up (&usb_kill_urb_queue);
 	usb_put_urb (urb);
 }
diff --git a/drivers/usb/core/urb.c b/drivers/usb/core/urb.c
index 1f68af9db3f7..b5e9948698bf 100644
--- a/drivers/usb/core/urb.c
+++ b/drivers/usb/core/urb.c
@@ -10,7 +10,6 @@
 
 #define to_urb(d) container_of(d, struct urb, kref)
 
-static DEFINE_SPINLOCK(usb_reject_lock);
 
 static void urb_destroy(struct kref *kref)
 {
@@ -131,9 +130,7 @@ void usb_anchor_urb(struct urb *urb, struct usb_anchor *anchor)
 	urb->anchor = anchor;
 
 	if (unlikely(anchor->poisoned)) {
-		spin_lock(&usb_reject_lock);
-		urb->reject++;
-		spin_unlock(&usb_reject_lock);
+		atomic_inc(&urb->reject);
 	}
 
 	spin_unlock_irqrestore(&anchor->lock, flags);
@@ -565,16 +562,12 @@ void usb_kill_urb(struct urb *urb)
 	might_sleep();
 	if (!(urb && urb->dev && urb->ep))
 		return;
-	spin_lock_irq(&usb_reject_lock);
-	++urb->reject;
-	spin_unlock_irq(&usb_reject_lock);
+	atomic_inc(&urb->reject);
 
 	usb_hcd_unlink_urb(urb, -ENOENT);
 	wait_event(usb_kill_urb_queue, atomic_read(&urb->use_count) == 0);
 
-	spin_lock_irq(&usb_reject_lock);
-	--urb->reject;
-	spin_unlock_irq(&usb_reject_lock);
+	atomic_dec(&urb->reject);
 }
 EXPORT_SYMBOL_GPL(usb_kill_urb);
 
@@ -606,9 +599,7 @@ void usb_poison_urb(struct urb *urb)
 	might_sleep();
 	if (!(urb && urb->dev && urb->ep))
 		return;
-	spin_lock_irq(&usb_reject_lock);
-	++urb->reject;
-	spin_unlock_irq(&usb_reject_lock);
+	atomic_inc(&urb->reject);
 
 	usb_hcd_unlink_urb(urb, -ENOENT);
 	wait_event(usb_kill_urb_queue, atomic_read(&urb->use_count) == 0);
@@ -617,14 +608,10 @@ EXPORT_SYMBOL_GPL(usb_poison_urb);
 
 void usb_unpoison_urb(struct urb *urb)
 {
-	unsigned long flags;
-
 	if (!urb)
 		return;
 
-	spin_lock_irqsave(&usb_reject_lock, flags);
-	--urb->reject;
-	spin_unlock_irqrestore(&usb_reject_lock, flags);
+	atomic_dec(&urb->reject);
 }
 EXPORT_SYMBOL_GPL(usb_unpoison_urb);
 
diff --git a/include/linux/usb.h b/include/linux/usb.h
index e9d63562325a..4e8654a18250 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -1340,7 +1340,7 @@ struct urb {
 	struct kref kref;		/* reference count of the URB */
 	void *hcpriv;			/* private data for host controller */
 	atomic_t use_count;		/* concurrent submissions counter */
-	u8 reject;			/* submissions will fail */
+	atomic_t reject;		/* submissions will fail */
 	int unlinked;			/* unlink error code */
 
 	/* public: documented fields in the urb that can be used by drivers */
-- 
cgit v1.2.3


From 856395d6e137b4e7194972cb7765f3de6a72ba61 Mon Sep 17 00:00:00 2001
From: Oliver Neukum <oliver@neukum.org>
Date: Thu, 18 Dec 2008 09:17:49 +0100
Subject: USB: extension of anchor API to unpoison an anchor

This extension allows unpoisoning an anchor allowing drivers that
resubmit URBs to reuse an anchor for methods like resume()

Signed-off-by: Oliver Neukum <oneukum@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/urb.c | 20 ++++++++++++++++++++
 include/linux/usb.h    |  1 +
 2 files changed, 21 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/usb/core/urb.c b/drivers/usb/core/urb.c
index b5e9948698bf..58bc5e3c2560 100644
--- a/drivers/usb/core/urb.c
+++ b/drivers/usb/core/urb.c
@@ -678,6 +678,26 @@ void usb_poison_anchored_urbs(struct usb_anchor *anchor)
 }
 EXPORT_SYMBOL_GPL(usb_poison_anchored_urbs);
 
+/**
+ * usb_unpoison_anchored_urbs - let an anchor be used successfully again
+ * @anchor: anchor the requests are bound to
+ *
+ * Reverses the effect of usb_poison_anchored_urbs
+ * the anchor can be used normally after it returns
+ */
+void usb_unpoison_anchored_urbs(struct usb_anchor *anchor)
+{
+	unsigned long flags;
+	struct urb *lazarus;
+
+	spin_lock_irqsave(&anchor->lock, flags);
+	list_for_each_entry(lazarus, &anchor->urb_list, anchor_list) {
+		usb_unpoison_urb(lazarus);
+	}
+	anchor->poisoned = 0;
+	spin_unlock_irqrestore(&anchor->lock, flags);
+}
+EXPORT_SYMBOL_GPL(usb_unpoison_anchored_urbs);
 /**
  * usb_unlink_anchored_urbs - asynchronously cancel transfer requests en masse
  * @anchor: anchor the requests are bound to
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 4e8654a18250..e89639896508 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -1485,6 +1485,7 @@ extern void usb_poison_urb(struct urb *urb);
 extern void usb_unpoison_urb(struct urb *urb);
 extern void usb_kill_anchored_urbs(struct usb_anchor *anchor);
 extern void usb_poison_anchored_urbs(struct usb_anchor *anchor);
+extern void usb_unpoison_anchored_urbs(struct usb_anchor *anchor);
 extern void usb_unlink_anchored_urbs(struct usb_anchor *anchor);
 extern void usb_anchor_urb(struct urb *urb, struct usb_anchor *anchor);
 extern void usb_unanchor_urb(struct urb *urb);
-- 
cgit v1.2.3


From 25ff1c316f6a763f1eefe7f8984b2d8c03888432 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 15 Dec 2008 12:43:41 -0500
Subject: USB: storage: add last-sector hacks

This patch (as1189b) adds some hacks to usb-storage for dealing with
the growing problems involving bad capacity values and last-sector
accesses:

	A new flag, US_FL_CAPACITY_OK, is created to indicate that
	the device is known to report its capacity correctly.  An
	unusual_devs entry for Linux's own File-backed Storage Gadget
	is added with this flag set, since g_file_storage always
	reports the correct capacity and since the capacity need
	not be even (it is determined by the size of the backing
	file).

	An entry in unusual_devs.h which has only the CAPACITY_OK
	flag set shouldn't prejudice libusual, since the device will
	work perfectly well with either usb-storage or ub.  So a
	new macro, COMPLIANT_DEV, is added to let libusual know
	about these entries.

	When a last-sector access succeeds and the total number of
	sectors is odd (the unexpected case, in which guessing that
	the number is even might cause trouble), a WARN is triggered.
	The kerneloops.org project will collect these warnings,
	allowing us to add CAPACITY_OK flags for the devices in
	question before implementing the default-to-even heuristic.
	If users want to prevent the stack dump produced by the WARN,
	they can disable the hack by adding an unusual_devs entry
	for their device with the CAPACITY_OK flag.

	When a last-sector access fails three times in a row and
	neither the FIX_CAPACITY nor the CAPACITY_OK flag is set,
	we assume the last-sector bug is present.  We replace the
	existing status and sense data with values that will cause
	the SCSI core to fail the access immediately rather than
	retry indefinitely.  This should fix the difficulties
	people have been having with Nokia phones.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/libusual.c     |   7 +++
 drivers/usb/storage/scsiglue.c     |   8 +++
 drivers/usb/storage/transport.c    | 110 +++++++++++++++++++++++++++++++++++++
 drivers/usb/storage/unusual_devs.h |  16 +++++-
 drivers/usb/storage/usb.c          |   6 ++
 drivers/usb/storage/usb.h          |   4 ++
 include/linux/usb_usual.h          |   6 +-
 7 files changed, 154 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/usb/storage/libusual.c b/drivers/usb/storage/libusual.c
index d617e8ae6b00..f970b27ba308 100644
--- a/drivers/usb/storage/libusual.c
+++ b/drivers/usb/storage/libusual.c
@@ -46,6 +46,12 @@ static int usu_probe_thread(void *arg);
 { USB_DEVICE_VER(id_vendor, id_product, bcdDeviceMin,bcdDeviceMax), \
   .driver_info = (flags)|(USB_US_TYPE_STOR<<24) }
 
+#define COMPLIANT_DEV(id_vendor, id_product, bcdDeviceMin, bcdDeviceMax, \
+		    vendorName, productName, useProtocol, useTransport, \
+		    initFunction, flags) \
+{ USB_DEVICE_VER(id_vendor, id_product, bcdDeviceMin, bcdDeviceMax), \
+  .driver_info = (flags) }
+
 #define USUAL_DEV(useProto, useTrans, useType) \
 { USB_INTERFACE_INFO(USB_CLASS_MASS_STORAGE, useProto, useTrans), \
   .driver_info = ((useType)<<24) }
@@ -57,6 +63,7 @@ struct usb_device_id storage_usb_ids [] = {
 
 #undef USUAL_DEV
 #undef UNUSUAL_DEV
+#undef COMPLIANT_DEV
 
 MODULE_DEVICE_TABLE(usb, storage_usb_ids);
 EXPORT_SYMBOL_GPL(storage_usb_ids);
diff --git a/drivers/usb/storage/scsiglue.c b/drivers/usb/storage/scsiglue.c
index e9d6c196a7ab..8d78084abf9f 100644
--- a/drivers/usb/storage/scsiglue.c
+++ b/drivers/usb/storage/scsiglue.c
@@ -208,6 +208,14 @@ static int slave_configure(struct scsi_device *sdev)
 		 * sector in a larger then 1 sector read, since the performance
 		 * impact is negible we set this flag for all USB disks */
 		sdev->last_sector_bug = 1;
+
+		/* Enable last-sector hacks for single-target devices using
+		 * the Bulk-only transport, unless we already know the
+		 * capacity will be decremented or is correct. */
+		if (!(us->fflags & (US_FL_FIX_CAPACITY | US_FL_CAPACITY_OK |
+					US_FL_SCM_MULT_TARG)) &&
+				us->protocol == US_PR_BULK)
+			us->use_last_sector_hacks = 1;
 	} else {
 
 		/* Non-disk-type devices don't need to blacklist any pages
diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
index 9cc30afd6d31..1d5438e6363b 100644
--- a/drivers/usb/storage/transport.c
+++ b/drivers/usb/storage/transport.c
@@ -57,6 +57,9 @@
 #include "scsiglue.h"
 #include "debug.h"
 
+#include <linux/blkdev.h>
+#include "../../scsi/sd.h"
+
 
 /***********************************************************************
  * Data transfer routines
@@ -511,6 +514,110 @@ int usb_stor_bulk_transfer_sg(struct us_data* us, unsigned int pipe,
  * Transport routines
  ***********************************************************************/
 
+/* There are so many devices that report the capacity incorrectly,
+ * this routine was written to counteract some of the resulting
+ * problems.
+ */
+static void last_sector_hacks(struct us_data *us, struct scsi_cmnd *srb)
+{
+	struct gendisk *disk;
+	struct scsi_disk *sdkp;
+	u32 sector;
+
+	/* To Report "Medium Error: Record Not Found */
+	static unsigned char record_not_found[18] = {
+		[0]	= 0x70,			/* current error */
+		[2]	= MEDIUM_ERROR,		/* = 0x03 */
+		[7]	= 0x0a,			/* additional length */
+		[12]	= 0x14			/* Record Not Found */
+	};
+
+	/* If last-sector problems can't occur, whether because the
+	 * capacity was already decremented or because the device is
+	 * known to report the correct capacity, then we don't need
+	 * to do anything.
+	 */
+	if (!us->use_last_sector_hacks)
+		return;
+
+	/* Was this command a READ(10) or a WRITE(10)? */
+	if (srb->cmnd[0] != READ_10 && srb->cmnd[0] != WRITE_10)
+		goto done;
+
+	/* Did this command access the last sector? */
+	sector = (srb->cmnd[2] << 24) | (srb->cmnd[3] << 16) |
+			(srb->cmnd[4] << 8) | (srb->cmnd[5]);
+	disk = srb->request->rq_disk;
+	if (!disk)
+		goto done;
+	sdkp = scsi_disk(disk);
+	if (!sdkp)
+		goto done;
+	if (sector + 1 != sdkp->capacity)
+		goto done;
+
+	if (srb->result == SAM_STAT_GOOD && scsi_get_resid(srb) == 0) {
+
+		/* The command succeeded.  If the capacity is odd
+		 * (i.e., if the sector number is even) then the
+		 * "always-even" heuristic would be wrong for this
+		 * device.  Issue a WARN() so that the kerneloops.org
+		 * project will be notified and we will then know to
+		 * mark the device with a CAPACITY_OK flag.  Hopefully
+		 * this will occur for only a few devices.
+		 *
+		 * Use the sign of us->last_sector_hacks to tell whether
+		 * the warning has already been issued; we don't need
+		 * more than one warning per device.
+		 */
+		if (!(sector & 1) && us->use_last_sector_hacks > 0) {
+			unsigned vid = le16_to_cpu(
+					us->pusb_dev->descriptor.idVendor);
+			unsigned pid = le16_to_cpu(
+					us->pusb_dev->descriptor.idProduct);
+			unsigned rev = le16_to_cpu(
+					us->pusb_dev->descriptor.bcdDevice);
+
+			WARN(1, "%s: Successful last sector success at %u, "
+					"device %04x:%04x:%04x\n",
+					sdkp->disk->disk_name, sector,
+					vid, pid, rev);
+			us->use_last_sector_hacks = -1;
+		}
+
+	} else {
+		/* The command failed.  Allow up to 3 retries in case this
+		 * is some normal sort of failure.  After that, assume the
+		 * capacity is wrong and we're trying to access the sector
+		 * beyond the end.  Replace the result code and sense data
+		 * with values that will cause the SCSI core to fail the
+		 * command immediately, instead of going into an infinite
+		 * (or even just a very long) retry loop.
+		 */
+		if (++us->last_sector_retries < 3)
+			return;
+		srb->result = SAM_STAT_CHECK_CONDITION;
+		memcpy(srb->sense_buffer, record_not_found,
+				sizeof(record_not_found));
+
+		/* In theory we might want to issue a WARN() here if the
+		 * capacity is even, since it could indicate the device
+		 * has the READ CAPACITY bug _and_ the real capacity is
+		 * odd.  But it could also indicate that the device
+		 * simply can't access its last sector, a failure mode
+		 * which is surprisingly common.  So no warning.
+		 */
+	}
+
+ done:
+	/* Don't reset the retry counter for TEST UNIT READY commands,
+	 * because they get issued after device resets which might be
+	 * caused by a failed last-sector access.
+	 */
+	if (srb->cmnd[0] != TEST_UNIT_READY)
+		us->last_sector_retries = 0;
+}
+
 /* Invoke the transport and basic error-handling/recovery methods
  *
  * This is used by the protocol layers to actually send the message to
@@ -544,6 +651,7 @@ void usb_stor_invoke_transport(struct scsi_cmnd *srb, struct us_data *us)
 	/* if the transport provided its own sense data, don't auto-sense */
 	if (result == USB_STOR_TRANSPORT_NO_SENSE) {
 		srb->result = SAM_STAT_CHECK_CONDITION;
+		last_sector_hacks(us, srb);
 		return;
 	}
 
@@ -705,6 +813,7 @@ void usb_stor_invoke_transport(struct scsi_cmnd *srb, struct us_data *us)
 			scsi_bufflen(srb) - scsi_get_resid(srb) < srb->underflow)
 		srb->result = (DID_ERROR << 16) | (SUGGEST_RETRY << 24);
 
+	last_sector_hacks(us, srb);
 	return;
 
 	/* Error and abort processing: try to resynchronize with the device
@@ -732,6 +841,7 @@ void usb_stor_invoke_transport(struct scsi_cmnd *srb, struct us_data *us)
 		us->transport_reset(us);
 	}
 	clear_bit(US_FLIDX_RESETTING, &us->dflags);
+	last_sector_hacks(us, srb);
 }
 
 /* Stop the current URB transfer */
diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index 0330ed53ec1c..035bbc5d8231 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -27,7 +27,8 @@
 
 /* IMPORTANT NOTE: This file must be included in another file which does
  * the following thing for it to work:
- * The macro UNUSUAL_DEV() must be defined before this file is included
+ * The UNUSUAL_DEV, COMPLIANT_DEV, and USUAL_DEV macros must be defined
+ * before this file is included.
  */
 
 /* If you edit this file, please try to keep it sorted first by VendorID,
@@ -46,6 +47,12 @@
  * <usb-storage@lists.one-eyed-alien.net>
  */
 
+/* Note: If you add an entry only in order to set the CAPACITY_OK flag,
+ * use the COMPLIANT_DEV macro instead of UNUSUAL_DEV.  This is
+ * because such entries mark devices which actually work correctly,
+ * as opposed to devices that do something strangely or wrongly.
+ */
+
 /* patch submitted by Vivian Bregier <Vivian.Bregier@imag.fr>
  */
 UNUSUAL_DEV(  0x03eb, 0x2002, 0x0100, 0x0100,
@@ -704,6 +711,13 @@ UNUSUAL_DEV(  0x0525, 0xa140, 0x0100, 0x0100,
 		US_SC_8070, US_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY ),
 
+/* Added by Alan Stern <stern@rowland.harvard.edu> */
+COMPLIANT_DEV(0x0525, 0xa4a5, 0x0000, 0x9999,
+		"Linux",
+		"File-backed Storage Gadget",
+		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		US_FL_CAPACITY_OK ),
+
 /* Yakumo Mega Image 37
  * Submitted by Stephan Fuhrmann <atomenergie@t-online.de> */
 UNUSUAL_DEV(  0x052b, 0x1801, 0x0100, 0x0100,
diff --git a/drivers/usb/storage/usb.c b/drivers/usb/storage/usb.c
index ce0b580db5ea..80e234bf4e50 100644
--- a/drivers/usb/storage/usb.c
+++ b/drivers/usb/storage/usb.c
@@ -134,6 +134,8 @@ static struct quirks_entry *quirks_list, *quirks_end;
 { USB_DEVICE_VER(id_vendor, id_product, bcdDeviceMin,bcdDeviceMax), \
   .driver_info = (flags)|(USB_US_TYPE_STOR<<24) }
 
+#define COMPLIANT_DEV	UNUSUAL_DEV
+
 #define USUAL_DEV(useProto, useTrans, useType) \
 { USB_INTERFACE_INFO(USB_CLASS_MASS_STORAGE, useProto, useTrans), \
   .driver_info = (USB_US_TYPE_STOR<<24) }
@@ -142,6 +144,7 @@ static struct usb_device_id storage_usb_ids [] = {
 
 #	include "unusual_devs.h"
 #undef UNUSUAL_DEV
+#undef COMPLIANT_DEV
 #undef USUAL_DEV
 	/* Terminating entry */
 	{ }
@@ -172,6 +175,8 @@ MODULE_DEVICE_TABLE (usb, storage_usb_ids);
 	.initFunction = init_function,	\
 }
 
+#define COMPLIANT_DEV	UNUSUAL_DEV
+
 #define USUAL_DEV(use_protocol, use_transport, use_type) \
 { \
 	.useProtocol = use_protocol,	\
@@ -181,6 +186,7 @@ MODULE_DEVICE_TABLE (usb, storage_usb_ids);
 static struct us_unusual_dev us_unusual_dev_list[] = {
 #	include "unusual_devs.h" 
 #	undef UNUSUAL_DEV
+#	undef COMPLIANT_DEV
 #	undef USUAL_DEV
 
 	/* Terminating entry */
diff --git a/drivers/usb/storage/usb.h b/drivers/usb/storage/usb.h
index e4674fc715e6..65e674e4be99 100644
--- a/drivers/usb/storage/usb.h
+++ b/drivers/usb/storage/usb.h
@@ -154,6 +154,10 @@ struct us_data {
 #ifdef CONFIG_PM
 	pm_hook			suspend_resume_hook;
 #endif
+
+	/* hacks for READ CAPACITY bug handling */
+	int			use_last_sector_hacks;
+	int			last_sector_retries;
 };
 
 /* Convert between us_data and the corresponding Scsi_Host */
diff --git a/include/linux/usb_usual.h b/include/linux/usb_usual.h
index 998e5cbbf29e..1eea1ab68dc4 100644
--- a/include/linux/usb_usual.h
+++ b/include/linux/usb_usual.h
@@ -53,8 +53,10 @@
 		/* Sets max_sectors to arch min */		\
 	US_FLAG(BULK_IGNORE_TAG,0x00004000)			\
 		/* Ignore tag mismatch in bulk operations */    \
-	US_FLAG(SANE_SENSE,     0x00008000)
-		/* Sane Sense (> 18 bytes) */
+	US_FLAG(SANE_SENSE,     0x00008000)			\
+		/* Sane Sense (> 18 bytes) */			\
+	US_FLAG(CAPACITY_OK,	0x00010000)			\
+		/* READ CAPACITY response is correct */
 
 #define US_FLAG(name, value)	US_FL_##name = value ,
 enum { US_DO_ALL_FLAGS };
-- 
cgit v1.2.3


From 338b67b0c1a97ca705023a8189cf41aa0828d294 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 14 Aug 2008 09:37:34 -0700
Subject: USB: remove warn() macro from usb.h

USB should not be having it's own printk macros, so remove warn() and
use the system-wide standard of dev_warn() wherever possible.  In the
few places that will not work out, use a basic printk().

Now that all in-tree users are gone, remove the macro.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/usb.h b/include/linux/usb.h
index e89639896508..28f68f587793 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -1744,8 +1744,6 @@ extern void usb_unregister_notify(struct notifier_block *nb);
 	format "\n" , ## arg)
 #define info(format, arg...) printk(KERN_INFO KBUILD_MODNAME ": " \
 	format "\n" , ## arg)
-#define warn(format, arg...) printk(KERN_WARNING KBUILD_MODNAME ": " \
-	format "\n" , ## arg)
 
 #endif  /* __KERNEL__ */
 
-- 
cgit v1.2.3


From 34c65d82e02147331701c7795e3144d511adf4e9 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 18 Aug 2008 13:21:04 -0700
Subject: USB: remove info() macro from usb.h

USB should not be having it's own printk macros, so remove info() and
use the system-wide standard of dev_info() wherever possible.

No one in the tree is using the macro, so it can now be removed.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/usb.h b/include/linux/usb.h
index 28f68f587793..85ee9be9361e 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -1742,8 +1742,6 @@ extern void usb_unregister_notify(struct notifier_block *nb);
 
 #define err(format, arg...) printk(KERN_ERR KBUILD_MODNAME ": " \
 	format "\n" , ## arg)
-#define info(format, arg...) printk(KERN_INFO KBUILD_MODNAME ": " \
-	format "\n" , ## arg)
 
 #endif  /* __KERNEL__ */
 
-- 
cgit v1.2.3


From 5e07878787ad07361571150230cc3a8d522ae046 Mon Sep 17 00:00:00 2001
From: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Date: Sat, 20 Dec 2008 16:57:39 -0800
Subject: debugfs: add helpers for exporting a size_t simple value
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In the same spirit as debugfs_create_*(), introduce helpers for
exporting size_t values over debugfs.

The only trick done is that the format verifier is kept at %llu
instead of %zu; otherwise type warnings would pop up:

format ‘%zu’ expects type ‘size_t’, but argument 2 has type ‘long long unsigned int’

There is no real way to fix this one--however, we can consider %llu
and %zu to be compatible if we consider that we are using the same for
validating in debugfs_create_{x,u}{8,16,32}().

Signed-off-by: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/debugfs/file.c       | 32 ++++++++++++++++++++++++++++++++
 include/linux/debugfs.h |  2 ++
 2 files changed, 34 insertions(+)

(limited to 'include/linux')

diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c
index 159a5efd6a8a..33a90120f6ad 100644
--- a/fs/debugfs/file.c
+++ b/fs/debugfs/file.c
@@ -294,6 +294,38 @@ struct dentry *debugfs_create_x32(const char *name, mode_t mode,
 }
 EXPORT_SYMBOL_GPL(debugfs_create_x32);
 
+
+static int debugfs_size_t_set(void *data, u64 val)
+{
+	*(size_t *)data = val;
+	return 0;
+}
+static int debugfs_size_t_get(void *data, u64 *val)
+{
+	*val = *(size_t *)data;
+	return 0;
+}
+DEFINE_SIMPLE_ATTRIBUTE(fops_size_t, debugfs_size_t_get, debugfs_size_t_set,
+			"%llu\n");	/* %llu and %zu are more or less the same */
+
+/**
+ * debugfs_create_size_t - create a debugfs file that is used to read and write an size_t value
+ * @name: a pointer to a string containing the name of the file to create.
+ * @mode: the permission that the file should have
+ * @parent: a pointer to the parent dentry for this file.  This should be a
+ *          directory dentry if set.  If this parameter is %NULL, then the
+ *          file will be created in the root of the debugfs filesystem.
+ * @value: a pointer to the variable that the file should read to and write
+ *         from.
+ */
+struct dentry *debugfs_create_size_t(const char *name, mode_t mode,
+				     struct dentry *parent, size_t *value)
+{
+	return debugfs_create_file(name, mode, parent, value, &fops_size_t);
+}
+EXPORT_SYMBOL_GPL(debugfs_create_size_t);
+
+
 static ssize_t read_file_bool(struct file *file, char __user *user_buf,
 			      size_t count, loff_t *ppos)
 {
diff --git a/include/linux/debugfs.h b/include/linux/debugfs.h
index e1a6c046cea3..23936b16426b 100644
--- a/include/linux/debugfs.h
+++ b/include/linux/debugfs.h
@@ -63,6 +63,8 @@ struct dentry *debugfs_create_x16(const char *name, mode_t mode,
 				  struct dentry *parent, u16 *value);
 struct dentry *debugfs_create_x32(const char *name, mode_t mode,
 				  struct dentry *parent, u32 *value);
+struct dentry *debugfs_create_size_t(const char *name, mode_t mode,
+				     struct dentry *parent, size_t *value);
 struct dentry *debugfs_create_bool(const char *name, mode_t mode,
 				  struct dentry *parent, u32 *value);
 
-- 
cgit v1.2.3


From ace22f0881e1333d0c55ddf484e5352fe03a806a Mon Sep 17 00:00:00 2001
From: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Date: Sat, 20 Dec 2008 16:57:33 -0800
Subject: wimax: headers for kernel API and user space interaction

Definitions for the user/kernel API protocol through generic
netlink. User space can copy it verbatim and use it.

Kernel API definition declares the main data types and calls for the
drivers to integrate into the WiMAX stack. Provides usage
documentation.

Signed-off-by: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/wimax.h | 234 +++++++++++++++++++++++
 include/net/wimax.h   | 520 ++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 754 insertions(+)
 create mode 100644 include/linux/wimax.h
 create mode 100644 include/net/wimax.h

(limited to 'include/linux')

diff --git a/include/linux/wimax.h b/include/linux/wimax.h
new file mode 100644
index 000000000000..c89de7f4e5b9
--- /dev/null
+++ b/include/linux/wimax.h
@@ -0,0 +1,234 @@
+/*
+ * Linux WiMax
+ * API for user space
+ *
+ *
+ * Copyright (C) 2007-2008 Intel Corporation. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ *   * Redistributions of source code must retain the above copyright
+ *     notice, this list of conditions and the following disclaimer.
+ *   * Redistributions in binary form must reproduce the above copyright
+ *     notice, this list of conditions and the following disclaimer in
+ *     the documentation and/or other materials provided with the
+ *     distribution.
+ *   * Neither the name of Intel Corporation nor the names of its
+ *     contributors may be used to endorse or promote products derived
+ *     from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *
+ * Intel Corporation <linux-wimax@intel.com>
+ * Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
+ *  - Initial implementation
+ *
+ *
+ * This file declares the user/kernel protocol that is spoken over
+ * Generic Netlink, as well as any type declaration that is to be used
+ * by kernel and user space.
+ *
+ * It is intended for user space to clone it verbatim to use it as a
+ * primary reference for definitions.
+ *
+ * Stuff intended for kernel usage as well as full protocol and stack
+ * documentation is rooted in include/net/wimax.h.
+ */
+
+#ifndef __LINUX__WIMAX_H__
+#define __LINUX__WIMAX_H__
+
+#include <linux/types.h>
+
+enum {
+	/**
+	 * Version of the interface (unsigned decimal, MMm, max 25.5)
+	 * M - Major: change if removing or modifying an existing call.
+	 * m - minor: change when adding a new call
+	 */
+	WIMAX_GNL_VERSION = 00,
+	/* Generic NetLink attributes */
+	WIMAX_GNL_ATTR_INVALID = 0x00,
+	WIMAX_GNL_ATTR_MAX = 10,
+};
+
+
+/*
+ * Generic NetLink operations
+ *
+ * Most of these map to an API call; _OP_ stands for operation, _RP_
+ * for reply and _RE_ for report (aka: signal).
+ */
+enum {
+	WIMAX_GNL_OP_MSG_FROM_USER,	/* User to kernel message */
+	WIMAX_GNL_OP_MSG_TO_USER,	/* Kernel to user message */
+	WIMAX_GNL_OP_RFKILL,	/* Run wimax_rfkill() */
+	WIMAX_GNL_OP_RESET,	/* Run wimax_rfkill() */
+	WIMAX_GNL_RE_STATE_CHANGE,	/* Report: status change */
+};
+
+
+/* Message from user / to user */
+enum {
+	WIMAX_GNL_MSG_IFIDX = 1,
+	WIMAX_GNL_MSG_PIPE_NAME,
+	WIMAX_GNL_MSG_DATA,
+};
+
+
+/*
+ * wimax_rfkill()
+ *
+ * The state of the radio (ON/OFF) is mapped to the rfkill subsystem's
+ * switch state (DISABLED/ENABLED).
+ */
+enum wimax_rf_state {
+	WIMAX_RF_OFF = 0,	/* Radio is off, rfkill on/enabled */
+	WIMAX_RF_ON = 1,	/* Radio is on, rfkill off/disabled */
+	WIMAX_RF_QUERY = 2,
+};
+
+/* Attributes */
+enum {
+	WIMAX_GNL_RFKILL_IFIDX = 1,
+	WIMAX_GNL_RFKILL_STATE,
+};
+
+
+/* Attributes for wimax_reset() */
+enum {
+	WIMAX_GNL_RESET_IFIDX = 1,
+};
+
+
+/*
+ * Attributes for the Report State Change
+ *
+ * For now we just have the old and new states; new attributes might
+ * be added later on.
+ */
+enum {
+	WIMAX_GNL_STCH_IFIDX = 1,
+	WIMAX_GNL_STCH_STATE_OLD,
+	WIMAX_GNL_STCH_STATE_NEW,
+};
+
+
+/**
+ * enum wimax_st - The different states of a WiMAX device
+ * @__WIMAX_ST_NULL: The device structure has been allocated and zeroed,
+ *     but still wimax_dev_add() hasn't been called. There is no state.
+ *
+ * @WIMAX_ST_DOWN: The device has been registered with the WiMAX and
+ *     networking stacks, but it is not initialized (normally that is
+ *     done with 'ifconfig DEV up' [or equivalent], which can upload
+ *     firmware and enable communications with the device).
+ *     In this state, the device is powered down and using as less
+ *     power as possible.
+ *     This state is the default after a call to wimax_dev_add(). It
+ *     is ok to have drivers move directly to %WIMAX_ST_UNINITIALIZED
+ *     or %WIMAX_ST_RADIO_OFF in _probe() after the call to
+ *     wimax_dev_add().
+ *     It is recommended that the driver leaves this state when
+ *     calling 'ifconfig DEV up' and enters it back on 'ifconfig DEV
+ *     down'.
+ *
+ * @__WIMAX_ST_QUIESCING: The device is being torn down, so no API
+ *     operations are allowed to proceed except the ones needed to
+ *     complete the device clean up process.
+ *
+ * @WIMAX_ST_UNINITIALIZED: [optional] Communication with the device
+ *     is setup, but the device still requires some configuration
+ *     before being operational.
+ *     Some WiMAX API calls might work.
+ *
+ * @WIMAX_ST_RADIO_OFF: The device is fully up; radio is off (wether
+ *     by hardware or software switches).
+ *     It is recommended to always leave the device in this state
+ *     after initialization.
+ *
+ * @WIMAX_ST_READY: The device is fully up and radio is on.
+ *
+ * @WIMAX_ST_SCANNING: [optional] The device has been instructed to
+ *     scan. In this state, the device cannot be actively connected to
+ *     a network.
+ *
+ * @WIMAX_ST_CONNECTING: The device is connecting to a network. This
+ *     state exists because in some devices, the connect process can
+ *     include a number of negotiations between user space, kernel
+ *     space and the device. User space needs to know what the device
+ *     is doing. If the connect sequence in a device is atomic and
+ *     fast, the device can transition directly to CONNECTED
+ *
+ * @WIMAX_ST_CONNECTED: The device is connected to a network.
+ *
+ * @__WIMAX_ST_INVALID: This is an invalid state used to mark the
+ *     maximum numeric value of states.
+ *
+ * Description:
+ *
+ * Transitions from one state to another one are atomic and can only
+ * be caused in kernel space with wimax_state_change(). To read the
+ * state, use wimax_state_get().
+ *
+ * States starting with __ are internal and shall not be used or
+ * referred to by drivers or userspace. They look ugly, but that's the
+ * point -- if any use is made non-internal to the stack, it is easier
+ * to catch on review.
+ *
+ * All API operations [with well defined exceptions] will take the
+ * device mutex before starting and then check the state. If the state
+ * is %__WIMAX_ST_NULL, %WIMAX_ST_DOWN, %WIMAX_ST_UNINITIALIZED or
+ * %__WIMAX_ST_QUIESCING, it will drop the lock and quit with
+ * -%EINVAL, -%ENOMEDIUM, -%ENOTCONN or -%ESHUTDOWN.
+ *
+ * The order of the definitions is important, so we can do numerical
+ * comparisons (eg: < %WIMAX_ST_RADIO_OFF means the device is not ready
+ * to operate).
+ */
+/*
+ * The allowed state transitions are described in the table below
+ * (states in rows can go to states in columns where there is an X):
+ *
+ *                                  UNINI   RADIO READY SCAN CONNEC CONNEC
+ *             NULL DOWN QUIESCING TIALIZED  OFF        NING  TING   TED
+ * NULL         -    x
+ * DOWN              -      x        x       x
+ * QUIESCING         x      -
+ * UNINITIALIZED            x        -       x
+ * RADIO_OFF                x                -     x
+ * READY                    x                x     -     x     x      x
+ * SCANNING                 x                x     x     -     x      x
+ * CONNECTING               x                x     x     x     -      x
+ * CONNECTED                x                x     x                  -
+ *
+ * This table not available in kernel-doc because the formatting messes it up.
+ */
+ enum wimax_st {
+	__WIMAX_ST_NULL = 0,
+	WIMAX_ST_DOWN,
+	__WIMAX_ST_QUIESCING,
+	WIMAX_ST_UNINITIALIZED,
+	WIMAX_ST_RADIO_OFF,
+	WIMAX_ST_READY,
+	WIMAX_ST_SCANNING,
+	WIMAX_ST_CONNECTING,
+	WIMAX_ST_CONNECTED,
+	__WIMAX_ST_INVALID			/* Always keep last */
+};
+
+
+#endif /* #ifndef __LINUX__WIMAX_H__ */
diff --git a/include/net/wimax.h b/include/net/wimax.h
new file mode 100644
index 000000000000..1602614fdaf9
--- /dev/null
+++ b/include/net/wimax.h
@@ -0,0 +1,520 @@
+/*
+ * Linux WiMAX
+ * Kernel space API for accessing WiMAX devices
+ *
+ *
+ * Copyright (C) 2007-2008 Intel Corporation <linux-wimax@intel.com>
+ * Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ *
+ *
+ * The WiMAX stack provides an API for controlling and managing the
+ * system's WiMAX devices. This API affects the control plane; the
+ * data plane is accessed via the network stack (netdev).
+ *
+ * Parts of the WiMAX stack API and notifications are exported to
+ * user space via Generic Netlink. In user space, libwimax (part of
+ * the wimax-tools package) provides a shim layer for accessing those
+ * calls.
+ *
+ * The API is standarized for all WiMAX devices and different drivers
+ * implement the backend support for it. However, device-specific
+ * messaging pipes are provided that can be used to issue commands and
+ * receive notifications in free form.
+ *
+ * Currently the messaging pipes are the only means of control as it
+ * is not known (due to the lack of more devices in the market) what
+ * will be a good abstraction layer. Expect this to change as more
+ * devices show in the market. This API is designed to be growable in
+ * order to address this problem.
+ *
+ * USAGE
+ *
+ * Embed a `struct wimax_dev` at the beginning of the the device's
+ * private structure, initialize and register it. For details, see
+ * `struct wimax_dev`s documentation.
+ *
+ * Once this is done, wimax-tools's libwimaxll can be used to
+ * communicate with the driver from user space. You user space
+ * application does not have to forcibily use libwimaxll and can talk
+ * the generic netlink protocol directly if desired.
+ *
+ * Remember this is a very low level API that will to provide all of
+ * WiMAX features. Other daemons and services running in user space
+ * are the expected clients of it. They offer a higher level API that
+ * applications should use (an example of this is the Intel's WiMAX
+ * Network Service for the i2400m).
+ *
+ * DESIGN
+ *
+ * Although not set on final stone, this very basic interface is
+ * mostly completed. Remember this is meant to grow as new common
+ * operations are decided upon. New operations will be added to the
+ * interface, intent being on keeping backwards compatibility as much
+ * as possible.
+ *
+ * This layer implements a set of calls to control a WiMAX device,
+ * exposing a frontend to the rest of the kernel and user space (via
+ * generic netlink) and a backend implementation in the driver through
+ * function pointers.
+ *
+ * WiMAX devices have a state, and a kernel-only API allows the
+ * drivers to manipulate that state. State transitions are atomic, and
+ * only some of them are allowed (see `enum wimax_st`).
+ *
+ * Most API calls will set the state automatically; in most cases
+ * drivers have to only report state changes due to external
+ * conditions.
+ *
+ * All API operations are 'atomic', serialized thorough a mutex in the
+ * `struct wimax_dev`.
+ *
+ * EXPORTING TO USER SPACE THROUGH GENERIC NETLINK
+ *
+ * The API is exported to user space using generic netlink (other
+ * methods can be added as needed).
+ *
+ * There is a Generic Netlink Family named "WiMAX", where interfaces
+ * supporting the WiMAX interface receive commands and broadcast their
+ * signals over a multicast group named "msg".
+ *
+ * Mapping to the source/destination interface is done by an interface
+ * index attribute.
+ *
+ * For user-to-kernel traffic (commands) we use a function call
+ * marshalling mechanism, where a message X with attributes A, B, C
+ * sent from user space to kernel space means executing the WiMAX API
+ * call wimax_X(A, B, C), sending the results back as a message.
+ *
+ * Kernel-to-user (notifications or signals) communication is sent
+ * over multicast groups. This allows to have multiple applications
+ * monitoring them.
+ *
+ * Each command/signal gets assigned it's own attribute policy. This
+ * way the validator will verify that all the attributes in there are
+ * only the ones that should be for each command/signal. Thing of an
+ * attribute mapping to a type+argumentname for each command/signal.
+ *
+ * If we had a single policy for *all* commands/signals, after running
+ * the validator we'd have to check "does this attribute belong in
+ * here"?  for each one. It can be done manually, but it's just easier
+ * to have the validator do that job with multiple policies. As well,
+ * it makes it easier to later expand each command/signal signature
+ * without affecting others and keeping the namespace more or less
+ * sane. Not that it is too complicated, but it makes it even easier.
+ *
+ * No state information is maintained in the kernel for each user
+ * space connection (the connection is stateless).
+ *
+ * TESTING FOR THE INTERFACE AND VERSIONING
+ *
+ * If network interface X is a WiMAX device, there will be a Generic
+ * Netlink family named "WiMAX X" and the device will present a
+ * "wimax" directory in it's network sysfs directory
+ * (/sys/class/net/DEVICE/wimax) [used by HAL].
+ *
+ * The inexistence of any of these means the device does not support
+ * this WiMAX API.
+ *
+ * By querying the generic netlink controller, versioning information
+ * and the multicast groups available can be found. Applications using
+ * the interface can either rely on that or use the generic netlink
+ * controller to figure out which generic netlink commands/signals are
+ * supported.
+ *
+ * NOTE: this versioning is a last resort to avoid hard
+ *    incompatibilities. It is the intention of the design of this
+ *    stack not to introduce backward incompatible changes.
+ *
+ * The version code has to fit in one byte (restrictions imposed by
+ * generic netlink); we use `version / 10` for the major version and
+ * `version % 10` for the minor. This gives 9 minors for each major
+ * and 25 majors.
+ *
+ * The version change protocol is as follow:
+ *
+ * - Major versions: needs to be increased if an existing message/API
+ *   call is changed or removed. Doesn't need to be changed if a new
+ *   message is added.
+ *
+ * - Minor version: needs to be increased if new messages/API calls are
+ *   being added or some other consideration that doesn't impact the
+ *   user-kernel interface too much (like some kind of bug fix) and
+ *   that is kind of left up in the air to common sense.
+ *
+ * User space code should not try to work if the major version it was
+ * compiled for differs from what the kernel offers. As well, if the
+ * minor version of the kernel interface is lower than the one user
+ * space is expecting (the one it was compiled for), the kernel
+ * might be missing API calls; user space shall be ready to handle
+ * said condition. Use the generic netlink controller operations to
+ * find which ones are supported and which not.
+ *
+ * libwimaxll:wimaxll_open() takes care of checking versions.
+ *
+ * THE OPERATIONS:
+ *
+ * Each operation is defined in its on file (drivers/net/wimax/op-*.c)
+ * for clarity. The parts needed for an operation are:
+ *
+ *  - a function pointer in `struct wimax_dev`: optional, as the
+ *    operation might be implemented by the stack and not by the
+ *    driver.
+ *
+ *    All function pointers are named wimax_dev->op_*(), and drivers
+ *    must implement them except where noted otherwise.
+ *
+ *  - When exported to user space, a `struct nla_policy` to define the
+ *    attributes of the generic netlink command and a `struct genl_ops`
+ *    to define the operation.
+ *
+ * All the declarations for the operation codes (WIMAX_GNL_OP_<NAME>)
+ * and generic netlink attributes (WIMAX_GNL_<NAME>_*) are declared in
+ * include/linux/wimax.h; this file is intended to be cloned by user
+ * space to gain access to those declarations.
+ *
+ * A few caveats to remember:
+ *
+ *  - Need to define attribute numbers starting in 1; otherwise it
+ *    fails.
+ *
+ *  - the `struct genl_family` requires a maximum attribute id; when
+ *    defining the `struct nla_policy` for each message, it has to have
+ *    an array size of WIMAX_GNL_ATTR_MAX+1.
+ *
+ * THE PIPE INTERFACE:
+ *
+ * This interface is kept intentionally simple. The driver can send
+ * and receive free-form messages to/from user space through a
+ * pipe. See drivers/net/wimax/op-msg.c for details.
+ *
+ * The kernel-to-user messages are sent with
+ * wimax_msg(). user-to-kernel messages are delivered via
+ * wimax_dev->op_msg_from_user().
+ *
+ * RFKILL:
+ *
+ * RFKILL support is built into the wimax_dev layer; the driver just
+ * needs to call wimax_report_rfkill_{hw,sw}() to inform of changes in
+ * the hardware or software RF kill switches. When the stack wants to
+ * turn the radio off, it will call wimax_dev->op_rfkill_sw_toggle(),
+ * which the driver implements.
+ *
+ * User space can set the software RF Kill switch by calling
+ * wimax_rfkill().
+ *
+ * The code for now only supports devices that don't require polling;
+ * If the device needs to be polled, create a self-rearming delayed
+ * work struct for polling or look into adding polled support to the
+ * WiMAX stack.
+ *
+ * When initializing the hardware (_probe), after calling
+ * wimax_dev_add(), query the device for it's RF Kill switches status
+ * and feed it back to the WiMAX stack using
+ * wimax_report_rfkill_{hw,sw}(). If any switch is missing, always
+ * report it as ON.
+ *
+ * NOTE: the wimax stack uses an inverted terminology to that of the
+ * RFKILL subsystem:
+ *
+ *  - ON: radio is ON, RFKILL is DISABLED or OFF.
+ *  - OFF: radio is OFF, RFKILL is ENABLED or ON.
+ *
+ * MISCELLANEOUS OPS:
+ *
+ * wimax_reset() can be used to reset the device to power on state; by
+ * default it issues a warm reset that maintains the same device
+ * node. If that is not possible, it falls back to a cold reset
+ * (device reconnect). The driver implements the backend to this
+ * through wimax_dev->op_reset().
+ */
+
+#ifndef __NET__WIMAX_H__
+#define __NET__WIMAX_H__
+#ifdef __KERNEL__
+
+#include <linux/wimax.h>
+#include <net/genetlink.h>
+#include <linux/netdevice.h>
+
+struct net_device;
+struct genl_info;
+struct wimax_dev;
+struct input_dev;
+
+/**
+ * struct wimax_dev - Generic WiMAX device
+ *
+ * @net_dev: [fill] Pointer to the &struct net_device this WiMAX
+ *     device implements.
+ *
+ * @op_msg_from_user: [fill] Driver-specific operation to
+ *     handle a raw message from user space to the driver. The
+ *     driver can send messages to user space using with
+ *     wimax_msg_to_user().
+ *
+ * @op_rfkill_sw_toggle: [fill] Driver-specific operation to act on
+ *     userspace (or any other agent) requesting the WiMAX device to
+ *     change the RF Kill software switch (WIMAX_RF_ON or
+ *     WIMAX_RF_OFF).
+ *     If such hardware support is not present, it is assumed the
+ *     radio cannot be switched off and it is always on (and the stack
+ *     will error out when trying to switch it off). In such case,
+ *     this function pointer can be left as NULL.
+ *
+ * @op_reset: [fill] Driver specific operation to reset the
+ *     device.
+ *     This operation should always attempt first a warm reset that
+ *     does not disconnect the device from the bus and return 0.
+ *     If that fails, it should resort to some sort of cold or bus
+ *     reset (even if it implies a bus disconnection and device
+ *     dissapearance). In that case, -ENODEV should be returned to
+ *     indicate the device is gone.
+ *     This operation has to be synchronous, and return only when the
+ *     reset is complete. In case of having had to resort to bus/cold
+ *     reset implying a device disconnection, the call is allowed to
+ *     return inmediately.
+ *     NOTE: wimax_dev->mutex is NOT locked when this op is being
+ *     called; however, wimax_dev->mutex_reset IS locked to ensure
+ *     serialization of calls to wimax_reset().
+ *     See wimax_reset()'s documentation.
+ *
+ * @name: [fill] A way to identify this device. We need to register a
+ *     name with many subsystems (input for RFKILL, workqueue
+ *     creation, etc). We can't use the network device name as that
+ *     might change and in some instances we don't know it yet (until
+ *     we don't call register_netdev()). So we generate an unique one
+ *     using the driver name and device bus id, place it here and use
+ *     it across the board. Recommended naming:
+ *     DRIVERNAME-BUSNAME:BUSID (dev->bus->name, dev->bus_id).
+ *
+ * @id_table_node: [private] link to the list of wimax devices kept by
+ *     id-table.c. Protected by it's own spinlock.
+ *
+ * @mutex: [private] Serializes all concurrent access and execution of
+ *     operations.
+ *
+ * @mutex_reset: [private] Serializes reset operations. Needs to be a
+ *     different mutex because as part of the reset operation, the
+ *     driver has to call back into the stack to do things such as
+ *     state change, that require wimax_dev->mutex.
+ *
+ * @state: [private] Current state of the WiMAX device.
+ *
+ * @rfkill: [private] integration into the RF-Kill infrastructure.
+ *
+ * @rfkill_input: [private] virtual input device to process the
+ *     hardware RF Kill switches.
+ *
+ * @rf_sw: [private] State of the software radio switch (OFF/ON)
+ *
+ * @rf_hw: [private] State of the hardware radio switch (OFF/ON)
+ *
+ * Description:
+ * This structure defines a common interface to access all WiMAX
+ * devices from different vendors and provides a common API as well as
+ * a free-form device-specific messaging channel.
+ *
+ * Usage:
+ *  1. Embed a &struct wimax_dev at *the beginning* the network
+ *     device structure so that netdev_priv() points to it.
+ *
+ *  2. memset() it to zero
+ *
+ *  3. Initialize with wimax_dev_init(). This will leave the WiMAX
+ *     device in the %__WIMAX_ST_NULL state.
+ *
+ *  4. Fill all the fields marked with [fill]; once called
+ *     wimax_dev_add(), those fields CANNOT be modified.
+ *
+ *  5. Call wimax_dev_add() *after* registering the network
+ *     device. This will leave the WiMAX device in the %WIMAX_ST_DOWN
+ *     state.
+ *     Protect the driver's net_device->open() against succeeding if
+ *     the wimax device state is lower than %WIMAX_ST_DOWN.
+ *
+ *  6. Select when the device is going to be turned on/initialized;
+ *     for example, it could be initialized on 'ifconfig up' (when the
+ *     netdev op 'open()' is called on the driver).
+ *
+ * When the device is initialized (at `ifconfig up` time, or right
+ * after calling wimax_dev_add() from _probe(), make sure the
+ * following steps are taken
+ *
+ *  a. Move the device to %WIMAX_ST_UNINITIALIZED. This is needed so
+ *     some API calls that shouldn't work until the device is ready
+ *     can be blocked.
+ *
+ *  b. Initialize the device. Make sure to turn the SW radio switch
+ *     off and move the device to state %WIMAX_ST_RADIO_OFF when
+ *     done. When just initialized, a device should be left in RADIO
+ *     OFF state until user space devices to turn it on.
+ *
+ *  c. Query the device for the state of the hardware rfkill switch
+ *     and call wimax_rfkill_report_hw() and wimax_rfkill_report_sw()
+ *     as needed. See below.
+ *
+ * wimax_dev_rm() undoes before unregistering the network device. Once
+ * wimax_dev_add() is called, the driver can get called on the
+ * wimax_dev->op_* function pointers
+ *
+ * CONCURRENCY:
+ *
+ * The stack provides a mutex for each device that will disallow API
+ * calls happening concurrently; thus, op calls into the driver
+ * through the wimax_dev->op*() function pointers will always be
+ * serialized and *never* concurrent.
+ *
+ * For locking, take wimax_dev->mutex is taken; (most) operations in
+ * the API have to check for wimax_dev_is_ready() to return 0 before
+ * continuing (this is done internally).
+ *
+ * REFERENCE COUNTING:
+ *
+ * The WiMAX device is reference counted by the associated network
+ * device. The only operation that can be used to reference the device
+ * is wimax_dev_get_by_genl_info(), and the reference it acquires has
+ * to be released with dev_put(wimax_dev->net_dev).
+ *
+ * RFKILL:
+ *
+ * At startup, both HW and SW radio switchess are assumed to be off.
+ *
+ * At initialization time [after calling wimax_dev_add()], have the
+ * driver query the device for the status of the software and hardware
+ * RF kill switches and call wimax_report_rfkill_hw() and
+ * wimax_rfkill_report_sw() to indicate their state. If any is
+ * missing, just call it to indicate it is ON (radio always on).
+ *
+ * Whenever the driver detects a change in the state of the RF kill
+ * switches, it should call wimax_report_rfkill_hw() or
+ * wimax_report_rfkill_sw() to report it to the stack.
+ */
+struct wimax_dev {
+	struct net_device *net_dev;
+	struct list_head id_table_node;
+	struct mutex mutex;		/* Protects all members and API calls */
+	struct mutex mutex_reset;
+	enum wimax_st state;
+
+	int (*op_msg_from_user)(struct wimax_dev *wimax_dev,
+				const char *,
+				const void *, size_t,
+				const struct genl_info *info);
+	int (*op_rfkill_sw_toggle)(struct wimax_dev *wimax_dev,
+				   enum wimax_rf_state);
+	int (*op_reset)(struct wimax_dev *wimax_dev);
+
+	struct rfkill *rfkill;
+	struct input_dev *rfkill_input;
+	unsigned rf_hw;
+	unsigned rf_sw;
+	char name[32];
+
+	struct dentry *debugfs_dentry;
+};
+
+
+
+/*
+ * WiMAX stack public API for device drivers
+ * -----------------------------------------
+ *
+ * These functions are not exported to user space.
+ */
+extern void wimax_dev_init(struct wimax_dev *);
+extern int wimax_dev_add(struct wimax_dev *, struct net_device *);
+extern void wimax_dev_rm(struct wimax_dev *);
+
+static inline
+struct wimax_dev *net_dev_to_wimax(struct net_device *net_dev)
+{
+	return netdev_priv(net_dev);
+}
+
+static inline
+struct device *wimax_dev_to_dev(struct wimax_dev *wimax_dev)
+{
+	return wimax_dev->net_dev->dev.parent;
+}
+
+extern void wimax_state_change(struct wimax_dev *, enum wimax_st);
+extern enum wimax_st wimax_state_get(struct wimax_dev *);
+
+/*
+ * Radio Switch state reporting.
+ *
+ * enum wimax_rf_state is declared in linux/wimax.h so the exports
+ * to user space can use it.
+ */
+extern void wimax_report_rfkill_hw(struct wimax_dev *, enum wimax_rf_state);
+extern void wimax_report_rfkill_sw(struct wimax_dev *, enum wimax_rf_state);
+
+
+/*
+ * Free-form messaging to/from user space
+ *
+ * Sending a message:
+ *
+ *   wimax_msg(wimax_dev, pipe_name, buf, buf_size, GFP_KERNEL);
+ *
+ * Broken up:
+ *
+ *   skb = wimax_msg_alloc(wimax_dev, pipe_name, buf_size, GFP_KERNEL);
+ *   ...fill up skb...
+ *   wimax_msg_send(wimax_dev, pipe_name, skb);
+ *
+ * Be sure not to modify skb->data in the middle (ie: don't use
+ * skb_push()/skb_pull()/skb_reserve() on the skb).
+ *
+ * "pipe_name" is any string, than can be interpreted as the name of
+ * the pipe or destinatary; the interpretation of it is driver
+ * specific, so the recipient can multiplex it as wished. It can be
+ * NULL, it won't be used - an example is using a "diagnostics" tag to
+ * send diagnostics information that a device-specific diagnostics
+ * tool would be interested in.
+ */
+extern struct sk_buff *wimax_msg_alloc(struct wimax_dev *, const char *,
+				       const void *, size_t, gfp_t);
+extern int wimax_msg_send(struct wimax_dev *, struct sk_buff *);
+extern int wimax_msg(struct wimax_dev *, const char *,
+		     const void *, size_t, gfp_t);
+
+extern const void *wimax_msg_data_len(struct sk_buff *, size_t *);
+extern const void *wimax_msg_data(struct sk_buff *);
+extern ssize_t wimax_msg_len(struct sk_buff *);
+
+
+/*
+ * WiMAX stack user space API
+ * --------------------------
+ *
+ * This API is what gets exported to user space for general
+ * operations. As well, they can be called from within the kernel,
+ * (with a properly referenced `struct wimax_dev`).
+ *
+ * Properly referenced means: the 'struct net_device' that embeds the
+ * device's control structure and (as such) the 'struct wimax_dev' is
+ * referenced by the caller.
+ */
+extern int wimax_rfkill(struct wimax_dev *, enum wimax_rf_state);
+extern int wimax_reset(struct wimax_dev *);
+
+#else
+/* You might be looking for linux/wimax.h */
+#error This file should not be included from user space.
+#endif /* #ifdef __KERNEL__ */
+#endif /* #ifndef __NET__WIMAX_H__ */
-- 
cgit v1.2.3


From ea912f4e7f264981faf8665cfb63d46d7f948117 Mon Sep 17 00:00:00 2001
From: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Date: Sat, 20 Dec 2008 16:57:35 -0800
Subject: wimax: debug macros and debug settings for the WiMAX stack

This file contains a simple debug framework that is used in the stack;
it allows the debug level to be controlled at compile-time (so the
debug code is optimized out) and at run-time (for what wasn't compiled
out).

This is eventually going to be moved to use dynamic_printk(). Just
need to find time to do it.

Signed-off-by: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/wimax/debug.h | 453 ++++++++++++++++++++++++++++++++++++++++++++
 net/wimax/debug-levels.h    |  42 ++++
 2 files changed, 495 insertions(+)
 create mode 100644 include/linux/wimax/debug.h
 create mode 100644 net/wimax/debug-levels.h

(limited to 'include/linux')

diff --git a/include/linux/wimax/debug.h b/include/linux/wimax/debug.h
new file mode 100644
index 000000000000..ba0c49399a83
--- /dev/null
+++ b/include/linux/wimax/debug.h
@@ -0,0 +1,453 @@
+/*
+ * Linux WiMAX
+ * Collection of tools to manage debug operations.
+ *
+ *
+ * Copyright (C) 2005-2007 Intel Corporation
+ * Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ *
+ *
+ * Don't #include this file directly, read on!
+ *
+ *
+ * EXECUTING DEBUGGING ACTIONS OR NOT
+ *
+ * The main thing this framework provides is decission power to take a
+ * debug action (like printing a message) if the current debug level
+ * allows it.
+ *
+ * The decission power is at two levels: at compile-time (what does
+ * not make it is compiled out) and at run-time. The run-time
+ * selection is done per-submodule (as they are declared by the user
+ * of the framework).
+ *
+ * A call to d_test(L) (L being the target debug level) returns true
+ * if the action should be taken because the current debug levels
+ * allow it (both compile and run time).
+ *
+ * It follows that a call to d_test() that can be determined to be
+ * always false at compile time will get the code depending on it
+ * compiled out by optimization.
+ *
+ *
+ * DEBUG LEVELS
+ *
+ * It is up to the caller to define how much a debugging level is.
+ *
+ * Convention sets 0 as "no debug" (so an action marked as debug level 0
+ * will always be taken). The increasing debug levels are used for
+ * increased verbosity.
+ *
+ *
+ * USAGE
+ *
+ * Group the code in modules and submodules inside each module [which
+ * in most cases maps to Linux modules and .c files that compose
+ * those].
+ *
+ *
+ * For each module, there is:
+ *
+ *  - a MODULENAME (single word, legal C identifier)
+ *
+ *  - a debug-levels.h header file that declares the list of
+ *    submodules and that is included by all .c files that use
+ *    the debugging tools. The file name can be anything.
+ *
+ *  - some (optional) .c code to manipulate the runtime debug levels
+ *    through debugfs.
+ *
+ * The debug-levels.h file would look like:
+ *
+ *     #ifndef __debug_levels__h__
+ *     #define __debug_levels__h__
+ *
+ *     #define D_MODULENAME modulename
+ *     #define D_MASTER 10
+ *
+ *     #include <linux/wimax/debug.h>
+ *
+ *     enum d_module {
+ *             D_SUBMODULE_DECLARE(submodule_1),
+ *             D_SUBMODULE_DECLARE(submodule_2),
+ *             ...
+ *             D_SUBMODULE_DECLARE(submodule_N)
+ *     };
+ *
+ *     #endif
+ *
+ * D_MASTER is the maximum compile-time debug level; any debug actions
+ * above this will be out. D_MODULENAME is the module name (legal C
+ * identifier), which has to be unique for each module (to avoid
+ * namespace collisions during linkage). Note those #defines need to
+ * be done before #including debug.h
+ *
+ * We declare N different submodules whose debug level can be
+ * independently controlled during runtime.
+ *
+ * In a .c file of the module (and only in one of them), define the
+ * following code:
+ *
+ *     struct d_level D_LEVEL[] = {
+ *             D_SUBMODULE_DEFINE(submodule_1),
+ *             D_SUBMODULE_DEFINE(submodule_2),
+ *             ...
+ *             D_SUBMODULE_DEFINE(submodule_N),
+ *     };
+ *     size_t D_LEVEL_SIZE = ARRAY_SIZE(D_LEVEL);
+ *
+ * Externs for d_level_MODULENAME and d_level_size_MODULENAME are used
+ * and declared in this file using the D_LEVEL and D_LEVEL_SIZE macros
+ * #defined also in this file.
+ *
+ * To manipulate from user space the levels, create a debugfs dentry
+ * and then register each submodule with:
+ *
+ *     result = d_level_register_debugfs("PREFIX_", submodule_X, parent);
+ *     if (result < 0)
+ *            goto error;
+ *
+ * Where PREFIX_ is a name of your chosing. This will create debugfs
+ * file with a single numeric value that can be use to tweak it. To
+ * remove the entires, just use debugfs_remove_recursive() on 'parent'.
+ *
+ * NOTE: remember that even if this will show attached to some
+ *     particular instance of a device, the settings are *global*.
+ *
+ *
+ * On each submodule (for example, .c files), the debug infrastructure
+ * should be included like this:
+ *
+ *     #define D_SUBMODULE submodule_x     // matches one in debug-levels.h
+ *     #include "debug-levels.h"
+ *
+ * after #including all your include files.
+ *
+ *
+ * Now you can use the d_*() macros below [d_test(), d_fnstart(),
+ * d_fnend(), d_printf(), d_dump()].
+ *
+ * If their debug level is greater than D_MASTER, they will be
+ * compiled out.
+ *
+ * If their debug level is lower or equal than D_MASTER but greater
+ * than the current debug level of their submodule, they'll be
+ * ignored.
+ *
+ * Otherwise, the action will be performed.
+ */
+#ifndef __debug__h__
+#define __debug__h__
+
+#include <linux/types.h>
+#include <linux/device.h>
+
+
+/* Backend stuff */
+
+/*
+ * Debug backend: generate a message header from a 'struct device'
+ *
+ * @head: buffer where to place the header
+ * @head_size: length of @head
+ * @dev: pointer to device used to generate a header from. If NULL,
+ *     an empty ("") header is generated.
+ */
+static inline
+void __d_head(char *head, size_t head_size,
+	      struct device *dev)
+{
+	if (dev == NULL)
+		head[0] = 0;
+	else if ((unsigned long)dev < 4096) {
+		printk(KERN_ERR "E: Corrupt dev %p\n", dev);
+		WARN_ON(1);
+	} else
+		snprintf(head, head_size, "%s %s: ",
+			 dev_driver_string(dev), dev->bus_id);
+}
+
+
+/*
+ * Debug backend: log some message if debugging is enabled
+ *
+ * @l: intended debug level
+ * @tag: tag to prefix the message with
+ * @dev: 'struct device' associated to this message
+ * @f: printf-like format and arguments
+ *
+ * Note this is optimized out if it doesn't pass the compile-time
+ * check; however, it is *always* compiled. This is useful to make
+ * sure the printf-like formats and variables are always checked and
+ * they don't get bit rot if you have all the debugging disabled.
+ */
+#define _d_printf(l, tag, dev, f, a...)					\
+do {									\
+	char head[64];							\
+	if (!d_test(l))							\
+		break;							\
+	__d_head(head, sizeof(head), dev);				\
+	printk(KERN_ERR "%s%s%s: " f, head, __func__, tag, ##a);	\
+} while (0)
+
+
+/*
+ * CPP sintatic sugar to generate A_B like symbol names when one of
+ * the arguments is a a preprocessor #define.
+ */
+#define __D_PASTE__(varname, modulename) varname##_##modulename
+#define __D_PASTE(varname, modulename) (__D_PASTE__(varname, modulename))
+#define _D_SUBMODULE_INDEX(_name) (D_SUBMODULE_DECLARE(_name))
+
+
+/*
+ * Store a submodule's runtime debug level and name
+ */
+struct d_level {
+	u8 level;
+	const char *name;
+};
+
+
+/*
+ * List of available submodules and their debug levels
+ *
+ * We call them d_level_MODULENAME and d_level_size_MODULENAME; the
+ * macros D_LEVEL and D_LEVEL_SIZE contain the name already for
+ * convenience.
+ *
+ * This array and the size are defined on some .c file that is part of
+ * the current module.
+ */
+#define D_LEVEL __D_PASTE(d_level, D_MODULENAME)
+#define D_LEVEL_SIZE __D_PASTE(d_level_size, D_MODULENAME)
+
+extern struct d_level D_LEVEL[];
+extern size_t D_LEVEL_SIZE;
+
+
+/*
+ * Frontend stuff
+ *
+ *
+ * Stuff you need to declare prior to using the actual "debug" actions
+ * (defined below).
+ */
+
+#ifndef D_MODULENAME
+#error D_MODULENAME is not defined in your debug-levels.h file
+/**
+ * D_MODULE - Name of the current module
+ *
+ * #define in your module's debug-levels.h, making sure it is
+ * unique. This has to be a legal C identifier.
+ */
+#define D_MODULENAME undefined_modulename
+#endif
+
+
+#ifndef D_MASTER
+#warning D_MASTER not defined, but debug.h included! [see docs]
+/**
+ * D_MASTER - Compile time maximum debug level
+ *
+ * #define in your debug-levels.h file to the maximum debug level the
+ * runtime code will be allowed to have. This allows you to provide a
+ * main knob.
+ *
+ * Anything above that level will be optimized out of the compile.
+ *
+ * Defaults to zero (no debug code compiled in).
+ *
+ * Maximum one definition per module (at the debug-levels.h file).
+ */
+#define D_MASTER 0
+#endif
+
+#ifndef D_SUBMODULE
+#error D_SUBMODULE not defined, but debug.h included! [see docs]
+/**
+ * D_SUBMODULE - Name of the current submodule
+ *
+ * #define in your submodule .c file before #including debug-levels.h
+ * to the name of the current submodule as previously declared and
+ * defined with D_SUBMODULE_DECLARE() (in your module's
+ * debug-levels.h) and D_SUBMODULE_DEFINE().
+ *
+ * This is used to provide runtime-control over the debug levels.
+ *
+ * Maximum one per .c file! Can be shared among different .c files
+ * (meaning they belong to the same submodule categorization).
+ */
+#define D_SUBMODULE undefined_module
+#endif
+
+
+/**
+ * D_SUBMODULE_DECLARE - Declare a submodule for runtime debug level control
+ *
+ * @_name: name of the submodule, restricted to the chars that make up a
+ *     valid C identifier ([a-zA-Z0-9_]).
+ *
+ * Declare in the module's debug-levels.h header file as:
+ *
+ * enum d_module {
+ *         D_SUBMODULE_DECLARE(submodule_1),
+ *         D_SUBMODULE_DECLARE(submodule_2),
+ *         D_SUBMODULE_DECLARE(submodule_3),
+ * };
+ *
+ * Some corresponding .c file needs to have a matching
+ * D_SUBMODULE_DEFINE().
+ */
+#define D_SUBMODULE_DECLARE(_name) __D_SUBMODULE_##_name
+
+
+/**
+ * D_SUBMODULE_DEFINE - Define a submodule for runtime debug level control
+ *
+ * @_name: name of the submodule, restricted to the chars that make up a
+ *     valid C identifier ([a-zA-Z0-9_]).
+ *
+ * Use once per module (in some .c file) as:
+ *
+ * static
+ * struct d_level d_level_SUBMODULENAME[] = {
+ *         D_SUBMODULE_DEFINE(submodule_1),
+ *         D_SUBMODULE_DEFINE(submodule_2),
+ *         D_SUBMODULE_DEFINE(submodule_3),
+ * };
+ * size_t d_level_size_SUBDMODULENAME = ARRAY_SIZE(d_level_SUBDMODULENAME);
+ *
+ * Matching D_SUBMODULE_DECLARE()s have to be present in a
+ * debug-levels.h header file.
+ */
+#define D_SUBMODULE_DEFINE(_name)		\
+[__D_SUBMODULE_##_name] = {			\
+	.level = 0,				\
+	.name = #_name				\
+}
+
+
+
+/* The actual "debug" operations */
+
+
+/**
+ * d_test - Returns true if debugging should be enabled
+ *
+ * @l: intended debug level (unsigned)
+ *
+ * If the master debug switch is enabled and the current settings are
+ * higher or equal to the requested level, then debugging
+ * output/actions should be enabled.
+ *
+ * NOTE:
+ *
+ * This needs to be coded so that it can be evaluated in compile
+ * time; this is why the ugly BUG_ON() is placed in there, so the
+ * D_MASTER evaluation compiles all out if it is compile-time false.
+ */
+#define d_test(l)							\
+({									\
+	unsigned __l = l;	/* type enforcer */			\
+	(D_MASTER) >= __l						\
+	&& ({								\
+		BUG_ON(_D_SUBMODULE_INDEX(D_SUBMODULE) >= D_LEVEL_SIZE);\
+		D_LEVEL[_D_SUBMODULE_INDEX(D_SUBMODULE)].level >= __l;	\
+	});								\
+})
+
+
+/**
+ * d_fnstart - log message at function start if debugging enabled
+ *
+ * @l: intended debug level
+ * @_dev: 'struct device' pointer, NULL if none (for context)
+ * @f: printf-like format and arguments
+ */
+#define d_fnstart(l, _dev, f, a...) _d_printf(l, " FNSTART", _dev, f, ## a)
+
+
+/**
+ * d_fnend - log message at function end if debugging enabled
+ *
+ * @l: intended debug level
+ * @_dev: 'struct device' pointer, NULL if none (for context)
+ * @f: printf-like format and arguments
+ */
+#define d_fnend(l, _dev, f, a...) _d_printf(l, " FNEND", _dev, f, ## a)
+
+
+/**
+ * d_printf - log message if debugging enabled
+ *
+ * @l: intended debug level
+ * @_dev: 'struct device' pointer, NULL if none (for context)
+ * @f: printf-like format and arguments
+ */
+#define d_printf(l, _dev, f, a...) _d_printf(l, "", _dev, f, ## a)
+
+
+/**
+ * d_dump - log buffer hex dump if debugging enabled
+ *
+ * @l: intended debug level
+ * @_dev: 'struct device' pointer, NULL if none (for context)
+ * @f: printf-like format and arguments
+ */
+#define d_dump(l, dev, ptr, size)			\
+do {							\
+	char head[64];					\
+	if (!d_test(l))					\
+		break;					\
+	__d_head(head, sizeof(head), dev);		\
+	print_hex_dump(KERN_ERR, head, 0, 16, 1,	\
+		       ((void *) ptr), (size), 0);	\
+} while (0)
+
+
+/**
+ * Export a submodule's debug level over debugfs as PREFIXSUBMODULE
+ *
+ * @prefix: string to prefix the name with
+ * @submodule: name of submodule (not a string, just the name)
+ * @dentry: debugfs parent dentry
+ *
+ * Returns: 0 if ok, < 0 errno on error.
+ *
+ * For removing, just use debugfs_remove_recursive() on the parent.
+ */
+#define d_level_register_debugfs(prefix, name, parent)			\
+({									\
+	int rc;								\
+	struct dentry *fd;						\
+	struct dentry *verify_parent_type = parent;			\
+	fd = debugfs_create_u8(						\
+		prefix #name, 0600, verify_parent_type,			\
+		&(D_LEVEL[__D_SUBMODULE_ ## name].level));		\
+	rc = PTR_ERR(fd);						\
+	if (IS_ERR(fd) && rc != -ENODEV)				\
+		printk(KERN_ERR "%s: Can't create debugfs entry %s: "	\
+		       "%d\n", __func__, prefix #name, rc);		\
+	else								\
+		rc = 0;							\
+	rc;								\
+})
+
+
+#endif /* #ifndef __debug__h__ */
diff --git a/net/wimax/debug-levels.h b/net/wimax/debug-levels.h
new file mode 100644
index 000000000000..1c29123a3aa9
--- /dev/null
+++ b/net/wimax/debug-levels.h
@@ -0,0 +1,42 @@
+/*
+ * Linux WiMAX Stack
+ * Debug levels control file for the wimax module
+ *
+ *
+ * Copyright (C) 2007-2008 Intel Corporation <linux-wimax@intel.com>
+ * Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+#ifndef __debug_levels__h__
+#define __debug_levels__h__
+
+/* Maximum compile and run time debug level for all submodules */
+#define D_MODULENAME wimax
+#define D_MASTER CONFIG_WIMAX_DEBUG_LEVEL
+
+#include <linux/wimax/debug.h>
+
+/* List of all the enabled modules */
+enum d_module {
+	D_SUBMODULE_DECLARE(debugfs),
+	D_SUBMODULE_DECLARE(id_table),
+	D_SUBMODULE_DECLARE(op_msg),
+	D_SUBMODULE_DECLARE(op_reset),
+	D_SUBMODULE_DECLARE(op_rfkill),
+	D_SUBMODULE_DECLARE(stack),
+};
+
+#endif /* #ifndef __debug_levels__h__ */
-- 
cgit v1.2.3


From ea24652d253eabfb83e955e55ce032228d9d99b9 Mon Sep 17 00:00:00 2001
From: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Date: Sat, 20 Dec 2008 16:57:43 -0800
Subject: i2400m: host/device procotol and core driver definitions

The wimax/i2400m.h defines the structures and constants for the
host-device protocols:

 - boot / firmware upload protocol

 - general data transport protocol

 - control protocol

It is done in such a way that can also be used verbatim by user space.

drivers/net/wimax/i2400m.h defines all the APIs used by the core,
bus-generic driver (i2400m) and the bus specific drivers
(i2400m-BUSNAME). It also gives a roadmap to the driver
implementation.

debug-levels.h adds the core driver's debug settings.

Signed-off-by: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/net/wimax/i2400m/debug-levels.h |  45 ++
 drivers/net/wimax/i2400m/i2400m.h       | 755 ++++++++++++++++++++++++++++++++
 include/linux/wimax/i2400m.h            | 512 ++++++++++++++++++++++
 3 files changed, 1312 insertions(+)
 create mode 100644 drivers/net/wimax/i2400m/debug-levels.h
 create mode 100644 drivers/net/wimax/i2400m/i2400m.h
 create mode 100644 include/linux/wimax/i2400m.h

(limited to 'include/linux')

diff --git a/drivers/net/wimax/i2400m/debug-levels.h b/drivers/net/wimax/i2400m/debug-levels.h
new file mode 100644
index 000000000000..3183baa16a52
--- /dev/null
+++ b/drivers/net/wimax/i2400m/debug-levels.h
@@ -0,0 +1,45 @@
+/*
+ * Intel Wireless WiMAX Connection 2400m
+ * Debug levels control file for the i2400m module
+ *
+ *
+ * Copyright (C) 2007-2008 Intel Corporation <linux-wimax@intel.com>
+ * Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version
+ * 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+#ifndef __debug_levels__h__
+#define __debug_levels__h__
+
+/* Maximum compile and run time debug level for all submodules */
+#define D_MODULENAME i2400m
+#define D_MASTER CONFIG_WIMAX_I2400M_DEBUG_LEVEL
+
+#include <linux/wimax/debug.h>
+
+/* List of all the enabled modules */
+enum d_module {
+	D_SUBMODULE_DECLARE(control),
+	D_SUBMODULE_DECLARE(driver),
+	D_SUBMODULE_DECLARE(debugfs),
+	D_SUBMODULE_DECLARE(fw),
+	D_SUBMODULE_DECLARE(netdev),
+	D_SUBMODULE_DECLARE(rfkill),
+	D_SUBMODULE_DECLARE(rx),
+	D_SUBMODULE_DECLARE(tx),
+};
+
+
+#endif /* #ifndef __debug_levels__h__ */
diff --git a/drivers/net/wimax/i2400m/i2400m.h b/drivers/net/wimax/i2400m/i2400m.h
new file mode 100644
index 000000000000..067c871cc226
--- /dev/null
+++ b/drivers/net/wimax/i2400m/i2400m.h
@@ -0,0 +1,755 @@
+/*
+ * Intel Wireless WiMAX Connection 2400m
+ * Declarations for bus-generic internal APIs
+ *
+ *
+ * Copyright (C) 2007-2008 Intel Corporation. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ *   * Redistributions of source code must retain the above copyright
+ *     notice, this list of conditions and the following disclaimer.
+ *   * Redistributions in binary form must reproduce the above copyright
+ *     notice, this list of conditions and the following disclaimer in
+ *     the documentation and/or other materials provided with the
+ *     distribution.
+ *   * Neither the name of Intel Corporation nor the names of its
+ *     contributors may be used to endorse or promote products derived
+ *     from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *
+ * Intel Corporation <linux-wimax@intel.com>
+ * Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
+ * Yanir Lubetkin <yanirx.lubetkin@intel.com>
+ *  - Initial implementation
+ *
+ *
+ * GENERAL DRIVER ARCHITECTURE
+ *
+ * The i2400m driver is split in the following two major parts:
+ *
+ *  - bus specific driver
+ *  - bus generic driver (this part)
+ *
+ * The bus specific driver sets up stuff specific to the bus the
+ * device is connected to (USB, SDIO, PCI, tam-tam...non-authoritative
+ * nor binding list) which is basically the device-model management
+ * (probe/disconnect, etc), moving data from device to kernel and
+ * back, doing the power saving details and reseting the device.
+ *
+ * For details on each bus-specific driver, see it's include file,
+ * i2400m-BUSNAME.h
+ *
+ * The bus-generic functionality break up is:
+ *
+ *  - Firmware upload: fw.c - takes care of uploading firmware to the
+ *        device. bus-specific driver just needs to provides a way to
+ *        execute boot-mode commands and to reset the device.
+ *
+ *  - RX handling: rx.c - receives data from the bus-specific code and
+ *        feeds it to the network or WiMAX stack or uses it to modify
+ *        the driver state. bus-specific driver only has to receive
+ *        frames and pass them to this module.
+ *
+ *  - TX handling: tx.c - manages the TX FIFO queue and provides means
+ *        for the bus-specific TX code to pull data from the FIFO
+ *        queue. bus-specific code just pulls frames from this module
+ *        to sends them to the device.
+ *
+ *  - netdev glue: netdev.c - interface with Linux networking
+ *        stack. Pass around data frames, and configure when the
+ *        device is up and running or shutdown (through ifconfig up /
+ *        down). Bus-generic only.
+ *
+ *  - control ops: control.c - implements various commmands for
+ *        controlling the device. bus-generic only.
+ *
+ *  - device model glue: driver.c - implements helpers for the
+ *        device-model glue done by the bus-specific layer
+ *        (setup/release the driver resources), turning the device on
+ *        and off, handling the device reboots/resets and a few simple
+ *        WiMAX stack ops.
+ *
+ * Code is also broken up in linux-glue / device-glue.
+ *
+ * Linux glue contains functions that deal mostly with gluing with the
+ * rest of the Linux kernel.
+ *
+ * Device-glue are functions that deal mostly with the way the device
+ * does things and talk the device's language.
+ *
+ * device-glue code is licensed BSD so other open source OSes can take
+ * it to implement their drivers.
+ *
+ *
+ * APIs AND HEADER FILES
+ *
+ * This bus generic code exports three APIs:
+ *
+ *  - HDI (host-device interface) definitions common to all busses
+ *    (include/linux/wimax/i2400m.h); these can be also used by user
+ *    space code.
+ *  - internal API for the bus-generic code
+ *  - external API for the bus-specific drivers
+ *
+ *
+ * LIFE CYCLE:
+ *
+ * When the bus-specific driver probes, it allocates a network device
+ * with enough space for it's data structue, that must contain a
+ * &struct i2400m at the top.
+ *
+ * On probe, it needs to fill the i2400m members marked as [fill], as
+ * well as i2400m->wimax_dev.net_dev and call i2400m_setup(). The
+ * i2400m driver will only register with the WiMAX and network stacks;
+ * the only access done to the device is to read the MAC address so we
+ * can register a network device. This calls i2400m_dev_start() to
+ * load firmware, setup communication with the device and configure it
+ * for operation.
+ *
+ * At this point, control and data communications are possible.
+ *
+ * On disconnect/driver unload, the bus-specific disconnect function
+ * calls i2400m_release() to undo i2400m_setup(). i2400m_dev_stop()
+ * shuts the firmware down and releases resources uses to communicate
+ * with the device.
+ *
+ * While the device is up, it might reset. The bus-specific driver has
+ * to catch that situation and call i2400m_dev_reset_handle() to deal
+ * with it (reset the internal driver structures and go back to square
+ * one).
+ */
+
+#ifndef __I2400M_H__
+#define __I2400M_H__
+
+#include <linux/usb.h>
+#include <linux/netdevice.h>
+#include <linux/completion.h>
+#include <linux/rwsem.h>
+#include <asm/atomic.h>
+#include <net/wimax.h>
+#include <linux/wimax/i2400m.h>
+#include <asm/byteorder.h>
+
+/* Misc constants */
+enum {
+	/* Firmware uploading */
+	I2400M_BOOT_RETRIES = 3,
+	/* Size of the Boot Mode Command buffer */
+	I2400M_BM_CMD_BUF_SIZE = 16 * 1024,
+	I2400M_BM_ACK_BUF_SIZE = 256,
+};
+
+
+/* Firmware version we request when pulling the fw image file */
+#define I2400M_FW_VERSION "1.3"
+
+
+/**
+ * i2400m_reset_type - methods to reset a device
+ *
+ * @I2400M_RT_WARM: Reset without device disconnection, device handles
+ *     are kept valid but state is back to power on, with firmware
+ *     re-uploaded.
+ * @I2400M_RT_COLD: Tell the device to disconnect itself from the bus
+ *     and reconnect. Renders all device handles invalid.
+ * @I2400M_RT_BUS: Tells the bus to reset the device; last measure
+ *     used when both types above don't work.
+ */
+enum i2400m_reset_type {
+	I2400M_RT_WARM,	/* first measure */
+	I2400M_RT_COLD,	/* second measure */
+	I2400M_RT_BUS,	/* call in artillery */
+};
+
+struct i2400m_reset_ctx;
+
+/**
+ * struct i2400m - descriptor for an Intel 2400m
+ *
+ * Members marked with [fill] must be filled out/initialized before
+ * calling i2400m_setup().
+ *
+ * @bus_tx_block_size: [fill] SDIO imposes a 256 block size, USB 16,
+ *     so we have a tx_blk_size variable that the bus layer sets to
+ *     tell the engine how much of that we need.
+ *
+ * @bus_pl_size_max: [fill] Maximum payload size.
+ *
+ * @bus_dev_start: [fill] Function called by the bus-generic code
+ *     [i2400m_dev_start()] to setup the bus-specific communications
+ *     to the the device. See LIFE CYCLE above.
+ *
+ *     NOTE: Doesn't need to upload the firmware, as that is taken
+ *     care of by the bus-generic code.
+ *
+ * @bus_dev_stop: [fill] Function called by the bus-generic code
+ *     [i2400m_dev_stop()] to shutdown the bus-specific communications
+ *     to the the device. See LIFE CYCLE above.
+ *
+ *     This function does not need to reset the device, just tear down
+ *     all the host resources created to  handle communication with
+ *     the device.
+ *
+ * @bus_tx_kick: [fill] Function called by the bus-generic code to let
+ *     the bus-specific code know that there is data available in the
+ *     TX FIFO for transmission to the device.
+ *
+ *     This function cannot sleep.
+ *
+ * @bus_reset: [fill] Function called by the bus-generic code to reset
+ *     the device in in various ways. Doesn't need to wait for the
+ *     reset to finish.
+ *
+ *     If warm or cold reset fail, this function is expected to do a
+ *     bus-specific reset (eg: USB reset) to get the device to a
+ *     working state (even if it implies device disconecction).
+ *
+ *     Note the warm reset is used by the firmware uploader to
+ *     reinitialize the device.
+ *
+ *     IMPORTANT: this is called very early in the device setup
+ *     process, so it cannot rely on common infrastructure being laid
+ *     out.
+ *
+ * @bus_bm_cmd_send: [fill] Function called to send a boot-mode
+ *     command. Flags are defined in 'enum i2400m_bm_cmd_flags'. This
+ *     is synchronous and has to return 0 if ok or < 0 errno code in
+ *     any error condition.
+ *
+ * @bus_bm_wait_for_ack: [fill] Function called to wait for a
+ *     boot-mode notification (that can be a response to a previously
+ *     issued command or an asynchronous one). Will read until all the
+ *     indicated size is read or timeout. Reading more or less data
+ *     than asked for is an error condition. Return 0 if ok, < 0 errno
+ *     code on error.
+ *
+ *     The caller to this function will check if the response is a
+ *     barker that indicates the device going into reset mode.
+ *
+ * @bus_fw_name: [fill] name of the firmware image (in most cases,
+ *     they are all the same for a single release, except that they
+ *     have the type of the bus embedded in the name (eg:
+ *     i2400m-fw-X-VERSION.sbcf, where X is the bus name).
+ *
+ * @bus_bm_mac_addr_impaired: [fill] Set to true if the device's MAC
+ *     address provided in boot mode is kind of broken and needs to
+ *     be re-read later on.
+ *
+ *
+ * @wimax_dev: WiMAX generic device for linkage into the kernel WiMAX
+ *     stack. Due to the way a net_device is allocated, we need to
+ *     force this to be the first field so that we can get from
+ *     netdev_priv() the right pointer.
+ *
+ * @state: device's state (as reported by it)
+ *
+ * @state_wq: waitqueue that is woken up whenever the state changes
+ *
+ * @tx_lock: spinlock to protect TX members
+ *
+ * @tx_buf: FIFO buffer for TX; we queue data here
+ *
+ * @tx_in: FIFO index for incoming data. Note this doesn't wrap around
+ *     and it is always greater than @tx_out.
+ *
+ * @tx_out: FIFO index for outgoing data
+ *
+ * @tx_msg: current TX message that is active in the FIFO for
+ *     appending payloads.
+ *
+ * @tx_sequence: current sequence number for TX messages from the
+ *     device to the host.
+ *
+ * @tx_msg_size: size of the current message being transmitted by the
+ *     bus-specific code.
+ *
+ * @tx_pl_num: total number of payloads sent
+ *
+ * @tx_pl_max: maximum number of payloads sent in a TX message
+ *
+ * @tx_pl_min: minimum number of payloads sent in a TX message
+ *
+ * @tx_num: number of TX messages sent
+ *
+ * @tx_size_acc: number of bytes in all TX messages sent
+ *     (this is different to net_dev's statistics as it also counts
+ *     control messages).
+ *
+ * @tx_size_min: smallest TX message sent.
+ *
+ * @tx_size_max: biggest TX message sent.
+ *
+ * @rx_lock: spinlock to protect RX members
+ *
+ * @rx_pl_num: total number of payloads received
+ *
+ * @rx_pl_max: maximum number of payloads received in a RX message
+ *
+ * @rx_pl_min: minimum number of payloads received in a RX message
+ *
+ * @rx_num: number of RX messages received
+ *
+ * @rx_size_acc: number of bytes in all RX messages received
+ *     (this is different to net_dev's statistics as it also counts
+ *     control messages).
+ *
+ * @rx_size_min: smallest RX message received.
+ *
+ * @rx_size_max: buggest RX message received.
+ *
+ * @init_mutex: Mutex used for serializing the device bringup
+ *     sequence; this way if the device reboots in the middle, we
+ *     don't try to do a bringup again while we are tearing down the
+ *     one that failed.
+ *
+ *     Can't reuse @msg_mutex because from within the bringup sequence
+ *     we need to send messages to the device and thus use @msg_mutex.
+ *
+ * @msg_mutex: mutex used to send control commands to the device (we
+ *     only allow one at a time, per host-device interface design).
+ *
+ * @msg_completion: used to wait for an ack to a control command sent
+ *     to the device.
+ *
+ * @ack_skb: used to store the actual ack to a control command if the
+ *     reception of the command was successful. Otherwise, a ERR_PTR()
+ *     errno code that indicates what failed with the ack reception.
+ *
+ *     Only valid after @msg_completion is woken up. Only updateable
+ *     if @msg_completion is armed. Only touched by
+ *     i2400m_msg_to_dev().
+ *
+ *     Protected by @rx_lock. In theory the command execution flow is
+ *     sequential, but in case the device sends an out-of-phase or
+ *     very delayed response, we need to avoid it trampling current
+ *     execution.
+ *
+ * @bm_cmd_buf: boot mode command buffer for composing firmware upload
+ *     commands.
+ *
+ *     USB can't r/w to stack, vmalloc, etc...as well, we end up
+ *     having to alloc/free a lot to compose commands, so we use these
+ *     for stagging and not having to realloc all the time.
+ *
+ *     This assumes the code always runs serialized. Only one thread
+ *     can call i2400m_bm_cmd() at the same time.
+ *
+ * @bm_ack_buf: boot mode acknoledge buffer for staging reception of
+ *     responses to commands.
+ *
+ *     See @bm_cmd_buf.
+ *
+ * @work_queue: work queue for processing device reports. This
+ *     workqueue cannot be used for processing TX or RX to the device,
+ *     as from it we'll process device reports, which might require
+ *     further communication with the device.
+ *
+ * @debugfs_dentry: hookup for debugfs files.
+ *     These have to be in a separate directory, a child of
+ *     (wimax_dev->debugfs_dentry) so they can be removed when the
+ *     module unloads, as we don't keep each dentry.
+ */
+struct i2400m {
+	struct wimax_dev wimax_dev;	/* FIRST! See doc */
+
+	unsigned updown:1;		/* Network device is up or down */
+	unsigned boot_mode:1;		/* is the device in boot mode? */
+	unsigned sboot:1;		/* signed or unsigned fw boot */
+	unsigned ready:1;		/* all probing steps done */
+	u8 trace_msg_from_user;		/* echo rx msgs to 'trace' pipe */
+					/* typed u8 so debugfs/u8 can tweak */
+	enum i2400m_system_state state;
+	wait_queue_head_t state_wq;	/* Woken up when on state updates */
+
+	size_t bus_tx_block_size;
+	size_t bus_pl_size_max;
+	int (*bus_dev_start)(struct i2400m *);
+	void (*bus_dev_stop)(struct i2400m *);
+	void (*bus_tx_kick)(struct i2400m *);
+	int (*bus_reset)(struct i2400m *, enum i2400m_reset_type);
+	ssize_t (*bus_bm_cmd_send)(struct i2400m *,
+				   const struct i2400m_bootrom_header *,
+				   size_t, int flags);
+	ssize_t (*bus_bm_wait_for_ack)(struct i2400m *,
+				       struct i2400m_bootrom_header *, size_t);
+	const char *bus_fw_name;
+	unsigned bus_bm_mac_addr_impaired:1;
+
+	spinlock_t tx_lock;		/* protect TX state */
+	void *tx_buf;
+	size_t tx_in, tx_out;
+	struct i2400m_msg_hdr *tx_msg;
+	size_t tx_sequence, tx_msg_size;
+	/* TX stats */
+	unsigned tx_pl_num, tx_pl_max, tx_pl_min,
+		tx_num, tx_size_acc, tx_size_min, tx_size_max;
+
+	/* RX stats */
+	spinlock_t rx_lock;		/* protect RX state */
+	unsigned rx_pl_num, rx_pl_max, rx_pl_min,
+		rx_num, rx_size_acc, rx_size_min, rx_size_max;
+
+	struct mutex msg_mutex;		/* serialize command execution */
+	struct completion msg_completion;
+	struct sk_buff *ack_skb;	/* protected by rx_lock */
+
+	void *bm_ack_buf;		/* for receiving acks over USB */
+	void *bm_cmd_buf;		/* for issuing commands over USB */
+
+	struct workqueue_struct *work_queue;
+
+	struct mutex init_mutex;	/* protect bringup seq */
+	struct i2400m_reset_ctx *reset_ctx;	/* protected by init_mutex */
+
+	struct work_struct wake_tx_ws;
+	struct sk_buff *wake_tx_skb;
+
+	struct dentry *debugfs_dentry;
+};
+
+
+/*
+ * Initialize a 'struct i2400m' from all zeroes
+ *
+ * This is a bus-generic API call.
+ */
+static inline
+void i2400m_init(struct i2400m *i2400m)
+{
+	wimax_dev_init(&i2400m->wimax_dev);
+
+	i2400m->boot_mode = 1;
+	init_waitqueue_head(&i2400m->state_wq);
+
+	spin_lock_init(&i2400m->tx_lock);
+	i2400m->tx_pl_min = UINT_MAX;
+	i2400m->tx_size_min = UINT_MAX;
+
+	spin_lock_init(&i2400m->rx_lock);
+	i2400m->rx_pl_min = UINT_MAX;
+	i2400m->rx_size_min = UINT_MAX;
+
+	mutex_init(&i2400m->msg_mutex);
+	init_completion(&i2400m->msg_completion);
+
+	mutex_init(&i2400m->init_mutex);
+	/* wake_tx_ws is initialized in i2400m_tx_setup() */
+}
+
+
+/*
+ * Bus-generic internal APIs
+ * -------------------------
+ */
+
+static inline
+struct i2400m *wimax_dev_to_i2400m(struct wimax_dev *wimax_dev)
+{
+	return container_of(wimax_dev, struct i2400m, wimax_dev);
+}
+
+static inline
+struct i2400m *net_dev_to_i2400m(struct net_device *net_dev)
+{
+	return wimax_dev_to_i2400m(netdev_priv(net_dev));
+}
+
+/*
+ * Boot mode support
+ */
+
+/**
+ * i2400m_bm_cmd_flags - flags to i2400m_bm_cmd()
+ *
+ * @I2400M_BM_CMD_RAW: send the command block as-is, without doing any
+ *     extra processing for adding CRC.
+ */
+enum i2400m_bm_cmd_flags {
+	I2400M_BM_CMD_RAW	= 1 << 2,
+};
+
+/**
+ * i2400m_bri - Boot-ROM indicators
+ *
+ * Flags for i2400m_bootrom_init() and i2400m_dev_bootstrap() [which
+ * are passed from things like i2400m_setup()]. Can be combined with
+ * |.
+ *
+ * @I2400M_BRI_SOFT: The device rebooted already and a reboot
+ *     barker received, proceed directly to ack the boot sequence.
+ * @I2400M_BRI_NO_REBOOT: Do not reboot the device and proceed
+ *     directly to wait for a reboot barker from the device.
+ * @I2400M_BRI_MAC_REINIT: We need to reinitialize the boot
+ *     rom after reading the MAC adress. This is quite a dirty hack,
+ *     if you ask me -- the device requires the bootrom to be
+ *     intialized after reading the MAC address.
+ */
+enum i2400m_bri {
+	I2400M_BRI_SOFT       = 1 << 1,
+	I2400M_BRI_NO_REBOOT  = 1 << 2,
+	I2400M_BRI_MAC_REINIT = 1 << 3,
+};
+
+extern void i2400m_bm_cmd_prepare(struct i2400m_bootrom_header *);
+extern int i2400m_dev_bootstrap(struct i2400m *, enum i2400m_bri);
+extern int i2400m_read_mac_addr(struct i2400m *);
+extern int i2400m_bootrom_init(struct i2400m *, enum i2400m_bri);
+
+/* Make/grok boot-rom header commands */
+
+static inline
+__le32 i2400m_brh_command(enum i2400m_brh_opcode opcode, unsigned use_checksum,
+			  unsigned direct_access)
+{
+	return cpu_to_le32(
+		I2400M_BRH_SIGNATURE
+		| (direct_access ? I2400M_BRH_DIRECT_ACCESS : 0)
+		| I2400M_BRH_RESPONSE_REQUIRED /* response always required */
+		| (use_checksum ? I2400M_BRH_USE_CHECKSUM : 0)
+		| (opcode & I2400M_BRH_OPCODE_MASK));
+}
+
+static inline
+void i2400m_brh_set_opcode(struct i2400m_bootrom_header *hdr,
+			   enum i2400m_brh_opcode opcode)
+{
+	hdr->command = cpu_to_le32(
+		(le32_to_cpu(hdr->command) & ~I2400M_BRH_OPCODE_MASK)
+		| (opcode & I2400M_BRH_OPCODE_MASK));
+}
+
+static inline
+unsigned i2400m_brh_get_opcode(const struct i2400m_bootrom_header *hdr)
+{
+	return le32_to_cpu(hdr->command) & I2400M_BRH_OPCODE_MASK;
+}
+
+static inline
+unsigned i2400m_brh_get_response(const struct i2400m_bootrom_header *hdr)
+{
+	return (le32_to_cpu(hdr->command) & I2400M_BRH_RESPONSE_MASK)
+		>> I2400M_BRH_RESPONSE_SHIFT;
+}
+
+static inline
+unsigned i2400m_brh_get_use_checksum(const struct i2400m_bootrom_header *hdr)
+{
+	return le32_to_cpu(hdr->command) & I2400M_BRH_USE_CHECKSUM;
+}
+
+static inline
+unsigned i2400m_brh_get_response_required(
+	const struct i2400m_bootrom_header *hdr)
+{
+	return le32_to_cpu(hdr->command) & I2400M_BRH_RESPONSE_REQUIRED;
+}
+
+static inline
+unsigned i2400m_brh_get_direct_access(const struct i2400m_bootrom_header *hdr)
+{
+	return le32_to_cpu(hdr->command) & I2400M_BRH_DIRECT_ACCESS;
+}
+
+static inline
+unsigned i2400m_brh_get_signature(const struct i2400m_bootrom_header *hdr)
+{
+	return (le32_to_cpu(hdr->command) & I2400M_BRH_SIGNATURE_MASK)
+		>> I2400M_BRH_SIGNATURE_SHIFT;
+}
+
+
+/*
+ * Driver / device setup and internal functions
+ */
+extern void i2400m_netdev_setup(struct net_device *net_dev);
+extern int i2400m_tx_setup(struct i2400m *);
+extern void i2400m_wake_tx_work(struct work_struct *);
+extern void i2400m_tx_release(struct i2400m *);
+
+extern void i2400m_net_rx(struct i2400m *, struct sk_buff *, unsigned,
+			  const void *, int);
+enum i2400m_pt;
+extern int i2400m_tx(struct i2400m *, const void *, size_t, enum i2400m_pt);
+
+#ifdef CONFIG_DEBUG_FS
+extern int i2400m_debugfs_add(struct i2400m *);
+extern void i2400m_debugfs_rm(struct i2400m *);
+#else
+static inline int i2400m_debugfs_add(struct i2400m *i2400m)
+{
+	return 0;
+}
+static inline void i2400m_debugfs_rm(struct i2400m *i2400m) {}
+#endif
+
+/* Called by _dev_start()/_dev_stop() to initialize the device itself */
+extern int i2400m_dev_initialize(struct i2400m *);
+extern void i2400m_dev_shutdown(struct i2400m *);
+
+extern struct attribute_group i2400m_dev_attr_group;
+
+extern int i2400m_schedule_work(struct i2400m *,
+				void (*)(struct work_struct *), gfp_t);
+
+/* HDI message's payload description handling */
+
+static inline
+size_t i2400m_pld_size(const struct i2400m_pld *pld)
+{
+	return I2400M_PLD_SIZE_MASK & le32_to_cpu(pld->val);
+}
+
+static inline
+enum i2400m_pt i2400m_pld_type(const struct i2400m_pld *pld)
+{
+	return (I2400M_PLD_TYPE_MASK & le32_to_cpu(pld->val))
+		>> I2400M_PLD_TYPE_SHIFT;
+}
+
+static inline
+void i2400m_pld_set(struct i2400m_pld *pld, size_t size,
+		    enum i2400m_pt type)
+{
+	pld->val = cpu_to_le32(
+		((type << I2400M_PLD_TYPE_SHIFT) & I2400M_PLD_TYPE_MASK)
+		|  (size & I2400M_PLD_SIZE_MASK));
+}
+
+
+/*
+ * API for the bus-specific drivers
+ * --------------------------------
+ */
+
+static inline
+struct i2400m *i2400m_get(struct i2400m *i2400m)
+{
+	dev_hold(i2400m->wimax_dev.net_dev);
+	return i2400m;
+}
+
+static inline
+void i2400m_put(struct i2400m *i2400m)
+{
+	dev_put(i2400m->wimax_dev.net_dev);
+}
+
+extern int i2400m_dev_reset_handle(struct i2400m *);
+
+/*
+ * _setup()/_release() are called by the probe/disconnect functions of
+ * the bus-specific drivers.
+ */
+extern int i2400m_setup(struct i2400m *, enum i2400m_bri bm_flags);
+extern void i2400m_release(struct i2400m *);
+
+extern int i2400m_rx(struct i2400m *, struct sk_buff *);
+extern struct i2400m_msg_hdr *i2400m_tx_msg_get(struct i2400m *, size_t *);
+extern void i2400m_tx_msg_sent(struct i2400m *);
+
+static const __le32 i2400m_NBOOT_BARKER[4] = {
+	__constant_cpu_to_le32(I2400M_NBOOT_BARKER),
+	__constant_cpu_to_le32(I2400M_NBOOT_BARKER),
+	__constant_cpu_to_le32(I2400M_NBOOT_BARKER),
+	__constant_cpu_to_le32(I2400M_NBOOT_BARKER)
+};
+
+static const __le32 i2400m_SBOOT_BARKER[4] = {
+	__constant_cpu_to_le32(I2400M_SBOOT_BARKER),
+	__constant_cpu_to_le32(I2400M_SBOOT_BARKER),
+	__constant_cpu_to_le32(I2400M_SBOOT_BARKER),
+	__constant_cpu_to_le32(I2400M_SBOOT_BARKER)
+};
+
+
+/*
+ * Utility functions
+ */
+
+static inline
+struct device *i2400m_dev(struct i2400m *i2400m)
+{
+	return i2400m->wimax_dev.net_dev->dev.parent;
+}
+
+/*
+ * Helper for scheduling simple work functions
+ *
+ * This struct can get any kind of payload attached (normally in the
+ * form of a struct where you pack the stuff you want to pass to the
+ * _work function).
+ */
+struct i2400m_work {
+	struct work_struct ws;
+	struct i2400m *i2400m;
+	u8 pl[0];
+};
+extern int i2400m_queue_work(struct i2400m *,
+			     void (*)(struct work_struct *), gfp_t,
+				const void *, size_t);
+
+extern int i2400m_msg_check_status(const struct i2400m_l3l4_hdr *,
+				   char *, size_t);
+extern int i2400m_msg_size_check(struct i2400m *,
+				 const struct i2400m_l3l4_hdr *, size_t);
+extern struct sk_buff *i2400m_msg_to_dev(struct i2400m *, const void *, size_t);
+extern void i2400m_msg_to_dev_cancel_wait(struct i2400m *, int);
+extern void i2400m_msg_ack_hook(struct i2400m *,
+				const struct i2400m_l3l4_hdr *, size_t);
+extern void i2400m_report_hook(struct i2400m *,
+			       const struct i2400m_l3l4_hdr *, size_t);
+extern int i2400m_cmd_enter_powersave(struct i2400m *);
+extern int i2400m_cmd_get_state(struct i2400m *);
+extern int i2400m_cmd_exit_idle(struct i2400m *);
+extern struct sk_buff *i2400m_get_device_info(struct i2400m *);
+extern int i2400m_firmware_check(struct i2400m *);
+extern int i2400m_set_init_config(struct i2400m *,
+				  const struct i2400m_tlv_hdr **, size_t);
+
+static inline
+struct usb_endpoint_descriptor *usb_get_epd(struct usb_interface *iface, int ep)
+{
+	return &iface->cur_altsetting->endpoint[ep].desc;
+}
+
+extern int i2400m_op_rfkill_sw_toggle(struct wimax_dev *,
+				      enum wimax_rf_state);
+extern void i2400m_report_tlv_rf_switches_status(
+	struct i2400m *, const struct i2400m_tlv_rf_switches_status *);
+
+
+/*
+ * Do a millisecond-sleep for allowing wireshark to dump all the data
+ * packets. Used only for debugging.
+ */
+static inline
+void __i2400m_msleep(unsigned ms)
+{
+#if 1
+#else
+	msleep(ms);
+#endif
+}
+
+/* Module parameters */
+
+extern int i2400m_idle_mode_disabled;
+
+
+#endif /* #ifndef __I2400M_H__ */
diff --git a/include/linux/wimax/i2400m.h b/include/linux/wimax/i2400m.h
new file mode 100644
index 000000000000..74198f5bb4dc
--- /dev/null
+++ b/include/linux/wimax/i2400m.h
@@ -0,0 +1,512 @@
+/*
+ * Intel Wireless WiMax Connection 2400m
+ * Host-Device protocol interface definitions
+ *
+ *
+ * Copyright (C) 2007-2008 Intel Corporation. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ *   * Redistributions of source code must retain the above copyright
+ *     notice, this list of conditions and the following disclaimer.
+ *   * Redistributions in binary form must reproduce the above copyright
+ *     notice, this list of conditions and the following disclaimer in
+ *     the documentation and/or other materials provided with the
+ *     distribution.
+ *   * Neither the name of Intel Corporation nor the names of its
+ *     contributors may be used to endorse or promote products derived
+ *     from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ *
+ * Intel Corporation <linux-wimax@intel.com>
+ * Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
+ *  - Initial implementation
+ *
+ *
+ * This header defines the data structures and constants used to
+ * communicate with the device.
+ *
+ * BOOTMODE/BOOTROM/FIRMWARE UPLOAD PROTOCOL
+ *
+ * The firmware upload protocol is quite simple and only requires a
+ * handful of commands. See drivers/net/wimax/i2400m/fw.c for more
+ * details.
+ *
+ * The BCF data structure is for the firmware file header.
+ *
+ *
+ * THE DATA / CONTROL PROTOCOL
+ *
+ * This is the normal protocol spoken with the device once the
+ * firmware is uploaded. It transports data payloads and control
+ * messages back and forth.
+ *
+ * It consists 'messages' that pack one or more payloads each. The
+ * format is described in detail in drivers/net/wimax/i2400m/rx.c and
+ * tx.c.
+ *
+ *
+ * THE L3L4 PROTOCOL
+ *
+ * The term L3L4 refers to Layer 3 (the device), Layer 4 (the
+ * driver/host software).
+ *
+ * This is the control protocol used by the host to control the i2400m
+ * device (scan, connect, disconnect...). This is sent to / received
+ * as control frames. These frames consist of a header and zero or
+ * more TLVs with information. We call each control frame a "message".
+ *
+ * Each message is composed of:
+ *
+ * HEADER
+ * [TLV0 + PAYLOAD0]
+ * [TLV1 + PAYLOAD1]
+ * [...]
+ * [TLVN + PAYLOADN]
+ *
+ * The HEADER is defined by 'struct i2400m_l3l4_hdr'. The payloads are
+ * defined by a TLV structure (Type Length Value) which is a 'header'
+ * (struct i2400m_tlv_hdr) and then the payload.
+ *
+ * All integers are represented as Little Endian.
+ *
+ * - REQUESTS AND EVENTS
+ *
+ * The requests can be clasified as follows:
+ *
+ *   COMMAND:  implies a request from the host to the device requesting
+ *             an action being performed. The device will reply with a
+ *             message (with the same type as the command), status and
+ *             no (TLV) payload. Execution of a command might cause
+ *             events (of different type) to be sent later on as
+ *             device's state changes.
+ *
+ *   GET/SET:  similar to COMMAND, but will not cause other
+ *             EVENTs. The reply, in the case of GET, will contain
+ *             TLVs with the requested information.
+ *
+ *   EVENT:    asynchronous messages sent from the device, maybe as a
+ *             consequence of previous COMMANDs but disassociated from
+ *             them.
+ *
+ * Only one request might be pending at the same time (ie: don't
+ * parallelize nor post another GET request before the previous
+ * COMMAND has been acknowledged with it's corresponding reply by the
+ * device).
+ *
+ * The different requests and their formats are described below:
+ *
+ *  I2400M_MT_*   Message types
+ *  I2400M_MS_*   Message status (for replies, events)
+ *  i2400m_tlv_*  TLVs
+ *
+ * data types are named 'struct i2400m_msg_OPNAME', OPNAME matching the
+ * operation.
+ */
+
+#ifndef __LINUX__WIMAX__I2400M_H__
+#define __LINUX__WIMAX__I2400M_H__
+
+#include <linux/types.h>
+
+
+/*
+ * Host Device Interface (HDI) common to all busses
+ */
+
+/* Boot-mode (firmware upload mode) commands */
+
+/* Header for the firmware file */
+struct i2400m_bcf_hdr {
+	__le32 module_type;
+	__le32 header_len;
+	__le32 header_version;
+	__le32 module_id;
+	__le32 module_vendor;
+	__le32 date;		/* BCD YYYMMDD */
+	__le32 size;
+	__le32 key_size;	/* in dwords */
+	__le32 modulus_size;	/* in dwords */
+	__le32 exponent_size;	/* in dwords */
+	__u8 reserved[88];
+} __attribute__ ((packed));
+
+/* Boot mode opcodes */
+enum i2400m_brh_opcode {
+	I2400M_BRH_READ = 1,
+	I2400M_BRH_WRITE = 2,
+	I2400M_BRH_JUMP = 3,
+	I2400M_BRH_SIGNED_JUMP = 8,
+	I2400M_BRH_HASH_PAYLOAD_ONLY = 9,
+};
+
+/* Boot mode command masks and stuff */
+enum i2400m_brh {
+	I2400M_BRH_SIGNATURE = 0xcbbc0000,
+	I2400M_BRH_SIGNATURE_MASK = 0xffff0000,
+	I2400M_BRH_SIGNATURE_SHIFT = 16,
+	I2400M_BRH_OPCODE_MASK = 0x0000000f,
+	I2400M_BRH_RESPONSE_MASK = 0x000000f0,
+	I2400M_BRH_RESPONSE_SHIFT = 4,
+	I2400M_BRH_DIRECT_ACCESS = 0x00000400,
+	I2400M_BRH_RESPONSE_REQUIRED = 0x00000200,
+	I2400M_BRH_USE_CHECKSUM = 0x00000100,
+};
+
+
+/* Constants for bcf->module_id */
+enum i2400m_bcf_mod_id {
+	/* Firmware file carries its own pokes -- pokes are a set of
+	 * magical values that have to be written in certain memory
+	 * addresses to get the device up and ready for firmware
+	 * download when it is in non-signed boot mode. */
+	I2400M_BCF_MOD_ID_POKES = 0x000000001,
+};
+
+
+/**
+ * i2400m_bootrom_header - Header for a boot-mode command
+ *
+ * @cmd: the above command descriptor
+ * @target_addr: where on the device memory should the action be performed.
+ * @data_size: for read/write, amount of data to be read/written
+ * @block_checksum: checksum value (if applicable)
+ * @payload: the beginning of data attached to this header
+ */
+struct i2400m_bootrom_header {
+	__le32 command;		/* Compose with enum i2400_brh */
+	__le32 target_addr;
+	__le32 data_size;
+	__le32 block_checksum;
+	char payload[0];
+} __attribute__ ((packed));
+
+
+/*
+ * Data / control protocol
+ */
+
+/* Packet types for the host-device interface */
+enum i2400m_pt {
+	I2400M_PT_DATA = 0,
+	I2400M_PT_CTRL,
+	I2400M_PT_TRACE,	/* For device debug */
+	I2400M_PT_RESET_WARM,	/* device reset */
+	I2400M_PT_RESET_COLD,	/* USB[transport] reset, like reconnect */
+	I2400M_PT_ILLEGAL
+};
+
+
+/*
+ * Payload for a data packet
+ *
+ * This is prefixed to each and every outgoing DATA type.
+ */
+struct i2400m_pl_data_hdr {
+	__le32 reserved;
+} __attribute__((packed));
+
+
+/* Misc constants */
+enum {
+	I2400M_PL_PAD = 16,	/* Payload data size alignment */
+	I2400M_PL_SIZE_MAX = 0x3EFF,
+	I2400M_MAX_PLS_IN_MSG = 60,
+	/* protocol barkers: sync sequences; for notifications they
+	 * are sent in groups of four. */
+	I2400M_H2D_PREVIEW_BARKER = 0xcafe900d,
+	I2400M_COLD_RESET_BARKER = 0xc01dc01d,
+	I2400M_WARM_RESET_BARKER = 0x50f750f7,
+	I2400M_NBOOT_BARKER = 0xdeadbeef,
+	I2400M_SBOOT_BARKER = 0x0ff1c1a1,
+	I2400M_ACK_BARKER = 0xfeedbabe,
+	I2400M_D2H_MSG_BARKER = 0xbeefbabe,
+};
+
+
+/*
+ * Hardware payload descriptor
+ *
+ * Bitfields encoded in a struct to enforce typing semantics.
+ *
+ * Look in rx.c and tx.c for a full description of the format.
+ */
+struct i2400m_pld {
+	__le32 val;
+} __attribute__ ((packed));
+
+#define I2400M_PLD_SIZE_MASK 0x00003fff
+#define I2400M_PLD_TYPE_SHIFT 16
+#define I2400M_PLD_TYPE_MASK 0x000f0000
+
+/*
+ * Header for a TX message or RX message
+ *
+ * @barker: preamble
+ * @size: used for management of the FIFO queue buffer; before
+ *     sending, this is converted to be a real preamble. This
+ *     indicates the real size of the TX message that starts at this
+ *     point. If the highest bit is set, then this message is to be
+ *     skipped.
+ * @sequence: sequence number of this message
+ * @offset: offset where the message itself starts -- see the comments
+ *     in the file header about message header and payload descriptor
+ *     alignment.
+ * @num_pls: number of payloads in this message
+ * @padding: amount of padding bytes at the end of the message to make
+ *           it be of block-size aligned
+ *
+ * Look in rx.c and tx.c for a full description of the format.
+ */
+struct i2400m_msg_hdr {
+	union {
+		__le32 barker;
+		__u32 size;	/* same size type as barker!! */
+	};
+	union {
+		__le32 sequence;
+		__u32 offset;	/* same size type as barker!! */
+	};
+	__le16 num_pls;
+	__le16 rsv1;
+	__le16 padding;
+	__le16 rsv2;
+	struct i2400m_pld pld[0];
+} __attribute__ ((packed));
+
+
+
+/*
+ * L3/L4 control protocol
+ */
+
+enum {
+	/* Interface version */
+	I2400M_L3L4_VERSION             = 0x0100,
+};
+
+/* Message types */
+enum i2400m_mt {
+	I2400M_MT_RESERVED              = 0x0000,
+	I2400M_MT_INVALID               = 0xffff,
+	I2400M_MT_REPORT_MASK		= 0x8000,
+
+	I2400M_MT_GET_SCAN_RESULT  	= 0x4202,
+	I2400M_MT_SET_SCAN_PARAM   	= 0x4402,
+	I2400M_MT_CMD_RF_CONTROL   	= 0x4602,
+	I2400M_MT_CMD_SCAN         	= 0x4603,
+	I2400M_MT_CMD_CONNECT      	= 0x4604,
+	I2400M_MT_CMD_DISCONNECT   	= 0x4605,
+	I2400M_MT_CMD_EXIT_IDLE   	= 0x4606,
+	I2400M_MT_GET_LM_VERSION   	= 0x5201,
+	I2400M_MT_GET_DEVICE_INFO  	= 0x5202,
+	I2400M_MT_GET_LINK_STATUS  	= 0x5203,
+	I2400M_MT_GET_STATISTICS   	= 0x5204,
+	I2400M_MT_GET_STATE        	= 0x5205,
+	I2400M_MT_GET_MEDIA_STATUS	= 0x5206,
+	I2400M_MT_SET_INIT_CONFIG	= 0x5404,
+	I2400M_MT_CMD_INIT	        = 0x5601,
+	I2400M_MT_CMD_TERMINATE		= 0x5602,
+	I2400M_MT_CMD_MODE_OF_OP	= 0x5603,
+	I2400M_MT_CMD_RESET_DEVICE	= 0x5604,
+	I2400M_MT_CMD_MONITOR_CONTROL   = 0x5605,
+	I2400M_MT_CMD_ENTER_POWERSAVE   = 0x5606,
+	I2400M_MT_GET_TLS_OPERATION_RESULT = 0x6201,
+	I2400M_MT_SET_EAP_SUCCESS       = 0x6402,
+	I2400M_MT_SET_EAP_FAIL          = 0x6403,
+	I2400M_MT_SET_EAP_KEY          	= 0x6404,
+	I2400M_MT_CMD_SEND_EAP_RESPONSE = 0x6602,
+	I2400M_MT_REPORT_SCAN_RESULT    = 0xc002,
+	I2400M_MT_REPORT_STATE		= 0xd002,
+	I2400M_MT_REPORT_POWERSAVE_READY = 0xd005,
+	I2400M_MT_REPORT_EAP_REQUEST    = 0xe002,
+	I2400M_MT_REPORT_EAP_RESTART    = 0xe003,
+	I2400M_MT_REPORT_ALT_ACCEPT    	= 0xe004,
+	I2400M_MT_REPORT_KEY_REQUEST 	= 0xe005,
+};
+
+
+/*
+ * Message Ack Status codes
+ *
+ * When a message is replied-to, this status is reported.
+ */
+enum i2400m_ms {
+	I2400M_MS_DONE_OK                  = 0,
+	I2400M_MS_DONE_IN_PROGRESS         = 1,
+	I2400M_MS_INVALID_OP               = 2,
+	I2400M_MS_BAD_STATE                = 3,
+	I2400M_MS_ILLEGAL_VALUE            = 4,
+	I2400M_MS_MISSING_PARAMS           = 5,
+	I2400M_MS_VERSION_ERROR            = 6,
+	I2400M_MS_ACCESSIBILITY_ERROR      = 7,
+	I2400M_MS_BUSY                     = 8,
+	I2400M_MS_CORRUPTED_TLV            = 9,
+	I2400M_MS_UNINITIALIZED            = 10,
+	I2400M_MS_UNKNOWN_ERROR            = 11,
+	I2400M_MS_PRODUCTION_ERROR         = 12,
+	I2400M_MS_NO_RF                    = 13,
+	I2400M_MS_NOT_READY_FOR_POWERSAVE  = 14,
+	I2400M_MS_THERMAL_CRITICAL         = 15,
+	I2400M_MS_MAX
+};
+
+
+/**
+ * i2400m_tlv - enumeration of the different types of TLVs
+ *
+ * TLVs stand for type-length-value and are the header for a payload
+ * composed of almost anything. Each payload has a type assigned
+ * and a length.
+ */
+enum i2400m_tlv {
+	I2400M_TLV_L4_MESSAGE_VERSIONS = 129,
+	I2400M_TLV_SYSTEM_STATE = 141,
+	I2400M_TLV_MEDIA_STATUS = 161,
+	I2400M_TLV_RF_OPERATION = 162,
+	I2400M_TLV_RF_STATUS = 163,
+	I2400M_TLV_DEVICE_RESET_TYPE = 132,
+	I2400M_TLV_CONFIG_IDLE_PARAMETERS = 601,
+};
+
+
+struct i2400m_tlv_hdr {
+	__le16 type;
+	__le16 length;		/* payload's */
+	__u8   pl[0];
+} __attribute__((packed));
+
+
+struct i2400m_l3l4_hdr {
+	__le16 type;
+	__le16 length;		/* payload's */
+	__le16 version;
+	__le16 resv1;
+	__le16 status;
+	__le16 resv2;
+	struct i2400m_tlv_hdr pl[0];
+} __attribute__((packed));
+
+
+/**
+ * i2400m_system_state - different states of the device
+ */
+enum i2400m_system_state {
+	I2400M_SS_UNINITIALIZED = 1,
+	I2400M_SS_INIT,
+	I2400M_SS_READY,
+	I2400M_SS_SCAN,
+	I2400M_SS_STANDBY,
+	I2400M_SS_CONNECTING,
+	I2400M_SS_WIMAX_CONNECTED,
+	I2400M_SS_DATA_PATH_CONNECTED,
+	I2400M_SS_IDLE,
+	I2400M_SS_DISCONNECTING,
+	I2400M_SS_OUT_OF_ZONE,
+	I2400M_SS_SLEEPACTIVE,
+	I2400M_SS_PRODUCTION,
+	I2400M_SS_CONFIG,
+	I2400M_SS_RF_OFF,
+	I2400M_SS_RF_SHUTDOWN,
+	I2400M_SS_DEVICE_DISCONNECT,
+	I2400M_SS_MAX,
+};
+
+
+/**
+ * i2400m_tlv_system_state - report on the state of the system
+ *
+ * @state: see enum i2400m_system_state
+ */
+struct i2400m_tlv_system_state {
+	struct i2400m_tlv_hdr hdr;
+	__le32 state;
+} __attribute__((packed));
+
+
+struct i2400m_tlv_l4_message_versions {
+	struct i2400m_tlv_hdr hdr;
+	__le16 major;
+	__le16 minor;
+	__le16 branch;
+	__le16 reserved;
+} __attribute__((packed));
+
+
+struct i2400m_tlv_detailed_device_info {
+	struct i2400m_tlv_hdr hdr;
+	__u8 reserved1[400];
+	__u8 mac_address[6];
+	__u8 reserved2[2];
+} __attribute__((packed));
+
+
+enum i2400m_rf_switch_status {
+	I2400M_RF_SWITCH_ON = 1,
+	I2400M_RF_SWITCH_OFF = 2,
+};
+
+struct i2400m_tlv_rf_switches_status {
+	struct i2400m_tlv_hdr hdr;
+	__u8 sw_rf_switch;	/* 1 ON, 2 OFF */
+	__u8 hw_rf_switch;	/* 1 ON, 2 OFF */
+	__u8 reserved[2];
+} __attribute__((packed));
+
+
+enum {
+	i2400m_rf_operation_on = 1,
+	i2400m_rf_operation_off = 2
+};
+
+struct i2400m_tlv_rf_operation {
+	struct i2400m_tlv_hdr hdr;
+	__le32 status;	/* 1 ON, 2 OFF */
+} __attribute__((packed));
+
+
+enum i2400m_tlv_reset_type {
+	I2400M_RESET_TYPE_COLD = 1,
+	I2400M_RESET_TYPE_WARM
+};
+
+struct i2400m_tlv_device_reset_type {
+	struct i2400m_tlv_hdr hdr;
+	__le32 reset_type;
+} __attribute__((packed));
+
+
+struct i2400m_tlv_config_idle_parameters {
+	struct i2400m_tlv_hdr hdr;
+	__le32 idle_timeout;	/* 100 to 300000 ms [5min], 100 increments
+				 * 0 disabled */
+	__le32 idle_paging_interval;	/* frames */
+} __attribute__((packed));
+
+
+enum i2400m_media_status {
+	I2400M_MEDIA_STATUS_LINK_UP = 1,
+	I2400M_MEDIA_STATUS_LINK_DOWN,
+	I2400M_MEDIA_STATUS_LINK_RENEW,
+};
+
+struct i2400m_tlv_media_status {
+	struct i2400m_tlv_hdr hdr;
+	__le32 media_status;
+} __attribute__((packed));
+
+#endif /* #ifndef __LINUX__WIMAX__I2400M_H__ */
-- 
cgit v1.2.3


From e30698743419d20dce03d033761f203b4d847ab0 Mon Sep 17 00:00:00 2001
From: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Date: Sat, 20 Dec 2008 16:57:59 -0800
Subject: wimax: export linux/wimax.h and linux/wimax/i2400m.h with
 headers_install

These two files are what user space can use to establish communication
with the WiMAX kernel API and to speak the Intel 2400m Wireless WiMAX
connection's control protocol.

Signed-off-by: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/Kbuild       | 2 ++
 include/linux/wimax/Kbuild | 1 +
 2 files changed, 3 insertions(+)
 create mode 100644 include/linux/wimax/Kbuild

(limited to 'include/linux')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index a3323f337e4d..12e9a2957caf 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -371,3 +371,5 @@ unifdef-y += xattr.h
 unifdef-y += xfrm.h
 
 objhdr-y += version.h
+header-y += wimax.h
+header-y += wimax/
diff --git a/include/linux/wimax/Kbuild b/include/linux/wimax/Kbuild
new file mode 100644
index 000000000000..3cb4f269bb09
--- /dev/null
+++ b/include/linux/wimax/Kbuild
@@ -0,0 +1 @@
+header-y += i2400m.h
-- 
cgit v1.2.3


From f7b7baae6b30ff04124259ff8d7c0c0d281320e6 Mon Sep 17 00:00:00 2001
From: Sheng Yang <sheng@linux.intel.com>
Date: Tue, 11 Nov 2008 17:17:46 +0800
Subject: PCI: add PCI Advanced Feature Capability defines

PCI Advanced Features Capability is introduced by "Conventional PCI
Advanced Caps ECN" (can be downloaded in pcisig.com).  Add defines for
the various AF capabilities, including function level reset (FLR).

Reviewed-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 include/linux/pci_regs.h | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h
index e5effd47ed74..7766488470e4 100644
--- a/include/linux/pci_regs.h
+++ b/include/linux/pci_regs.h
@@ -210,6 +210,7 @@
 #define  PCI_CAP_ID_AGP3	0x0E	/* AGP Target PCI-PCI bridge */
 #define  PCI_CAP_ID_EXP 	0x10	/* PCI Express */
 #define  PCI_CAP_ID_MSIX	0x11	/* MSI-X */
+#define  PCI_CAP_ID_AF		0x13	/* PCI Advanced Features */
 #define PCI_CAP_LIST_NEXT	1	/* Next capability in the list */
 #define PCI_CAP_FLAGS		2	/* Capability defined flags (16 bits) */
 #define PCI_CAP_SIZEOF		4
@@ -316,6 +317,17 @@
 #define  PCI_CHSWP_EXT		0x40	/* ENUM# status - extraction */
 #define  PCI_CHSWP_INS		0x80	/* ENUM# status - insertion */
 
+/* PCI Advanced Feature registers */
+
+#define PCI_AF_LENGTH		2
+#define PCI_AF_CAP		3
+#define  PCI_AF_CAP_TP		0x01
+#define  PCI_AF_CAP_FLR		0x02
+#define PCI_AF_CTRL		4
+#define  PCI_AF_CTRL_FLR	0x01
+#define PCI_AF_STATUS		5
+#define  PCI_AF_STATUS_TP	0x01
+
 /* PCI-X registers */
 
 #define PCI_X_CMD		2	/* Modes & Features */
-- 
cgit v1.2.3


From 8b62091e20215730be1b94b7cd135a78a3e692ca Mon Sep 17 00:00:00 2001
From: Andrew Patterson <andrew.patterson@hp.com>
Date: Mon, 10 Nov 2008 15:30:40 -0700
Subject: ACPI/PCI: include missing acpi.h file in pci-acpi.h.

The pci-acpi.h file will not compile without including linux/acpi.h.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 include/linux/pci-acpi.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/pci-acpi.h b/include/linux/pci-acpi.h
index 8837928fbf33..a9e4c34e9389 100644
--- a/include/linux/pci-acpi.h
+++ b/include/linux/pci-acpi.h
@@ -8,6 +8,8 @@
 #ifndef _PCI_ACPI_H_
 #define _PCI_ACPI_H_
 
+#include <linux/acpi.h>
+
 #define OSC_QUERY_TYPE			0
 #define OSC_SUPPORT_TYPE 		1
 #define OSC_CONTROL_TYPE		2
-- 
cgit v1.2.3


From 990a7ac5645883a833a11b900bb6f25b65dea65b Mon Sep 17 00:00:00 2001
From: Andrew Patterson <andrew.patterson@hp.com>
Date: Mon, 10 Nov 2008 15:30:45 -0700
Subject: ACPI/PCI: call _OSC support during root bridge discovery

Add pci_acpi_osc_support() and call it when a PCI bridge is added.  This
allows us to avoid having every individual PCI root bridge driver call
_OSC support for every root bridge in their probe functions, a
significant savings in boot time.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/acpi/pci_root.c  |  9 +++++++++
 drivers/pci/pci-acpi.c   | 24 +++++++++++++++++++-----
 include/linux/pci-acpi.h |  1 +
 3 files changed, 29 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index 642554b1b60c..de4d57114fe4 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -31,6 +31,7 @@
 #include <linux/spinlock.h>
 #include <linux/pm.h>
 #include <linux/pci.h>
+#include <linux/pci-acpi.h>
 #include <linux/acpi.h>
 #include <acpi/acpi_bus.h>
 #include <acpi/acpi_drivers.h>
@@ -193,6 +194,7 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	unsigned long long value = 0;
 	acpi_handle handle = NULL;
 	struct acpi_device *child;
+	u32 flags;
 
 
 	if (!device)
@@ -210,6 +212,13 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 
 	device->ops.bind = acpi_pci_bind;
 
+	/*
+	 * All supported architectures that use ACPI have support for
+	 * PCI domains, so we indicate this in _OSC support capabilities.
+	 */
+	flags = OSC_PCI_SEGMENT_GROUPS_SUPPORT;
+	pci_acpi_osc_support(device->handle, flags);
+
 	/* 
 	 * Segment
 	 * -------
diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c
index 2ed3f10d0860..8a1f02c3c915 100644
--- a/drivers/pci/pci-acpi.c
+++ b/drivers/pci/pci-acpi.c
@@ -143,28 +143,42 @@ static acpi_status __acpi_query_osc(u32 flags, struct acpi_osc_data *osc_data,
 	return status;
 }
 
-static acpi_status acpi_query_osc(acpi_handle handle,
-				  u32 level, void *context, void **retval)
+/*
+ * pci_acpi_osc_support: Invoke _OSC indicating support for the given feature
+ * @flags: Bitmask of flags to support
+ *
+ * See the ACPI spec for the definition of the flags
+ */
+int pci_acpi_osc_support(acpi_handle handle, u32 flags)
 {
+	u32 dummy;
 	acpi_status status;
-	struct acpi_osc_data *osc_data;
-	u32 flags = (unsigned long)context, dummy;
 	acpi_handle tmp;
+	struct acpi_osc_data *osc_data;
+	int rc = 0;
 
 	status = acpi_get_handle(handle, "_OSC", &tmp);
 	if (ACPI_FAILURE(status))
-		return AE_OK;
+		return -ENOTTY;
 
 	mutex_lock(&pci_acpi_lock);
 	osc_data = acpi_get_osc_data(handle);
 	if (!osc_data) {
 		printk(KERN_ERR "acpi osc data array is full\n");
+		rc = -ENOMEM;
 		goto out;
 	}
 
 	__acpi_query_osc(flags, osc_data, &dummy);
 out:
 	mutex_unlock(&pci_acpi_lock);
+	return rc;
+}
+
+static acpi_status acpi_query_osc(acpi_handle handle, u32 level,
+				  void *context, void **retval)
+{
+	pci_acpi_osc_support(handle, (unsigned long)context);
 	return AE_OK;
 }
 
diff --git a/include/linux/pci-acpi.h b/include/linux/pci-acpi.h
index a9e4c34e9389..424f06f84cab 100644
--- a/include/linux/pci-acpi.h
+++ b/include/linux/pci-acpi.h
@@ -51,6 +51,7 @@
 #ifdef CONFIG_ACPI
 extern acpi_status pci_osc_control_set(acpi_handle handle, u32 flags);
 extern acpi_status __pci_osc_support_set(u32 flags, const char *hid);
+int pci_acpi_osc_support(acpi_handle handle, u32 flags);
 static inline acpi_status pci_osc_support_set(u32 flags)
 {
 	return __pci_osc_support_set(flags, PCI_ROOT_HID_STRING);
-- 
cgit v1.2.3


From 0ef5f8f6159e44b4faa997be08d1a3bcbf44ad08 Mon Sep 17 00:00:00 2001
From: Andrew Patterson <andrew.patterson@hp.com>
Date: Mon, 10 Nov 2008 15:30:50 -0700
Subject: ACPI/PCI: PCI extended config _OSC support called when root bridge
 added

The _OSC capability OSC_EXT_PCI_CONFIG_SUPPORT is set when the root
bridge is added with pci_acpi_osc_support() if we can access PCI
extended config space.

This adds the function pci_ext_cfg_avail which returns true if we can
access PCI extended config space (offset greater than 0xff). It
currently only returns false if arch=x86 and raw_pci_ext_ops is not set
(which might happen if pci=nommcfg is set on the kernel command-line).

Signed-off-by: Andrew Patterson <andrew.patterson@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 arch/x86/pci/common.c   |  8 ++++++++
 drivers/acpi/pci_root.c | 10 ++++++++--
 drivers/pci/pci.c       | 13 +++++++++++++
 include/linux/pci.h     |  2 ++
 4 files changed, 31 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c
index 62ddb73e09ed..9ab8509f7b15 100644
--- a/arch/x86/pci/common.c
+++ b/arch/x86/pci/common.c
@@ -562,6 +562,14 @@ void pcibios_disable_device (struct pci_dev *dev)
 		pcibios_disable_irq(dev);
 }
 
+int pci_ext_cfg_avail(struct pci_dev *dev)
+{
+	if (raw_pci_ext_ops)
+		return 1;
+	else
+		return 0;
+}
+
 struct pci_bus * __devinit pci_scan_bus_on_node(int busno, struct pci_ops *ops, int node)
 {
 	struct pci_bus *bus = NULL;
diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index de4d57114fe4..96e68e841539 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -194,7 +194,7 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	unsigned long long value = 0;
 	acpi_handle handle = NULL;
 	struct acpi_device *child;
-	u32 flags;
+	u32 flags, base_flags;
 
 
 	if (!device)
@@ -216,7 +216,7 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	 * All supported architectures that use ACPI have support for
 	 * PCI domains, so we indicate this in _OSC support capabilities.
 	 */
-	flags = OSC_PCI_SEGMENT_GROUPS_SUPPORT;
+	flags = base_flags = OSC_PCI_SEGMENT_GROUPS_SUPPORT;
 	pci_acpi_osc_support(device->handle, flags);
 
 	/* 
@@ -344,6 +344,12 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	list_for_each_entry(child, &device->children, node)
 		acpi_pci_bridge_scan(child);
 
+	/* Indicate support for various _OSC capabilities. */
+	if (pci_ext_cfg_avail(root->bus->self))
+		flags |= OSC_EXT_PCI_CONFIG_SUPPORT;
+	if (flags != base_flags)
+		pci_acpi_osc_support(device->handle, flags);
+
       end:
 	if (result) {
 		if (!list_empty(&root->node))
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 3c2fa2fdc9cd..48fa860276d4 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -2084,6 +2084,19 @@ static void __devinit pci_no_domains(void)
 #endif
 }
 
+/**
+ * pci_ext_cfg_enabled - can we access extended PCI config space?
+ * @dev: The PCI device of the root bridge.
+ *
+ * Returns 1 if we can access PCI extended config space (offsets
+ * greater than 0xff). This is the default implementation. Architecture
+ * implementations can override this.
+ */
+int __attribute__ ((weak)) pci_ext_cfg_avail(struct pci_dev *dev)
+{
+	return 1;
+}
+
 static int __devinit pci_init(void)
 {
 	struct pci_dev *dev = NULL;
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 4bb156ba854a..6fd47654ca4e 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -1140,6 +1140,8 @@ static inline void pci_mmcfg_early_init(void) { }
 static inline void pci_mmcfg_late_init(void) { }
 #endif
 
+int pci_ext_cfg_avail(struct pci_dev *dev);
+
 #ifdef CONFIG_HAS_IOMEM
 static inline void __iomem *pci_ioremap_bar(struct pci_dev *pdev, int bar)
 {
-- 
cgit v1.2.3


From 3e1b16002af29758b6bc9c38939d43838d9335bc Mon Sep 17 00:00:00 2001
From: Andrew Patterson <andrew.patterson@hp.com>
Date: Mon, 10 Nov 2008 15:30:55 -0700
Subject: ACPI/PCI: PCIe ASPM _OSC support capabilities called when root bridge
 added

The _OSC capabilities OSC_ACTIVE_STATE_PWR_SUPPORT and
OSC_CLOCK_PWR_CAPABILITY_SUPPORT are set when the root bridge is added
with pci_acpi_osc_support(), so we no longer need to do it in the ASPM
driver.  Also add the function pcie_aspm_enabled, which returns true if
pcie_aspm=off is not on the kernel command-line.

Signed-off-by: Andrew Patterson <andrew.patterson@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/acpi/pci_root.c |  3 +++
 drivers/pci/pcie/aspm.c | 27 +++++++++------------------
 include/linux/pci.h     |  9 +++++++++
 3 files changed, 21 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index 96e68e841539..9fe026b1c9d0 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -347,6 +347,9 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	/* Indicate support for various _OSC capabilities. */
 	if (pci_ext_cfg_avail(root->bus->self))
 		flags |= OSC_EXT_PCI_CONFIG_SUPPORT;
+	if (pcie_aspm_enabled())
+		flags |= OSC_ACTIVE_STATE_PWR_SUPPORT |
+			OSC_CLOCK_PWR_CAPABILITY_SUPPORT;
 	if (flags != base_flags)
 		pci_acpi_osc_support(device->handle, flags);
 
diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 9aad608bcf3f..e361c7dc726f 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -857,24 +857,15 @@ void pcie_no_aspm(void)
 		aspm_disabled = 1;
 }
 
-#ifdef CONFIG_ACPI
-#include <acpi/acpi_bus.h>
-#include <linux/pci-acpi.h>
-static void pcie_aspm_platform_init(void)
-{
-	pcie_osc_support_set(OSC_ACTIVE_STATE_PWR_SUPPORT|
-		OSC_CLOCK_PWR_CAPABILITY_SUPPORT);
-}
-#else
-static inline void pcie_aspm_platform_init(void) { }
-#endif
-
-static int __init pcie_aspm_init(void)
+/**
+ * pcie_aspm_enabled - is PCIe ASPM enabled?
+ *
+ * Returns true if ASPM has not been disabled by the command-line option
+ * pcie_aspm=off.
+ **/
+int pcie_aspm_enabled(void)
 {
-	if (aspm_disabled)
-		return 0;
-	pcie_aspm_platform_init();
-	return 0;
+       return !aspm_disabled;
 }
+EXPORT_SYMBOL(pcie_aspm_enabled);
 
-fs_initcall(pcie_aspm_init);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 6fd47654ca4e..eae97a2bf603 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -791,6 +791,15 @@ extern void msi_remove_pci_irq_vectors(struct pci_dev *dev);
 extern void pci_restore_msi_state(struct pci_dev *dev);
 #endif
 
+#ifndef CONFIG_PCIEASPM
+static inline int pcie_aspm_enabled(void)
+{
+	return 0;
+}
+#else
+extern int pcie_aspm_enabled(void);
+#endif
+
 #ifdef CONFIG_HT_IRQ
 /* The functions a driver should call */
 int  ht_create_irq(struct pci_dev *dev, int idx);
-- 
cgit v1.2.3


From 07ae95f988a34465bdcb384bfa73c03424fe2312 Mon Sep 17 00:00:00 2001
From: Andrew Patterson <andrew.patterson@hp.com>
Date: Mon, 10 Nov 2008 15:31:05 -0700
Subject: ACPI/PCI: PCI MSI _OSC support capabilities called when root bridge
 added

The _OSC capability OSC_MSI_SUPPORT is set when the root bridge is added
with pci_acpi_osc_support(), so we no longer need to do it in the PCI
MSI driver.  Also adds the function pci_msi_enabled, which returns true
if pci=nomsi is not on the kernel command-line.

Signed-off-by: Andrew Patterson <andrew.patterson@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/acpi/pci_root.c |  2 ++
 drivers/pci/msi.c       | 31 +++++++++++--------------------
 drivers/pci/pci.c       |  2 --
 drivers/pci/pci.h       |  2 --
 include/linux/pci.h     |  5 +++++
 5 files changed, 18 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index 9fe026b1c9d0..5b38a026d122 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -350,6 +350,8 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	if (pcie_aspm_enabled())
 		flags |= OSC_ACTIVE_STATE_PWR_SUPPORT |
 			OSC_CLOCK_PWR_CAPABILITY_SUPPORT;
+	if (pci_msi_enabled())
+		flags |= OSC_MSI_SUPPORT;
 	if (flags != base_flags)
 		pci_acpi_osc_support(device->handle, flags);
 
diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index 11a51f8ed3b3..b4a90badd0a6 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -776,28 +776,19 @@ void pci_no_msi(void)
 	pci_msi_enable = 0;
 }
 
-void pci_msi_init_pci_dev(struct pci_dev *dev)
-{
-	INIT_LIST_HEAD(&dev->msi_list);
-}
-
-#ifdef CONFIG_ACPI
-#include <linux/acpi.h>
-#include <linux/pci-acpi.h>
-static void __devinit msi_acpi_init(void)
+/**
+ * pci_msi_enabled - is MSI enabled?
+ *
+ * Returns true if MSI has not been disabled by the command-line option
+ * pci=nomsi.
+ **/
+int pci_msi_enabled(void)
 {
-	if (acpi_pci_disabled)
-		return;
-	pci_osc_support_set(OSC_MSI_SUPPORT);
-	pcie_osc_support_set(OSC_MSI_SUPPORT);
+	return pci_msi_enable;
 }
-#else
-static inline void msi_acpi_init(void) { }
-#endif /* CONFIG_ACPI */
+EXPORT_SYMBOL(pci_msi_enabled);
 
-void __devinit msi_init(void)
+void pci_msi_init_pci_dev(struct pci_dev *dev)
 {
-	if (!pci_msi_enable)
-		return;
-	msi_acpi_init();
+	INIT_LIST_HEAD(&dev->msi_list);
 }
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 48fa860276d4..2cfa41e367a7 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -2105,8 +2105,6 @@ static int __devinit pci_init(void)
 		pci_fixup_device(pci_fixup_final, dev);
 	}
 
-	msi_init();
-
 	return 0;
 }
 
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index d3e65e29df51..9162e242b99e 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -102,11 +102,9 @@ extern unsigned int pci_pm_d3_delay;
 #ifdef CONFIG_PCI_MSI
 void pci_no_msi(void);
 extern void pci_msi_init_pci_dev(struct pci_dev *dev);
-extern void __devinit msi_init(void);
 #else
 static inline void pci_no_msi(void) { }
 static inline void pci_msi_init_pci_dev(struct pci_dev *dev) { }
-static inline void msi_init(void) { }
 #endif
 
 #ifdef CONFIG_PCIEAER
diff --git a/include/linux/pci.h b/include/linux/pci.h
index eae97a2bf603..59a3dc2059d3 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -779,6 +779,10 @@ static inline void msi_remove_pci_irq_vectors(struct pci_dev *dev)
 
 static inline void pci_restore_msi_state(struct pci_dev *dev)
 { }
+static inline int pci_msi_enabled(void)
+{
+	return 0;
+}
 #else
 extern int pci_enable_msi(struct pci_dev *dev);
 extern void pci_msi_shutdown(struct pci_dev *dev);
@@ -789,6 +793,7 @@ extern void pci_msix_shutdown(struct pci_dev *dev);
 extern void pci_disable_msix(struct pci_dev *dev);
 extern void msi_remove_pci_irq_vectors(struct pci_dev *dev);
 extern void pci_restore_msi_state(struct pci_dev *dev);
+extern int pci_msi_enabled(void);
 #endif
 
 #ifndef CONFIG_PCIEASPM
-- 
cgit v1.2.3


From 23616941914917cf25b94789856b5326b68d8ee8 Mon Sep 17 00:00:00 2001
From: Andrew Patterson <andrew.patterson@hp.com>
Date: Mon, 10 Nov 2008 15:31:10 -0700
Subject: ACPI/PCI: remove obsolete _OSC capability support functions

The acpi_query_osc, __pci_osc_support_set, pci_osc_support_set, and
pcie_osc_support_set functions have been obsoleted in favor of setting
these capabilities during root bridge discovery with
pci_acpi_osc_support.  There are no longer any callers of these
functions, so remove them.

Signed-off-by: Andrew Patterson <andrew.patterson@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci-acpi.c   | 25 -------------------------
 include/linux/pci-acpi.h | 11 -----------
 2 files changed, 36 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c
index 8a1f02c3c915..8f923ee5177f 100644
--- a/drivers/pci/pci-acpi.c
+++ b/drivers/pci/pci-acpi.c
@@ -175,31 +175,6 @@ out:
 	return rc;
 }
 
-static acpi_status acpi_query_osc(acpi_handle handle, u32 level,
-				  void *context, void **retval)
-{
-	pci_acpi_osc_support(handle, (unsigned long)context);
-	return AE_OK;
-}
-
-/**
- * __pci_osc_support_set - register OS support to Firmware
- * @flags: OS support bits
- * @hid: hardware ID
- *
- * Update OS support fields and doing a _OSC Query to obtain an update
- * from Firmware on supported control bits.
- **/
-acpi_status __pci_osc_support_set(u32 flags, const char *hid)
-{
-	if (!(flags & OSC_SUPPORT_MASKS))
-		return AE_TYPE;
-
-	acpi_get_devices(hid, acpi_query_osc,
-			 (void *)(unsigned long)flags, NULL);
-	return AE_OK;
-}
-
 /**
  * pci_osc_control_set - commit requested control to Firmware
  * @handle: acpi_handle for the target ACPI object
diff --git a/include/linux/pci-acpi.h b/include/linux/pci-acpi.h
index 424f06f84cab..871e096e0fbc 100644
--- a/include/linux/pci-acpi.h
+++ b/include/linux/pci-acpi.h
@@ -50,16 +50,7 @@
 
 #ifdef CONFIG_ACPI
 extern acpi_status pci_osc_control_set(acpi_handle handle, u32 flags);
-extern acpi_status __pci_osc_support_set(u32 flags, const char *hid);
 int pci_acpi_osc_support(acpi_handle handle, u32 flags);
-static inline acpi_status pci_osc_support_set(u32 flags)
-{
-	return __pci_osc_support_set(flags, PCI_ROOT_HID_STRING);
-}
-static inline acpi_status pcie_osc_support_set(u32 flags)
-{
-	return __pci_osc_support_set(flags, PCI_EXPRESS_ROOT_HID_STRING);
-}
 static inline acpi_handle acpi_find_root_bridge_handle(struct pci_dev *pdev)
 {
 	/* Find root host bridge */
@@ -76,8 +67,6 @@ typedef u32 		acpi_status;
 #endif    
 static inline acpi_status pci_osc_control_set(acpi_handle handle, u32 flags)
 {return AE_ERROR;}
-static inline acpi_status pci_osc_support_set(u32 flags) {return AE_ERROR;} 
-static inline acpi_status pcie_osc_support_set(u32 flags) {return AE_ERROR;}
 static inline acpi_handle acpi_find_root_bridge_handle(struct pci_dev *pdev)
 { return NULL; }
 #endif
-- 
cgit v1.2.3


From e8de1481fd7126ee9e93d6889da6f00c05e1e019 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Wed, 22 Oct 2008 19:55:31 -0700
Subject: resource: allow MMIO exclusivity for device drivers

Device drivers that use pci_request_regions() (and similar APIs) have a
reasonable expectation that they are the only ones accessing their device.
As part of the e1000e hunt, we were afraid that some userland (X or some
bootsplash stuff) was mapping the MMIO region that the driver thought it
had exclusively via /dev/mem or via various sysfs resource mappings.

This patch adds the option for device drivers to cause their reserved
regions to the "banned from /dev/mem use" list, so now both kernel memory
and device-exclusive MMIO regions are banned.
NOTE: This is only active when CONFIG_STRICT_DEVMEM is set.

In addition to the config option, a kernel parameter iomem=relaxed is
provided for the cases where developers want to diagnose, in the field,
drivers issues from userspace.

Reviewed-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 Documentation/kernel-parameters.txt |   4 ++
 arch/x86/mm/init_32.c               |   2 +
 arch/x86/mm/init_64.c               |   2 +
 drivers/net/e1000e/netdev.c         |   2 +-
 drivers/pci/pci-sysfs.c             |   3 +
 drivers/pci/pci.c                   | 107 ++++++++++++++++++++++++++++++++----
 include/linux/ioport.h              |  11 +++-
 include/linux/pci.h                 |   3 +
 kernel/resource.c                   |  61 +++++++++++++++++++-
 9 files changed, 176 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 0b3f6711d2f1..0072fabb1dd1 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -918,6 +918,10 @@ and is between 256 and 4096 characters. It is defined in the file
 
 	inttest=	[IA64]
 
+	iomem=		Disable strict checking of access to MMIO memory
+		strict	regions from userspace.
+		relaxed
+
 	iommu=		[x86]
 		off
 		force
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 544d724caeee..88f1b10de3be 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -328,6 +328,8 @@ int devmem_is_allowed(unsigned long pagenr)
 {
 	if (pagenr <= 256)
 		return 1;
+	if (iomem_is_exclusive(pagenr << PAGE_SHIFT))
+		return 0;
 	if (!page_is_ram(pagenr))
 		return 1;
 	return 0;
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 54c437e96541..23f68e77ad1f 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -888,6 +888,8 @@ int devmem_is_allowed(unsigned long pagenr)
 {
 	if (pagenr <= 256)
 		return 1;
+	if (iomem_is_exclusive(pagenr << PAGE_SHIFT))
+		return 0;
 	if (!page_is_ram(pagenr))
 		return 1;
 	return 0;
diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c
index d4639facd1bd..91817d0afcaf 100644
--- a/drivers/net/e1000e/netdev.c
+++ b/drivers/net/e1000e/netdev.c
@@ -4807,7 +4807,7 @@ static int __devinit e1000_probe(struct pci_dev *pdev,
 		}
 	}
 
-	err = pci_request_selected_regions(pdev,
+	err = pci_request_selected_regions_exclusive(pdev,
 	                                  pci_select_bars(pdev, IORESOURCE_MEM),
 	                                  e1000e_driver_name);
 	if (err)
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index 388440e0d222..d5cdccf27a69 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -620,6 +620,9 @@ pci_mmap_resource(struct kobject *kobj, struct bin_attribute *attr,
 	vma->vm_pgoff += start >> PAGE_SHIFT;
 	mmap_type = res->flags & IORESOURCE_MEM ? pci_mmap_mem : pci_mmap_io;
 
+	if (res->flags & IORESOURCE_MEM && iomem_is_exclusive(start))
+		return -EINVAL;
+
 	return pci_mmap_page_range(pdev, vma, mmap_type, write_combine);
 }
 
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 2cfa41e367a7..47663dc0daf7 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -1395,7 +1395,8 @@ void pci_release_region(struct pci_dev *pdev, int bar)
  *	Returns 0 on success, or %EBUSY on error.  A warning
  *	message is also printed on failure.
  */
-int pci_request_region(struct pci_dev *pdev, int bar, const char *res_name)
+static int __pci_request_region(struct pci_dev *pdev, int bar, const char *res_name,
+									int exclusive)
 {
 	struct pci_devres *dr;
 
@@ -1408,8 +1409,9 @@ int pci_request_region(struct pci_dev *pdev, int bar, const char *res_name)
 			goto err_out;
 	}
 	else if (pci_resource_flags(pdev, bar) & IORESOURCE_MEM) {
-		if (!request_mem_region(pci_resource_start(pdev, bar),
-				        pci_resource_len(pdev, bar), res_name))
+		if (!__request_mem_region(pci_resource_start(pdev, bar),
+					pci_resource_len(pdev, bar), res_name,
+					exclusive))
 			goto err_out;
 	}
 
@@ -1427,6 +1429,47 @@ err_out:
 	return -EBUSY;
 }
 
+/**
+ *	pci_request_region - Reserved PCI I/O and memory resource
+ *	@pdev: PCI device whose resources are to be reserved
+ *	@bar: BAR to be reserved
+ *	@res_name: Name to be associated with resource.
+ *
+ *	Mark the PCI region associated with PCI device @pdev BR @bar as
+ *	being reserved by owner @res_name.  Do not access any
+ *	address inside the PCI regions unless this call returns
+ *	successfully.
+ *
+ *	Returns 0 on success, or %EBUSY on error.  A warning
+ *	message is also printed on failure.
+ */
+int pci_request_region(struct pci_dev *pdev, int bar, const char *res_name)
+{
+	return __pci_request_region(pdev, bar, res_name, 0);
+}
+
+/**
+ *	pci_request_region_exclusive - Reserved PCI I/O and memory resource
+ *	@pdev: PCI device whose resources are to be reserved
+ *	@bar: BAR to be reserved
+ *	@res_name: Name to be associated with resource.
+ *
+ *	Mark the PCI region associated with PCI device @pdev BR @bar as
+ *	being reserved by owner @res_name.  Do not access any
+ *	address inside the PCI regions unless this call returns
+ *	successfully.
+ *
+ *	Returns 0 on success, or %EBUSY on error.  A warning
+ *	message is also printed on failure.
+ *
+ *	The key difference that _exclusive makes it that userspace is
+ *	explicitly not allowed to map the resource via /dev/mem or
+ * 	sysfs.
+ */
+int pci_request_region_exclusive(struct pci_dev *pdev, int bar, const char *res_name)
+{
+	return __pci_request_region(pdev, bar, res_name, IORESOURCE_EXCLUSIVE);
+}
 /**
  * pci_release_selected_regions - Release selected PCI I/O and memory resources
  * @pdev: PCI device whose resources were previously reserved
@@ -1444,20 +1487,14 @@ void pci_release_selected_regions(struct pci_dev *pdev, int bars)
 			pci_release_region(pdev, i);
 }
 
-/**
- * pci_request_selected_regions - Reserve selected PCI I/O and memory resources
- * @pdev: PCI device whose resources are to be reserved
- * @bars: Bitmask of BARs to be requested
- * @res_name: Name to be associated with resource
- */
-int pci_request_selected_regions(struct pci_dev *pdev, int bars,
-				 const char *res_name)
+int __pci_request_selected_regions(struct pci_dev *pdev, int bars,
+				 const char *res_name, int excl)
 {
 	int i;
 
 	for (i = 0; i < 6; i++)
 		if (bars & (1 << i))
-			if(pci_request_region(pdev, i, res_name))
+			if (__pci_request_region(pdev, i, res_name, excl))
 				goto err_out;
 	return 0;
 
@@ -1469,6 +1506,26 @@ err_out:
 	return -EBUSY;
 }
 
+
+/**
+ * pci_request_selected_regions - Reserve selected PCI I/O and memory resources
+ * @pdev: PCI device whose resources are to be reserved
+ * @bars: Bitmask of BARs to be requested
+ * @res_name: Name to be associated with resource
+ */
+int pci_request_selected_regions(struct pci_dev *pdev, int bars,
+				 const char *res_name)
+{
+	return __pci_request_selected_regions(pdev, bars, res_name, 0);
+}
+
+int pci_request_selected_regions_exclusive(struct pci_dev *pdev,
+				 int bars, const char *res_name)
+{
+	return __pci_request_selected_regions(pdev, bars, res_name,
+			IORESOURCE_EXCLUSIVE);
+}
+
 /**
  *	pci_release_regions - Release reserved PCI I/O and memory resources
  *	@pdev: PCI device whose resources were previously reserved by pci_request_regions
@@ -1501,6 +1558,29 @@ int pci_request_regions(struct pci_dev *pdev, const char *res_name)
 	return pci_request_selected_regions(pdev, ((1 << 6) - 1), res_name);
 }
 
+/**
+ *	pci_request_regions_exclusive - Reserved PCI I/O and memory resources
+ *	@pdev: PCI device whose resources are to be reserved
+ *	@res_name: Name to be associated with resource.
+ *
+ *	Mark all PCI regions associated with PCI device @pdev as
+ *	being reserved by owner @res_name.  Do not access any
+ *	address inside the PCI regions unless this call returns
+ *	successfully.
+ *
+ *	pci_request_regions_exclusive() will mark the region so that
+ * 	/dev/mem and the sysfs MMIO access will not be allowed.
+ *
+ *	Returns 0 on success, or %EBUSY on error.  A warning
+ *	message is also printed on failure.
+ */
+int pci_request_regions_exclusive(struct pci_dev *pdev, const char *res_name)
+{
+	return pci_request_selected_regions_exclusive(pdev,
+					((1 << 6) - 1), res_name);
+}
+
+
 /**
  * pci_set_master - enables bus-mastering for device dev
  * @dev: the PCI device to enable
@@ -2149,10 +2229,13 @@ EXPORT_SYMBOL(pci_find_capability);
 EXPORT_SYMBOL(pci_bus_find_capability);
 EXPORT_SYMBOL(pci_release_regions);
 EXPORT_SYMBOL(pci_request_regions);
+EXPORT_SYMBOL(pci_request_regions_exclusive);
 EXPORT_SYMBOL(pci_release_region);
 EXPORT_SYMBOL(pci_request_region);
+EXPORT_SYMBOL(pci_request_region_exclusive);
 EXPORT_SYMBOL(pci_release_selected_regions);
 EXPORT_SYMBOL(pci_request_selected_regions);
+EXPORT_SYMBOL(pci_request_selected_regions_exclusive);
 EXPORT_SYMBOL(pci_set_master);
 EXPORT_SYMBOL(pci_set_mwi);
 EXPORT_SYMBOL(pci_try_set_mwi);
diff --git a/include/linux/ioport.h b/include/linux/ioport.h
index 041e95aac2bf..f6bb2ca8e3ba 100644
--- a/include/linux/ioport.h
+++ b/include/linux/ioport.h
@@ -49,6 +49,7 @@ struct resource_list {
 #define IORESOURCE_SIZEALIGN	0x00020000	/* size indicates alignment */
 #define IORESOURCE_STARTALIGN	0x00040000	/* start field is alignment */
 
+#define IORESOURCE_EXCLUSIVE	0x08000000	/* Userland may not map this resource */
 #define IORESOURCE_DISABLED	0x10000000
 #define IORESOURCE_UNSET	0x20000000
 #define IORESOURCE_AUTO		0x40000000
@@ -133,13 +134,16 @@ static inline unsigned long resource_type(struct resource *res)
 }
 
 /* Convenience shorthand with allocation */
-#define request_region(start,n,name)	__request_region(&ioport_resource, (start), (n), (name))
-#define request_mem_region(start,n,name) __request_region(&iomem_resource, (start), (n), (name))
+#define request_region(start,n,name)	__request_region(&ioport_resource, (start), (n), (name), 0)
+#define __request_mem_region(start,n,name, excl) __request_region(&iomem_resource, (start), (n), (name), excl)
+#define request_mem_region(start,n,name) __request_region(&iomem_resource, (start), (n), (name), 0)
+#define request_mem_region_exclusive(start,n,name) \
+	__request_region(&iomem_resource, (start), (n), (name), IORESOURCE_EXCLUSIVE)
 #define rename_region(region, newname) do { (region)->name = (newname); } while (0)
 
 extern struct resource * __request_region(struct resource *,
 					resource_size_t start,
-					resource_size_t n, const char *name);
+					resource_size_t n, const char *name, int relaxed);
 
 /* Compatibility cruft */
 #define release_region(start,n)	__release_region(&ioport_resource, (start), (n))
@@ -175,6 +179,7 @@ extern struct resource * __devm_request_region(struct device *dev,
 extern void __devm_release_region(struct device *dev, struct resource *parent,
 				  resource_size_t start, resource_size_t n);
 extern int iomem_map_sanity_check(resource_size_t addr, unsigned long size);
+extern int iomem_is_exclusive(u64 addr);
 
 #endif /* __ASSEMBLY__ */
 #endif	/* _LINUX_IOPORT_H */
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 59a3dc2059d3..bfcb39ca8879 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -686,10 +686,13 @@ void pci_fixup_irqs(u8 (*)(struct pci_dev *, u8 *),
 		    int (*)(struct pci_dev *, u8, u8));
 #define HAVE_PCI_REQ_REGIONS	2
 int __must_check pci_request_regions(struct pci_dev *, const char *);
+int __must_check pci_request_regions_exclusive(struct pci_dev *, const char *);
 void pci_release_regions(struct pci_dev *);
 int __must_check pci_request_region(struct pci_dev *, int, const char *);
+int __must_check pci_request_region_exclusive(struct pci_dev *, int, const char *);
 void pci_release_region(struct pci_dev *, int);
 int pci_request_selected_regions(struct pci_dev *, int, const char *);
+int pci_request_selected_regions_exclusive(struct pci_dev *, int, const char *);
 void pci_release_selected_regions(struct pci_dev *, int);
 
 /* drivers/pci/bus.c */
diff --git a/kernel/resource.c b/kernel/resource.c
index e633106b12f6..ca6a1536b205 100644
--- a/kernel/resource.c
+++ b/kernel/resource.c
@@ -623,7 +623,7 @@ resource_size_t resource_alignment(struct resource *res)
  */
 struct resource * __request_region(struct resource *parent,
 				   resource_size_t start, resource_size_t n,
-				   const char *name)
+				   const char *name, int flags)
 {
 	struct resource *res = kzalloc(sizeof(*res), GFP_KERNEL);
 
@@ -634,6 +634,7 @@ struct resource * __request_region(struct resource *parent,
 	res->start = start;
 	res->end = start + n - 1;
 	res->flags = IORESOURCE_BUSY;
+	res->flags |= flags;
 
 	write_lock(&resource_lock);
 
@@ -679,7 +680,7 @@ int __check_region(struct resource *parent, resource_size_t start,
 {
 	struct resource * res;
 
-	res = __request_region(parent, start, n, "check-region");
+	res = __request_region(parent, start, n, "check-region", 0);
 	if (!res)
 		return -EBUSY;
 
@@ -776,7 +777,7 @@ struct resource * __devm_request_region(struct device *dev,
 	dr->start = start;
 	dr->n = n;
 
-	res = __request_region(parent, start, n, name);
+	res = __request_region(parent, start, n, name, 0);
 	if (res)
 		devres_add(dev, dr);
 	else
@@ -876,3 +877,57 @@ int iomem_map_sanity_check(resource_size_t addr, unsigned long size)
 
 	return err;
 }
+
+#ifdef CONFIG_STRICT_DEVMEM
+static int strict_iomem_checks = 1;
+#else
+static int strict_iomem_checks;
+#endif
+
+/*
+ * check if an address is reserved in the iomem resource tree
+ * returns 1 if reserved, 0 if not reserved.
+ */
+int iomem_is_exclusive(u64 addr)
+{
+	struct resource *p = &iomem_resource;
+	int err = 0;
+	loff_t l;
+	int size = PAGE_SIZE;
+
+	if (!strict_iomem_checks)
+		return 0;
+
+	addr = addr & PAGE_MASK;
+
+	read_lock(&resource_lock);
+	for (p = p->child; p ; p = r_next(NULL, p, &l)) {
+		/*
+		 * We can probably skip the resources without
+		 * IORESOURCE_IO attribute?
+		 */
+		if (p->start >= addr + size)
+			break;
+		if (p->end < addr)
+			continue;
+		if (p->flags & IORESOURCE_BUSY &&
+		     p->flags & IORESOURCE_EXCLUSIVE) {
+			err = 1;
+			break;
+		}
+	}
+	read_unlock(&resource_lock);
+
+	return err;
+}
+
+static int __init strict_iomem(char *str)
+{
+	if (strstr(str, "relaxed"))
+		strict_iomem_checks = 0;
+	if (strstr(str, "strict"))
+		strict_iomem_checks = 1;
+	return 1;
+}
+
+__setup("iomem=", strict_iomem);
-- 
cgit v1.2.3


From 57c2cf71c12318b72ebaa5720d210476b6bac4d4 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 11 Dec 2008 11:24:23 -0700
Subject: PCI: add pci_swizzle_interrupt_pin()

This patch adds pci_swizzle_interrupt_pin(), which implements the
INTx swizzling algorithm specified in Table 9-1 of the "PCI-to-PCI
Bridge Architecture Specification," revision 1.2.

There are many architecture-specific implementations of this
swizzle that can be replaced by this common one.

Reviewed-by: David Howells <dhowells@redhat.com>
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c   | 16 +++++++++++++++-
 include/linux/pci.h |  1 +
 2 files changed, 16 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index bd52ca4c2893..d4d71fae6233 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -1366,6 +1366,20 @@ void pci_enable_ari(struct pci_dev *dev)
 	bridge->ari_enabled = 1;
 }
 
+/**
+ * pci_swizzle_interrupt_pin - swizzle INTx for device behind bridge
+ * @dev: the PCI device
+ * @pin: the INTx pin (1=INTA, 2=INTB, 3=INTD, 4=INTD)
+ *
+ * Perform INTx swizzling for a device behind one level of bridge.  This is
+ * required by section 9.1 of the PCI-to-PCI bridge specification for devices
+ * behind bridges on add-in cards.
+ */
+u8 pci_swizzle_interrupt_pin(struct pci_dev *dev, u8 pin)
+{
+	return (((pin - 1) + PCI_SLOT(dev->devfn)) % 4) + 1;
+}
+
 int
 pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge)
 {
@@ -1376,7 +1390,7 @@ pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge)
 		return -1;
 
 	while (dev->bus->self) {
-		pin = (((pin - 1) + PCI_SLOT(dev->devfn)) % 4) + 1;
+		pin = pci_swizzle_interrupt_pin(dev, pin);
 		dev = dev->bus->self;
 	}
 	*bridge = dev;
diff --git a/include/linux/pci.h b/include/linux/pci.h
index bfcb39ca8879..58357d14f94c 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -532,6 +532,7 @@ int __must_check pci_bus_add_device(struct pci_dev *dev);
 void pci_read_bridge_bases(struct pci_bus *child);
 struct resource *pci_find_parent_resource(const struct pci_dev *dev,
 					  struct resource *res);
+u8 pci_swizzle_interrupt_pin(struct pci_dev *dev, u8 pin);
 int pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge);
 extern struct pci_dev *pci_dev_get(struct pci_dev *dev);
 extern void pci_dev_put(struct pci_dev *dev);
-- 
cgit v1.2.3


From 1684f5ddd4c0c754f52c78eaa2c5c69ad09fb18c Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Mon, 1 Dec 2008 14:30:30 -0800
Subject: PCI: uninline pci_ioremap_bar()

It's too large to be inlined.

Acked-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c   | 16 ++++++++++++++++
 include/linux/pci.h | 15 +--------------
 2 files changed, 17 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 1fb7cff4cdae..9354dd63f035 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -56,6 +56,22 @@ unsigned char pci_bus_max_busnr(struct pci_bus* bus)
 }
 EXPORT_SYMBOL_GPL(pci_bus_max_busnr);
 
+#ifdef CONFIG_HAS_IOMEM
+void __iomem *pci_ioremap_bar(struct pci_dev *pdev, int bar)
+{
+	/*
+	 * Make sure the BAR is actually a memory resource, not an IO resource
+	 */
+	if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM)) {
+		WARN_ON(1);
+		return NULL;
+	}
+	return ioremap_nocache(pci_resource_start(pdev, bar),
+				     pci_resource_len(pdev, bar));
+}
+EXPORT_SYMBOL_GPL(pci_ioremap_bar);
+#endif
+
 #if 0
 /**
  * pci_max_busnr - returns maximum PCI bus number
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 58357d14f94c..0d8bc920c2e5 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -1160,20 +1160,7 @@ static inline void pci_mmcfg_late_init(void) { }
 
 int pci_ext_cfg_avail(struct pci_dev *dev);
 
-#ifdef CONFIG_HAS_IOMEM
-static inline void __iomem *pci_ioremap_bar(struct pci_dev *pdev, int bar)
-{
-	/*
-	 * Make sure the BAR is actually a memory resource, not an IO resource
-	 */
-	if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM)) {
-		WARN_ON(1);
-		return NULL;
-	}
-	return ioremap_nocache(pci_resource_start(pdev, bar),
-				     pci_resource_len(pdev, bar));
-}
-#endif
+void __iomem *pci_ioremap_bar(struct pci_dev *pdev, int bar);
 
 #endif /* __KERNEL__ */
 #endif /* LINUX_PCI_H */
-- 
cgit v1.2.3


From 14add80b5120966fe0659d61815b9e9b4b68fdc5 Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Sat, 22 Nov 2008 02:38:52 +0800
Subject: PCI: remove unnecessary arg of pci_update_resource()

This cleanup removes unnecessary argument 'struct resource *res' in
pci_update_resource(), so it takes same arguments as other companion
functions (pci_assign_resource(), etc.).

Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c       | 4 ++--
 drivers/pci/setup-res.c | 7 ++++---
 include/linux/pci.h     | 2 +-
 3 files changed, 7 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 9354dd63f035..c3ef2e78fc58 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -393,8 +393,8 @@ pci_restore_bars(struct pci_dev *dev)
 		return;
 	}
 
-	for (i = 0; i < numres; i ++)
-		pci_update_resource(dev, &dev->resource[i], i);
+	for (i = 0; i < numres; i++)
+		pci_update_resource(dev, i);
 }
 
 static struct pci_platform_pm_ops *pci_platform_pm;
diff --git a/drivers/pci/setup-res.c b/drivers/pci/setup-res.c
index 4e375632499a..3c5203ff53c7 100644
--- a/drivers/pci/setup-res.c
+++ b/drivers/pci/setup-res.c
@@ -26,11 +26,12 @@
 #include "pci.h"
 
 
-void pci_update_resource(struct pci_dev *dev, struct resource *res, int resno)
+void pci_update_resource(struct pci_dev *dev, int resno)
 {
 	struct pci_bus_region region;
 	u32 new, check, mask;
 	int reg;
+	struct resource *res = dev->resource + resno;
 
 	/*
 	 * Ignore resources for unimplemented BARs and unused resource slots
@@ -162,7 +163,7 @@ int pci_assign_resource(struct pci_dev *dev, int resno)
 	} else {
 		res->flags &= ~IORESOURCE_STARTALIGN;
 		if (resno < PCI_BRIDGE_RESOURCES)
-			pci_update_resource(dev, res, resno);
+			pci_update_resource(dev, resno);
 	}
 
 	return ret;
@@ -197,7 +198,7 @@ int pci_assign_resource_fixed(struct pci_dev *dev, int resno)
 		dev_err(&dev->dev, "BAR %d: can't allocate %s resource %pR\n",
 			resno, res->flags & IORESOURCE_IO ? "I/O" : "mem", res);
 	} else if (resno < PCI_BRIDGE_RESOURCES) {
-		pci_update_resource(dev, res, resno);
+		pci_update_resource(dev, resno);
 	}
 
 	return ret;
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 0d8bc920c2e5..c5e02f324e13 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -648,7 +648,7 @@ int pcie_get_readrq(struct pci_dev *dev);
 int pcie_set_readrq(struct pci_dev *dev, int rq);
 int pci_reset_function(struct pci_dev *dev);
 int pci_execute_reset_function(struct pci_dev *dev);
-void pci_update_resource(struct pci_dev *dev, struct resource *res, int resno);
+void pci_update_resource(struct pci_dev *dev, int resno);
 int __must_check pci_assign_resource(struct pci_dev *dev, int i);
 int pci_select_bars(struct pci_dev *dev, unsigned long flags);
 
-- 
cgit v1.2.3


From fde09c6d8f92de0c9f75698a75f0989f2234c517 Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Sat, 22 Nov 2008 02:39:32 +0800
Subject: PCI: define PCI resource names in an 'enum'

This patch moves all definitions of the PCI resource names to an 'enum',
and also replaces some hard-coded resource variables with symbol
names. This change eases introduction of device specific resources.

Reviewed-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci-sysfs.c |  4 +++-
 drivers/pci/probe.c     |  2 +-
 drivers/pci/proc.c      |  7 ++++---
 include/linux/pci.h     | 37 ++++++++++++++++++++++++-------------
 4 files changed, 32 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index d2f1354fd189..ea54cedcdfc6 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -102,11 +102,13 @@ resource_show(struct device * dev, struct device_attribute *attr, char * buf)
 	struct pci_dev * pci_dev = to_pci_dev(dev);
 	char * str = buf;
 	int i;
-	int max = 7;
+	int max;
 	resource_size_t start, end;
 
 	if (pci_dev->subordinate)
 		max = DEVICE_COUNT_RESOURCE;
+	else
+		max = PCI_BRIDGE_RESOURCES;
 
 	for (i = 0; i < max; i++) {
 		struct resource *res =  &pci_dev->resource[i];
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index 5dcf2b65e3f9..e1cf5d50ed4d 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -423,7 +423,7 @@ static struct pci_bus *pci_alloc_child_bus(struct pci_bus *parent,
 	child->subordinate = 0xff;
 
 	/* Set up default resource pointers and names.. */
-	for (i = 0; i < 4; i++) {
+	for (i = 0; i < PCI_BRIDGE_RESOURCE_NUM; i++) {
 		child->resource[i] = &bridge->resource[PCI_BRIDGE_RESOURCES+i];
 		child->resource[i]->name = child->name;
 	}
diff --git a/drivers/pci/proc.c b/drivers/pci/proc.c
index 7fb086d39617..593bb844b8db 100644
--- a/drivers/pci/proc.c
+++ b/drivers/pci/proc.c
@@ -361,15 +361,16 @@ static int show_device(struct seq_file *m, void *v)
 			dev->vendor,
 			dev->device,
 			dev->irq);
-	/* Here should be 7 and not PCI_NUM_RESOURCES as we need to preserve compatibility */
-	for (i=0; i<7; i++) {
+
+	/* only print standard and ROM resources to preserve compatibility */
+	for (i = 0; i <= PCI_ROM_RESOURCE; i++) {
 		resource_size_t start, end;
 		pci_resource_to_user(dev, i, &dev->resource[i], &start, &end);
 		seq_printf(m, "\t%16llx",
 			(unsigned long long)(start |
 			(dev->resource[i].flags & PCI_REGION_FLAG_MASK)));
 	}
-	for (i=0; i<7; i++) {
+	for (i = 0; i <= PCI_ROM_RESOURCE; i++) {
 		resource_size_t start, end;
 		pci_resource_to_user(dev, i, &dev->resource[i], &start, &end);
 		seq_printf(m, "\t%16llx",
diff --git a/include/linux/pci.h b/include/linux/pci.h
index c5e02f324e13..da1c22bab40e 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -82,7 +82,30 @@ enum pci_mmap_state {
 #define PCI_DMA_FROMDEVICE	2
 #define PCI_DMA_NONE		3
 
-#define DEVICE_COUNT_RESOURCE	12
+/*
+ *  For PCI devices, the region numbers are assigned this way:
+ */
+enum {
+	/* #0-5: standard PCI resources */
+	PCI_STD_RESOURCES,
+	PCI_STD_RESOURCE_END = 5,
+
+	/* #6: expansion ROM resource */
+	PCI_ROM_RESOURCE,
+
+	/* resources assigned to buses behind the bridge */
+#define PCI_BRIDGE_RESOURCE_NUM 4
+
+	PCI_BRIDGE_RESOURCES,
+	PCI_BRIDGE_RESOURCE_END = PCI_BRIDGE_RESOURCES +
+				  PCI_BRIDGE_RESOURCE_NUM - 1,
+
+	/* total resources associated with a PCI device */
+	PCI_NUM_RESOURCES,
+
+	/* preserve this for compatibility */
+	DEVICE_COUNT_RESOURCE
+};
 
 typedef int __bitwise pci_power_t;
 
@@ -274,18 +297,6 @@ static inline void pci_add_saved_cap(struct pci_dev *pci_dev,
 	hlist_add_head(&new_cap->next, &pci_dev->saved_cap_space);
 }
 
-/*
- *  For PCI devices, the region numbers are assigned this way:
- *
- *	0-5	standard PCI regions
- *	6	expansion ROM
- *	7-10	bridges: address space assigned to buses behind the bridge
- */
-
-#define PCI_ROM_RESOURCE	6
-#define PCI_BRIDGE_RESOURCES	7
-#define PCI_NUM_RESOURCES	11
-
 #ifndef PCI_BUS_NUM_RESOURCES
 #define PCI_BUS_NUM_RESOURCES	16
 #endif
-- 
cgit v1.2.3


From e8c331e963c58b83db24b7d0e39e8c07f687dbc6 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 17 Dec 2008 12:09:12 +0900
Subject: PCI hotplug: introduce functions for ACPI slot detection

Some ACPI related PCI hotplug code can be shared among PCI hotplug
drivers. This patch introduces the following functions in
drivers/pci/hotplug/acpi_pcihp.c to share the code, and changes
acpiphp and pciehp to use them.

- int acpi_pci_detect_ejectable(struct pci_bus *pbus)
  This checks if the specified PCI bus has ejectable slots.

- int acpi_pci_check_ejectable(struct pci_bus *pbus, acpi_handle handle)
  This checks if the specified handle is ejectable ACPI PCI slot. The
  'pbus' parameter is needed to check if 'handle' is PCI related ACPI
  object.

This patch also introduces the following inline function in
include/linux/pci-acpi.h, which is useful to get ACPI handle of the
PCI bridge from struct pci_bus of the bridge's secondary bus.

- static inline acpi_handle acpi_pci_get_bridge_handle(struct pci_bus *pbus)
  This returns ACPI handle of the PCI bridge which generates PCI bus
  specified by 'pbus'.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/acpi_pcihp.c   |  69 ++++++++++++++++++++++++
 drivers/pci/hotplug/acpiphp_glue.c | 107 ++++++-------------------------------
 drivers/pci/hotplug/pciehp_acpi.c  |  55 ++-----------------
 include/linux/pci-acpi.h           |   9 ++++
 include/linux/pci_hotplug.h        |   2 +
 5 files changed, 100 insertions(+), 142 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/hotplug/acpi_pcihp.c b/drivers/pci/hotplug/acpi_pcihp.c
index e17ef54f0efc..c62ab8d240aa 100644
--- a/drivers/pci/hotplug/acpi_pcihp.c
+++ b/drivers/pci/hotplug/acpi_pcihp.c
@@ -501,5 +501,74 @@ int acpi_root_bridge(acpi_handle handle)
 }
 EXPORT_SYMBOL_GPL(acpi_root_bridge);
 
+
+static int is_ejectable(acpi_handle handle)
+{
+	acpi_status status;
+	acpi_handle tmp;
+	unsigned long long removable;
+	status = acpi_get_handle(handle, "_ADR", &tmp);
+	if (ACPI_FAILURE(status))
+		return 0;
+	status = acpi_get_handle(handle, "_EJ0", &tmp);
+	if (ACPI_SUCCESS(status))
+		return 1;
+	status = acpi_evaluate_integer(handle, "_RMV", NULL, &removable);
+	if (ACPI_SUCCESS(status) && removable)
+		return 1;
+	return 0;
+}
+
+/**
+ * acpi_pcihp_check_ejectable - check if handle is ejectable ACPI PCI slot
+ * @pbus: the PCI bus of the PCI slot corresponding to 'handle'
+ * @handle: ACPI handle to check
+ *
+ * Return 1 if handle is ejectable PCI slot, 0 otherwise.
+ */
+int acpi_pci_check_ejectable(struct pci_bus *pbus, acpi_handle handle)
+{
+	acpi_handle bridge_handle, parent_handle;
+
+	if (!(bridge_handle = acpi_pci_get_bridge_handle(pbus)))
+		return 0;
+	if ((ACPI_FAILURE(acpi_get_parent(handle, &parent_handle))))
+		return 0;
+	if (bridge_handle != parent_handle)
+		return 0;
+	return is_ejectable(handle);
+}
+EXPORT_SYMBOL_GPL(acpi_pci_check_ejectable);
+
+static acpi_status
+check_hotplug(acpi_handle handle, u32 lvl, void *context, void **rv)
+{
+	int *found = (int *)context;
+	if (is_ejectable(handle)) {
+		*found = 1;
+		return AE_CTRL_TERMINATE;
+	}
+	return AE_OK;
+}
+
+/**
+ * acpi_pci_detect_ejectable - check if the PCI bus has ejectable slots
+ * @pbus - PCI bus to scan
+ *
+ * Returns 1 if the PCI bus has ACPI based ejectable slots, 0 otherwise.
+ */
+int acpi_pci_detect_ejectable(struct pci_bus *pbus)
+{
+	acpi_handle handle;
+	int found = 0;
+
+	if (!(handle = acpi_pci_get_bridge_handle(pbus)))
+		return 0;
+	acpi_walk_namespace(ACPI_TYPE_DEVICE, handle, (u32)1,
+			    check_hotplug, (void *)&found, NULL);
+	return found;
+}
+EXPORT_SYMBOL_GPL(acpi_pci_detect_ejectable);
+
 module_param(debug_acpi, bool, 0644);
 MODULE_PARM_DESC(debug_acpi, "Debugging mode for ACPI enabled or not");
diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index 7a5760426897..f09b1010d477 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -46,6 +46,7 @@
 #include <linux/kernel.h>
 #include <linux/pci.h>
 #include <linux/pci_hotplug.h>
+#include <linux/pci-acpi.h>
 #include <linux/mutex.h>
 
 #include "../pci.h"
@@ -62,68 +63,6 @@ static void acpiphp_sanitize_bus(struct pci_bus *bus);
 static void acpiphp_set_hpp_values(acpi_handle handle, struct pci_bus *bus);
 static void handle_hotplug_event_func(acpi_handle handle, u32 type, void *context);
 
-
-/*
- * initialization & terminatation routines
- */
-
-/**
- * is_ejectable - determine if a slot is ejectable
- * @handle: handle to acpi namespace
- *
- * Ejectable slot should satisfy at least these conditions:
- *
- *  1. has _ADR method
- *  2. has _EJ0 method or _RMV method
- *
- * optionally
- *
- *  1. has _STA method
- *  2. has _PS0 method
- *  3. has _PS3 method
- *  4. ..
- */
-static int is_ejectable(acpi_handle handle)
-{
-	acpi_status status;
-	acpi_handle tmp;
-	unsigned long long removable;
-
-	status = acpi_get_handle(handle, "_ADR", &tmp);
-	if (ACPI_FAILURE(status))
-		return 0;
-
-	status = acpi_get_handle(handle, "_EJ0", &tmp);
-	if (ACPI_SUCCESS(status))
-		return 1;
-
-	status = acpi_get_handle(handle, "_RMV", &tmp);
-	if (ACPI_SUCCESS(status)) {
-		status = acpi_evaluate_integer(handle, "_RMV", NULL,
-					       &removable);
-		if (ACPI_SUCCESS(status) && removable)
-			return 1;
-	}
-
-	return 0;
-}
-
-
-/* callback routine to check for the existence of ejectable slots */
-static acpi_status
-is_ejectable_slot(acpi_handle handle, u32 lvl, void *context, void **rv)
-{
-	int *count = (int *)context;
-
-	if (is_ejectable(handle)) {
-		(*count)++;
-		/* only one ejectable slot is enough */
-		return AE_CTRL_TERMINATE;
-	} else {
-		return AE_OK;
-	}
-}
-
 /* callback routine to check for the existence of a pci dock device */
 static acpi_status
 is_pci_dock_device(acpi_handle handle, u32 lvl, void *context, void **rv)
@@ -138,9 +77,6 @@ is_pci_dock_device(acpi_handle handle, u32 lvl, void *context, void **rv)
 	}
 }
 
-
-
-
 /*
  * the _DCK method can do funny things... and sometimes not
  * hah-hah funny.
@@ -191,8 +127,9 @@ register_slot(acpi_handle handle, u32 lvl, void *context, void **rv)
 	acpi_status status = AE_OK;
 	unsigned long long adr, sun;
 	int device, function, retval;
+	struct pci_bus *pbus = bridge->pci_bus;
 
-	if (!is_ejectable(handle) && !is_dock_device(handle))
+	if (!acpi_pci_check_ejectable(pbus, handle) && !is_dock_device(handle))
 		return AE_OK;
 
 	acpi_evaluate_integer(handle, "_ADR", NULL, &adr);
@@ -258,8 +195,7 @@ register_slot(acpi_handle handle, u32 lvl, void *context, void **rv)
 		bridge->nr_slots++;
 
 		dbg("found ACPI PCI Hotplug slot %llu at PCI %04x:%02x:%02x\n",
-				slot->sun, pci_domain_nr(bridge->pci_bus),
-				bridge->pci_bus->number, slot->device);
+		    slot->sun, pci_domain_nr(pbus), pbus->number, device);
 		retval = acpiphp_register_hotplug_slot(slot);
 		if (retval) {
 			if (retval == -EBUSY)
@@ -276,8 +212,7 @@ register_slot(acpi_handle handle, u32 lvl, void *context, void **rv)
 	list_add_tail(&newfunc->sibling, &slot->funcs);
 
 	/* associate corresponding pci_dev */
-	newfunc->pci_dev = pci_get_slot(bridge->pci_bus,
-					 PCI_DEVFN(device, function));
+	newfunc->pci_dev = pci_get_slot(pbus, PCI_DEVFN(device, function));
 	if (newfunc->pci_dev) {
 		slot->flags |= (SLOT_ENABLED | SLOT_POWEREDON);
 	}
@@ -326,27 +261,15 @@ register_slot(acpi_handle handle, u32 lvl, void *context, void **rv)
 
 
 /* see if it's worth looking at this bridge */
-static int detect_ejectable_slots(acpi_handle *bridge_handle)
+static int detect_ejectable_slots(struct pci_bus *pbus)
 {
-	acpi_status status;
-	int count;
-
-	count = 0;
-
-	/* only check slots defined directly below bridge object */
-	status = acpi_walk_namespace(ACPI_TYPE_DEVICE, bridge_handle, (u32)1,
-				     is_ejectable_slot, (void *)&count, NULL);
-
-	/*
-	 * we also need to add this bridge if there is a dock bridge or
-	 * other pci device on a dock station (removable)
-	 */
-	if (!count)
-		status = acpi_walk_namespace(ACPI_TYPE_DEVICE, bridge_handle,
-				(u32)1, is_pci_dock_device, (void *)&count,
-				NULL);
-
-	return count;
+	int found = acpi_pci_detect_ejectable(pbus);
+	if (!found) {
+		acpi_handle bridge_handle = acpi_pci_get_bridge_handle(pbus);
+		acpi_walk_namespace(ACPI_TYPE_DEVICE, bridge_handle, (u32)1,
+				    is_pci_dock_device, (void *)&found, NULL);
+	}
+	return found;
 }
 
 
@@ -556,7 +479,7 @@ find_p2p_bridge(acpi_handle handle, u32 lvl, void *context, void **rv)
 		goto out;
 
 	/* check if this bridge has ejectable slots */
-	if ((detect_ejectable_slots(handle) > 0)) {
+	if ((detect_ejectable_slots(dev->subordinate) > 0)) {
 		dbg("found PCI-to-PCI bridge at PCI %s\n", pci_name(dev));
 		add_p2p_bridge(handle, dev);
 	}
@@ -617,7 +540,7 @@ static int add_bridge(acpi_handle handle)
 	}
 
 	/* check if this bridge has ejectable slots */
-	if (detect_ejectable_slots(handle) > 0) {
+	if (detect_ejectable_slots(pci_bus) > 0) {
 		dbg("found PCI host-bus bridge with hot-pluggable slots\n");
 		add_host_bridge(handle, pci_bus);
 	}
diff --git a/drivers/pci/hotplug/pciehp_acpi.c b/drivers/pci/hotplug/pciehp_acpi.c
index 88a5c57f2e5b..438d795f9fe3 100644
--- a/drivers/pci/hotplug/pciehp_acpi.c
+++ b/drivers/pci/hotplug/pciehp_acpi.c
@@ -24,6 +24,8 @@
  */
 
 #include <linux/acpi.h>
+#include <linux/pci.h>
+#include <linux/pci_hotplug.h>
 #include "pciehp.h"
 
 #define PCIEHP_DETECT_PCIE	(0)
@@ -41,59 +43,11 @@ MODULE_PARM_DESC(pciehp_detect_mode,
 	 "  auto(default) - Auto select mode. Use acpi option if duplicate\n"
 	 "                  slot ids are found. Otherwise, use pcie option\n");
 
-static int is_ejectable(acpi_handle handle)
-{
-	acpi_status status;
-	acpi_handle tmp;
-	unsigned long long removable;
-	status = acpi_get_handle(handle, "_ADR", &tmp);
-	if (ACPI_FAILURE(status))
-		return 0;
-	status = acpi_get_handle(handle, "_EJ0", &tmp);
-	if (ACPI_SUCCESS(status))
-		return 1;
-	status = acpi_evaluate_integer(handle, "_RMV", NULL, &removable);
-	if (ACPI_SUCCESS(status) && removable)
-		return 1;
-	return 0;
-}
-
-static acpi_status
-check_hotplug(acpi_handle handle, u32 lvl, void *context, void **rv)
-{
-	int *found = (int *)context;
-	if (is_ejectable(handle)) {
-		*found = 1;
-		return AE_CTRL_TERMINATE;
-	}
-	return AE_OK;
-}
-
-static int pciehp_detect_acpi_slot(struct pci_bus *pbus)
-{
-	acpi_handle handle;
-	struct pci_dev *pdev = pbus->self;
-	int found = 0;
-
-	if (!pdev){
-		int seg = pci_domain_nr(pbus), busnr = pbus->number;
-		handle = acpi_get_pci_rootbridge_handle(seg, busnr);
-	} else
-		handle = DEVICE_ACPI_HANDLE(&(pdev->dev));
-
-	if (!handle)
-		return 0;
-
-	acpi_walk_namespace(ACPI_TYPE_DEVICE, handle, (u32)1,
-			    check_hotplug, (void *)&found, NULL);
-	return found;
-}
-
 int pciehp_acpi_slot_detection_check(struct pci_dev *dev)
 {
 	if (slot_detection_mode != PCIEHP_DETECT_ACPI)
 		return 0;
-	if (pciehp_detect_acpi_slot(dev->subordinate))
+	if (acpi_pci_detect_ejectable(dev->subordinate))
 		return 0;
 	return -ENODEV;
 }
@@ -135,6 +89,7 @@ static int __init dummy_probe(struct pcie_device *dev,
 	u32 slot_cap;
 	struct slot *slot, *tmp;
 	struct pci_dev *pdev = dev->port;
+	struct pci_bus *pbus = pdev->subordinate;
 	if (!(slot = kzalloc(sizeof(*slot), GFP_KERNEL)))
 		return -ENOMEM;
 	/* Note: pciehp_detect_mode != PCIEHP_DETECT_ACPI here */
@@ -149,7 +104,7 @@ static int __init dummy_probe(struct pcie_device *dev,
 			dup_slot_id++;
 	}
 	list_add_tail(&slot->slot_list, &dummy_slots);
-	if (!acpi_slot_detected && pciehp_detect_acpi_slot(pdev->subordinate))
+	if (!acpi_slot_detected && acpi_pci_detect_ejectable(pbus))
 		acpi_slot_detected = 1;
 	return -ENODEV;         /* dummy driver always returns error */
 }
diff --git a/include/linux/pci-acpi.h b/include/linux/pci-acpi.h
index 871e096e0fbc..042c166f65d5 100644
--- a/include/linux/pci-acpi.h
+++ b/include/linux/pci-acpi.h
@@ -60,6 +60,15 @@ static inline acpi_handle acpi_find_root_bridge_handle(struct pci_dev *pdev)
 	return acpi_get_pci_rootbridge_handle(pci_domain_nr(pdev->bus),
 			pdev->bus->number);
 }
+
+static inline acpi_handle acpi_pci_get_bridge_handle(struct pci_bus *pbus)
+{
+	int seg = pci_domain_nr(pbus), busnr = pbus->number;
+	struct pci_dev *bridge = pbus->self;
+	if (bridge)
+		return DEVICE_ACPI_HANDLE(&(bridge->dev));
+	return acpi_get_pci_rootbridge_handle(seg, busnr);
+}
 #else
 #if !defined(AE_ERROR)
 typedef u32 		acpi_status;
diff --git a/include/linux/pci_hotplug.h b/include/linux/pci_hotplug.h
index a00bd1a0f156..f7cc204fab07 100644
--- a/include/linux/pci_hotplug.h
+++ b/include/linux/pci_hotplug.h
@@ -228,6 +228,8 @@ extern acpi_status acpi_get_hp_params_from_firmware(struct pci_bus *bus,
 				struct hotplug_params *hpp);
 int acpi_get_hp_hw_control_from_firmware(struct pci_dev *dev, u32 flags);
 int acpi_root_bridge(acpi_handle handle);
+int acpi_pci_check_ejectable(struct pci_bus *pbus, acpi_handle handle);
+int acpi_pci_detect_ejectable(struct pci_bus *pbus);
 #endif
 #endif
 
-- 
cgit v1.2.3


From 68feac87de15edfc2c700d2d81b814288c93d003 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Tue, 16 Dec 2008 21:36:55 -0700
Subject: PCI: add pci_common_swizzle() for INTx swizzling

This patch adds pci_common_swizzle(), which swizzles INTx values all the
way up to a root bridge.

This common implementation can replace several architecture-specific
ones.  This should someday be combined with pci_get_interrupt_pin(),
but I left it separate for now to make reviewing easier.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c   | 20 ++++++++++++++++++++
 include/linux/pci.h |  1 +
 2 files changed, 21 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 9d2aa6366fd0..c824dc8d617c 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -1421,6 +1421,26 @@ pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge)
 	return pin;
 }
 
+/**
+ * pci_common_swizzle - swizzle INTx all the way to root bridge
+ * @dev: the PCI device
+ * @pinp: pointer to the INTx pin value (1=INTA, 2=INTB, 3=INTD, 4=INTD)
+ *
+ * Perform INTx swizzling for a device.  This traverses through all PCI-to-PCI
+ * bridges all the way up to a PCI root bus.
+ */
+u8 pci_common_swizzle(struct pci_dev *dev, u8 *pinp)
+{
+	u8 pin = *pinp;
+
+	while (dev->bus->self) {
+		pin = pci_swizzle_interrupt_pin(dev, pin);
+		dev = dev->bus->self;
+	}
+	*pinp = pin;
+	return PCI_SLOT(dev->devfn);
+}
+
 /**
  *	pci_release_region - Release a PCI bar
  *	@pdev: PCI device whose resources were previously reserved by pci_request_region
diff --git a/include/linux/pci.h b/include/linux/pci.h
index da1c22bab40e..170f9ae2d8a0 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -545,6 +545,7 @@ struct resource *pci_find_parent_resource(const struct pci_dev *dev,
 					  struct resource *res);
 u8 pci_swizzle_interrupt_pin(struct pci_dev *dev, u8 pin);
 int pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge);
+u8 pci_common_swizzle(struct pci_dev *dev, u8 *pinp);
 extern struct pci_dev *pci_dev_get(struct pci_dev *dev);
 extern void pci_dev_put(struct pci_dev *dev);
 extern void pci_remove_bus(struct pci_bus *b);
-- 
cgit v1.2.3


From 287d19ce2e67c15e79a187b3bdcbbea1a0a51a7d Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@vyatta.com>
Date: Thu, 18 Dec 2008 09:17:16 -0800
Subject: PCI: revise VPD access interface

Change PCI VPD API which was only used by sysfs to something usable
in drivers.
   * move iteration over multiple words to the low level
   * use conventional types for arguments
   * add exportable wrapper

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/access.c    | 156 ++++++++++++++++++++++++++++++------------------
 drivers/pci/pci-sysfs.c |  38 +++---------
 drivers/pci/pci.h       |   6 +-
 include/linux/pci.h     |   4 ++
 4 files changed, 114 insertions(+), 90 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/access.c b/drivers/pci/access.c
index 98ddba94b5b9..86ec4ad44bcd 100644
--- a/drivers/pci/access.c
+++ b/drivers/pci/access.c
@@ -66,6 +66,39 @@ EXPORT_SYMBOL(pci_bus_write_config_byte);
 EXPORT_SYMBOL(pci_bus_write_config_word);
 EXPORT_SYMBOL(pci_bus_write_config_dword);
 
+
+/**
+ * pci_read_vpd - Read one entry from Vital Product Data
+ * @dev:	pci device struct
+ * @pos:	offset in vpd space
+ * @count:	number of bytes to read
+ * @buf:	pointer to where to store result
+ *
+ */
+ssize_t pci_read_vpd(struct pci_dev *dev, loff_t pos, size_t count, void *buf)
+{
+	if (!dev->vpd || !dev->vpd->ops)
+		return -ENODEV;
+	return dev->vpd->ops->read(dev, pos, count, buf);
+}
+EXPORT_SYMBOL(pci_read_vpd);
+
+/**
+ * pci_write_vpd - Write entry to Vital Product Data
+ * @dev:	pci device struct
+ * @pos:	offset in vpd space
+ * @count:	number of bytes to read
+ * @val:	value to write
+ *
+ */
+ssize_t pci_write_vpd(struct pci_dev *dev, loff_t pos, size_t count, const void *buf)
+{
+	if (!dev->vpd || !dev->vpd->ops)
+		return -ENODEV;
+	return dev->vpd->ops->write(dev, pos, count, buf);
+}
+EXPORT_SYMBOL(pci_write_vpd);
+
 /*
  * The following routines are to prevent the user from accessing PCI config
  * space when it's unsafe to do so.  Some devices require this during BIST and
@@ -176,19 +209,17 @@ static int pci_vpd_pci22_wait(struct pci_dev *dev)
 	}
 }
 
-static int pci_vpd_pci22_read(struct pci_dev *dev, int pos, int size,
-			      char *buf)
+static ssize_t pci_vpd_pci22_read(struct pci_dev *dev, loff_t pos, size_t count,
+				  void *arg)
 {
 	struct pci_vpd_pci22 *vpd =
 		container_of(dev->vpd, struct pci_vpd_pci22, base);
-	u32 val;
-	int ret = 0;
-	int begin, end, i;
+	int ret;
+	loff_t end = pos + count;
+	u8 *buf = arg;
 
-	if (pos < 0 || pos > vpd->base.len || size > vpd->base.len  - pos)
+	if (pos < 0 || pos > vpd->base.len || end > vpd->base.len)
 		return -EINVAL;
-	if (size == 0)
-		return 0;
 
 	if (mutex_lock_killable(&vpd->lock))
 		return -EINTR;
@@ -196,73 +227,84 @@ static int pci_vpd_pci22_read(struct pci_dev *dev, int pos, int size,
 	ret = pci_vpd_pci22_wait(dev);
 	if (ret < 0)
 		goto out;
-	ret = pci_user_write_config_word(dev, vpd->cap + PCI_VPD_ADDR,
-					 pos & ~3);
-	if (ret < 0)
-		goto out;
 
-	vpd->busy = true;
-	vpd->flag = PCI_VPD_ADDR_F;
-	ret = pci_vpd_pci22_wait(dev);
-	if (ret < 0)
-		goto out;
-	ret = pci_user_read_config_dword(dev, vpd->cap + PCI_VPD_DATA,
-					 &val);
+	while (pos < end) {
+		u32 val;
+		unsigned int i, skip;
+
+		ret = pci_user_write_config_word(dev, vpd->cap + PCI_VPD_ADDR,
+						 pos & ~3);
+		if (ret < 0)
+			break;
+		vpd->busy = true;
+		vpd->flag = PCI_VPD_ADDR_F;
+		ret = pci_vpd_pci22_wait(dev);
+		if (ret < 0)
+			break;
+
+		ret = pci_user_read_config_dword(dev, vpd->cap + PCI_VPD_DATA, &val);
+		if (ret < 0)
+			break;
+
+		skip = pos & 3;
+		for (i = 0;  i < sizeof(u32); i++) {
+			if (i >= skip) {
+				*buf++ = val;
+				if (++pos == end)
+					break;
+			}
+			val >>= 8;
+		}
+	}
 out:
 	mutex_unlock(&vpd->lock);
-	if (ret < 0)
-		return ret;
-
-	/* Convert to bytes */
-	begin = pos & 3;
-	end = min(4, begin + size);
-	for (i = 0; i < end; ++i) {
-		if (i >= begin)
-			*buf++ = val;
-		val >>= 8;
-	}
-	return end - begin;
+	return ret ? ret : count;
 }
 
-static int pci_vpd_pci22_write(struct pci_dev *dev, int pos, int size,
-			       const char *buf)
+static ssize_t pci_vpd_pci22_write(struct pci_dev *dev, loff_t pos, size_t count,
+				   const void *arg)
 {
 	struct pci_vpd_pci22 *vpd =
 		container_of(dev->vpd, struct pci_vpd_pci22, base);
-	u32 val;
+	const u8 *buf = arg;
+	loff_t end = pos + count;
 	int ret = 0;
 
-	if (pos < 0 || pos > vpd->base.len || pos & 3 ||
-	    size > vpd->base.len - pos || size < 4)
+	if (pos < 0 || (pos & 3) || (count & 3) || end > vpd->base.len)
 		return -EINVAL;
 
-	val = (u8) *buf++;
-	val |= ((u8) *buf++) << 8;
-	val |= ((u8) *buf++) << 16;
-	val |= ((u32)(u8) *buf++) << 24;
-
 	if (mutex_lock_killable(&vpd->lock))
 		return -EINTR;
+
 	ret = pci_vpd_pci22_wait(dev);
 	if (ret < 0)
 		goto out;
-	ret = pci_user_write_config_dword(dev, vpd->cap + PCI_VPD_DATA,
-					  val);
-	if (ret < 0)
-		goto out;
-	ret = pci_user_write_config_word(dev, vpd->cap + PCI_VPD_ADDR,
-					 pos | PCI_VPD_ADDR_F);
-	if (ret < 0)
-		goto out;
-	vpd->busy = true;
-	vpd->flag = 0;
-	ret = pci_vpd_pci22_wait(dev);
+
+	while (pos < end) {
+		u32 val;
+
+		val = *buf++;
+		val |= *buf++ << 8;
+		val |= *buf++ << 16;
+		val |= *buf++ << 24;
+
+		ret = pci_user_write_config_dword(dev, vpd->cap + PCI_VPD_DATA, val);
+		if (ret < 0)
+			break;
+		ret = pci_user_write_config_word(dev, vpd->cap + PCI_VPD_ADDR,
+						 pos | PCI_VPD_ADDR_F);
+		if (ret < 0)
+			break;
+
+		vpd->busy = true;
+		vpd->flag = 0;
+		ret = pci_vpd_pci22_wait(dev);
+
+		pos += sizeof(u32);
+	}
 out:
 	mutex_unlock(&vpd->lock);
-	if (ret < 0)
-		return ret;
-
-	return 4;
+	return ret ? ret : count;
 }
 
 static void pci_vpd_pci22_release(struct pci_dev *dev)
@@ -270,7 +312,7 @@ static void pci_vpd_pci22_release(struct pci_dev *dev)
 	kfree(container_of(dev->vpd, struct pci_vpd_pci22, base));
 }
 
-static struct pci_vpd_ops pci_vpd_pci22_ops = {
+static const struct pci_vpd_ops pci_vpd_pci22_ops = {
 	.read = pci_vpd_pci22_read,
 	.write = pci_vpd_pci22_write,
 	.release = pci_vpd_pci22_release,
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index ea54cedcdfc6..c23619fb6c4b 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -371,55 +371,33 @@ pci_write_config(struct kobject *kobj, struct bin_attribute *bin_attr,
 }
 
 static ssize_t
-pci_read_vpd(struct kobject *kobj, struct bin_attribute *bin_attr,
-	     char *buf, loff_t off, size_t count)
+read_vpd_attr(struct kobject *kobj, struct bin_attribute *bin_attr,
+	      char *buf, loff_t off, size_t count)
 {
 	struct pci_dev *dev =
 		to_pci_dev(container_of(kobj, struct device, kobj));
-	int end;
-	int ret;
 
 	if (off > bin_attr->size)
 		count = 0;
 	else if (count > bin_attr->size - off)
 		count = bin_attr->size - off;
-	end = off + count;
-
-	while (off < end) {
-		ret = dev->vpd->ops->read(dev, off, end - off, buf);
-		if (ret < 0)
-			return ret;
-		buf += ret;
-		off += ret;
-	}
 
-	return count;
+	return pci_read_vpd(dev, off, count, buf);
 }
 
 static ssize_t
-pci_write_vpd(struct kobject *kobj, struct bin_attribute *bin_attr,
-	      char *buf, loff_t off, size_t count)
+write_vpd_attr(struct kobject *kobj, struct bin_attribute *bin_attr,
+	       char *buf, loff_t off, size_t count)
 {
 	struct pci_dev *dev =
 		to_pci_dev(container_of(kobj, struct device, kobj));
-	int end;
-	int ret;
 
 	if (off > bin_attr->size)
 		count = 0;
 	else if (count > bin_attr->size - off)
 		count = bin_attr->size - off;
-	end = off + count;
-
-	while (off < end) {
-		ret = dev->vpd->ops->write(dev, off, end - off, buf);
-		if (ret < 0)
-			return ret;
-		buf += ret;
-		off += ret;
-	}
 
-	return count;
+	return pci_write_vpd(dev, off, count, buf);
 }
 
 #ifdef HAVE_PCI_LEGACY
@@ -845,8 +823,8 @@ static int pci_create_capabilities_sysfs(struct pci_dev *dev)
 		attr->size = dev->vpd->len;
 		attr->attr.name = "vpd";
 		attr->attr.mode = S_IRUSR | S_IWUSR;
-		attr->read = pci_read_vpd;
-		attr->write = pci_write_vpd;
+		attr->read = read_vpd_attr;
+		attr->write = write_vpd_attr;
 		retval = sysfs_create_bin_file(&dev->dev.kobj, attr);
 		if (retval) {
 			kfree(dev->vpd->attr);
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index 65deed8bfc06..211fd418f48f 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -56,14 +56,14 @@ extern int pci_user_write_config_word(struct pci_dev *dev, int where, u16 val);
 extern int pci_user_write_config_dword(struct pci_dev *dev, int where, u32 val);
 
 struct pci_vpd_ops {
-	int (*read)(struct pci_dev *dev, int pos, int size, char *buf);
-	int (*write)(struct pci_dev *dev, int pos, int size, const char *buf);
+	ssize_t (*read)(struct pci_dev *dev, loff_t pos, size_t count, void *buf);
+	ssize_t (*write)(struct pci_dev *dev, loff_t pos, size_t count, const void *buf);
 	void (*release)(struct pci_dev *dev);
 };
 
 struct pci_vpd {
 	unsigned int len;
-	struct pci_vpd_ops *ops;
+	const struct pci_vpd_ops *ops;
 	struct bin_attribute *attr; /* descriptor for sysfs VPD entry */
 };
 
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 170f9ae2d8a0..76079e106895 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -687,6 +687,10 @@ int pci_back_from_sleep(struct pci_dev *dev);
 /* Functions for PCI Hotplug drivers to use */
 int pci_bus_find_capability(struct pci_bus *bus, unsigned int devfn, int cap);
 
+/* Vital product data routines */
+ssize_t pci_read_vpd(struct pci_dev *dev, loff_t pos, size_t count, void *buf);
+ssize_t pci_write_vpd(struct pci_dev *dev, loff_t pos, size_t count, const void *buf);
+
 /* Helper functions for low-level code (drivers/pci/setup-[bus,res].c) */
 void pci_bus_assign_resources(struct pci_bus *bus);
 void pci_bus_size_bridges(struct pci_bus *bus);
-- 
cgit v1.2.3


From db5679437a2b938c9127480a3923633721583a4f Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@vyatta.com>
Date: Thu, 18 Dec 2008 09:17:16 -0800
Subject: PCI: add interface to set visible size of VPD

The VPD on all devices may not be 32K. Unfortunately, there is no
generic way to find the size, so this adds a simple API hook
to reset it.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/access.c | 23 +++++++++++++++++++++++
 include/linux/pci.h  |  1 +
 2 files changed, 24 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/pci/access.c b/drivers/pci/access.c
index 86ec4ad44bcd..381444794778 100644
--- a/drivers/pci/access.c
+++ b/drivers/pci/access.c
@@ -339,6 +339,29 @@ int pci_vpd_pci22_init(struct pci_dev *dev)
 	return 0;
 }
 
+/**
+ * pci_vpd_truncate - Set available Vital Product Data size
+ * @dev:	pci device struct
+ * @size:	available memory in bytes
+ *
+ * Adjust size of available VPD area.
+ */
+int pci_vpd_truncate(struct pci_dev *dev, size_t size)
+{
+	if (!dev->vpd)
+		return -EINVAL;
+
+	/* limited by the access method */
+	if (size > dev->vpd->len)
+		return -EINVAL;
+
+	dev->vpd->len = size;
+	dev->vpd->attr->size = size;
+
+	return 0;
+}
+EXPORT_SYMBOL(pci_vpd_truncate);
+
 /**
  * pci_block_user_cfg_access - Block userspace PCI config reads/writes
  * @dev:	pci device struct
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 76079e106895..7cbecef19bb6 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -690,6 +690,7 @@ int pci_bus_find_capability(struct pci_bus *bus, unsigned int devfn, int cap);
 /* Vital product data routines */
 ssize_t pci_read_vpd(struct pci_dev *dev, loff_t pos, size_t count, void *buf);
 ssize_t pci_write_vpd(struct pci_dev *dev, loff_t pos, size_t count, const void *buf);
+int pci_vpd_truncate(struct pci_dev *dev, size_t size);
 
 /* Helper functions for low-level code (drivers/pci/setup-[bus,res].c) */
 void pci_bus_assign_resources(struct pci_bus *bus);
-- 
cgit v1.2.3


From 322162a71bd9fc4edb1b11236e7bc8aa27ccac22 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Fri, 19 Dec 2008 15:19:02 +0900
Subject: PCI: pciehp: cleanup register and field definitions

Clean up register definitions related to PCI Express Hot plug.

  - Add register definitions into include/linux/pci_regs.h, and use
    them instead of pciehp's locally definied register definitions.
  - Remove pciehp's locally defined register definitions
  - Remove unused register definitions in pciehp.
  - Some minor cleanups.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/pciehp_hpc.c | 328 ++++++++++++---------------------------
 include/linux/pci_regs.h         |  64 +++++++-
 2 files changed, 156 insertions(+), 236 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c
index 22b88cb17a07..71a8012886b0 100644
--- a/drivers/pci/hotplug/pciehp_hpc.c
+++ b/drivers/pci/hotplug/pciehp_hpc.c
@@ -42,42 +42,6 @@
 
 static atomic_t pciehp_num_controllers = ATOMIC_INIT(0);
 
-struct ctrl_reg {
-	u8 cap_id;
-	u8 nxt_ptr;
-	u16 cap_reg;
-	u32 dev_cap;
-	u16 dev_ctrl;
-	u16 dev_status;
-	u32 lnk_cap;
-	u16 lnk_ctrl;
-	u16 lnk_status;
-	u32 slot_cap;
-	u16 slot_ctrl;
-	u16 slot_status;
-	u16 root_ctrl;
-	u16 rsvp;
-	u32 root_status;
-} __attribute__ ((packed));
-
-/* offsets to the controller registers based on the above structure layout */
-enum ctrl_offsets {
-	PCIECAPID	=	offsetof(struct ctrl_reg, cap_id),
-	NXTCAPPTR	=	offsetof(struct ctrl_reg, nxt_ptr),
-	CAPREG		=	offsetof(struct ctrl_reg, cap_reg),
-	DEVCAP		=	offsetof(struct ctrl_reg, dev_cap),
-	DEVCTRL		=	offsetof(struct ctrl_reg, dev_ctrl),
-	DEVSTATUS	=	offsetof(struct ctrl_reg, dev_status),
-	LNKCAP		=	offsetof(struct ctrl_reg, lnk_cap),
-	LNKCTRL		=	offsetof(struct ctrl_reg, lnk_ctrl),
-	LNKSTATUS	=	offsetof(struct ctrl_reg, lnk_status),
-	SLOTCAP		=	offsetof(struct ctrl_reg, slot_cap),
-	SLOTCTRL	=	offsetof(struct ctrl_reg, slot_ctrl),
-	SLOTSTATUS	=	offsetof(struct ctrl_reg, slot_status),
-	ROOTCTRL	=	offsetof(struct ctrl_reg, root_ctrl),
-	ROOTSTATUS	=	offsetof(struct ctrl_reg, root_status),
-};
-
 static inline int pciehp_readw(struct controller *ctrl, int reg, u16 *value)
 {
 	struct pci_dev *dev = ctrl->pci_dev;
@@ -102,95 +66,9 @@ static inline int pciehp_writel(struct controller *ctrl, int reg, u32 value)
 	return pci_write_config_dword(dev, ctrl->cap_base + reg, value);
 }
 
-/* Field definitions in PCI Express Capabilities Register */
-#define CAP_VER			0x000F
-#define DEV_PORT_TYPE		0x00F0
-#define SLOT_IMPL		0x0100
-#define MSG_NUM			0x3E00
-
-/* Device or Port Type */
-#define NAT_ENDPT		0x00
-#define LEG_ENDPT		0x01
-#define ROOT_PORT		0x04
-#define UP_STREAM		0x05
-#define	DN_STREAM		0x06
-#define PCIE_PCI_BRDG		0x07
-#define PCI_PCIE_BRDG		0x10
-
-/* Field definitions in Device Capabilities Register */
-#define DATTN_BUTTN_PRSN	0x1000
-#define DATTN_LED_PRSN		0x2000
-#define DPWR_LED_PRSN		0x4000
-
-/* Field definitions in Link Capabilities Register */
-#define MAX_LNK_SPEED		0x000F
-#define MAX_LNK_WIDTH		0x03F0
-#define LINK_ACTIVE_REPORTING	0x00100000
-
-/* Link Width Encoding */
-#define LNK_X1		0x01
-#define LNK_X2		0x02
-#define LNK_X4		0x04
-#define LNK_X8		0x08
-#define LNK_X12		0x0C
-#define LNK_X16		0x10
-#define LNK_X32		0x20
-
-/*Field definitions of Link Status Register */
-#define LNK_SPEED	0x000F
-#define NEG_LINK_WD	0x03F0
-#define LNK_TRN_ERR	0x0400
-#define	LNK_TRN		0x0800
-#define SLOT_CLK_CONF	0x1000
-#define LINK_ACTIVE	0x2000
-
-/* Field definitions in Slot Capabilities Register */
-#define ATTN_BUTTN_PRSN	0x00000001
-#define	PWR_CTRL_PRSN	0x00000002
-#define MRL_SENS_PRSN	0x00000004
-#define ATTN_LED_PRSN	0x00000008
-#define PWR_LED_PRSN	0x00000010
-#define HP_SUPR_RM_SUP	0x00000020
-#define HP_CAP		0x00000040
-#define SLOT_PWR_VALUE	0x000003F8
-#define SLOT_PWR_LIMIT	0x00000C00
-#define PSN		0xFFF80000	/* PSN: Physical Slot Number */
-
-/* Field definitions in Slot Control Register */
-#define ATTN_BUTTN_ENABLE		0x0001
-#define PWR_FAULT_DETECT_ENABLE		0x0002
-#define MRL_DETECT_ENABLE		0x0004
-#define PRSN_DETECT_ENABLE		0x0008
-#define CMD_CMPL_INTR_ENABLE		0x0010
-#define HP_INTR_ENABLE			0x0020
-#define ATTN_LED_CTRL			0x00C0
-#define PWR_LED_CTRL			0x0300
-#define PWR_CTRL			0x0400
-#define EMI_CTRL			0x0800
-
-/* Attention indicator and Power indicator states */
-#define LED_ON		0x01
-#define LED_BLINK	0x10
-#define LED_OFF		0x11
-
 /* Power Control Command */
 #define POWER_ON	0
-#define POWER_OFF	0x0400
-
-/* EMI Status defines */
-#define EMI_DISENGAGED	0
-#define EMI_ENGAGED	1
-
-/* Field definitions in Slot Status Register */
-#define ATTN_BUTTN_PRESSED	0x0001
-#define PWR_FAULT_DETECTED	0x0002
-#define MRL_SENS_CHANGED	0x0004
-#define PRSN_DETECT_CHANGED	0x0008
-#define CMD_COMPLETED		0x0010
-#define MRL_STATE		0x0020
-#define PRSN_STATE		0x0040
-#define EMI_STATE		0x0080
-#define EMI_STATUS_BIT		7
+#define POWER_OFF	PCI_EXP_SLTCTL_PCC
 
 static irqreturn_t pcie_isr(int irq, void *dev_id);
 static void start_int_poll_timer(struct controller *ctrl, int sec);
@@ -253,22 +131,20 @@ static inline void pciehp_free_irq(struct controller *ctrl)
 static int pcie_poll_cmd(struct controller *ctrl)
 {
 	u16 slot_status;
-	int timeout = 1000;
+	int err, timeout = 1000;
 
-	if (!pciehp_readw(ctrl, SLOTSTATUS, &slot_status)) {
-		if (slot_status & CMD_COMPLETED) {
-			pciehp_writew(ctrl, SLOTSTATUS, CMD_COMPLETED);
-			return 1;
-		}
+	err = pciehp_readw(ctrl, PCI_EXP_SLTSTA, &slot_status);
+	if (!err && (slot_status & PCI_EXP_SLTSTA_CC)) {
+		pciehp_writew(ctrl, PCI_EXP_SLTSTA, PCI_EXP_SLTSTA_CC);
+		return 1;
 	}
 	while (timeout > 0) {
 		msleep(10);
 		timeout -= 10;
-		if (!pciehp_readw(ctrl, SLOTSTATUS, &slot_status)) {
-			if (slot_status & CMD_COMPLETED) {
-				pciehp_writew(ctrl, SLOTSTATUS, CMD_COMPLETED);
-				return 1;
-			}
+		err = pciehp_readw(ctrl, PCI_EXP_SLTSTA, &slot_status);
+		if (!err && (slot_status & PCI_EXP_SLTSTA_CC)) {
+			pciehp_writew(ctrl, PCI_EXP_SLTSTA, PCI_EXP_SLTSTA_CC);
+			return 1;
 		}
 	}
 	return 0;	/* timeout */
@@ -302,14 +178,14 @@ static int pcie_write_cmd(struct controller *ctrl, u16 cmd, u16 mask)
 
 	mutex_lock(&ctrl->ctrl_lock);
 
-	retval = pciehp_readw(ctrl, SLOTSTATUS, &slot_status);
+	retval = pciehp_readw(ctrl, PCI_EXP_SLTSTA, &slot_status);
 	if (retval) {
 		ctrl_err(ctrl, "%s: Cannot read SLOTSTATUS register\n",
 			 __func__);
 		goto out;
 	}
 
-	if (slot_status & CMD_COMPLETED) {
+	if (slot_status & PCI_EXP_SLTSTA_CC) {
 		if (!ctrl->no_cmd_complete) {
 			/*
 			 * After 1 sec and CMD_COMPLETED still not set, just
@@ -332,7 +208,7 @@ static int pcie_write_cmd(struct controller *ctrl, u16 cmd, u16 mask)
 		}
 	}
 
-	retval = pciehp_readw(ctrl, SLOTCTRL, &slot_ctrl);
+	retval = pciehp_readw(ctrl, PCI_EXP_SLTCTL, &slot_ctrl);
 	if (retval) {
 		ctrl_err(ctrl, "%s: Cannot read SLOTCTRL register\n", __func__);
 		goto out;
@@ -342,7 +218,7 @@ static int pcie_write_cmd(struct controller *ctrl, u16 cmd, u16 mask)
 	slot_ctrl |= (cmd & mask);
 	ctrl->cmd_busy = 1;
 	smp_mb();
-	retval = pciehp_writew(ctrl, SLOTCTRL, slot_ctrl);
+	retval = pciehp_writew(ctrl, PCI_EXP_SLTCTL, slot_ctrl);
 	if (retval)
 		ctrl_err(ctrl, "Cannot write to SLOTCTRL register\n");
 
@@ -356,8 +232,8 @@ static int pcie_write_cmd(struct controller *ctrl, u16 cmd, u16 mask)
 		 * completed interrupt is not enabled, we need to poll
 		 * command completed event.
 		 */
-		if (!(slot_ctrl & HP_INTR_ENABLE) ||
-		    !(slot_ctrl & CMD_CMPL_INTR_ENABLE))
+		if (!(slot_ctrl & PCI_EXP_SLTCTL_HPIE) ||
+		    !(slot_ctrl & PCI_EXP_SLTCTL_CCIE))
 			poll = 1;
                 pcie_wait_cmd(ctrl, poll);
 	}
@@ -370,9 +246,9 @@ static inline int check_link_active(struct controller *ctrl)
 {
 	u16 link_status;
 
-	if (pciehp_readw(ctrl, LNKSTATUS, &link_status))
+	if (pciehp_readw(ctrl, PCI_EXP_LNKSTA, &link_status))
 		return 0;
-	return !!(link_status & LINK_ACTIVE);
+	return !!(link_status & PCI_EXP_LNKSTA_DLLLA);
 }
 
 static void pcie_wait_link_active(struct controller *ctrl)
@@ -412,14 +288,15 @@ static int hpc_check_lnk_status(struct controller *ctrl)
         } else
                 msleep(1000);
 
-	retval = pciehp_readw(ctrl, LNKSTATUS, &lnk_status);
+	retval = pciehp_readw(ctrl, PCI_EXP_LNKSTA, &lnk_status);
 	if (retval) {
 		ctrl_err(ctrl, "Cannot read LNKSTATUS register\n");
 		return retval;
 	}
 
 	ctrl_dbg(ctrl, "%s: lnk_status = %x\n", __func__, lnk_status);
-	if ((lnk_status & LNK_TRN) || !(lnk_status & NEG_LINK_WD)) {
+	if ((lnk_status & PCI_EXP_LNKSTA_LT) ||
+	    !(lnk_status & PCI_EXP_LNKSTA_NLW)) {
 		ctrl_err(ctrl, "Link Training Error occurs \n");
 		retval = -1;
 		return retval;
@@ -435,16 +312,16 @@ static int hpc_get_attention_status(struct slot *slot, u8 *status)
 	u8 atten_led_state;
 	int retval = 0;
 
-	retval = pciehp_readw(ctrl, SLOTCTRL, &slot_ctrl);
+	retval = pciehp_readw(ctrl, PCI_EXP_SLTCTL, &slot_ctrl);
 	if (retval) {
 		ctrl_err(ctrl, "%s: Cannot read SLOTCTRL register\n", __func__);
 		return retval;
 	}
 
 	ctrl_dbg(ctrl, "%s: SLOTCTRL %x, value read %x\n",
-		 __func__, ctrl->cap_base + SLOTCTRL, slot_ctrl);
+		 __func__, ctrl->cap_base + PCI_EXP_SLTCTL, slot_ctrl);
 
-	atten_led_state = (slot_ctrl & ATTN_LED_CTRL) >> 6;
+	atten_led_state = (slot_ctrl & PCI_EXP_SLTCTL_AIC) >> 6;
 
 	switch (atten_led_state) {
 	case 0:
@@ -474,15 +351,15 @@ static int hpc_get_power_status(struct slot *slot, u8 *status)
 	u8 pwr_state;
 	int	retval = 0;
 
-	retval = pciehp_readw(ctrl, SLOTCTRL, &slot_ctrl);
+	retval = pciehp_readw(ctrl, PCI_EXP_SLTCTL, &slot_ctrl);
 	if (retval) {
 		ctrl_err(ctrl, "%s: Cannot read SLOTCTRL register\n", __func__);
 		return retval;
 	}
 	ctrl_dbg(ctrl, "%s: SLOTCTRL %x value read %x\n",
-		 __func__, ctrl->cap_base + SLOTCTRL, slot_ctrl);
+		 __func__, ctrl->cap_base + PCI_EXP_SLTCTL, slot_ctrl);
 
-	pwr_state = (slot_ctrl & PWR_CTRL) >> 10;
+	pwr_state = (slot_ctrl & PCI_EXP_SLTCTL_PCC) >> 10;
 
 	switch (pwr_state) {
 	case 0:
@@ -503,17 +380,15 @@ static int hpc_get_latch_status(struct slot *slot, u8 *status)
 {
 	struct controller *ctrl = slot->ctrl;
 	u16 slot_status;
-	int retval = 0;
+	int retval;
 
-	retval = pciehp_readw(ctrl, SLOTSTATUS, &slot_status);
+	retval = pciehp_readw(ctrl, PCI_EXP_SLTSTA, &slot_status);
 	if (retval) {
 		ctrl_err(ctrl, "%s: Cannot read SLOTSTATUS register\n",
 			 __func__);
 		return retval;
 	}
-
-	*status = (((slot_status & MRL_STATE) >> 5) == 0) ? 0 : 1;
-
+	*status = !!(slot_status & PCI_EXP_SLTSTA_MRLSS);
 	return 0;
 }
 
@@ -521,18 +396,15 @@ static int hpc_get_adapter_status(struct slot *slot, u8 *status)
 {
 	struct controller *ctrl = slot->ctrl;
 	u16 slot_status;
-	u8 card_state;
-	int retval = 0;
+	int retval;
 
-	retval = pciehp_readw(ctrl, SLOTSTATUS, &slot_status);
+	retval = pciehp_readw(ctrl, PCI_EXP_SLTSTA, &slot_status);
 	if (retval) {
 		ctrl_err(ctrl, "%s: Cannot read SLOTSTATUS register\n",
 			 __func__);
 		return retval;
 	}
-	card_state = (u8)((slot_status & PRSN_STATE) >> 6);
-	*status = (card_state == 1) ? 1 : 0;
-
+	*status = !!(slot_status & PCI_EXP_SLTSTA_PDS);
 	return 0;
 }
 
@@ -540,32 +412,28 @@ static int hpc_query_power_fault(struct slot *slot)
 {
 	struct controller *ctrl = slot->ctrl;
 	u16 slot_status;
-	u8 pwr_fault;
-	int retval = 0;
+	int retval;
 
-	retval = pciehp_readw(ctrl, SLOTSTATUS, &slot_status);
+	retval = pciehp_readw(ctrl, PCI_EXP_SLTSTA, &slot_status);
 	if (retval) {
 		ctrl_err(ctrl, "Cannot check for power fault\n");
 		return retval;
 	}
-	pwr_fault = (u8)((slot_status & PWR_FAULT_DETECTED) >> 1);
-
-	return pwr_fault;
+	return !!(slot_status & PCI_EXP_SLTSTA_PFD);
 }
 
 static int hpc_get_emi_status(struct slot *slot, u8 *status)
 {
 	struct controller *ctrl = slot->ctrl;
 	u16 slot_status;
-	int retval = 0;
+	int retval;
 
-	retval = pciehp_readw(ctrl, SLOTSTATUS, &slot_status);
+	retval = pciehp_readw(ctrl, PCI_EXP_SLTSTA, &slot_status);
 	if (retval) {
 		ctrl_err(ctrl, "Cannot check EMI status\n");
 		return retval;
 	}
-	*status = (slot_status & EMI_STATE) >> EMI_STATUS_BIT;
-
+	*status = !!(slot_status & PCI_EXP_SLTSTA_EIS);
 	return retval;
 }
 
@@ -575,8 +443,8 @@ static int hpc_toggle_emi(struct slot *slot)
 	u16 cmd_mask;
 	int rc;
 
-	slot_cmd = EMI_CTRL;
-	cmd_mask = EMI_CTRL;
+	slot_cmd = PCI_EXP_SLTCTL_EIC;
+	cmd_mask = PCI_EXP_SLTCTL_EIC;
 	rc = pcie_write_cmd(slot->ctrl, slot_cmd, cmd_mask);
 	slot->last_emi_toggle = get_seconds();
 
@@ -590,7 +458,7 @@ static int hpc_set_attention_status(struct slot *slot, u8 value)
 	u16 cmd_mask;
 	int rc;
 
-	cmd_mask = ATTN_LED_CTRL;
+	cmd_mask = PCI_EXP_SLTCTL_AIC;
 	switch (value) {
 		case 0 :	/* turn off */
 			slot_cmd = 0x00C0;
@@ -606,7 +474,7 @@ static int hpc_set_attention_status(struct slot *slot, u8 value)
 	}
 	rc = pcie_write_cmd(ctrl, slot_cmd, cmd_mask);
 	ctrl_dbg(ctrl, "%s: SLOTCTRL %x write cmd %x\n",
-		 __func__, ctrl->cap_base + SLOTCTRL, slot_cmd);
+		 __func__, ctrl->cap_base + PCI_EXP_SLTCTL, slot_cmd);
 
 	return rc;
 }
@@ -618,10 +486,10 @@ static void hpc_set_green_led_on(struct slot *slot)
 	u16 cmd_mask;
 
 	slot_cmd = 0x0100;
-	cmd_mask = PWR_LED_CTRL;
+	cmd_mask = PCI_EXP_SLTCTL_PIC;
 	pcie_write_cmd(ctrl, slot_cmd, cmd_mask);
 	ctrl_dbg(ctrl, "%s: SLOTCTRL %x write cmd %x\n",
-		 __func__, ctrl->cap_base + SLOTCTRL, slot_cmd);
+		 __func__, ctrl->cap_base + PCI_EXP_SLTCTL, slot_cmd);
 }
 
 static void hpc_set_green_led_off(struct slot *slot)
@@ -631,10 +499,10 @@ static void hpc_set_green_led_off(struct slot *slot)
 	u16 cmd_mask;
 
 	slot_cmd = 0x0300;
-	cmd_mask = PWR_LED_CTRL;
+	cmd_mask = PCI_EXP_SLTCTL_PIC;
 	pcie_write_cmd(ctrl, slot_cmd, cmd_mask);
 	ctrl_dbg(ctrl, "%s: SLOTCTRL %x write cmd %x\n",
-		 __func__, ctrl->cap_base + SLOTCTRL, slot_cmd);
+		 __func__, ctrl->cap_base + PCI_EXP_SLTCTL, slot_cmd);
 }
 
 static void hpc_set_green_led_blink(struct slot *slot)
@@ -644,10 +512,10 @@ static void hpc_set_green_led_blink(struct slot *slot)
 	u16 cmd_mask;
 
 	slot_cmd = 0x0200;
-	cmd_mask = PWR_LED_CTRL;
+	cmd_mask = PCI_EXP_SLTCTL_PIC;
 	pcie_write_cmd(ctrl, slot_cmd, cmd_mask);
 	ctrl_dbg(ctrl, "%s: SLOTCTRL %x write cmd %x\n",
-		 __func__, ctrl->cap_base + SLOTCTRL, slot_cmd);
+		 __func__, ctrl->cap_base + PCI_EXP_SLTCTL, slot_cmd);
 }
 
 static int hpc_power_on_slot(struct slot * slot)
@@ -661,15 +529,15 @@ static int hpc_power_on_slot(struct slot * slot)
 	ctrl_dbg(ctrl, "%s: slot->hp_slot %x\n", __func__, slot->hp_slot);
 
 	/* Clear sticky power-fault bit from previous power failures */
-	retval = pciehp_readw(ctrl, SLOTSTATUS, &slot_status);
+	retval = pciehp_readw(ctrl, PCI_EXP_SLTSTA, &slot_status);
 	if (retval) {
 		ctrl_err(ctrl, "%s: Cannot read SLOTSTATUS register\n",
 			 __func__);
 		return retval;
 	}
-	slot_status &= PWR_FAULT_DETECTED;
+	slot_status &= PCI_EXP_SLTSTA_PFD;
 	if (slot_status) {
-		retval = pciehp_writew(ctrl, SLOTSTATUS, slot_status);
+		retval = pciehp_writew(ctrl, PCI_EXP_SLTSTA, slot_status);
 		if (retval) {
 			ctrl_err(ctrl,
 				 "%s: Cannot write to SLOTSTATUS register\n",
@@ -679,13 +547,13 @@ static int hpc_power_on_slot(struct slot * slot)
 	}
 
 	slot_cmd = POWER_ON;
-	cmd_mask = PWR_CTRL;
+	cmd_mask = PCI_EXP_SLTCTL_PCC;
 	/* Enable detection that we turned off at slot power-off time */
 	if (!pciehp_poll_mode) {
-		slot_cmd |= (PWR_FAULT_DETECT_ENABLE | MRL_DETECT_ENABLE |
-			     PRSN_DETECT_ENABLE);
-		cmd_mask |= (PWR_FAULT_DETECT_ENABLE | MRL_DETECT_ENABLE |
-			     PRSN_DETECT_ENABLE);
+		slot_cmd |= (PCI_EXP_SLTCTL_PFDE | PCI_EXP_SLTCTL_MRLSCE |
+			     PCI_EXP_SLTCTL_PDCE);
+		cmd_mask |= (PCI_EXP_SLTCTL_PFDE | PCI_EXP_SLTCTL_MRLSCE |
+			     PCI_EXP_SLTCTL_PDCE);
 	}
 
 	retval = pcie_write_cmd(ctrl, slot_cmd, cmd_mask);
@@ -695,7 +563,7 @@ static int hpc_power_on_slot(struct slot * slot)
 		return -1;
 	}
 	ctrl_dbg(ctrl, "%s: SLOTCTRL %x write cmd %x\n",
-		 __func__, ctrl->cap_base + SLOTCTRL, slot_cmd);
+		 __func__, ctrl->cap_base + PCI_EXP_SLTCTL, slot_cmd);
 
 	return retval;
 }
@@ -752,7 +620,7 @@ static int hpc_power_off_slot(struct slot * slot)
 	changed = pcie_mask_bad_dllp(ctrl);
 
 	slot_cmd = POWER_OFF;
-	cmd_mask = PWR_CTRL;
+	cmd_mask = PCI_EXP_SLTCTL_PCC;
 	/*
 	 * If we get MRL or presence detect interrupts now, the isr
 	 * will notice the sticky power-fault bit too and issue power
@@ -761,10 +629,10 @@ static int hpc_power_off_slot(struct slot * slot)
 	 * till the slot is powered on again.
 	 */
 	if (!pciehp_poll_mode) {
-		slot_cmd &= ~(PWR_FAULT_DETECT_ENABLE | MRL_DETECT_ENABLE |
-			      PRSN_DETECT_ENABLE);
-		cmd_mask |= (PWR_FAULT_DETECT_ENABLE | MRL_DETECT_ENABLE |
-			     PRSN_DETECT_ENABLE);
+		slot_cmd &= ~(PCI_EXP_SLTCTL_PFDE | PCI_EXP_SLTCTL_MRLSCE |
+			      PCI_EXP_SLTCTL_PDCE);
+		cmd_mask |= (PCI_EXP_SLTCTL_PFDE | PCI_EXP_SLTCTL_MRLSCE |
+			     PCI_EXP_SLTCTL_PDCE);
 	}
 
 	retval = pcie_write_cmd(ctrl, slot_cmd, cmd_mask);
@@ -774,7 +642,7 @@ static int hpc_power_off_slot(struct slot * slot)
 		goto out;
 	}
 	ctrl_dbg(ctrl, "%s: SLOTCTRL %x write cmd %x\n",
-		 __func__, ctrl->cap_base + SLOTCTRL, slot_cmd);
+		 __func__, ctrl->cap_base + PCI_EXP_SLTCTL, slot_cmd);
  out:
 	if (changed)
 		pcie_unmask_bad_dllp(ctrl);
@@ -795,19 +663,19 @@ static irqreturn_t pcie_isr(int irq, void *dev_id)
 	 */
 	intr_loc = 0;
 	do {
-		if (pciehp_readw(ctrl, SLOTSTATUS, &detected)) {
+		if (pciehp_readw(ctrl, PCI_EXP_SLTSTA, &detected)) {
 			ctrl_err(ctrl, "%s: Cannot read SLOTSTATUS\n",
 				 __func__);
 			return IRQ_NONE;
 		}
 
-		detected &= (ATTN_BUTTN_PRESSED | PWR_FAULT_DETECTED |
-			     MRL_SENS_CHANGED | PRSN_DETECT_CHANGED |
-			     CMD_COMPLETED);
+		detected &= (PCI_EXP_SLTSTA_ABP | PCI_EXP_SLTSTA_PFD |
+			     PCI_EXP_SLTSTA_MRLSC | PCI_EXP_SLTSTA_PDC |
+			     PCI_EXP_SLTSTA_CC);
 		intr_loc |= detected;
 		if (!intr_loc)
 			return IRQ_NONE;
-		if (detected && pciehp_writew(ctrl, SLOTSTATUS, detected)) {
+		if (detected && pciehp_writew(ctrl, PCI_EXP_SLTSTA, detected)) {
 			ctrl_err(ctrl, "%s: Cannot write to SLOTSTATUS\n",
 				 __func__);
 			return IRQ_NONE;
@@ -817,31 +685,31 @@ static irqreturn_t pcie_isr(int irq, void *dev_id)
 	ctrl_dbg(ctrl, "%s: intr_loc %x\n", __func__, intr_loc);
 
 	/* Check Command Complete Interrupt Pending */
-	if (intr_loc & CMD_COMPLETED) {
+	if (intr_loc & PCI_EXP_SLTSTA_CC) {
 		ctrl->cmd_busy = 0;
 		smp_mb();
 		wake_up(&ctrl->queue);
 	}
 
-	if (!(intr_loc & ~CMD_COMPLETED))
+	if (!(intr_loc & ~PCI_EXP_SLTSTA_CC))
 		return IRQ_HANDLED;
 
 	p_slot = pciehp_find_slot(ctrl, ctrl->slot_device_offset);
 
 	/* Check MRL Sensor Changed */
-	if (intr_loc & MRL_SENS_CHANGED)
+	if (intr_loc & PCI_EXP_SLTSTA_MRLSC)
 		pciehp_handle_switch_change(p_slot);
 
 	/* Check Attention Button Pressed */
-	if (intr_loc & ATTN_BUTTN_PRESSED)
+	if (intr_loc & PCI_EXP_SLTSTA_ABP)
 		pciehp_handle_attention_button(p_slot);
 
 	/* Check Presence Detect Changed */
-	if (intr_loc & PRSN_DETECT_CHANGED)
+	if (intr_loc & PCI_EXP_SLTSTA_PDC)
 		pciehp_handle_presence_change(p_slot);
 
 	/* Check Power Fault Detected */
-	if (intr_loc & PWR_FAULT_DETECTED)
+	if (intr_loc & PCI_EXP_SLTSTA_PFD)
 		pciehp_handle_power_fault(p_slot);
 
 	return IRQ_HANDLED;
@@ -854,7 +722,7 @@ static int hpc_get_max_lnk_speed(struct slot *slot, enum pci_bus_speed *value)
 	u32	lnk_cap;
 	int retval = 0;
 
-	retval = pciehp_readl(ctrl, LNKCAP, &lnk_cap);
+	retval = pciehp_readl(ctrl, PCI_EXP_LNKCAP, &lnk_cap);
 	if (retval) {
 		ctrl_err(ctrl, "%s: Cannot read LNKCAP register\n", __func__);
 		return retval;
@@ -883,13 +751,13 @@ static int hpc_get_max_lnk_width(struct slot *slot,
 	u32	lnk_cap;
 	int retval = 0;
 
-	retval = pciehp_readl(ctrl, LNKCAP, &lnk_cap);
+	retval = pciehp_readl(ctrl, PCI_EXP_LNKCAP, &lnk_cap);
 	if (retval) {
 		ctrl_err(ctrl, "%s: Cannot read LNKCAP register\n", __func__);
 		return retval;
 	}
 
-	switch ((lnk_cap & 0x03F0) >> 4){
+	switch ((lnk_cap & PCI_EXP_LNKSTA_NLW) >> 4){
 	case 0:
 		lnk_wdth = PCIE_LNK_WIDTH_RESRV;
 		break;
@@ -932,14 +800,14 @@ static int hpc_get_cur_lnk_speed(struct slot *slot, enum pci_bus_speed *value)
 	int retval = 0;
 	u16 lnk_status;
 
-	retval = pciehp_readw(ctrl, LNKSTATUS, &lnk_status);
+	retval = pciehp_readw(ctrl, PCI_EXP_LNKSTA, &lnk_status);
 	if (retval) {
 		ctrl_err(ctrl, "%s: Cannot read LNKSTATUS register\n",
 			 __func__);
 		return retval;
 	}
 
-	switch (lnk_status & 0x0F) {
+	switch (lnk_status & PCI_EXP_LNKSTA_CLS) {
 	case 1:
 		lnk_speed = PCIE_2PT5GB;
 		break;
@@ -962,14 +830,14 @@ static int hpc_get_cur_lnk_width(struct slot *slot,
 	int retval = 0;
 	u16 lnk_status;
 
-	retval = pciehp_readw(ctrl, LNKSTATUS, &lnk_status);
+	retval = pciehp_readw(ctrl, PCI_EXP_LNKSTA, &lnk_status);
 	if (retval) {
 		ctrl_err(ctrl, "%s: Cannot read LNKSTATUS register\n",
 			 __func__);
 		return retval;
 	}
 
-	switch ((lnk_status & 0x03F0) >> 4){
+	switch ((lnk_status & PCI_EXP_LNKSTA_NLW) >> 4){
 	case 0:
 		lnk_wdth = PCIE_LNK_WIDTH_RESRV;
 		break;
@@ -1035,18 +903,19 @@ int pcie_enable_notification(struct controller *ctrl)
 {
 	u16 cmd, mask;
 
-	cmd = PRSN_DETECT_ENABLE;
+	cmd = PCI_EXP_SLTCTL_PDCE;
 	if (ATTN_BUTTN(ctrl))
-		cmd |= ATTN_BUTTN_ENABLE;
+		cmd |= PCI_EXP_SLTCTL_ABPE;
 	if (POWER_CTRL(ctrl))
-		cmd |= PWR_FAULT_DETECT_ENABLE;
+		cmd |= PCI_EXP_SLTCTL_PFDE;
 	if (MRL_SENS(ctrl))
-		cmd |= MRL_DETECT_ENABLE;
+		cmd |= PCI_EXP_SLTCTL_MRLSCE;
 	if (!pciehp_poll_mode)
-		cmd |= HP_INTR_ENABLE | CMD_CMPL_INTR_ENABLE;
+		cmd |= PCI_EXP_SLTCTL_HPIE | PCI_EXP_SLTCTL_CCIE;
 
-	mask = PRSN_DETECT_ENABLE | ATTN_BUTTN_ENABLE | MRL_DETECT_ENABLE |
-	       PWR_FAULT_DETECT_ENABLE | HP_INTR_ENABLE | CMD_CMPL_INTR_ENABLE;
+	mask = (PCI_EXP_SLTCTL_PDCE | PCI_EXP_SLTCTL_ABPE |
+		PCI_EXP_SLTCTL_MRLSCE | PCI_EXP_SLTCTL_PFDE |
+		PCI_EXP_SLTCTL_HPIE | PCI_EXP_SLTCTL_CCIE);
 
 	if (pcie_write_cmd(ctrl, cmd, mask)) {
 		ctrl_err(ctrl, "Cannot enable software notification\n");
@@ -1058,8 +927,9 @@ int pcie_enable_notification(struct controller *ctrl)
 static void pcie_disable_notification(struct controller *ctrl)
 {
 	u16 mask;
-	mask = PRSN_DETECT_ENABLE | ATTN_BUTTN_ENABLE | MRL_DETECT_ENABLE |
-	       PWR_FAULT_DETECT_ENABLE | HP_INTR_ENABLE | CMD_CMPL_INTR_ENABLE;
+	mask = (PCI_EXP_SLTCTL_PDCE | PCI_EXP_SLTCTL_ABPE |
+		PCI_EXP_SLTCTL_MRLSCE | PCI_EXP_SLTCTL_PFDE |
+		PCI_EXP_SLTCTL_HPIE | PCI_EXP_SLTCTL_CCIE);
 	if (pcie_write_cmd(ctrl, 0, mask))
 		ctrl_warn(ctrl, "Cannot disable software notification\n");
 }
@@ -1156,9 +1026,9 @@ static inline void dbg_ctrl(struct controller *ctrl)
 		  EMI(ctrl)        ? "yes" : "no");
 	ctrl_info(ctrl, "  Command Completed    : %3s\n",
 		  NO_CMD_CMPL(ctrl) ? "no" : "yes");
-	pciehp_readw(ctrl, SLOTSTATUS, &reg16);
+	pciehp_readw(ctrl, PCI_EXP_SLTSTA, &reg16);
 	ctrl_info(ctrl, "Slot Status            : 0x%04x\n", reg16);
-	pciehp_readw(ctrl, SLOTCTRL, &reg16);
+	pciehp_readw(ctrl, PCI_EXP_SLTCTL, &reg16);
 	ctrl_info(ctrl, "Slot Control           : 0x%04x\n", reg16);
 }
 
@@ -1182,7 +1052,7 @@ struct controller *pcie_init(struct pcie_device *dev)
 		ctrl_err(ctrl, "Cannot find PCI Express capability\n");
 		goto abort_ctrl;
 	}
-	if (pciehp_readl(ctrl, SLOTCAP, &slot_cap)) {
+	if (pciehp_readl(ctrl, PCI_EXP_SLTCAP, &slot_cap)) {
 		ctrl_err(ctrl, "Cannot read SLOTCAP register\n");
 		goto abort_ctrl;
 	}
@@ -1207,17 +1077,17 @@ struct controller *pcie_init(struct pcie_device *dev)
 	    ctrl->no_cmd_complete = 1;
 
         /* Check if Data Link Layer Link Active Reporting is implemented */
-        if (pciehp_readl(ctrl, LNKCAP, &link_cap)) {
+        if (pciehp_readl(ctrl, PCI_EXP_LNKCAP, &link_cap)) {
                 ctrl_err(ctrl, "%s: Cannot read LNKCAP register\n", __func__);
                 goto abort_ctrl;
         }
-        if (link_cap & LINK_ACTIVE_REPORTING) {
+        if (link_cap & PCI_EXP_LNKCAP_DLLLARC) {
                 ctrl_dbg(ctrl, "Link Active Reporting supported\n");
                 ctrl->link_active_reporting = 1;
         }
 
 	/* Clear all remaining event bits in Slot Status register */
-	if (pciehp_writew(ctrl, SLOTSTATUS, 0x1f))
+	if (pciehp_writew(ctrl, PCI_EXP_SLTSTA, 0x1f))
 		goto abort_ctrl;
 
 	/* Disable sotfware notification */
diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h
index 7766488470e4..027815b4635e 100644
--- a/include/linux/pci_regs.h
+++ b/include/linux/pci_regs.h
@@ -411,20 +411,70 @@
 #define  PCI_EXP_DEVSTA_AUXPD	0x10	/* AUX Power Detected */
 #define  PCI_EXP_DEVSTA_TRPND	0x20	/* Transactions Pending */
 #define PCI_EXP_LNKCAP		12	/* Link Capabilities */
-#define  PCI_EXP_LNKCAP_ASPMS	0xc00	/* ASPM Support */
-#define  PCI_EXP_LNKCAP_L0SEL	0x7000	/* L0s Exit Latency */
-#define  PCI_EXP_LNKCAP_L1EL	0x38000	/* L1 Exit Latency */
-#define  PCI_EXP_LNKCAP_CLKPM	0x40000	/* L1 Clock Power Management */
+#define  PCI_EXP_LNKCAP_SLS	0x0000000f /* Supported Link Speeds */
+#define  PCI_EXP_LNKCAP_MLW	0x000003f0 /* Maximum Link Width */
+#define  PCI_EXP_LNKCAP_ASPMS	0x00000c00 /* ASPM Support */
+#define  PCI_EXP_LNKCAP_L0SEL	0x00007000 /* L0s Exit Latency */
+#define  PCI_EXP_LNKCAP_L1EL	0x00038000 /* L1 Exit Latency */
+#define  PCI_EXP_LNKCAP_CLKPM	0x00040000 /* L1 Clock Power Management */
+#define  PCI_EXP_LNKCAP_SDERC	0x00080000 /* Suprise Down Error Reporting Capable */
+#define  PCI_EXP_LNKCAP_DLLLARC	0x00100000 /* Data Link Layer Link Active Reporting Capable */
+#define  PCI_EXP_LNKCAP_LBNC	0x00200000 /* Link Bandwidth Notification Capability */
+#define  PCI_EXP_LNKCAP_PN	0xff000000 /* Port Number */
 #define PCI_EXP_LNKCTL		16	/* Link Control */
-#define  PCI_EXP_LNKCTL_RL	0x20	/* Retrain Link */
-#define  PCI_EXP_LNKCTL_CCC	0x40	/* Common Clock COnfiguration */
+#define  PCI_EXP_LNKCTL_ASPMC	0x0003	/* ASPM Control */
+#define  PCI_EXP_LNKCTL_RCB	0x0008	/* Read Completion Boundary */
+#define  PCI_EXP_LNKCTL_LD	0x0010	/* Link Disable */
+#define  PCI_EXP_LNKCTL_RL	0x0020	/* Retrain Link */
+#define  PCI_EXP_LNKCTL_CCC	0x0040	/* Common Clock Configuration */
+#define  PCI_EXP_LNKCTL_ES	0x0080	/* Extended Synch */
 #define  PCI_EXP_LNKCTL_CLKREQ_EN 0x100	/* Enable clkreq */
+#define  PCI_EXP_LNKCTL_HAWD	0x0200	/* Hardware Autonomous Width Disable */
+#define  PCI_EXP_LNKCTL_LBMIE	0x0400	/* Link Bandwidth Management Interrupt Enable */
+#define  PCI_EXP_LNKCTL_LABIE	0x0800	/* Lnk Autonomous Bandwidth Interrupt Enable */
 #define PCI_EXP_LNKSTA		18	/* Link Status */
-#define  PCI_EXP_LNKSTA_LT	0x800	/* Link Training */
+#define  PCI_EXP_LNKSTA_CLS	0x000f	/* Current Link Speed */
+#define  PCI_EXP_LNKSTA_NLW	0x03f0	/* Nogotiated Link Width */
+#define  PCI_EXP_LNKSTA_LT	0x0800	/* Link Training */
 #define  PCI_EXP_LNKSTA_SLC	0x1000	/* Slot Clock Configuration */
+#define  PCI_EXP_LNKSTA_DLLLA	0x2000	/* Data Link Layer Link Active */
+#define  PCI_EXP_LNKSTA_LBMS	0x4000	/* Link Bandwidth Management Status */
+#define  PCI_EXP_LNKSTA_LABS	0x8000	/* Link Autonomous Bandwidth Status */
 #define PCI_EXP_SLTCAP		20	/* Slot Capabilities */
+#define  PCI_EXP_SLTCAP_ABP	0x00000001 /* Attention Button Present */
+#define  PCI_EXP_SLTCAP_PCP	0x00000002 /* Power Controller Present */
+#define  PCI_EXP_SLTCAP_MRLSP	0x00000004 /* MRL Sensor Present */
+#define  PCI_EXP_SLTCAP_AIP	0x00000008 /* Attention Indicator Present */
+#define  PCI_EXP_SLTCAP_PIP	0x00000010 /* Power Indicator Present */
+#define  PCI_EXP_SLTCAP_HPS	0x00000020 /* Hot-Plug Surprise */
+#define  PCI_EXP_SLTCAP_HPC	0x00000040 /* Hot-Plug Capable */
+#define  PCI_EXP_SLTCAP_SPLV	0x00007f80 /* Slot Power Limit Value */
+#define  PCI_EXP_SLTCAP_SPLS	0x00018000 /* Slot Power Limit Scale */
+#define  PCI_EXP_SLTCAP_EIP	0x00020000 /* Electromechanical Interlock Present */
+#define  PCI_EXP_SLTCAP_NCCS	0x00040000 /* No Command Completed Support */
+#define  PCI_EXP_SLTCAP_PSN	0xfff80000 /* Physical Slot Number */
 #define PCI_EXP_SLTCTL		24	/* Slot Control */
+#define  PCI_EXP_SLTCTL_ABPE	0x0001	/* Attention Button Pressed Enable */
+#define  PCI_EXP_SLTCTL_PFDE	0x0002	/* Power Fault Detected Enable */
+#define  PCI_EXP_SLTCTL_MRLSCE	0x0004	/* MRL Sensor Changed Enable */
+#define  PCI_EXP_SLTCTL_PDCE	0x0008	/* Presence Detect Changed Enable */
+#define  PCI_EXP_SLTCTL_CCIE	0x0010	/* Command Completed Interrupt Enable */
+#define  PCI_EXP_SLTCTL_HPIE	0x0020	/* Hot-Plug Interrupt Enable */
+#define  PCI_EXP_SLTCTL_AIC	0x00c0	/* Attention Indicator Control */
+#define  PCI_EXP_SLTCTL_PIC	0x0300	/* Power Indicator Control */
+#define  PCI_EXP_SLTCTL_PCC	0x0400	/* Power Controller Control */
+#define  PCI_EXP_SLTCTL_EIC	0x0800	/* Electromechanical Interlock Control */
+#define  PCI_EXP_SLTCTL_DLLSCE	0x1000	/* Data Link Layer State Changed Enable */
 #define PCI_EXP_SLTSTA		26	/* Slot Status */
+#define  PCI_EXP_SLTSTA_ABP	0x0001	/* Attention Button Pressed */
+#define  PCI_EXP_SLTSTA_PFD	0x0002	/* Power Fault Detected */
+#define  PCI_EXP_SLTSTA_MRLSC	0x0004	/* MRL Sensor Changed */
+#define  PCI_EXP_SLTSTA_PDC	0x0008	/* Presence Detect Changed */
+#define  PCI_EXP_SLTSTA_CC	0x0010	/* Command Completed */
+#define  PCI_EXP_SLTSTA_MRLSS	0x0020	/* MRL Sensor State */
+#define  PCI_EXP_SLTSTA_PDS	0x0040	/* Presence Detect State */
+#define  PCI_EXP_SLTSTA_EIS	0x0080	/* Electromechanical Interlock Status */
+#define  PCI_EXP_SLTSTA_DLLSC	0x0100	/* Data Link Layer State Changed */
 #define PCI_EXP_RTCTL		28	/* Root Control */
 #define  PCI_EXP_RTCTL_SECEE	0x01	/* System Error on Correctable Error */
 #define  PCI_EXP_RTCTL_SENFEE	0x02	/* System Error on Non-Fatal Error */
-- 
cgit v1.2.3


From 6a479079c07211bf348ac8a79754f26bea258f26 Mon Sep 17 00:00:00 2001
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Tue, 23 Dec 2008 03:08:29 +0000
Subject: PCI: Add pci_clear_master() as opposite of pci_set_master()

During an online device reset it may be useful to disable bus-mastering.
pci_disable_device() does that, and far more besides, so is not suitable
for an online reset.

Add pci_clear_master() which does just this.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Reviewed-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 Documentation/PCI/pci.txt |  3 ++-
 drivers/pci/pci.c         | 39 ++++++++++++++++++++++++++++-----------
 include/linux/pci.h       |  1 +
 3 files changed, 31 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/PCI/pci.txt b/Documentation/PCI/pci.txt
index fd4907a2968c..7f6de6ea5b47 100644
--- a/Documentation/PCI/pci.txt
+++ b/Documentation/PCI/pci.txt
@@ -294,7 +294,8 @@ NOTE: pci_enable_device() can fail! Check the return value.
 
 pci_set_master() will enable DMA by setting the bus master bit
 in the PCI_COMMAND register. It also fixes the latency timer value if
-it's set to something bogus by the BIOS.
+it's set to something bogus by the BIOS.  pci_clear_master() will
+disable DMA by clearing the bus master bit.
 
 If the PCI device can use the PCI Memory-Write-Invalidate transaction,
 call pci_set_mwi().  This enables the PCI_COMMAND bit for Mem-Wr-Inval
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index c824dc8d617c..f3fd55df67db 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -1667,6 +1667,22 @@ int pci_request_regions_exclusive(struct pci_dev *pdev, const char *res_name)
 					((1 << 6) - 1), res_name);
 }
 
+static void __pci_set_master(struct pci_dev *dev, bool enable)
+{
+	u16 old_cmd, cmd;
+
+	pci_read_config_word(dev, PCI_COMMAND, &old_cmd);
+	if (enable)
+		cmd = old_cmd | PCI_COMMAND_MASTER;
+	else
+		cmd = old_cmd & ~PCI_COMMAND_MASTER;
+	if (cmd != old_cmd) {
+		dev_dbg(&dev->dev, "%s bus mastering\n",
+			enable ? "enabling" : "disabling");
+		pci_write_config_word(dev, PCI_COMMAND, cmd);
+	}
+	dev->is_busmaster = enable;
+}
 
 /**
  * pci_set_master - enables bus-mastering for device dev
@@ -1675,21 +1691,21 @@ int pci_request_regions_exclusive(struct pci_dev *pdev, const char *res_name)
  * Enables bus-mastering on the device and calls pcibios_set_master()
  * to do the needed arch specific settings.
  */
-void
-pci_set_master(struct pci_dev *dev)
+void pci_set_master(struct pci_dev *dev)
 {
-	u16 cmd;
-
-	pci_read_config_word(dev, PCI_COMMAND, &cmd);
-	if (! (cmd & PCI_COMMAND_MASTER)) {
-		dev_dbg(&dev->dev, "enabling bus mastering\n");
-		cmd |= PCI_COMMAND_MASTER;
-		pci_write_config_word(dev, PCI_COMMAND, cmd);
-	}
-	dev->is_busmaster = 1;
+	__pci_set_master(dev, true);
 	pcibios_set_master(dev);
 }
 
+/**
+ * pci_clear_master - disables bus-mastering for device dev
+ * @dev: the PCI device to disable
+ */
+void pci_clear_master(struct pci_dev *dev)
+{
+	__pci_set_master(dev, false);
+}
+
 #ifdef PCI_DISABLE_MWI
 int pci_set_mwi(struct pci_dev *dev)
 {
@@ -2346,6 +2362,7 @@ EXPORT_SYMBOL(pci_release_selected_regions);
 EXPORT_SYMBOL(pci_request_selected_regions);
 EXPORT_SYMBOL(pci_request_selected_regions_exclusive);
 EXPORT_SYMBOL(pci_set_master);
+EXPORT_SYMBOL(pci_clear_master);
 EXPORT_SYMBOL(pci_set_mwi);
 EXPORT_SYMBOL(pci_try_set_mwi);
 EXPORT_SYMBOL(pci_clear_mwi);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 7cbecef19bb6..0f6d2bb1df9c 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -642,6 +642,7 @@ static inline int pci_is_managed(struct pci_dev *pdev)
 
 void pci_disable_device(struct pci_dev *dev);
 void pci_set_master(struct pci_dev *dev);
+void pci_clear_master(struct pci_dev *dev);
 int pci_set_pcie_reset_state(struct pci_dev *dev, enum pcie_reset_state state);
 #define HAVE_PCI_SET_MWI
 int __must_check pci_set_mwi(struct pci_dev *dev);
-- 
cgit v1.2.3


From 16cf0ebc35dd63f72628ba1246132a6fd17bced2 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 5 Jan 2009 14:50:27 +0100
Subject: x86/PCI: Do not use interrupt links for devices using MSI-X

pcibios_enable_device() and pcibios_disable_device() don't handle
IRQs for devices that have MSI enabled and it should treat the
devices with MSI-X enabled in the same way.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 arch/x86/pci/common.c | 4 ++--
 include/linux/pci.h   | 9 +++++++++
 2 files changed, 11 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c
index 9ab8509f7b15..82d22fc601ae 100644
--- a/arch/x86/pci/common.c
+++ b/arch/x86/pci/common.c
@@ -551,14 +551,14 @@ int pcibios_enable_device(struct pci_dev *dev, int mask)
 	if ((err = pci_enable_resources(dev, mask)) < 0)
 		return err;
 
-	if (!dev->msi_enabled)
+	if (!pci_dev_msi_enabled(dev))
 		return pcibios_enable_irq(dev);
 	return 0;
 }
 
 void pcibios_disable_device (struct pci_dev *dev)
 {
-	if (!dev->msi_enabled && pcibios_disable_irq)
+	if (!pci_dev_msi_enabled(dev) && pcibios_disable_irq)
 		pcibios_disable_irq(dev);
 }
 
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 0f6d2bb1df9c..80f8b8b65fde 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -336,6 +336,15 @@ struct pci_bus {
 #define pci_bus_b(n)	list_entry(n, struct pci_bus, node)
 #define to_pci_bus(n)	container_of(n, struct pci_bus, dev)
 
+#ifdef CONFIG_PCI_MSI
+static inline bool pci_dev_msi_enabled(struct pci_dev *pci_dev)
+{
+	return pci_dev->msi_enabled || pci_dev->msix_enabled;
+}
+#else
+static inline bool pci_dev_msi_enabled(struct pci_dev *pci_dev) { return false; }
+#endif
+
 /*
  * Error values that may be returned by PCI functions.
  */
-- 
cgit v1.2.3


From 940fbf411e5fb42aee8ab7dd814b24080951dbfc Mon Sep 17 00:00:00 2001
From: Detlef Riekenberg <wine.dev@web.de>
Date: Wed, 7 Jan 2009 10:11:44 +0100
Subject: linux/types.h: Don't depend on __GNUC__ for __le64/__be64

The typedefs for __u64 and __s64 where fixed to be available for other
compiler on May 2 2008 by H.  Peter Anvin (in commit edfa5cfa3dc5)

Acked-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Detlef Riekenberg <wine.dev@web.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/types.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/types.h b/include/linux/types.h
index 3b864f2d9560..712ca53bc348 100644
--- a/include/linux/types.h
+++ b/include/linux/types.h
@@ -176,10 +176,9 @@ typedef __u16 __bitwise __le16;
 typedef __u16 __bitwise __be16;
 typedef __u32 __bitwise __le32;
 typedef __u32 __bitwise __be32;
-#if defined(__GNUC__)
 typedef __u64 __bitwise __le64;
 typedef __u64 __bitwise __be64;
-#endif
+
 typedef __u16 __bitwise __sum16;
 typedef __u32 __bitwise __wsum;
 
-- 
cgit v1.2.3


From 8d1a0a13edecfdcb47fee3238ed4a2af2a2867f9 Mon Sep 17 00:00:00 2001
From: Anders Larsen <al@alarsen.net>
Date: Thu, 1 Jan 2009 17:17:35 +0100
Subject: qnx: include <linux/types.h> for definitions of __[us]{8,16,32,64}
 types

On 2008-12-30 11:32:33, Sam Ravnborg wrote:
> We have added a few additional validation checks of the userspace headers:
...
> 3) We should include <linux/types.h> and not <asm/types.h>
> 4) If we use a __[us]{8,16,32,64} type then we must include <linux/types.h>

Satisfy these requirements for the linux/qnx*.h headers.

Signed-off-by: Anders Larsen <al@alarsen.net>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
---
 include/linux/qnx4_fs.h  | 4 +---
 include/linux/qnxtypes.h | 5 ++---
 2 files changed, 3 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/qnx4_fs.h b/include/linux/qnx4_fs.h
index 34a196ee7941..787d19ea9f46 100644
--- a/include/linux/qnx4_fs.h
+++ b/include/linux/qnx4_fs.h
@@ -2,14 +2,12 @@
  *  Name                         : qnx4_fs.h
  *  Author                       : Richard Frowijn
  *  Function                     : qnx4 global filesystem definitions
- *  Version                      : 1.0.2
- *  Last modified                : 2000-01-31
- *
  *  History                      : 23-03-1998 created
  */
 #ifndef _LINUX_QNX4_FS_H
 #define _LINUX_QNX4_FS_H
 
+#include <linux/types.h>
 #include <linux/qnxtypes.h>
 #include <linux/magic.h>
 
diff --git a/include/linux/qnxtypes.h b/include/linux/qnxtypes.h
index a3eb1137857b..bebbe5cc4fb8 100644
--- a/include/linux/qnxtypes.h
+++ b/include/linux/qnxtypes.h
@@ -2,9 +2,6 @@
  *  Name                         : qnxtypes.h
  *  Author                       : Richard Frowijn
  *  Function                     : standard qnx types
- *  Version                      : 1.0.2
- *  Last modified                : 2000-01-06
- *
  *  History                      : 22-03-1998 created
  *
  */
@@ -12,6 +9,8 @@
 #ifndef _QNX4TYPES_H
 #define _QNX4TYPES_H
 
+#include <linux/types.h>
+
 typedef __le16 qnx4_nxtnt_t;
 typedef __u8  qnx4_ftype_t;
 
-- 
cgit v1.2.3


From 14f0ca8eaea42a5b5a69cfcb699665dd2618db5f Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@amd.com>
Date: Wed, 7 Jan 2009 21:50:22 +0100
Subject: oprofile: make new cpu buffer functions part of the api

This patch creates the new functions

 oprofile_write_reserve()
 oprofile_add_data()
 oprofile_write_commit()

and makes them part of the oprofile api.

Signed-off-by: Robert Richter <robert.richter@amd.com>
---
 arch/x86/oprofile/op_model_amd.c | 57 +++++++++++++++++-----------------------
 drivers/oprofile/cpu_buffer.c    | 17 +++++++++---
 drivers/oprofile/cpu_buffer.h    |  8 +-----
 include/linux/oprofile.h         | 18 +++++++++++++
 4 files changed, 57 insertions(+), 43 deletions(-)

(limited to 'include/linux')

diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
index cf310aeb462c..8fdf06e4edf9 100644
--- a/arch/x86/oprofile/op_model_amd.c
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -22,7 +22,6 @@
 
 #include "op_x86_model.h"
 #include "op_counter.h"
-#include "../../../drivers/oprofile/cpu_buffer.h"
 
 #define NUM_COUNTERS 4
 #define NUM_CONTROLS 4
@@ -61,14 +60,6 @@ static unsigned long reset_value[NUM_COUNTERS];
 #define IBS_OP_LOW_VALID_BIT		(1ULL<<18)	/* bit 18 */
 #define IBS_OP_LOW_ENABLE		(1ULL<<17)	/* bit 17 */
 
-/*
- * The function interface needs to be fixed, something like add
- * data. Should then be added to linux/oprofile.h.
- */
-extern
-void oprofile_add_data(struct op_entry *entry, struct pt_regs * const regs,
-		       unsigned long pc, int code, int size);
-
 #define IBS_FETCH_SIZE	6
 #define IBS_OP_SIZE	12
 
@@ -174,16 +165,16 @@ op_amd_handle_ibs(struct pt_regs * const regs,
 		rdmsr(MSR_AMD64_IBSFETCHCTL, low, high);
 		if (high & IBS_FETCH_HIGH_VALID_BIT) {
 			rdmsrl(MSR_AMD64_IBSFETCHLINAD, msr);
-			oprofile_add_data(&entry, regs, msr, IBS_FETCH_CODE,
-					  IBS_FETCH_SIZE);
-			op_cpu_buffer_add_data(&entry, (u32)msr);
-			op_cpu_buffer_add_data(&entry, (u32)(msr >> 32));
-			op_cpu_buffer_add_data(&entry, low);
-			op_cpu_buffer_add_data(&entry, high);
+			oprofile_write_reserve(&entry, regs, msr,
+					       IBS_FETCH_CODE, IBS_FETCH_SIZE);
+			oprofile_add_data(&entry, (u32)msr);
+			oprofile_add_data(&entry, (u32)(msr >> 32));
+			oprofile_add_data(&entry, low);
+			oprofile_add_data(&entry, high);
 			rdmsrl(MSR_AMD64_IBSFETCHPHYSAD, msr);
-			op_cpu_buffer_add_data(&entry, (u32)msr);
-			op_cpu_buffer_add_data(&entry, (u32)(msr >> 32));
-			op_cpu_buffer_write_commit(&entry);
+			oprofile_add_data(&entry, (u32)msr);
+			oprofile_add_data(&entry, (u32)(msr >> 32));
+			oprofile_write_commit(&entry);
 
 			/* reenable the IRQ */
 			high &= ~IBS_FETCH_HIGH_VALID_BIT;
@@ -197,26 +188,26 @@ op_amd_handle_ibs(struct pt_regs * const regs,
 		rdmsr(MSR_AMD64_IBSOPCTL, low, high);
 		if (low & IBS_OP_LOW_VALID_BIT) {
 			rdmsrl(MSR_AMD64_IBSOPRIP, msr);
-			oprofile_add_data(&entry, regs, msr, IBS_OP_CODE,
-					  IBS_OP_SIZE);
-			op_cpu_buffer_add_data(&entry, (u32)msr);
-			op_cpu_buffer_add_data(&entry, (u32)(msr >> 32));
+			oprofile_write_reserve(&entry, regs, msr,
+					       IBS_OP_CODE, IBS_OP_SIZE);
+			oprofile_add_data(&entry, (u32)msr);
+			oprofile_add_data(&entry, (u32)(msr >> 32));
 			rdmsrl(MSR_AMD64_IBSOPDATA, msr);
-			op_cpu_buffer_add_data(&entry, (u32)msr);
-			op_cpu_buffer_add_data(&entry, (u32)(msr >> 32));
+			oprofile_add_data(&entry, (u32)msr);
+			oprofile_add_data(&entry, (u32)(msr >> 32));
 			rdmsrl(MSR_AMD64_IBSOPDATA2, msr);
-			op_cpu_buffer_add_data(&entry, (u32)msr);
-			op_cpu_buffer_add_data(&entry, (u32)(msr >> 32));
+			oprofile_add_data(&entry, (u32)msr);
+			oprofile_add_data(&entry, (u32)(msr >> 32));
 			rdmsrl(MSR_AMD64_IBSOPDATA3, msr);
-			op_cpu_buffer_add_data(&entry, (u32)msr);
-			op_cpu_buffer_add_data(&entry, (u32)(msr >> 32));
+			oprofile_add_data(&entry, (u32)msr);
+			oprofile_add_data(&entry, (u32)(msr >> 32));
 			rdmsrl(MSR_AMD64_IBSDCLINAD, msr);
-			op_cpu_buffer_add_data(&entry, (u32)msr);
-			op_cpu_buffer_add_data(&entry, (u32)(msr >> 32));
+			oprofile_add_data(&entry, (u32)msr);
+			oprofile_add_data(&entry, (u32)(msr >> 32));
 			rdmsrl(MSR_AMD64_IBSDCPHYSAD, msr);
-			op_cpu_buffer_add_data(&entry, (u32)msr);
-			op_cpu_buffer_add_data(&entry, (u32)(msr >> 32));
-			op_cpu_buffer_write_commit(&entry);
+			oprofile_add_data(&entry, (u32)msr);
+			oprofile_add_data(&entry, (u32)(msr >> 32));
+			oprofile_write_commit(&entry);
 
 			/* reenable the IRQ */
 			high = 0;
diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c
index b846af632c81..2e03b6d796d3 100644
--- a/drivers/oprofile/cpu_buffer.c
+++ b/drivers/oprofile/cpu_buffer.c
@@ -364,10 +364,11 @@ void oprofile_add_sample(struct pt_regs * const regs, unsigned long event)
 /*
  * Add samples with data to the ring buffer.
  *
- * Use op_cpu_buffer_add_data(&entry, val) to add data and
- * op_cpu_buffer_write_commit(&entry) to commit the sample.
+ * Use oprofile_add_data(&entry, val) to add data and
+ * oprofile_write_commit(&entry) to commit the sample.
  */
-void oprofile_add_data(struct op_entry *entry, struct pt_regs * const regs,
+void
+oprofile_write_reserve(struct op_entry *entry, struct pt_regs * const regs,
 		       unsigned long pc, int code, int size)
 {
 	struct op_sample *sample;
@@ -395,6 +396,16 @@ fail:
 	cpu_buf->sample_lost_overflow++;
 }
 
+int oprofile_add_data(struct op_entry *entry, unsigned long val)
+{
+	return op_cpu_buffer_add_data(entry, val);
+}
+
+int oprofile_write_commit(struct op_entry *entry)
+{
+	return op_cpu_buffer_write_commit(entry);
+}
+
 void oprofile_add_pc(unsigned long pc, int is_kernel, unsigned long event)
 {
 	struct oprofile_cpu_buffer *cpu_buf = &__get_cpu_var(cpu_buffer);
diff --git a/drivers/oprofile/cpu_buffer.h b/drivers/oprofile/cpu_buffer.h
index 525cc4d13d8d..63f81c44846a 100644
--- a/drivers/oprofile/cpu_buffer.h
+++ b/drivers/oprofile/cpu_buffer.h
@@ -35,13 +35,7 @@ struct op_sample {
 	unsigned long data[0];
 };
 
-struct op_entry {
-	struct ring_buffer_event *event;
-	struct op_sample *sample;
-	unsigned long irq_flags;
-	unsigned long size;
-	unsigned long *data;
-};
+struct op_entry;
 
 struct oprofile_cpu_buffer {
 	unsigned long buffer_size;
diff --git a/include/linux/oprofile.h b/include/linux/oprofile.h
index 1ce9fe572e51..1d9518bc4c58 100644
--- a/include/linux/oprofile.h
+++ b/include/linux/oprofile.h
@@ -164,4 +164,22 @@ void oprofile_put_buff(unsigned long *buf, unsigned int start,
 unsigned long oprofile_get_cpu_buffer_size(void);
 void oprofile_cpu_buffer_inc_smpl_lost(void);
  
+/* cpu buffer functions */
+
+struct op_sample;
+
+struct op_entry {
+	struct ring_buffer_event *event;
+	struct op_sample *sample;
+	unsigned long irq_flags;
+	unsigned long size;
+	unsigned long *data;
+};
+
+void oprofile_write_reserve(struct op_entry *entry,
+			    struct pt_regs * const regs,
+			    unsigned long pc, int code, int size);
+int oprofile_add_data(struct op_entry *entry, unsigned long val);
+int oprofile_write_commit(struct op_entry *entry);
+
 #endif /* OPROFILE_H */
-- 
cgit v1.2.3


From 87df4de8073f922a1f643b9fa6ba0412d5529ecf Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Mon, 15 Dec 2008 19:42:03 +0200
Subject: nfsd: last_byte_offset

refactor the nfs4 server lock code to use last_byte_offset
to compute the last byte covered by the lock.  Check for overflow
so that the last byte is set to NFS4_MAX_UINT64 if offset + len
wraps around.

Also, use NFS4_MAX_UINT64 for ~(u64)0 where appropriate.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4state.c  | 42 ++++++++++++++++++++++++++----------------
 include/linux/nfs4.h |  2 ++
 2 files changed, 28 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index e62d0e3df8b3..88db7d3ec120 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -2416,6 +2416,26 @@ out:
 #define LOCK_HASH_SIZE             (1 << LOCK_HASH_BITS)
 #define LOCK_HASH_MASK             (LOCK_HASH_SIZE - 1)
 
+static inline u64
+end_offset(u64 start, u64 len)
+{
+	u64 end;
+
+	end = start + len;
+	return end >= start ? end: NFS4_MAX_UINT64;
+}
+
+/* last octet in a range */
+static inline u64
+last_byte_offset(u64 start, u64 len)
+{
+	u64 end;
+
+	BUG_ON(!len);
+	end = start + len;
+	return end > start ? end - 1: NFS4_MAX_UINT64;
+}
+
 #define lockownerid_hashval(id) \
         ((id) & LOCK_HASH_MASK)
 
@@ -2519,8 +2539,8 @@ nfs4_set_lock_denied(struct file_lock *fl, struct nfsd4_lock_denied *deny)
 		deny->ld_clientid.cl_id = 0;
 	}
 	deny->ld_start = fl->fl_start;
-	deny->ld_length = ~(u64)0;
-	if (fl->fl_end != ~(u64)0)
+	deny->ld_length = NFS4_MAX_UINT64;
+	if (fl->fl_end != NFS4_MAX_UINT64)
 		deny->ld_length = fl->fl_end - fl->fl_start + 1;        
 	deny->ld_type = NFS4_READ_LT;
 	if (fl->fl_type != F_RDLCK)
@@ -2617,7 +2637,7 @@ out:
 static int
 check_lock_length(u64 offset, u64 length)
 {
-	return ((length == 0)  || ((length != ~(u64)0) &&
+	return ((length == 0)  || ((length != NFS4_MAX_UINT64) &&
 	     LOFF_OVERFLOW(offset, length)));
 }
 
@@ -2737,11 +2757,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	file_lock.fl_lmops = &nfsd_posix_mng_ops;
 
 	file_lock.fl_start = lock->lk_offset;
-	if ((lock->lk_length == ~(u64)0) || 
-			LOFF_OVERFLOW(lock->lk_offset, lock->lk_length))
-		file_lock.fl_end = ~(u64)0;
-	else
-		file_lock.fl_end = lock->lk_offset + lock->lk_length - 1;
+	file_lock.fl_end = last_byte_offset(lock->lk_offset, lock->lk_length);
 	nfs4_transform_lock_offset(&file_lock);
 
 	/*
@@ -2858,10 +2874,7 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	file_lock.fl_lmops = &nfsd_posix_mng_ops;
 
 	file_lock.fl_start = lockt->lt_offset;
-	if ((lockt->lt_length == ~(u64)0) || LOFF_OVERFLOW(lockt->lt_offset, lockt->lt_length))
-		file_lock.fl_end = ~(u64)0;
-	else
-		file_lock.fl_end = lockt->lt_offset + lockt->lt_length - 1;
+	file_lock.fl_end = last_byte_offset(lockt->lt_offset, lockt->lt_length);
 
 	nfs4_transform_lock_offset(&file_lock);
 
@@ -2917,10 +2930,7 @@ nfsd4_locku(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	file_lock.fl_lmops = &nfsd_posix_mng_ops;
 	file_lock.fl_start = locku->lu_offset;
 
-	if ((locku->lu_length == ~(u64)0) || LOFF_OVERFLOW(locku->lu_offset, locku->lu_length))
-		file_lock.fl_end = ~(u64)0;
-	else
-		file_lock.fl_end = locku->lu_offset + locku->lu_length - 1;
+	file_lock.fl_end = last_byte_offset(locku->lu_offset, locku->lu_length);
 	nfs4_transform_lock_offset(&file_lock);
 
 	/*
diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index ea0366769484..b912311a56b1 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -88,6 +88,8 @@
 #define NFS4_ACE_GENERIC_EXECUTE              0x001200A0
 #define NFS4_ACE_MASK_ALL                     0x001F01FF
 
+#define NFS4_MAX_UINT64	(~(u64)0)
+
 enum nfs4_acl_whotype {
 	NFS4_ACL_WHO_NAMED = 0,
 	NFS4_ACL_WHO_OWNER,
-- 
cgit v1.2.3


From db43910cb42285a99f45f7e0a0a32e32d0b61dcf Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Mon, 15 Dec 2008 19:42:24 +0200
Subject: nfsd: get rid of NFSD_VERSION

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 include/linux/nfsd/nfsd.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h
index 21269405ffe2..e19f45991b2e 100644
--- a/include/linux/nfsd/nfsd.h
+++ b/include/linux/nfsd/nfsd.h
@@ -23,7 +23,6 @@
 /*
  * nfsd version
  */
-#define NFSD_VERSION		"0.5"
 #define NFSD_SUPPORTED_MINOR_VERSION	0
 
 /*
-- 
cgit v1.2.3


From 5886188dc7ba9a76babcd37452f44079a9a77f71 Mon Sep 17 00:00:00 2001
From: Benjamin Krill <ben@codiert.org>
Date: Wed, 7 Jan 2009 10:32:38 +0100
Subject: serial: Add driver for the Cell Network Processor serial port NWP
 device

Add support for the nwp serial device which is connected to a DCR bus. It
uses the of_serial device driver to determine necessary properties from
the device tree.  The supported device is added as serial port number 85.

NWP stands for network processor and it is part of the QPACE - Quantum
Chromodynamics Parallel Computing on the Cell Broadband Engine project.
The implementation is a lightweight uart implementation with the focus
to consume as little resources as possible and it is connected to a
DCR bus.

Signed-off-by: Benjamin Krill <ben@codiert.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 drivers/serial/Kconfig      |  19 +-
 drivers/serial/Makefile     |   1 +
 drivers/serial/nwpserial.c  | 475 ++++++++++++++++++++++++++++++++++++++++++++
 drivers/serial/of_serial.c  |  19 ++
 include/linux/nwpserial.h   |  18 ++
 include/linux/serial_core.h |   3 +
 6 files changed, 534 insertions(+), 1 deletion(-)
 create mode 100644 drivers/serial/nwpserial.c
 create mode 100644 include/linux/nwpserial.h

(limited to 'include/linux')

diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
index b695ab3142d8..64fcf8c20821 100644
--- a/drivers/serial/Kconfig
+++ b/drivers/serial/Kconfig
@@ -1320,13 +1320,30 @@ config SERIAL_NETX_CONSOLE
 config SERIAL_OF_PLATFORM
 	tristate "Serial port on Open Firmware platform bus"
 	depends on PPC_OF
-	depends on SERIAL_8250
+	depends on SERIAL_8250 || SERIAL_OF_PLATFORM_NWPSERIAL
 	help
 	  If you have a PowerPC based system that has serial ports
 	  on a platform specific bus, you should enable this option.
 	  Currently, only 8250 compatible ports are supported, but
 	  others can easily be added.
 
+config SERIAL_OF_PLATFORM_NWPSERIAL
+	tristate "NWP serial port driver"
+	depends on PPC_OF && PPC_DCR
+	select SERIAL_OF_PLATFORM
+	select SERIAL_CORE_CONSOLE
+	select SERIAL_CORE
+	help
+	  This driver supports the cell network processor nwp serial
+	  device.
+
+config SERIAL_OF_PLATFORM_NWPSERIAL_CONSOLE
+	bool "Console on NWP serial port"
+	depends on SERIAL_OF_PLATFORM_NWPSERIAL=y
+	select SERIAL_CORE_CONSOLE
+	help
+	  Support for Console on the NWP serial ports.
+
 config SERIAL_QE
 	tristate "Freescale QUICC Engine serial port support"
 	depends on QUICC_ENGINE
diff --git a/drivers/serial/Makefile b/drivers/serial/Makefile
index dfe775ac45b2..8844c0a03929 100644
--- a/drivers/serial/Makefile
+++ b/drivers/serial/Makefile
@@ -72,6 +72,7 @@ obj-$(CONFIG_SERIAL_ATMEL) += atmel_serial.o
 obj-$(CONFIG_SERIAL_UARTLITE) += uartlite.o
 obj-$(CONFIG_SERIAL_NETX) += netx-serial.o
 obj-$(CONFIG_SERIAL_OF_PLATFORM) += of_serial.o
+obj-$(CONFIG_SERIAL_OF_PLATFORM_NWPSERIAL) += nwpserial.o
 obj-$(CONFIG_SERIAL_KS8695) += serial_ks8695.o
 obj-$(CONFIG_KGDB_SERIAL_CONSOLE) += kgdboc.o
 obj-$(CONFIG_SERIAL_QE) += ucc_uart.o
diff --git a/drivers/serial/nwpserial.c b/drivers/serial/nwpserial.c
new file mode 100644
index 000000000000..32f3eaf0d262
--- /dev/null
+++ b/drivers/serial/nwpserial.c
@@ -0,0 +1,475 @@
+/*
+ *  Serial Port driver for a NWP uart device
+ *
+ *    Copyright (C) 2008 IBM Corp., Benjamin Krill <ben@codiert.org>
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License
+ *  as published by the Free Software Foundation; either version
+ *  2 of the License, or (at your option) any later version.
+ *
+ */
+#include <linux/init.h>
+#include <linux/console.h>
+#include <linux/serial.h>
+#include <linux/serial_reg.h>
+#include <linux/serial_core.h>
+#include <linux/tty.h>
+#include <linux/irqreturn.h>
+#include <linux/mutex.h>
+#include <linux/of_platform.h>
+#include <linux/of_device.h>
+#include <linux/nwpserial.h>
+#include <asm/prom.h>
+#include <asm/dcr.h>
+
+#define NWPSERIAL_NR               2
+
+#define NWPSERIAL_STATUS_RXVALID 0x1
+#define NWPSERIAL_STATUS_TXFULL  0x2
+
+struct nwpserial_port {
+	struct uart_port port;
+	dcr_host_t dcr_host;
+	unsigned int ier;
+	unsigned int mcr;
+};
+
+static DEFINE_MUTEX(nwpserial_mutex);
+static struct nwpserial_port nwpserial_ports[NWPSERIAL_NR];
+
+static void wait_for_bits(struct nwpserial_port *up, int bits)
+{
+	unsigned int status, tmout = 10000;
+
+	/* Wait up to 10ms for the character(s) to be sent. */
+	do {
+		status = dcr_read(up->dcr_host, UART_LSR);
+
+		if (--tmout == 0)
+			break;
+		udelay(1);
+	} while ((status & bits) != bits);
+}
+
+#ifdef CONFIG_SERIAL_OF_PLATFORM_NWPSERIAL_CONSOLE
+static void nwpserial_console_putchar(struct uart_port *port, int c)
+{
+	struct nwpserial_port *up;
+	up = container_of(port, struct nwpserial_port, port);
+	/* check if tx buffer is full */
+	wait_for_bits(up, UART_LSR_THRE);
+	dcr_write(up->dcr_host, UART_TX, c);
+	up->port.icount.tx++;
+}
+
+static void
+nwpserial_console_write(struct console *co, const char *s, unsigned int count)
+{
+	struct nwpserial_port *up = &nwpserial_ports[co->index];
+	unsigned long flags;
+	int locked = 1;
+
+	if (oops_in_progress)
+		locked = spin_trylock_irqsave(&up->port.lock, flags);
+	else
+		spin_lock_irqsave(&up->port.lock, flags);
+
+	/* save and disable interrupt */
+	up->ier = dcr_read(up->dcr_host, UART_IER);
+	dcr_write(up->dcr_host, UART_IER, up->ier & ~UART_IER_RDI);
+
+	uart_console_write(&up->port, s, count, nwpserial_console_putchar);
+
+	/* wait for transmitter to become emtpy */
+	while ((dcr_read(up->dcr_host, UART_LSR) & UART_LSR_THRE) == 0)
+		cpu_relax();
+
+	/* restore interrupt state */
+	dcr_write(up->dcr_host, UART_IER, up->ier);
+
+	if (locked)
+		spin_unlock_irqrestore(&up->port.lock, flags);
+}
+
+static struct uart_driver nwpserial_reg;
+static struct console nwpserial_console = {
+	.name		= "ttySQ",
+	.write		= nwpserial_console_write,
+	.device		= uart_console_device,
+	.flags		= CON_PRINTBUFFER,
+	.index		= -1,
+	.data		= &nwpserial_reg,
+};
+#define NWPSERIAL_CONSOLE	(&nwpserial_console)
+#else
+#define NWPSERIAL_CONSOLE	NULL
+#endif /* CONFIG_SERIAL_OF_PLATFORM_NWPSERIAL_CONSOLE */
+
+/**************************************************************************/
+
+static int nwpserial_request_port(struct uart_port *port)
+{
+	return 0;
+}
+
+static void nwpserial_release_port(struct uart_port *port)
+{
+	/* N/A */
+}
+
+static void nwpserial_config_port(struct uart_port *port, int flags)
+{
+	port->type = PORT_NWPSERIAL;
+}
+
+static irqreturn_t nwpserial_interrupt(int irq, void *dev_id)
+{
+	struct nwpserial_port *up = dev_id;
+	struct tty_struct *tty = up->port.info->port.tty;
+	irqreturn_t ret;
+	unsigned int iir;
+	unsigned char ch;
+
+	spin_lock(&up->port.lock);
+
+	/* check if the uart was the interrupt source. */
+	iir = dcr_read(up->dcr_host, UART_IIR);
+	if (!iir) {
+		ret = IRQ_NONE;
+		goto out;
+	}
+
+	do {
+		up->port.icount.rx++;
+		ch = dcr_read(up->dcr_host, UART_RX);
+		if (up->port.ignore_status_mask != NWPSERIAL_STATUS_RXVALID)
+			tty_insert_flip_char(tty, ch, TTY_NORMAL);
+	} while (dcr_read(up->dcr_host, UART_RX) & UART_LSR_DR);
+
+	tty_flip_buffer_push(tty);
+	ret = IRQ_HANDLED;
+
+out:
+	spin_unlock(&up->port.lock);
+	return ret;
+}
+
+static int nwpserial_startup(struct uart_port *port)
+{
+	struct nwpserial_port *up;
+	int err;
+
+	up = container_of(port, struct nwpserial_port, port);
+
+	/* disable flow control by default */
+	up->mcr = dcr_read(up->dcr_host, UART_MCR) & ~UART_MCR_AFE;
+	dcr_write(up->dcr_host, UART_MCR, up->mcr);
+
+	/* register interrupt handler */
+	err = request_irq(up->port.irq, nwpserial_interrupt,
+			IRQF_SHARED, "nwpserial", up);
+	if (err)
+		return err;
+
+	/* enable interrupts */
+	up->ier = UART_IER_RDI;
+	dcr_write(up->dcr_host, UART_IER, up->ier);
+
+	/* enable receiving */
+	up->port.ignore_status_mask &= ~NWPSERIAL_STATUS_RXVALID;
+
+	return 0;
+}
+
+static void nwpserial_shutdown(struct uart_port *port)
+{
+	struct nwpserial_port *up;
+	up = container_of(port, struct nwpserial_port, port);
+
+	/* disable receiving */
+	up->port.ignore_status_mask |= NWPSERIAL_STATUS_RXVALID;
+
+	/* disable interrupts from this port */
+	up->ier = 0;
+	dcr_write(up->dcr_host, UART_IER, up->ier);
+
+	/* free irq */
+	free_irq(up->port.irq, port);
+}
+
+static int nwpserial_verify_port(struct uart_port *port,
+			struct serial_struct *ser)
+{
+	return -EINVAL;
+}
+
+static const char *nwpserial_type(struct uart_port *port)
+{
+	return port->type == PORT_NWPSERIAL ? "nwpserial" : NULL;
+}
+
+static void nwpserial_set_termios(struct uart_port *port,
+			struct ktermios *termios, struct ktermios *old)
+{
+	struct nwpserial_port *up;
+	up = container_of(port, struct nwpserial_port, port);
+
+	up->port.read_status_mask = NWPSERIAL_STATUS_RXVALID
+				| NWPSERIAL_STATUS_TXFULL;
+
+	up->port.ignore_status_mask = 0;
+	/* ignore all characters if CREAD is not set */
+	if ((termios->c_cflag & CREAD) == 0)
+		up->port.ignore_status_mask |= NWPSERIAL_STATUS_RXVALID;
+
+	/* Copy back the old hardware settings */
+	if (old)
+		tty_termios_copy_hw(termios, old);
+}
+
+static void nwpserial_break_ctl(struct uart_port *port, int ctl)
+{
+	/* N/A */
+}
+
+static void nwpserial_enable_ms(struct uart_port *port)
+{
+	/* N/A */
+}
+
+static void nwpserial_stop_rx(struct uart_port *port)
+{
+	struct nwpserial_port *up;
+	up = container_of(port, struct nwpserial_port, port);
+	/* don't forward any more data (like !CREAD) */
+	up->port.ignore_status_mask = NWPSERIAL_STATUS_RXVALID;
+}
+
+static void nwpserial_putchar(struct nwpserial_port *up, unsigned char c)
+{
+	/* check if tx buffer is full */
+	wait_for_bits(up, UART_LSR_THRE);
+	dcr_write(up->dcr_host, UART_TX, c);
+	up->port.icount.tx++;
+}
+
+static void nwpserial_start_tx(struct uart_port *port)
+{
+	struct nwpserial_port *up;
+	struct circ_buf *xmit;
+	up = container_of(port, struct nwpserial_port, port);
+	xmit  = &up->port.info->xmit;
+
+	if (port->x_char) {
+		nwpserial_putchar(up, up->port.x_char);
+		port->x_char = 0;
+	}
+
+	while (!(uart_circ_empty(xmit) || uart_tx_stopped(&up->port))) {
+		nwpserial_putchar(up, xmit->buf[xmit->tail]);
+		xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE-1);
+	}
+}
+
+static unsigned int nwpserial_get_mctrl(struct uart_port *port)
+{
+	return 0;
+}
+
+static void nwpserial_set_mctrl(struct uart_port *port, unsigned int mctrl)
+{
+	/* N/A */
+}
+
+static void nwpserial_stop_tx(struct uart_port *port)
+{
+	/* N/A */
+}
+
+static unsigned int nwpserial_tx_empty(struct uart_port *port)
+{
+	struct nwpserial_port *up;
+	unsigned long flags;
+	int ret;
+	up = container_of(port, struct nwpserial_port, port);
+
+	spin_lock_irqsave(&up->port.lock, flags);
+	ret = dcr_read(up->dcr_host, UART_LSR);
+	spin_unlock_irqrestore(&up->port.lock, flags);
+
+	return ret & UART_LSR_TEMT ? TIOCSER_TEMT : 0;
+}
+
+static struct uart_ops nwpserial_pops = {
+	.tx_empty     = nwpserial_tx_empty,
+	.set_mctrl    = nwpserial_set_mctrl,
+	.get_mctrl    = nwpserial_get_mctrl,
+	.stop_tx      = nwpserial_stop_tx,
+	.start_tx     = nwpserial_start_tx,
+	.stop_rx      = nwpserial_stop_rx,
+	.enable_ms    = nwpserial_enable_ms,
+	.break_ctl    = nwpserial_break_ctl,
+	.startup      = nwpserial_startup,
+	.shutdown     = nwpserial_shutdown,
+	.set_termios  = nwpserial_set_termios,
+	.type         = nwpserial_type,
+	.release_port = nwpserial_release_port,
+	.request_port = nwpserial_request_port,
+	.config_port  = nwpserial_config_port,
+	.verify_port  = nwpserial_verify_port,
+};
+
+static struct uart_driver nwpserial_reg = {
+	.owner       = THIS_MODULE,
+	.driver_name = "nwpserial",
+	.dev_name    = "ttySQ",
+	.major       = TTY_MAJOR,
+	.minor       = 68,
+	.nr          = NWPSERIAL_NR,
+	.cons        = NWPSERIAL_CONSOLE,
+};
+
+int nwpserial_register_port(struct uart_port *port)
+{
+	struct nwpserial_port *up = NULL;
+	int ret = -1;
+	int i;
+	static int first = 1;
+	int dcr_len;
+	int dcr_base;
+	struct device_node *dn;
+
+	mutex_lock(&nwpserial_mutex);
+
+	dn = to_of_device(port->dev)->node;
+	if (dn == NULL)
+		goto out;
+
+	/* get dcr base. */
+	dcr_base = dcr_resource_start(dn, 0);
+
+	/* find matching entry */
+	for (i = 0; i < NWPSERIAL_NR; i++)
+		if (nwpserial_ports[i].port.iobase == dcr_base) {
+			up = &nwpserial_ports[i];
+			break;
+		}
+
+	/* we didn't find a mtching entry, search for a free port */
+	if (up == NULL)
+		for (i = 0; i < NWPSERIAL_NR; i++)
+			if (nwpserial_ports[i].port.type == PORT_UNKNOWN &&
+				nwpserial_ports[i].port.iobase == 0) {
+				up = &nwpserial_ports[i];
+				break;
+			}
+
+	if (up == NULL) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	if (first)
+		uart_register_driver(&nwpserial_reg);
+	first = 0;
+
+	up->port.membase      = port->membase;
+	up->port.irq          = port->irq;
+	up->port.uartclk      = port->uartclk;
+	up->port.fifosize     = port->fifosize;
+	up->port.regshift     = port->regshift;
+	up->port.iotype       = port->iotype;
+	up->port.flags        = port->flags;
+	up->port.mapbase      = port->mapbase;
+	up->port.private_data = port->private_data;
+
+	if (port->dev)
+		up->port.dev = port->dev;
+
+	if (up->port.iobase != dcr_base) {
+		up->port.ops          = &nwpserial_pops;
+		up->port.fifosize     = 16;
+
+		spin_lock_init(&up->port.lock);
+
+		up->port.iobase = dcr_base;
+		dcr_len = dcr_resource_len(dn, 0);
+
+		up->dcr_host = dcr_map(dn, dcr_base, dcr_len);
+		if (!DCR_MAP_OK(up->dcr_host)) {
+			printk(KERN_ERR "Cannot map DCR resources for NWPSERIAL");
+			goto out;
+		}
+	}
+
+	ret = uart_add_one_port(&nwpserial_reg, &up->port);
+	if (ret == 0)
+		ret = up->port.line;
+
+out:
+	mutex_unlock(&nwpserial_mutex);
+
+	return ret;
+}
+EXPORT_SYMBOL(nwpserial_register_port);
+
+void nwpserial_unregister_port(int line)
+{
+	struct nwpserial_port *up = &nwpserial_ports[line];
+	mutex_lock(&nwpserial_mutex);
+	uart_remove_one_port(&nwpserial_reg, &up->port);
+
+	up->port.type = PORT_UNKNOWN;
+
+	mutex_unlock(&nwpserial_mutex);
+}
+EXPORT_SYMBOL(nwpserial_unregister_port);
+
+#ifdef CONFIG_SERIAL_OF_PLATFORM_NWPSERIAL_CONSOLE
+static int __init nwpserial_console_init(void)
+{
+	struct nwpserial_port *up = NULL;
+	struct device_node *dn;
+	const char *name;
+	int dcr_base;
+	int dcr_len;
+	int i;
+
+	/* search for a free port */
+	for (i = 0; i < NWPSERIAL_NR; i++)
+		if (nwpserial_ports[i].port.type == PORT_UNKNOWN) {
+			up = &nwpserial_ports[i];
+			break;
+		}
+
+	if (up == NULL)
+		return -1;
+
+	name = of_get_property(of_chosen, "linux,stdout-path", NULL);
+	if (name == NULL)
+		return -1;
+
+	dn = of_find_node_by_path(name);
+	if (!dn)
+		return -1;
+
+	spin_lock_init(&up->port.lock);
+	up->port.ops = &nwpserial_pops;
+	up->port.type = PORT_NWPSERIAL;
+	up->port.fifosize = 16;
+
+	dcr_base = dcr_resource_start(dn, 0);
+	dcr_len = dcr_resource_len(dn, 0);
+	up->port.iobase = dcr_base;
+
+	up->dcr_host = dcr_map(dn, dcr_base, dcr_len);
+	if (!DCR_MAP_OK(up->dcr_host)) {
+		printk("Cannot map DCR resources for SERIAL");
+		return -1;
+	}
+	register_console(&nwpserial_console);
+	return 0;
+}
+console_initcall(nwpserial_console_init);
+#endif /* CONFIG_SERIAL_OF_PLATFORM_NWPSERIAL_CONSOLE */
diff --git a/drivers/serial/of_serial.c b/drivers/serial/of_serial.c
index 8fa0ff561e9f..a821e3a3d664 100644
--- a/drivers/serial/of_serial.c
+++ b/drivers/serial/of_serial.c
@@ -14,6 +14,7 @@
 #include <linux/serial_core.h>
 #include <linux/serial_8250.h>
 #include <linux/of_platform.h>
+#include <linux/nwpserial.h>
 
 #include <asm/prom.h>
 
@@ -99,9 +100,16 @@ static int __devinit of_platform_serial_probe(struct of_device *ofdev,
 		goto out;
 
 	switch (port_type) {
+#ifdef CONFIG_SERIAL_8250
 	case PORT_8250 ... PORT_MAX_8250:
 		ret = serial8250_register_port(&port);
 		break;
+#endif
+#ifdef CONFIG_SERIAL_OF_PLATFORM_NWPSERIAL
+	case PORT_NWPSERIAL:
+		ret = nwpserial_register_port(&port);
+		break;
+#endif
 	default:
 		/* need to add code for these */
 	case PORT_UNKNOWN:
@@ -129,9 +137,16 @@ static int of_platform_serial_remove(struct of_device *ofdev)
 {
 	struct of_serial_info *info = ofdev->dev.driver_data;
 	switch (info->type) {
+#ifdef CONFIG_SERIAL_8250
 	case PORT_8250 ... PORT_MAX_8250:
 		serial8250_unregister_port(info->line);
 		break;
+#endif
+#ifdef CONFIG_SERIAL_OF_PLATFORM_NWPSERIAL
+	case PORT_NWPSERIAL:
+		nwpserial_unregister_port(info->line);
+		break;
+#endif
 	default:
 		/* need to add code for these */
 		break;
@@ -148,6 +163,10 @@ static struct of_device_id __devinitdata of_platform_serial_table[] = {
 	{ .type = "serial", .compatible = "ns16450",  .data = (void *)PORT_16450, },
 	{ .type = "serial", .compatible = "ns16550",  .data = (void *)PORT_16550, },
 	{ .type = "serial", .compatible = "ns16750",  .data = (void *)PORT_16750, },
+#ifdef CONFIG_SERIAL_OF_PLATFORM_NWPSERIAL
+	{ .type = "serial", .compatible = "ibm,qpace-nwp-serial",
+					.data = (void *)PORT_NWPSERIAL, },
+#endif
 	{ .type = "serial",			      .data = (void *)PORT_UNKNOWN, },
 	{ /* end of list */ },
 };
diff --git a/include/linux/nwpserial.h b/include/linux/nwpserial.h
new file mode 100644
index 000000000000..9acb21572eaf
--- /dev/null
+++ b/include/linux/nwpserial.h
@@ -0,0 +1,18 @@
+/*
+ *  Serial Port driver for a NWP uart device
+ *
+ *    Copyright (C) 2008 IBM Corp., Benjamin Krill <ben@codiert.org>
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License
+ *  as published by the Free Software Foundation; either version
+ *  2 of the License, or (at your option) any later version.
+ *
+ */
+#ifndef _NWPSERIAL_H
+#define _NWPSERIAL_H
+
+int nwpserial_register_port(struct uart_port *port);
+void nwpserial_unregister_port(int line);
+
+#endif /* _NWPSERIAL_H */
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index b4199841f1fc..90bbbf0b1161 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -161,6 +161,9 @@
 
 #define PORT_S3C6400	84
 
+/* NWPSERIAL */
+#define PORT_NWPSERIAL	85
+
 #ifdef __KERNEL__
 
 #include <linux/compiler.h>
-- 
cgit v1.2.3


From 8feae13110d60cc6287afabc2887366b0eb226c2 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Thu, 8 Jan 2009 12:04:47 +0000
Subject: NOMMU: Make VMAs per MM as for MMU-mode linux

Make VMAs per mm_struct as for MMU-mode linux.  This solves two problems:

 (1) In SYSV SHM where nattch for a segment does not reflect the number of
     shmat's (and forks) done.

 (2) In mmap() where the VMA's vm_mm is set to point to the parent mm by an
     exec'ing process when VM_EXECUTABLE is specified, regardless of the fact
     that a VMA might be shared and already have its vm_mm assigned to another
     process or a dead process.

A new struct (vm_region) is introduced to track a mapped region and to remember
the circumstances under which it may be shared and the vm_list_struct structure
is discarded as it's no longer required.

This patch makes the following additional changes:

 (1) Regions are now allocated with alloc_pages() rather than kmalloc() and
     with no recourse to __GFP_COMP, so the pages are not composite.  Instead,
     each page has a reference on it held by the region.  Anything else that is
     interested in such a page will have to get a reference on it to retain it.
     When the pages are released due to unmapping, each page is passed to
     put_page() and will be freed when the page usage count reaches zero.

 (2) Excess pages are trimmed after an allocation as the allocation must be
     made as a power-of-2 quantity of pages.

 (3) VMAs are added to the parent MM's R/B tree and mmap lists.  As an MM may
     end up with overlapping VMAs within the tree, the VMA struct address is
     appended to the sort key.

 (4) Non-anonymous VMAs are now added to the backing inode's prio list.

 (5) Holes may be punched in anonymous VMAs with munmap(), releasing parts of
     the backing region.  The VMA and region structs will be split if
     necessary.

 (6) sys_shmdt() only releases one attachment to a SYSV IPC shared memory
     segment instead of all the attachments at that addresss.  Multiple
     shmat()'s return the same address under NOMMU-mode instead of different
     virtual addresses as under MMU-mode.

 (7) Core dumping for ELF-FDPIC requires fewer exceptions for NOMMU-mode.

 (8) /proc/maps is now the global list of mapped regions, and may list bits
     that aren't actually mapped anywhere.

 (9) /proc/meminfo gains a line (tagged "MmapCopy") that indicates the amount
     of RAM currently allocated by mmap to hold mappable regions that can't be
     mapped directly.  These are copies of the backing device or file if not
     anonymous.

These changes make NOMMU mode more similar to MMU mode.  The downside is that
NOMMU mode requires some extra memory to track things over NOMMU without this
patch (VMAs are no longer shared, and there are now region structs).

Signed-off-by: David Howells <dhowells@redhat.com>
Tested-by: Mike Frysinger <vapier.adi@gmail.com>
Acked-by: Paul Mundt <lethal@linux-sh.org>
---
 Documentation/nommu-mmap.txt     |  18 +-
 arch/arm/include/asm/mmu.h       |   1 -
 arch/blackfin/include/asm/mmu.h  |   1 -
 arch/blackfin/kernel/ptrace.c    |   6 +-
 arch/blackfin/kernel/traps.c     |  11 +-
 arch/frv/kernel/ptrace.c         |  11 +-
 arch/h8300/include/asm/mmu.h     |   1 -
 arch/m68knommu/include/asm/mmu.h |   1 -
 arch/sh/include/asm/mmu.h        |   1 -
 fs/binfmt_elf_fdpic.c            |  27 +-
 fs/proc/internal.h               |   2 -
 fs/proc/meminfo.c                |   6 +
 fs/proc/nommu.c                  |  71 ++-
 fs/proc/task_nommu.c             | 108 +++--
 include/asm-frv/mmu.h            |   1 -
 include/asm-m32r/mmu.h           |   1 -
 include/linux/mm.h               |  18 +-
 include/linux/mm_types.h         |  18 +-
 ipc/shm.c                        |  12 +
 kernel/fork.c                    |   4 +-
 lib/Kconfig.debug                |   7 +
 mm/mmap.c                        |  10 +
 mm/nommu.c                       | 960 +++++++++++++++++++++++++++------------
 23 files changed, 860 insertions(+), 436 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/nommu-mmap.txt b/Documentation/nommu-mmap.txt
index 7714f57caad5..02b89dcf38ac 100644
--- a/Documentation/nommu-mmap.txt
+++ b/Documentation/nommu-mmap.txt
@@ -109,12 +109,18 @@ and it's also much more restricted in the latter case:
 FURTHER NOTES ON NO-MMU MMAP
 ============================
 
- (*) A request for a private mapping of less than a page in size may not return
-     a page-aligned buffer. This is because the kernel calls kmalloc() to
-     allocate the buffer, not get_free_page().
-
- (*) A list of all the mappings on the system is visible through /proc/maps in
-     no-MMU mode.
+ (*) A request for a private mapping of a file may return a buffer that is not
+     page-aligned.  This is because XIP may take place, and the data may not be
+     paged aligned in the backing store.
+
+ (*) A request for an anonymous mapping will always be page aligned.  If
+     possible the size of the request should be a power of two otherwise some
+     of the space may be wasted as the kernel must allocate a power-of-2
+     granule but will only discard the excess if appropriately configured as
+     this has an effect on fragmentation.
+
+ (*) A list of all the private copy and anonymous mappings on the system is
+     visible through /proc/maps in no-MMU mode.
 
  (*) A list of all the mappings in use by a process is visible through
      /proc/<pid>/maps in no-MMU mode.
diff --git a/arch/arm/include/asm/mmu.h b/arch/arm/include/asm/mmu.h
index 53099d4ee421..b561584d04a1 100644
--- a/arch/arm/include/asm/mmu.h
+++ b/arch/arm/include/asm/mmu.h
@@ -24,7 +24,6 @@ typedef struct {
  *  modified for 2.6 by Hyok S. Choi <hyok.choi@samsung.com>
  */
 typedef struct {
-	struct vm_list_struct	*vmlist;
 	unsigned long		end_brk;
 } mm_context_t;
 
diff --git a/arch/blackfin/include/asm/mmu.h b/arch/blackfin/include/asm/mmu.h
index 757e43906ed4..dbfd686360e6 100644
--- a/arch/blackfin/include/asm/mmu.h
+++ b/arch/blackfin/include/asm/mmu.h
@@ -10,7 +10,6 @@ struct sram_list_struct {
 };
 
 typedef struct {
-	struct vm_list_struct *vmlist;
 	unsigned long end_brk;
 	unsigned long stack_start;
 
diff --git a/arch/blackfin/kernel/ptrace.c b/arch/blackfin/kernel/ptrace.c
index d2d388536630..594e325b40e4 100644
--- a/arch/blackfin/kernel/ptrace.c
+++ b/arch/blackfin/kernel/ptrace.c
@@ -160,15 +160,15 @@ put_reg(struct task_struct *task, int regno, unsigned long data)
 static inline int is_user_addr_valid(struct task_struct *child,
 				     unsigned long start, unsigned long len)
 {
-	struct vm_list_struct *vml;
+	struct vm_area_struct *vma;
 	struct sram_list_struct *sraml;
 
 	/* overflow */
 	if (start + len < start)
 		return -EIO;
 
-	for (vml = child->mm->context.vmlist; vml; vml = vml->next)
-		if (start >= vml->vma->vm_start && start + len < vml->vma->vm_end)
+	vma = find_vma(child->mm, start);
+	if (vma && start >= vma->vm_start && start + len <= vma->vm_end)
 			return 0;
 
 	for (sraml = child->mm->context.sram_list; sraml; sraml = sraml->next)
diff --git a/arch/blackfin/kernel/traps.c b/arch/blackfin/kernel/traps.c
index 17d8e4172896..5b0667da8d05 100644
--- a/arch/blackfin/kernel/traps.c
+++ b/arch/blackfin/kernel/traps.c
@@ -32,6 +32,7 @@
 #include <linux/module.h>
 #include <linux/kallsyms.h>
 #include <linux/fs.h>
+#include <linux/rbtree.h>
 #include <asm/traps.h>
 #include <asm/cacheflush.h>
 #include <asm/cplb.h>
@@ -83,6 +84,7 @@ static void decode_address(char *buf, unsigned long address)
 	struct mm_struct *mm;
 	unsigned long flags, offset;
 	unsigned char in_atomic = (bfin_read_IPEND() & 0x10) || in_atomic();
+	struct rb_node *n;
 
 #ifdef CONFIG_KALLSYMS
 	unsigned long symsize;
@@ -128,9 +130,10 @@ static void decode_address(char *buf, unsigned long address)
 		if (!mm)
 			continue;
 
-		vml = mm->context.vmlist;
-		while (vml) {
-			struct vm_area_struct *vma = vml->vma;
+		for (n = rb_first(&mm->mm_rb); n; n = rb_next(n)) {
+			struct vm_area_struct *vma;
+
+			vma = rb_entry(n, struct vm_area_struct, vm_rb);
 
 			if (address >= vma->vm_start && address < vma->vm_end) {
 				char _tmpbuf[256];
@@ -176,8 +179,6 @@ static void decode_address(char *buf, unsigned long address)
 
 				goto done;
 			}
-
-			vml = vml->next;
 		}
 		if (!in_atomic)
 			mmput(mm);
diff --git a/arch/frv/kernel/ptrace.c b/arch/frv/kernel/ptrace.c
index 709e9bdc6126..5e7d401d21e7 100644
--- a/arch/frv/kernel/ptrace.c
+++ b/arch/frv/kernel/ptrace.c
@@ -69,7 +69,8 @@ static inline int put_reg(struct task_struct *task, int regno,
 }
 
 /*
- * check that an address falls within the bounds of the target process's memory mappings
+ * check that an address falls within the bounds of the target process's memory
+ * mappings
  */
 static inline int is_user_addr_valid(struct task_struct *child,
 				     unsigned long start, unsigned long len)
@@ -79,11 +80,11 @@ static inline int is_user_addr_valid(struct task_struct *child,
 		return -EIO;
 	return 0;
 #else
-	struct vm_list_struct *vml;
+	struct vm_area_struct *vma;
 
-	for (vml = child->mm->context.vmlist; vml; vml = vml->next)
-		if (start >= vml->vma->vm_start && start + len <= vml->vma->vm_end)
-			return 0;
+	vma = find_vma(child->mm, start);
+	if (vma && start >= vma->vm_start && start + len <= vma->vm_end)
+		return 0;
 
 	return -EIO;
 #endif
diff --git a/arch/h8300/include/asm/mmu.h b/arch/h8300/include/asm/mmu.h
index 2ce06ea46104..31309969df70 100644
--- a/arch/h8300/include/asm/mmu.h
+++ b/arch/h8300/include/asm/mmu.h
@@ -4,7 +4,6 @@
 /* Copyright (C) 2002, David McCullough <davidm@snapgear.com> */
 
 typedef struct {
-	struct vm_list_struct	*vmlist;
 	unsigned long		end_brk;
 } mm_context_t;
 
diff --git a/arch/m68knommu/include/asm/mmu.h b/arch/m68knommu/include/asm/mmu.h
index 5fa6b68353ba..e2da1e6f09fe 100644
--- a/arch/m68knommu/include/asm/mmu.h
+++ b/arch/m68knommu/include/asm/mmu.h
@@ -4,7 +4,6 @@
 /* Copyright (C) 2002, David McCullough <davidm@snapgear.com> */
 
 typedef struct {
-	struct vm_list_struct	*vmlist;
 	unsigned long		end_brk;
 } mm_context_t;
 
diff --git a/arch/sh/include/asm/mmu.h b/arch/sh/include/asm/mmu.h
index fdcb93bc6d11..6c43625bb1a5 100644
--- a/arch/sh/include/asm/mmu.h
+++ b/arch/sh/include/asm/mmu.h
@@ -9,7 +9,6 @@ typedef struct {
 	mm_context_id_t		id;
 	void			*vdso;
 #else
-	struct vm_list_struct	*vmlist;
 	unsigned long		end_brk;
 #endif
 #ifdef CONFIG_BINFMT_ELF_FDPIC
diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c
index aa5b43205e37..22baf1b13493 100644
--- a/fs/binfmt_elf_fdpic.c
+++ b/fs/binfmt_elf_fdpic.c
@@ -1567,11 +1567,9 @@ end_coredump:
 static int elf_fdpic_dump_segments(struct file *file, size_t *size,
 			   unsigned long *limit, unsigned long mm_flags)
 {
-	struct vm_list_struct *vml;
-
-	for (vml = current->mm->context.vmlist; vml; vml = vml->next) {
-	struct vm_area_struct *vma = vml->vma;
+	struct vm_area_struct *vma;
 
+	for (vma = current->mm->mmap; vma; vma = vma->vm_next) {
 		if (!maydump(vma, mm_flags))
 			continue;
 
@@ -1617,9 +1615,6 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs,
 	elf_fpxregset_t *xfpu = NULL;
 #endif
 	int thread_status_size = 0;
-#ifndef CONFIG_MMU
-	struct vm_list_struct *vml;
-#endif
 	elf_addr_t *auxv;
 	unsigned long mm_flags;
 
@@ -1685,13 +1680,7 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs,
 	fill_prstatus(prstatus, current, signr);
 	elf_core_copy_regs(&prstatus->pr_reg, regs);
 
-#ifdef CONFIG_MMU
 	segs = current->mm->map_count;
-#else
-	segs = 0;
-	for (vml = current->mm->context.vmlist; vml; vml = vml->next)
-	    segs++;
-#endif
 #ifdef ELF_CORE_EXTRA_PHDRS
 	segs += ELF_CORE_EXTRA_PHDRS;
 #endif
@@ -1766,20 +1755,10 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs,
 	mm_flags = current->mm->flags;
 
 	/* write program headers for segments dump */
-	for (
-#ifdef CONFIG_MMU
-		vma = current->mm->mmap; vma; vma = vma->vm_next
-#else
-			vml = current->mm->context.vmlist; vml; vml = vml->next
-#endif
-	     ) {
+	for (vma = current->mm->mmap; vma; vma = vma->vm_next) {
 		struct elf_phdr phdr;
 		size_t sz;
 
-#ifndef CONFIG_MMU
-		vma = vml->vma;
-#endif
-
 		sz = vma->vm_end - vma->vm_start;
 
 		phdr.p_type = PT_LOAD;
diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index 3e8aeb8b61ce..cd53ff838498 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -41,8 +41,6 @@ do {						\
 	(vmi)->used = 0;			\
 	(vmi)->largest_chunk = 0;		\
 } while(0)
-
-extern int nommu_vma_show(struct seq_file *, struct vm_area_struct *);
 #endif
 
 extern int proc_tid_stat(struct seq_file *m, struct pid_namespace *ns,
diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
index b1675c4e66da..43d23948384a 100644
--- a/fs/proc/meminfo.c
+++ b/fs/proc/meminfo.c
@@ -73,6 +73,9 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
 		"HighFree:       %8lu kB\n"
 		"LowTotal:       %8lu kB\n"
 		"LowFree:        %8lu kB\n"
+#endif
+#ifndef CONFIG_MMU
+		"MmapCopy:       %8lu kB\n"
 #endif
 		"SwapTotal:      %8lu kB\n"
 		"SwapFree:       %8lu kB\n"
@@ -115,6 +118,9 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
 		K(i.freehigh),
 		K(i.totalram-i.totalhigh),
 		K(i.freeram-i.freehigh),
+#endif
+#ifndef CONFIG_MMU
+		K((unsigned long) atomic_read(&mmap_pages_allocated)),
 #endif
 		K(i.totalswap),
 		K(i.freeswap),
diff --git a/fs/proc/nommu.c b/fs/proc/nommu.c
index 3f87d2632947..b446d7ad0b0d 100644
--- a/fs/proc/nommu.c
+++ b/fs/proc/nommu.c
@@ -33,33 +33,33 @@
 #include "internal.h"
 
 /*
- * display a single VMA to a sequenced file
+ * display a single region to a sequenced file
  */
-int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma)
+static int nommu_region_show(struct seq_file *m, struct vm_region *region)
 {
 	unsigned long ino = 0;
 	struct file *file;
 	dev_t dev = 0;
 	int flags, len;
 
-	flags = vma->vm_flags;
-	file = vma->vm_file;
+	flags = region->vm_flags;
+	file = region->vm_file;
 
 	if (file) {
-		struct inode *inode = vma->vm_file->f_path.dentry->d_inode;
+		struct inode *inode = region->vm_file->f_path.dentry->d_inode;
 		dev = inode->i_sb->s_dev;
 		ino = inode->i_ino;
 	}
 
 	seq_printf(m,
 		   "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n",
-		   vma->vm_start,
-		   vma->vm_end,
+		   region->vm_start,
+		   region->vm_end,
 		   flags & VM_READ ? 'r' : '-',
 		   flags & VM_WRITE ? 'w' : '-',
 		   flags & VM_EXEC ? 'x' : '-',
 		   flags & VM_MAYSHARE ? flags & VM_SHARED ? 'S' : 's' : 'p',
-		   ((loff_t)vma->vm_pgoff) << PAGE_SHIFT,
+		   ((loff_t)region->vm_pgoff) << PAGE_SHIFT,
 		   MAJOR(dev), MINOR(dev), ino, &len);
 
 	if (file) {
@@ -75,61 +75,54 @@ int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma)
 }
 
 /*
- * display a list of all the VMAs the kernel knows about
+ * display a list of all the REGIONs the kernel knows about
  * - nommu kernals have a single flat list
  */
-static int nommu_vma_list_show(struct seq_file *m, void *v)
+static int nommu_region_list_show(struct seq_file *m, void *_p)
 {
-	struct vm_area_struct *vma;
+	struct rb_node *p = _p;
 
-	vma = rb_entry((struct rb_node *) v, struct vm_area_struct, vm_rb);
-	return nommu_vma_show(m, vma);
+	return nommu_region_show(m, rb_entry(p, struct vm_region, vm_rb));
 }
 
-static void *nommu_vma_list_start(struct seq_file *m, loff_t *_pos)
+static void *nommu_region_list_start(struct seq_file *m, loff_t *_pos)
 {
-	struct rb_node *_rb;
+	struct rb_node *p;
 	loff_t pos = *_pos;
-	void *next = NULL;
 
-	down_read(&nommu_vma_sem);
+	down_read(&nommu_region_sem);
 
-	for (_rb = rb_first(&nommu_vma_tree); _rb; _rb = rb_next(_rb)) {
-		if (pos == 0) {
-			next = _rb;
-			break;
-		}
-		pos--;
-	}
-
-	return next;
+	for (p = rb_first(&nommu_region_tree); p; p = rb_next(p))
+		if (pos-- == 0)
+			return p;
+	return NULL;
 }
 
-static void nommu_vma_list_stop(struct seq_file *m, void *v)
+static void nommu_region_list_stop(struct seq_file *m, void *v)
 {
-	up_read(&nommu_vma_sem);
+	up_read(&nommu_region_sem);
 }
 
-static void *nommu_vma_list_next(struct seq_file *m, void *v, loff_t *pos)
+static void *nommu_region_list_next(struct seq_file *m, void *v, loff_t *pos)
 {
 	(*pos)++;
 	return rb_next((struct rb_node *) v);
 }
 
-static const struct seq_operations proc_nommu_vma_list_seqop = {
-	.start	= nommu_vma_list_start,
-	.next	= nommu_vma_list_next,
-	.stop	= nommu_vma_list_stop,
-	.show	= nommu_vma_list_show
+static struct seq_operations proc_nommu_region_list_seqop = {
+	.start	= nommu_region_list_start,
+	.next	= nommu_region_list_next,
+	.stop	= nommu_region_list_stop,
+	.show	= nommu_region_list_show
 };
 
-static int proc_nommu_vma_list_open(struct inode *inode, struct file *file)
+static int proc_nommu_region_list_open(struct inode *inode, struct file *file)
 {
-	return seq_open(file, &proc_nommu_vma_list_seqop);
+	return seq_open(file, &proc_nommu_region_list_seqop);
 }
 
-static const struct file_operations proc_nommu_vma_list_operations = {
-	.open    = proc_nommu_vma_list_open,
+static const struct file_operations proc_nommu_region_list_operations = {
+	.open    = proc_nommu_region_list_open,
 	.read    = seq_read,
 	.llseek  = seq_lseek,
 	.release = seq_release,
@@ -137,7 +130,7 @@ static const struct file_operations proc_nommu_vma_list_operations = {
 
 static int __init proc_nommu_init(void)
 {
-	proc_create("maps", S_IRUGO, NULL, &proc_nommu_vma_list_operations);
+	proc_create("maps", S_IRUGO, NULL, &proc_nommu_region_list_operations);
 	return 0;
 }
 
diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c
index d4a8be32b902..ca4a48d0d311 100644
--- a/fs/proc/task_nommu.c
+++ b/fs/proc/task_nommu.c
@@ -15,25 +15,25 @@
  */
 void task_mem(struct seq_file *m, struct mm_struct *mm)
 {
-	struct vm_list_struct *vml;
+	struct vm_area_struct *vma;
+	struct rb_node *p;
 	unsigned long bytes = 0, sbytes = 0, slack = 0;
         
 	down_read(&mm->mmap_sem);
-	for (vml = mm->context.vmlist; vml; vml = vml->next) {
-		if (!vml->vma)
-			continue;
+	for (p = rb_first(&mm->mm_rb); p; p = rb_next(p)) {
+		vma = rb_entry(p, struct vm_area_struct, vm_rb);
 
-		bytes += kobjsize(vml);
+		bytes += kobjsize(vma);
 		if (atomic_read(&mm->mm_count) > 1 ||
-		    atomic_read(&vml->vma->vm_usage) > 1
-		    ) {
-			sbytes += kobjsize((void *) vml->vma->vm_start);
-			sbytes += kobjsize(vml->vma);
+		    vma->vm_region ||
+		    vma->vm_flags & VM_MAYSHARE) {
+			sbytes += kobjsize((void *) vma->vm_start);
+			if (vma->vm_region)
+				sbytes += kobjsize(vma->vm_region);
 		} else {
-			bytes += kobjsize((void *) vml->vma->vm_start);
-			bytes += kobjsize(vml->vma);
-			slack += kobjsize((void *) vml->vma->vm_start) -
-				(vml->vma->vm_end - vml->vma->vm_start);
+			bytes += kobjsize((void *) vma->vm_start);
+			slack += kobjsize((void *) vma->vm_start) -
+				(vma->vm_end - vma->vm_start);
 		}
 	}
 
@@ -70,13 +70,14 @@ void task_mem(struct seq_file *m, struct mm_struct *mm)
 
 unsigned long task_vsize(struct mm_struct *mm)
 {
-	struct vm_list_struct *tbp;
+	struct vm_area_struct *vma;
+	struct rb_node *p;
 	unsigned long vsize = 0;
 
 	down_read(&mm->mmap_sem);
-	for (tbp = mm->context.vmlist; tbp; tbp = tbp->next) {
-		if (tbp->vma)
-			vsize += kobjsize((void *) tbp->vma->vm_start);
+	for (p = rb_first(&mm->mm_rb); p; p = rb_next(p)) {
+		vma = rb_entry(p, struct vm_area_struct, vm_rb);
+		vsize += vma->vm_region->vm_end - vma->vm_region->vm_start;
 	}
 	up_read(&mm->mmap_sem);
 	return vsize;
@@ -85,16 +86,15 @@ unsigned long task_vsize(struct mm_struct *mm)
 int task_statm(struct mm_struct *mm, int *shared, int *text,
 	       int *data, int *resident)
 {
-	struct vm_list_struct *tbp;
+	struct vm_area_struct *vma;
+	struct rb_node *p;
 	int size = kobjsize(mm);
 
 	down_read(&mm->mmap_sem);
-	for (tbp = mm->context.vmlist; tbp; tbp = tbp->next) {
-		size += kobjsize(tbp);
-		if (tbp->vma) {
-			size += kobjsize(tbp->vma);
-			size += kobjsize((void *) tbp->vma->vm_start);
-		}
+	for (p = rb_first(&mm->mm_rb); p; p = rb_next(p)) {
+		vma = rb_entry(p, struct vm_area_struct, vm_rb);
+		size += kobjsize(vma);
+		size += kobjsize((void *) vma->vm_start);
 	}
 
 	size += (*text = mm->end_code - mm->start_code);
@@ -104,21 +104,63 @@ int task_statm(struct mm_struct *mm, int *shared, int *text,
 	return size;
 }
 
+/*
+ * display a single VMA to a sequenced file
+ */
+static int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma)
+{
+	unsigned long ino = 0;
+	struct file *file;
+	dev_t dev = 0;
+	int flags, len;
+
+	flags = vma->vm_flags;
+	file = vma->vm_file;
+
+	if (file) {
+		struct inode *inode = vma->vm_file->f_path.dentry->d_inode;
+		dev = inode->i_sb->s_dev;
+		ino = inode->i_ino;
+	}
+
+	seq_printf(m,
+		   "%08lx-%08lx %c%c%c%c %08lx %02x:%02x %lu %n",
+		   vma->vm_start,
+		   vma->vm_end,
+		   flags & VM_READ ? 'r' : '-',
+		   flags & VM_WRITE ? 'w' : '-',
+		   flags & VM_EXEC ? 'x' : '-',
+		   flags & VM_MAYSHARE ? flags & VM_SHARED ? 'S' : 's' : 'p',
+		   vma->vm_pgoff << PAGE_SHIFT,
+		   MAJOR(dev), MINOR(dev), ino, &len);
+
+	if (file) {
+		len = 25 + sizeof(void *) * 6 - len;
+		if (len < 1)
+			len = 1;
+		seq_printf(m, "%*c", len, ' ');
+		seq_path(m, &file->f_path, "");
+	}
+
+	seq_putc(m, '\n');
+	return 0;
+}
+
 /*
  * display mapping lines for a particular process's /proc/pid/maps
  */
-static int show_map(struct seq_file *m, void *_vml)
+static int show_map(struct seq_file *m, void *_p)
 {
-	struct vm_list_struct *vml = _vml;
+	struct rb_node *p = _p;
 
-	return nommu_vma_show(m, vml->vma);
+	return nommu_vma_show(m, rb_entry(p, struct vm_area_struct, vm_rb));
 }
 
 static void *m_start(struct seq_file *m, loff_t *pos)
 {
 	struct proc_maps_private *priv = m->private;
-	struct vm_list_struct *vml;
 	struct mm_struct *mm;
+	struct rb_node *p;
 	loff_t n = *pos;
 
 	/* pin the task and mm whilst we play with them */
@@ -134,9 +176,9 @@ static void *m_start(struct seq_file *m, loff_t *pos)
 	}
 
 	/* start from the Nth VMA */
-	for (vml = mm->context.vmlist; vml; vml = vml->next)
+	for (p = rb_first(&mm->mm_rb); p; p = rb_next(p))
 		if (n-- == 0)
-			return vml;
+			return p;
 	return NULL;
 }
 
@@ -152,12 +194,12 @@ static void m_stop(struct seq_file *m, void *_vml)
 	}
 }
 
-static void *m_next(struct seq_file *m, void *_vml, loff_t *pos)
+static void *m_next(struct seq_file *m, void *_p, loff_t *pos)
 {
-	struct vm_list_struct *vml = _vml;
+	struct rb_node *p = _p;
 
 	(*pos)++;
-	return vml ? vml->next : NULL;
+	return p ? rb_next(p) : NULL;
 }
 
 static const struct seq_operations proc_pid_maps_ops = {
diff --git a/include/asm-frv/mmu.h b/include/asm-frv/mmu.h
index 22c03714fb14..86ca0e86e7d2 100644
--- a/include/asm-frv/mmu.h
+++ b/include/asm-frv/mmu.h
@@ -22,7 +22,6 @@ typedef struct {
 	unsigned long	dtlb_ptd_mapping;	/* [DAMR5] PTD mapping for dtlb cached PGE */
 
 #else
-	struct vm_list_struct	*vmlist;
 	unsigned long		end_brk;
 
 #endif
diff --git a/include/asm-m32r/mmu.h b/include/asm-m32r/mmu.h
index d9bd724479cf..150cb92bb666 100644
--- a/include/asm-m32r/mmu.h
+++ b/include/asm-m32r/mmu.h
@@ -4,7 +4,6 @@
 #if !defined(CONFIG_MMU)
 
 typedef struct {
-	struct vm_list_struct	*vmlist;
 	unsigned long		end_brk;
 } mm_context_t;
 
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 4a3d28c86443..b91a73fd1bcc 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -56,19 +56,9 @@ extern unsigned long mmap_min_addr;
 
 extern struct kmem_cache *vm_area_cachep;
 
-/*
- * This struct defines the per-mm list of VMAs for uClinux. If CONFIG_MMU is
- * disabled, then there's a single shared list of VMAs maintained by the
- * system, and mm's subscribe to these individually
- */
-struct vm_list_struct {
-	struct vm_list_struct	*next;
-	struct vm_area_struct	*vma;
-};
-
 #ifndef CONFIG_MMU
-extern struct rb_root nommu_vma_tree;
-extern struct rw_semaphore nommu_vma_sem;
+extern struct rb_root nommu_region_tree;
+extern struct rw_semaphore nommu_region_sem;
 
 extern unsigned int kobjsize(const void *objp);
 #endif
@@ -1061,6 +1051,7 @@ extern void memmap_init_zone(unsigned long, int, unsigned long,
 				unsigned long, enum memmap_context);
 extern void setup_per_zone_pages_min(void);
 extern void mem_init(void);
+extern void __init mmap_init(void);
 extern void show_mem(void);
 extern void si_meminfo(struct sysinfo * val);
 extern void si_meminfo_node(struct sysinfo *val, int nid);
@@ -1072,6 +1063,9 @@ extern void setup_per_cpu_pageset(void);
 static inline void setup_per_cpu_pageset(void) {}
 #endif
 
+/* nommu.c */
+extern atomic_t mmap_pages_allocated;
+
 /* prio_tree.c */
 void vma_prio_tree_add(struct vm_area_struct *, struct vm_area_struct *old);
 void vma_prio_tree_insert(struct vm_area_struct *, struct prio_tree_root *);
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 9cfc9b627fdd..1c1e0d3a1714 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -96,6 +96,22 @@ struct page {
 #endif /* WANT_PAGE_VIRTUAL */
 };
 
+/*
+ * A region containing a mapping of a non-memory backed file under NOMMU
+ * conditions.  These are held in a global tree and are pinned by the VMAs that
+ * map parts of them.
+ */
+struct vm_region {
+	struct rb_node	vm_rb;		/* link in global region tree */
+	unsigned long	vm_flags;	/* VMA vm_flags */
+	unsigned long	vm_start;	/* start address of region */
+	unsigned long	vm_end;		/* region initialised to here */
+	unsigned long	vm_pgoff;	/* the offset in vm_file corresponding to vm_start */
+	struct file	*vm_file;	/* the backing file or NULL */
+
+	atomic_t	vm_usage;	/* region usage count */
+};
+
 /*
  * This struct defines a memory VMM memory area. There is one of these
  * per VM-area/task.  A VM area is any part of the process virtual memory
@@ -152,7 +168,7 @@ struct vm_area_struct {
 	unsigned long vm_truncate_count;/* truncate_count or restart_addr */
 
 #ifndef CONFIG_MMU
-	atomic_t vm_usage;		/* refcount (VMAs shared if !MMU) */
+	struct vm_region *vm_region;	/* NOMMU mapping region */
 #endif
 #ifdef CONFIG_NUMA
 	struct mempolicy *vm_policy;	/* NUMA policy for the VMA */
diff --git a/ipc/shm.c b/ipc/shm.c
index b125b560240e..d0ab5527bf45 100644
--- a/ipc/shm.c
+++ b/ipc/shm.c
@@ -990,6 +990,7 @@ asmlinkage long sys_shmdt(char __user *shmaddr)
 	 */
 	vma = find_vma(mm, addr);
 
+#ifdef CONFIG_MMU
 	while (vma) {
 		next = vma->vm_next;
 
@@ -1034,6 +1035,17 @@ asmlinkage long sys_shmdt(char __user *shmaddr)
 		vma = next;
 	}
 
+#else /* CONFIG_MMU */
+	/* under NOMMU conditions, the exact address to be destroyed must be
+	 * given */
+	retval = -EINVAL;
+	if (vma->vm_start == addr && vma->vm_ops == &shm_vm_ops) {
+		do_munmap(mm, vma->vm_start, vma->vm_end - vma->vm_start);
+		retval = 0;
+	}
+
+#endif
+
 	up_write(&mm->mmap_sem);
 	return retval;
 }
diff --git a/kernel/fork.c b/kernel/fork.c
index 7b8f2a78be3d..0bce4a43bb37 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1481,12 +1481,10 @@ void __init proc_caches_init(void)
 	fs_cachep = kmem_cache_create("fs_cache",
 			sizeof(struct fs_struct), 0,
 			SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
-	vm_area_cachep = kmem_cache_create("vm_area_struct",
-			sizeof(struct vm_area_struct), 0,
-			SLAB_PANIC, NULL);
 	mm_cachep = kmem_cache_create("mm_struct",
 			sizeof(struct mm_struct), ARCH_MIN_MMSTRUCT_ALIGN,
 			SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
+	mmap_init();
 }
 
 /*
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 2e75478e9c69..d0a32aab03ff 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -512,6 +512,13 @@ config DEBUG_VIRTUAL
 
 	  If unsure, say N.
 
+config DEBUG_NOMMU_REGIONS
+	bool "Debug the global anon/private NOMMU mapping region tree"
+	depends on DEBUG_KERNEL && !MMU
+	help
+	  This option causes the global tree of anonymous and private mapping
+	  regions to be regularly checked for invalid topology.
+
 config DEBUG_WRITECOUNT
 	bool "Debug filesystem writers count"
 	depends on DEBUG_KERNEL
diff --git a/mm/mmap.c b/mm/mmap.c
index a910c045cfd4..749623196cb9 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -2472,3 +2472,13 @@ void mm_drop_all_locks(struct mm_struct *mm)
 
 	mutex_unlock(&mm_all_locks_mutex);
 }
+
+/*
+ * initialise the VMA slab
+ */
+void __init mmap_init(void)
+{
+	vm_area_cachep = kmem_cache_create("vm_area_struct",
+			sizeof(struct vm_area_struct), 0,
+			SLAB_PANIC, NULL);
+}
diff --git a/mm/nommu.c b/mm/nommu.c
index 23f355bbe262..0d363dfcf10e 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -6,7 +6,7 @@
  *
  *  See Documentation/nommu-mmap.txt
  *
- *  Copyright (c) 2004-2005 David Howells <dhowells@redhat.com>
+ *  Copyright (c) 2004-2008 David Howells <dhowells@redhat.com>
  *  Copyright (c) 2000-2003 David McCullough <davidm@snapgear.com>
  *  Copyright (c) 2000-2001 D Jeff Dionne <jeff@uClinux.org>
  *  Copyright (c) 2002      Greg Ungerer <gerg@snapgear.com>
@@ -33,6 +33,28 @@
 #include <asm/uaccess.h>
 #include <asm/tlb.h>
 #include <asm/tlbflush.h>
+#include "internal.h"
+
+static inline __attribute__((format(printf, 1, 2)))
+void no_printk(const char *fmt, ...)
+{
+}
+
+#if 0
+#define kenter(FMT, ...) \
+	printk(KERN_DEBUG "==> %s("FMT")\n", __func__, ##__VA_ARGS__)
+#define kleave(FMT, ...) \
+	printk(KERN_DEBUG "<== %s()"FMT"\n", __func__, ##__VA_ARGS__)
+#define kdebug(FMT, ...) \
+	printk(KERN_DEBUG "xxx" FMT"yyy\n", ##__VA_ARGS__)
+#else
+#define kenter(FMT, ...) \
+	no_printk(KERN_DEBUG "==> %s("FMT")\n", __func__, ##__VA_ARGS__)
+#define kleave(FMT, ...) \
+	no_printk(KERN_DEBUG "<== %s()"FMT"\n", __func__, ##__VA_ARGS__)
+#define kdebug(FMT, ...) \
+	no_printk(KERN_DEBUG FMT"\n", ##__VA_ARGS__)
+#endif
 
 #include "internal.h"
 
@@ -46,12 +68,15 @@ int sysctl_overcommit_ratio = 50; /* default is 50% */
 int sysctl_max_map_count = DEFAULT_MAX_MAP_COUNT;
 int heap_stack_gap = 0;
 
+atomic_t mmap_pages_allocated;
+
 EXPORT_SYMBOL(mem_map);
 EXPORT_SYMBOL(num_physpages);
 
-/* list of shareable VMAs */
-struct rb_root nommu_vma_tree = RB_ROOT;
-DECLARE_RWSEM(nommu_vma_sem);
+/* list of mapped, potentially shareable regions */
+static struct kmem_cache *vm_region_jar;
+struct rb_root nommu_region_tree = RB_ROOT;
+DECLARE_RWSEM(nommu_region_sem);
 
 struct vm_operations_struct generic_file_vm_ops = {
 };
@@ -400,129 +425,174 @@ asmlinkage unsigned long sys_brk(unsigned long brk)
 	return mm->brk = brk;
 }
 
-#ifdef DEBUG
-static void show_process_blocks(void)
+/*
+ * initialise the VMA and region record slabs
+ */
+void __init mmap_init(void)
 {
-	struct vm_list_struct *vml;
-
-	printk("Process blocks %d:", current->pid);
-
-	for (vml = &current->mm->context.vmlist; vml; vml = vml->next) {
-		printk(" %p: %p", vml, vml->vma);
-		if (vml->vma)
-			printk(" (%d @%lx #%d)",
-			       kobjsize((void *) vml->vma->vm_start),
-			       vml->vma->vm_start,
-			       atomic_read(&vml->vma->vm_usage));
-		printk(vml->next ? " ->" : ".\n");
-	}
+	vm_region_jar = kmem_cache_create("vm_region_jar",
+					  sizeof(struct vm_region), 0,
+					  SLAB_PANIC, NULL);
+	vm_area_cachep = kmem_cache_create("vm_area_struct",
+					   sizeof(struct vm_area_struct), 0,
+					   SLAB_PANIC, NULL);
 }
-#endif /* DEBUG */
 
 /*
- * add a VMA into a process's mm_struct in the appropriate place in the list
- * - should be called with mm->mmap_sem held writelocked
+ * validate the region tree
+ * - the caller must hold the region lock
  */
-static void add_vma_to_mm(struct mm_struct *mm, struct vm_list_struct *vml)
+#ifdef CONFIG_DEBUG_NOMMU_REGIONS
+static noinline void validate_nommu_regions(void)
 {
-	struct vm_list_struct **ppv;
+	struct vm_region *region, *last;
+	struct rb_node *p, *lastp;
 
-	for (ppv = &current->mm->context.vmlist; *ppv; ppv = &(*ppv)->next)
-		if ((*ppv)->vma->vm_start > vml->vma->vm_start)
-			break;
+	lastp = rb_first(&nommu_region_tree);
+	if (!lastp)
+		return;
+
+	last = rb_entry(lastp, struct vm_region, vm_rb);
+	if (unlikely(last->vm_end <= last->vm_start))
+		BUG();
+
+	while ((p = rb_next(lastp))) {
+		region = rb_entry(p, struct vm_region, vm_rb);
+		last = rb_entry(lastp, struct vm_region, vm_rb);
+
+		if (unlikely(region->vm_end <= region->vm_start))
+			BUG();
+		if (unlikely(region->vm_start < last->vm_end))
+			BUG();
 
-	vml->next = *ppv;
-	*ppv = vml;
+		lastp = p;
+	}
 }
+#else
+#define validate_nommu_regions() do {} while(0)
+#endif
 
 /*
- * look up the first VMA in which addr resides, NULL if none
- * - should be called with mm->mmap_sem at least held readlocked
+ * add a region into the global tree
  */
-struct vm_area_struct *find_vma(struct mm_struct *mm, unsigned long addr)
+static void add_nommu_region(struct vm_region *region)
 {
-	struct vm_list_struct *loop, *vml;
+	struct vm_region *pregion;
+	struct rb_node **p, *parent;
 
-	/* search the vm_start ordered list */
-	vml = NULL;
-	for (loop = mm->context.vmlist; loop; loop = loop->next) {
-		if (loop->vma->vm_start > addr)
-			break;
-		vml = loop;
+	validate_nommu_regions();
+
+	BUG_ON(region->vm_start & ~PAGE_MASK);
+
+	parent = NULL;
+	p = &nommu_region_tree.rb_node;
+	while (*p) {
+		parent = *p;
+		pregion = rb_entry(parent, struct vm_region, vm_rb);
+		if (region->vm_start < pregion->vm_start)
+			p = &(*p)->rb_left;
+		else if (region->vm_start > pregion->vm_start)
+			p = &(*p)->rb_right;
+		else if (pregion == region)
+			return;
+		else
+			BUG();
 	}
 
-	if (vml && vml->vma->vm_end > addr)
-		return vml->vma;
+	rb_link_node(&region->vm_rb, parent, p);
+	rb_insert_color(&region->vm_rb, &nommu_region_tree);
 
-	return NULL;
+	validate_nommu_regions();
 }
-EXPORT_SYMBOL(find_vma);
 
 /*
- * find a VMA
- * - we don't extend stack VMAs under NOMMU conditions
+ * delete a region from the global tree
  */
-struct vm_area_struct *find_extend_vma(struct mm_struct *mm, unsigned long addr)
+static void delete_nommu_region(struct vm_region *region)
 {
-	return find_vma(mm, addr);
-}
+	BUG_ON(!nommu_region_tree.rb_node);
 
-int expand_stack(struct vm_area_struct *vma, unsigned long address)
-{
-	return -ENOMEM;
+	validate_nommu_regions();
+	rb_erase(&region->vm_rb, &nommu_region_tree);
+	validate_nommu_regions();
 }
 
 /*
- * look up the first VMA exactly that exactly matches addr
- * - should be called with mm->mmap_sem at least held readlocked
+ * free a contiguous series of pages
  */
-static inline struct vm_area_struct *find_vma_exact(struct mm_struct *mm,
-						    unsigned long addr)
+static void free_page_series(unsigned long from, unsigned long to)
 {
-	struct vm_list_struct *vml;
-
-	/* search the vm_start ordered list */
-	for (vml = mm->context.vmlist; vml; vml = vml->next) {
-		if (vml->vma->vm_start == addr)
-			return vml->vma;
-		if (vml->vma->vm_start > addr)
-			break;
+	for (; from < to; from += PAGE_SIZE) {
+		struct page *page = virt_to_page(from);
+
+		kdebug("- free %lx", from);
+		atomic_dec(&mmap_pages_allocated);
+		if (page_count(page) != 1)
+			kdebug("free page %p [%d]", page, page_count(page));
+		put_page(page);
 	}
-
-	return NULL;
 }
 
 /*
- * find a VMA in the global tree
+ * release a reference to a region
+ * - the caller must hold the region semaphore, which this releases
+ * - the region may not have been added to the tree yet, in which case vm_end
+ *   will equal vm_start
  */
-static inline struct vm_area_struct *find_nommu_vma(unsigned long start)
+static void __put_nommu_region(struct vm_region *region)
+	__releases(nommu_region_sem)
 {
-	struct vm_area_struct *vma;
-	struct rb_node *n = nommu_vma_tree.rb_node;
+	kenter("%p{%d}", region, atomic_read(&region->vm_usage));
 
-	while (n) {
-		vma = rb_entry(n, struct vm_area_struct, vm_rb);
+	BUG_ON(!nommu_region_tree.rb_node);
 
-		if (start < vma->vm_start)
-			n = n->rb_left;
-		else if (start > vma->vm_start)
-			n = n->rb_right;
-		else
-			return vma;
+	if (atomic_dec_and_test(&region->vm_usage)) {
+		if (region->vm_end > region->vm_start)
+			delete_nommu_region(region);
+		up_write(&nommu_region_sem);
+
+		if (region->vm_file)
+			fput(region->vm_file);
+
+		/* IO memory and memory shared directly out of the pagecache
+		 * from ramfs/tmpfs mustn't be released here */
+		if (region->vm_flags & VM_MAPPED_COPY) {
+			kdebug("free series");
+			free_page_series(region->vm_start, region->vm_end);
+		}
+		kmem_cache_free(vm_region_jar, region);
+	} else {
+		up_write(&nommu_region_sem);
 	}
+}
 
-	return NULL;
+/*
+ * release a reference to a region
+ */
+static void put_nommu_region(struct vm_region *region)
+{
+	down_write(&nommu_region_sem);
+	__put_nommu_region(region);
 }
 
 /*
- * add a VMA in the global tree
+ * add a VMA into a process's mm_struct in the appropriate place in the list
+ * and tree and add to the address space's page tree also if not an anonymous
+ * page
+ * - should be called with mm->mmap_sem held writelocked
  */
-static void add_nommu_vma(struct vm_area_struct *vma)
+static void add_vma_to_mm(struct mm_struct *mm, struct vm_area_struct *vma)
 {
-	struct vm_area_struct *pvma;
+	struct vm_area_struct *pvma, **pp;
 	struct address_space *mapping;
-	struct rb_node **p = &nommu_vma_tree.rb_node;
-	struct rb_node *parent = NULL;
+	struct rb_node **p, *parent;
+
+	kenter(",%p", vma);
+
+	BUG_ON(!vma->vm_region);
+
+	mm->map_count++;
+	vma->vm_mm = mm;
 
 	/* add the VMA to the mapping */
 	if (vma->vm_file) {
@@ -533,42 +603,62 @@ static void add_nommu_vma(struct vm_area_struct *vma)
 		flush_dcache_mmap_unlock(mapping);
 	}
 
-	/* add the VMA to the master list */
+	/* add the VMA to the tree */
+	parent = NULL;
+	p = &mm->mm_rb.rb_node;
 	while (*p) {
 		parent = *p;
 		pvma = rb_entry(parent, struct vm_area_struct, vm_rb);
 
-		if (vma->vm_start < pvma->vm_start) {
+		/* sort by: start addr, end addr, VMA struct addr in that order
+		 * (the latter is necessary as we may get identical VMAs) */
+		if (vma->vm_start < pvma->vm_start)
 			p = &(*p)->rb_left;
-		}
-		else if (vma->vm_start > pvma->vm_start) {
+		else if (vma->vm_start > pvma->vm_start)
 			p = &(*p)->rb_right;
-		}
-		else {
-			/* mappings are at the same address - this can only
-			 * happen for shared-mem chardevs and shared file
-			 * mappings backed by ramfs/tmpfs */
-			BUG_ON(!(pvma->vm_flags & VM_SHARED));
-
-			if (vma < pvma)
-				p = &(*p)->rb_left;
-			else if (vma > pvma)
-				p = &(*p)->rb_right;
-			else
-				BUG();
-		}
+		else if (vma->vm_end < pvma->vm_end)
+			p = &(*p)->rb_left;
+		else if (vma->vm_end > pvma->vm_end)
+			p = &(*p)->rb_right;
+		else if (vma < pvma)
+			p = &(*p)->rb_left;
+		else if (vma > pvma)
+			p = &(*p)->rb_right;
+		else
+			BUG();
 	}
 
 	rb_link_node(&vma->vm_rb, parent, p);
-	rb_insert_color(&vma->vm_rb, &nommu_vma_tree);
+	rb_insert_color(&vma->vm_rb, &mm->mm_rb);
+
+	/* add VMA to the VMA list also */
+	for (pp = &mm->mmap; (pvma = *pp); pp = &(*pp)->vm_next) {
+		if (pvma->vm_start > vma->vm_start)
+			break;
+		if (pvma->vm_start < vma->vm_start)
+			continue;
+		if (pvma->vm_end < vma->vm_end)
+			break;
+	}
+
+	vma->vm_next = *pp;
+	*pp = vma;
 }
 
 /*
- * delete a VMA from the global list
+ * delete a VMA from its owning mm_struct and address space
  */
-static void delete_nommu_vma(struct vm_area_struct *vma)
+static void delete_vma_from_mm(struct vm_area_struct *vma)
 {
+	struct vm_area_struct **pp;
 	struct address_space *mapping;
+	struct mm_struct *mm = vma->vm_mm;
+
+	kenter("%p", vma);
+
+	mm->map_count--;
+	if (mm->mmap_cache == vma)
+		mm->mmap_cache = NULL;
 
 	/* remove the VMA from the mapping */
 	if (vma->vm_file) {
@@ -579,8 +669,115 @@ static void delete_nommu_vma(struct vm_area_struct *vma)
 		flush_dcache_mmap_unlock(mapping);
 	}
 
-	/* remove from the master list */
-	rb_erase(&vma->vm_rb, &nommu_vma_tree);
+	/* remove from the MM's tree and list */
+	rb_erase(&vma->vm_rb, &mm->mm_rb);
+	for (pp = &mm->mmap; *pp; pp = &(*pp)->vm_next) {
+		if (*pp == vma) {
+			*pp = vma->vm_next;
+			break;
+		}
+	}
+
+	vma->vm_mm = NULL;
+}
+
+/*
+ * destroy a VMA record
+ */
+static void delete_vma(struct mm_struct *mm, struct vm_area_struct *vma)
+{
+	kenter("%p", vma);
+	if (vma->vm_ops && vma->vm_ops->close)
+		vma->vm_ops->close(vma);
+	if (vma->vm_file) {
+		fput(vma->vm_file);
+		if (vma->vm_flags & VM_EXECUTABLE)
+			removed_exe_file_vma(mm);
+	}
+	put_nommu_region(vma->vm_region);
+	kmem_cache_free(vm_area_cachep, vma);
+}
+
+/*
+ * look up the first VMA in which addr resides, NULL if none
+ * - should be called with mm->mmap_sem at least held readlocked
+ */
+struct vm_area_struct *find_vma(struct mm_struct *mm, unsigned long addr)
+{
+	struct vm_area_struct *vma;
+	struct rb_node *n = mm->mm_rb.rb_node;
+
+	/* check the cache first */
+	vma = mm->mmap_cache;
+	if (vma && vma->vm_start <= addr && vma->vm_end > addr)
+		return vma;
+
+	/* trawl the tree (there may be multiple mappings in which addr
+	 * resides) */
+	for (n = rb_first(&mm->mm_rb); n; n = rb_next(n)) {
+		vma = rb_entry(n, struct vm_area_struct, vm_rb);
+		if (vma->vm_start > addr)
+			return NULL;
+		if (vma->vm_end > addr) {
+			mm->mmap_cache = vma;
+			return vma;
+		}
+	}
+
+	return NULL;
+}
+EXPORT_SYMBOL(find_vma);
+
+/*
+ * find a VMA
+ * - we don't extend stack VMAs under NOMMU conditions
+ */
+struct vm_area_struct *find_extend_vma(struct mm_struct *mm, unsigned long addr)
+{
+	return find_vma(mm, addr);
+}
+
+/*
+ * expand a stack to a given address
+ * - not supported under NOMMU conditions
+ */
+int expand_stack(struct vm_area_struct *vma, unsigned long address)
+{
+	return -ENOMEM;
+}
+
+/*
+ * look up the first VMA exactly that exactly matches addr
+ * - should be called with mm->mmap_sem at least held readlocked
+ */
+static struct vm_area_struct *find_vma_exact(struct mm_struct *mm,
+					     unsigned long addr,
+					     unsigned long len)
+{
+	struct vm_area_struct *vma;
+	struct rb_node *n = mm->mm_rb.rb_node;
+	unsigned long end = addr + len;
+
+	/* check the cache first */
+	vma = mm->mmap_cache;
+	if (vma && vma->vm_start == addr && vma->vm_end == end)
+		return vma;
+
+	/* trawl the tree (there may be multiple mappings in which addr
+	 * resides) */
+	for (n = rb_first(&mm->mm_rb); n; n = rb_next(n)) {
+		vma = rb_entry(n, struct vm_area_struct, vm_rb);
+		if (vma->vm_start < addr)
+			continue;
+		if (vma->vm_start > addr)
+			return NULL;
+		if (vma->vm_end == end) {
+			mm->mmap_cache = vma;
+			return vma;
+		}
+	}
+
+	return NULL;
 }
 
 /*
@@ -595,7 +792,7 @@ static int validate_mmap_request(struct file *file,
 				 unsigned long pgoff,
 				 unsigned long *_capabilities)
 {
-	unsigned long capabilities;
+	unsigned long capabilities, rlen;
 	unsigned long reqprot = prot;
 	int ret;
 
@@ -615,12 +812,12 @@ static int validate_mmap_request(struct file *file,
 		return -EINVAL;
 
 	/* Careful about overflows.. */
-	len = PAGE_ALIGN(len);
-	if (!len || len > TASK_SIZE)
+	rlen = PAGE_ALIGN(len);
+	if (!rlen || rlen > TASK_SIZE)
 		return -ENOMEM;
 
 	/* offset overflow? */
-	if ((pgoff + (len >> PAGE_SHIFT)) < pgoff)
+	if ((pgoff + (rlen >> PAGE_SHIFT)) < pgoff)
 		return -EOVERFLOW;
 
 	if (file) {
@@ -794,9 +991,10 @@ static unsigned long determine_vm_flags(struct file *file,
 }
 
 /*
- * set up a shared mapping on a file
+ * set up a shared mapping on a file (the driver or filesystem provides and
+ * pins the storage)
  */
-static int do_mmap_shared_file(struct vm_area_struct *vma, unsigned long len)
+static int do_mmap_shared_file(struct vm_area_struct *vma)
 {
 	int ret;
 
@@ -814,10 +1012,14 @@ static int do_mmap_shared_file(struct vm_area_struct *vma, unsigned long len)
 /*
  * set up a private mapping or an anonymous shared mapping
  */
-static int do_mmap_private(struct vm_area_struct *vma, unsigned long len)
+static int do_mmap_private(struct vm_area_struct *vma,
+			   struct vm_region *region,
+			   unsigned long len)
 {
+	struct page *pages;
+	unsigned long total, point, n, rlen;
 	void *base;
-	int ret;
+	int ret, order;
 
 	/* invoke the file's mapping function so that it can keep track of
 	 * shared mappings on devices or memory
@@ -836,23 +1038,46 @@ static int do_mmap_private(struct vm_area_struct *vma, unsigned long len)
 		 * make a private copy of the data and map that instead */
 	}
 
+	rlen = PAGE_ALIGN(len);
+
 	/* allocate some memory to hold the mapping
 	 * - note that this may not return a page-aligned address if the object
 	 *   we're allocating is smaller than a page
 	 */
-	base = kmalloc(len, GFP_KERNEL|__GFP_COMP);
-	if (!base)
+	order = get_order(rlen);
+	kdebug("alloc order %d for %lx", order, len);
+
+	pages = alloc_pages(GFP_KERNEL, order);
+	if (!pages)
 		goto enomem;
 
-	vma->vm_start = (unsigned long) base;
-	vma->vm_end = vma->vm_start + len;
-	vma->vm_flags |= VM_MAPPED_COPY;
+	/* we allocated a power-of-2 sized page set, so we need to trim off the
+	 * excess */
+	total = 1 << order;
+	atomic_add(total, &mmap_pages_allocated);
+
+	point = rlen >> PAGE_SHIFT;
+	while (total > point) {
+		order = ilog2(total - point);
+		n = 1 << order;
+		kdebug("shave %lu/%lu @%lu", n, total - point, total);
+		atomic_sub(n, &mmap_pages_allocated);
+		total -= n;
+		set_page_refcounted(pages + total);
+		__free_pages(pages + total, order);
+	}
+
+	total = rlen >> PAGE_SHIFT;
+	for (point = 1; point < total; point++)
+		set_page_refcounted(&pages[point]);
 
-#ifdef WARN_ON_SLACK
-	if (len + WARN_ON_SLACK <= kobjsize(result))
-		printk("Allocation of %lu bytes from process %d has %lu bytes of slack\n",
-		       len, current->pid, kobjsize(result) - len);
-#endif
+	base = page_address(pages);
+	region->vm_flags = vma->vm_flags |= VM_MAPPED_COPY;
+	region->vm_start = (unsigned long) base;
+	region->vm_end   = region->vm_start + rlen;
+
+	vma->vm_start = region->vm_start;
+	vma->vm_end   = region->vm_start + len;
 
 	if (vma->vm_file) {
 		/* read the contents of a file into the copy */
@@ -864,26 +1089,27 @@ static int do_mmap_private(struct vm_area_struct *vma, unsigned long len)
 
 		old_fs = get_fs();
 		set_fs(KERNEL_DS);
-		ret = vma->vm_file->f_op->read(vma->vm_file, base, len, &fpos);
+		ret = vma->vm_file->f_op->read(vma->vm_file, base, rlen, &fpos);
 		set_fs(old_fs);
 
 		if (ret < 0)
 			goto error_free;
 
 		/* clear the last little bit */
-		if (ret < len)
-			memset(base + ret, 0, len - ret);
+		if (ret < rlen)
+			memset(base + ret, 0, rlen - ret);
 
 	} else {
 		/* if it's an anonymous mapping, then just clear it */
-		memset(base, 0, len);
+		memset(base, 0, rlen);
 	}
 
 	return 0;
 
 error_free:
-	kfree(base);
-	vma->vm_start = 0;
+	free_page_series(region->vm_start, region->vm_end);
+	region->vm_start = vma->vm_start = 0;
+	region->vm_end   = vma->vm_end = 0;
 	return ret;
 
 enomem:
@@ -903,13 +1129,14 @@ unsigned long do_mmap_pgoff(struct file *file,
 			    unsigned long flags,
 			    unsigned long pgoff)
 {
-	struct vm_list_struct *vml = NULL;
-	struct vm_area_struct *vma = NULL;
+	struct vm_area_struct *vma;
+	struct vm_region *region;
 	struct rb_node *rb;
-	unsigned long capabilities, vm_flags;
-	void *result;
+	unsigned long capabilities, vm_flags, result;
 	int ret;
 
+	kenter(",%lx,%lx,%lx,%lx,%lx", addr, len, prot, flags, pgoff);
+
 	if (!(flags & MAP_FIXED))
 		addr = round_hint_to_min(addr);
 
@@ -917,73 +1144,120 @@ unsigned long do_mmap_pgoff(struct file *file,
 	 * mapping */
 	ret = validate_mmap_request(file, addr, len, prot, flags, pgoff,
 				    &capabilities);
-	if (ret < 0)
+	if (ret < 0) {
+		kleave(" = %d [val]", ret);
 		return ret;
+	}
 
 	/* we've determined that we can make the mapping, now translate what we
 	 * now know into VMA flags */
 	vm_flags = determine_vm_flags(file, prot, flags, capabilities);
 
-	/* we're going to need to record the mapping if it works */
-	vml = kzalloc(sizeof(struct vm_list_struct), GFP_KERNEL);
-	if (!vml)
-		goto error_getting_vml;
+	/* we're going to need to record the mapping */
+	region = kmem_cache_zalloc(vm_region_jar, GFP_KERNEL);
+	if (!region)
+		goto error_getting_region;
+
+	vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL);
+	if (!vma)
+		goto error_getting_vma;
+
+	atomic_set(&region->vm_usage, 1);
+	region->vm_flags = vm_flags;
+	region->vm_pgoff = pgoff;
 
-	down_write(&nommu_vma_sem);
+	INIT_LIST_HEAD(&vma->anon_vma_node);
+	vma->vm_flags = vm_flags;
+	vma->vm_pgoff = pgoff;
 
-	/* if we want to share, we need to check for VMAs created by other
+	if (file) {
+		region->vm_file = file;
+		get_file(file);
+		vma->vm_file = file;
+		get_file(file);
+		if (vm_flags & VM_EXECUTABLE) {
+			added_exe_file_vma(current->mm);
+			vma->vm_mm = current->mm;
+		}
+	}
+
+	down_write(&nommu_region_sem);
+
+	/* if we want to share, we need to check for regions created by other
 	 * mmap() calls that overlap with our proposed mapping
-	 * - we can only share with an exact match on most regular files
+	 * - we can only share with a superset match on most regular files
 	 * - shared mappings on character devices and memory backed files are
 	 *   permitted to overlap inexactly as far as we are concerned for in
 	 *   these cases, sharing is handled in the driver or filesystem rather
 	 *   than here
 	 */
 	if (vm_flags & VM_MAYSHARE) {
-		unsigned long pglen = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
-		unsigned long vmpglen;
+		struct vm_region *pregion;
+		unsigned long pglen, rpglen, pgend, rpgend, start;
 
-		/* suppress VMA sharing for shared regions */
-		if (vm_flags & VM_SHARED &&
-		    capabilities & BDI_CAP_MAP_DIRECT)
-			goto dont_share_VMAs;
+		pglen = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+		pgend = pgoff + pglen;
 
-		for (rb = rb_first(&nommu_vma_tree); rb; rb = rb_next(rb)) {
-			vma = rb_entry(rb, struct vm_area_struct, vm_rb);
+		for (rb = rb_first(&nommu_region_tree); rb; rb = rb_next(rb)) {
+			pregion = rb_entry(rb, struct vm_region, vm_rb);
 
-			if (!(vma->vm_flags & VM_MAYSHARE))
+			if (!(pregion->vm_flags & VM_MAYSHARE))
 				continue;
 
 			/* search for overlapping mappings on the same file */
-			if (vma->vm_file->f_path.dentry->d_inode != file->f_path.dentry->d_inode)
+			if (pregion->vm_file->f_path.dentry->d_inode !=
+			    file->f_path.dentry->d_inode)
 				continue;
 
-			if (vma->vm_pgoff >= pgoff + pglen)
+			if (pregion->vm_pgoff >= pgend)
 				continue;
 
-			vmpglen = vma->vm_end - vma->vm_start + PAGE_SIZE - 1;
-			vmpglen >>= PAGE_SHIFT;
-			if (pgoff >= vma->vm_pgoff + vmpglen)
+			rpglen = pregion->vm_end - pregion->vm_start;
+			rpglen = (rpglen + PAGE_SIZE - 1) >> PAGE_SHIFT;
+			rpgend = pregion->vm_pgoff + rpglen;
+			if (pgoff >= rpgend)
 				continue;
 
-			/* handle inexactly overlapping matches between mappings */
-			if (vma->vm_pgoff != pgoff || vmpglen != pglen) {
+			/* handle inexactly overlapping matches between
+			 * mappings */
+			if ((pregion->vm_pgoff != pgoff || rpglen != pglen) &&
+			    !(pgoff >= pregion->vm_pgoff && pgend <= rpgend)) {
+				/* new mapping is not a subset of the region */
 				if (!(capabilities & BDI_CAP_MAP_DIRECT))
 					goto sharing_violation;
 				continue;
 			}
 
-			/* we've found a VMA we can share */
-			atomic_inc(&vma->vm_usage);
-
-			vml->vma = vma;
-			result = (void *) vma->vm_start;
-			goto shared;
+			/* we've found a region we can share */
+			atomic_inc(&pregion->vm_usage);
+			vma->vm_region = pregion;
+			start = pregion->vm_start;
+			start += (pgoff - pregion->vm_pgoff) << PAGE_SHIFT;
+			vma->vm_start = start;
+			vma->vm_end = start + len;
+
+			if (pregion->vm_flags & VM_MAPPED_COPY) {
+				kdebug("share copy");
+				vma->vm_flags |= VM_MAPPED_COPY;
+			} else {
+				kdebug("share mmap");
+				ret = do_mmap_shared_file(vma);
+				if (ret < 0) {
+					vma->vm_region = NULL;
+					vma->vm_start = 0;
+					vma->vm_end = 0;
+					atomic_dec(&pregion->vm_usage);
+					pregion = NULL;
+					goto error_just_free;
+				}
+			}
+			fput(region->vm_file);
+			kmem_cache_free(vm_region_jar, region);
+			region = pregion;
+			result = start;
+			goto share;
 		}
 
-	dont_share_VMAs:
-		vma = NULL;
-
 		/* obtain the address at which to make a shared mapping
 		 * - this is the hook for quasi-memory character devices to
 		 *   tell us the location of a shared mapping
@@ -994,102 +1268,93 @@ unsigned long do_mmap_pgoff(struct file *file,
 			if (IS_ERR((void *) addr)) {
 				ret = addr;
 				if (ret != (unsigned long) -ENOSYS)
-					goto error;
+					goto error_just_free;
 
 				/* the driver refused to tell us where to site
 				 * the mapping so we'll have to attempt to copy
 				 * it */
 				ret = (unsigned long) -ENODEV;
 				if (!(capabilities & BDI_CAP_MAP_COPY))
-					goto error;
+					goto error_just_free;
 
 				capabilities &= ~BDI_CAP_MAP_DIRECT;
+			} else {
+				vma->vm_start = region->vm_start = addr;
+				vma->vm_end = region->vm_end = addr + len;
 			}
 		}
 	}
 
-	/* we're going to need a VMA struct as well */
-	vma = kzalloc(sizeof(struct vm_area_struct), GFP_KERNEL);
-	if (!vma)
-		goto error_getting_vma;
-
-	INIT_LIST_HEAD(&vma->anon_vma_node);
-	atomic_set(&vma->vm_usage, 1);
-	if (file) {
-		get_file(file);
-		if (vm_flags & VM_EXECUTABLE) {
-			added_exe_file_vma(current->mm);
-			vma->vm_mm = current->mm;
-		}
-	}
-	vma->vm_file	= file;
-	vma->vm_flags	= vm_flags;
-	vma->vm_start	= addr;
-	vma->vm_end	= addr + len;
-	vma->vm_pgoff	= pgoff;
-
-	vml->vma = vma;
+	vma->vm_region = region;
 
 	/* set up the mapping */
 	if (file && vma->vm_flags & VM_SHARED)
-		ret = do_mmap_shared_file(vma, len);
+		ret = do_mmap_shared_file(vma);
 	else
-		ret = do_mmap_private(vma, len);
+		ret = do_mmap_private(vma, region, len);
 	if (ret < 0)
-		goto error;
+		goto error_put_region;
+
+	add_nommu_region(region);
 
 	/* okay... we have a mapping; now we have to register it */
-	result = (void *) vma->vm_start;
+	result = vma->vm_start;
 
 	current->mm->total_vm += len >> PAGE_SHIFT;
 
-	add_nommu_vma(vma);
+share:
+	add_vma_to_mm(current->mm, vma);
 
- shared:
-	add_vma_to_mm(current->mm, vml);
-
-	up_write(&nommu_vma_sem);
+	up_write(&nommu_region_sem);
 
 	if (prot & PROT_EXEC)
-		flush_icache_range((unsigned long) result,
-				   (unsigned long) result + len);
+		flush_icache_range(result, result + len);
 
-#ifdef DEBUG
-	printk("do_mmap:\n");
-	show_process_blocks();
-#endif
+	kleave(" = %lx", result);
+	return result;
 
-	return (unsigned long) result;
-
- error:
-	up_write(&nommu_vma_sem);
-	kfree(vml);
+error_put_region:
+	__put_nommu_region(region);
 	if (vma) {
 		if (vma->vm_file) {
 			fput(vma->vm_file);
 			if (vma->vm_flags & VM_EXECUTABLE)
 				removed_exe_file_vma(vma->vm_mm);
 		}
-		kfree(vma);
+		kmem_cache_free(vm_area_cachep, vma);
 	}
+	kleave(" = %d [pr]", ret);
 	return ret;
 
- sharing_violation:
-	up_write(&nommu_vma_sem);
-	printk("Attempt to share mismatched mappings\n");
-	kfree(vml);
-	return -EINVAL;
+error_just_free:
+	up_write(&nommu_region_sem);
+error:
+	fput(region->vm_file);
+	kmem_cache_free(vm_region_jar, region);
+	fput(vma->vm_file);
+	if (vma->vm_flags & VM_EXECUTABLE)
+		removed_exe_file_vma(vma->vm_mm);
+	kmem_cache_free(vm_area_cachep, vma);
+	kleave(" = %d", ret);
+	return ret;
+
+sharing_violation:
+	up_write(&nommu_region_sem);
+	printk(KERN_WARNING "Attempt to share mismatched mappings\n");
+	ret = -EINVAL;
+	goto error;
 
- error_getting_vma:
-	up_write(&nommu_vma_sem);
-	kfree(vml);
-	printk("Allocation of vma for %lu byte allocation from process %d failed\n",
+error_getting_vma:
+	kmem_cache_free(vm_region_jar, region);
+	printk(KERN_WARNING "Allocation of vma for %lu byte allocation"
+	       " from process %d failed\n",
 	       len, current->pid);
 	show_free_areas();
 	return -ENOMEM;
 
- error_getting_vml:
-	printk("Allocation of vml for %lu byte allocation from process %d failed\n",
+error_getting_region:
+	printk(KERN_WARNING "Allocation of vm region for %lu byte allocation"
+	       " from process %d failed\n",
 	       len, current->pid);
 	show_free_areas();
 	return -ENOMEM;
@@ -1097,77 +1362,180 @@ unsigned long do_mmap_pgoff(struct file *file,
 EXPORT_SYMBOL(do_mmap_pgoff);
 
 /*
- * handle mapping disposal for uClinux
+ * split a vma into two pieces at address 'addr', a new vma is allocated either
+ * for the first part or the tail.
  */
-static void put_vma(struct mm_struct *mm, struct vm_area_struct *vma)
+int split_vma(struct mm_struct *mm, struct vm_area_struct *vma,
+	      unsigned long addr, int new_below)
 {
-	if (vma) {
-		down_write(&nommu_vma_sem);
+	struct vm_area_struct *new;
+	struct vm_region *region;
+	unsigned long npages;
 
-		if (atomic_dec_and_test(&vma->vm_usage)) {
-			delete_nommu_vma(vma);
+	kenter("");
 
-			if (vma->vm_ops && vma->vm_ops->close)
-				vma->vm_ops->close(vma);
+	/* we're only permitted to split anonymous regions that have a single
+	 * owner */
+	if (vma->vm_file ||
+	    atomic_read(&vma->vm_region->vm_usage) != 1)
+		return -ENOMEM;
 
-			/* IO memory and memory shared directly out of the pagecache from
-			 * ramfs/tmpfs mustn't be released here */
-			if (vma->vm_flags & VM_MAPPED_COPY)
-				kfree((void *) vma->vm_start);
+	if (mm->map_count >= sysctl_max_map_count)
+		return -ENOMEM;
 
-			if (vma->vm_file) {
-				fput(vma->vm_file);
-				if (vma->vm_flags & VM_EXECUTABLE)
-					removed_exe_file_vma(mm);
-			}
-			kfree(vma);
-		}
+	region = kmem_cache_alloc(vm_region_jar, GFP_KERNEL);
+	if (!region)
+		return -ENOMEM;
+
+	new = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL);
+	if (!new) {
+		kmem_cache_free(vm_region_jar, region);
+		return -ENOMEM;
+	}
+
+	/* most fields are the same, copy all, and then fixup */
+	*new = *vma;
+	*region = *vma->vm_region;
+	new->vm_region = region;
+
+	npages = (addr - vma->vm_start) >> PAGE_SHIFT;
+
+	if (new_below) {
+		region->vm_end = new->vm_end = addr;
+	} else {
+		region->vm_start = new->vm_start = addr;
+		region->vm_pgoff = new->vm_pgoff += npages;
+	}
 
-		up_write(&nommu_vma_sem);
+	if (new->vm_ops && new->vm_ops->open)
+		new->vm_ops->open(new);
+
+	delete_vma_from_mm(vma);
+	down_write(&nommu_region_sem);
+	delete_nommu_region(vma->vm_region);
+	if (new_below) {
+		vma->vm_region->vm_start = vma->vm_start = addr;
+		vma->vm_region->vm_pgoff = vma->vm_pgoff += npages;
+	} else {
+		vma->vm_region->vm_end = vma->vm_end = addr;
 	}
+	add_nommu_region(vma->vm_region);
+	add_nommu_region(new->vm_region);
+	up_write(&nommu_region_sem);
+	add_vma_to_mm(mm, vma);
+	add_vma_to_mm(mm, new);
+	return 0;
 }
 
 /*
- * release a mapping
- * - under NOMMU conditions the parameters must match exactly to the mapping to
- *   be removed
+ * shrink a VMA by removing the specified chunk from either the beginning or
+ * the end
  */
-int do_munmap(struct mm_struct *mm, unsigned long addr, size_t len)
+static int shrink_vma(struct mm_struct *mm,
+		      struct vm_area_struct *vma,
+		      unsigned long from, unsigned long to)
 {
-	struct vm_list_struct *vml, **parent;
-	unsigned long end = addr + len;
+	struct vm_region *region;
 
-#ifdef DEBUG
-	printk("do_munmap:\n");
-#endif
+	kenter("");
 
-	for (parent = &mm->context.vmlist; *parent; parent = &(*parent)->next) {
-		if ((*parent)->vma->vm_start > addr)
-			break;
-		if ((*parent)->vma->vm_start == addr &&
-		    ((len == 0) || ((*parent)->vma->vm_end == end)))
-			goto found;
-	}
+	/* adjust the VMA's pointers, which may reposition it in the MM's tree
+	 * and list */
+	delete_vma_from_mm(vma);
+	if (from > vma->vm_start)
+		vma->vm_end = from;
+	else
+		vma->vm_start = to;
+	add_vma_to_mm(mm, vma);
 
-	printk("munmap of non-mmaped memory by process %d (%s): %p\n",
-	       current->pid, current->comm, (void *) addr);
-	return -EINVAL;
+	/* cut the backing region down to size */
+	region = vma->vm_region;
+	BUG_ON(atomic_read(&region->vm_usage) != 1);
 
- found:
-	vml = *parent;
+	down_write(&nommu_region_sem);
+	delete_nommu_region(region);
+	if (from > region->vm_start)
+		region->vm_end = from;
+	else
+		region->vm_start = to;
+	add_nommu_region(region);
+	up_write(&nommu_region_sem);
 
-	put_vma(mm, vml->vma);
+	free_page_series(from, to);
+	return 0;
+}
 
-	*parent = vml->next;
-	kfree(vml);
+/*
+ * release a mapping
+ * - under NOMMU conditions the chunk to be unmapped must be backed by a single
+ *   VMA, though it need not cover the whole VMA
+ */
+int do_munmap(struct mm_struct *mm, unsigned long start, size_t len)
+{
+	struct vm_area_struct *vma;
+	struct rb_node *rb;
+	unsigned long end = start + len;
+	int ret;
 
-	update_hiwater_vm(mm);
-	mm->total_vm -= len >> PAGE_SHIFT;
+	kenter(",%lx,%zx", start, len);
 
-#ifdef DEBUG
-	show_process_blocks();
-#endif
+	if (len == 0)
+		return -EINVAL;
+
+	/* find the first potentially overlapping VMA */
+	vma = find_vma(mm, start);
+	if (!vma) {
+		printk(KERN_WARNING
+		       "munmap of memory not mmapped by process %d (%s):"
+		       " 0x%lx-0x%lx\n",
+		       current->pid, current->comm, start, start + len - 1);
+		return -EINVAL;
+	}
 
+	/* we're allowed to split an anonymous VMA but not a file-backed one */
+	if (vma->vm_file) {
+		do {
+			if (start > vma->vm_start) {
+				kleave(" = -EINVAL [miss]");
+				return -EINVAL;
+			}
+			if (end == vma->vm_end)
+				goto erase_whole_vma;
+			rb = rb_next(&vma->vm_rb);
+			vma = rb_entry(rb, struct vm_area_struct, vm_rb);
+		} while (rb);
+		kleave(" = -EINVAL [split file]");
+		return -EINVAL;
+	} else {
+		/* the chunk must be a subset of the VMA found */
+		if (start == vma->vm_start && end == vma->vm_end)
+			goto erase_whole_vma;
+		if (start < vma->vm_start || end > vma->vm_end) {
+			kleave(" = -EINVAL [superset]");
+			return -EINVAL;
+		}
+		if (start & ~PAGE_MASK) {
+			kleave(" = -EINVAL [unaligned start]");
+			return -EINVAL;
+		}
+		if (end != vma->vm_end && end & ~PAGE_MASK) {
+			kleave(" = -EINVAL [unaligned split]");
+			return -EINVAL;
+		}
+		if (start != vma->vm_start && end != vma->vm_end) {
+			ret = split_vma(mm, vma, start, 1);
+			if (ret < 0) {
+				kleave(" = %d [split]", ret);
+				return ret;
+			}
+		}
+		return shrink_vma(mm, vma, start, end);
+	}
+
+erase_whole_vma:
+	delete_vma_from_mm(vma);
+	delete_vma(mm, vma);
+	kleave(" = 0");
 	return 0;
 }
 EXPORT_SYMBOL(do_munmap);
@@ -1184,29 +1552,26 @@ asmlinkage long sys_munmap(unsigned long addr, size_t len)
 }
 
 /*
- * Release all mappings
+ * release all the mappings made in a process's VM space
  */
-void exit_mmap(struct mm_struct * mm)
+void exit_mmap(struct mm_struct *mm)
 {
-	struct vm_list_struct *tmp;
+	struct vm_area_struct *vma;
 
-	if (mm) {
-#ifdef DEBUG
-		printk("Exit_mmap:\n");
-#endif
+	if (!mm)
+		return;
 
-		mm->total_vm = 0;
+	kenter("");
 
-		while ((tmp = mm->context.vmlist)) {
-			mm->context.vmlist = tmp->next;
-			put_vma(mm, tmp->vma);
-			kfree(tmp);
-		}
+	mm->total_vm = 0;
 
-#ifdef DEBUG
-		show_process_blocks();
-#endif
+	while ((vma = mm->mmap)) {
+		mm->mmap = vma->vm_next;
+		delete_vma_from_mm(vma);
+		delete_vma(mm, vma);
 	}
+
+	kleave("");
 }
 
 unsigned long do_brk(unsigned long addr, unsigned long len)
@@ -1219,8 +1584,8 @@ unsigned long do_brk(unsigned long addr, unsigned long len)
  * time (controlled by the MREMAP_MAYMOVE flag and available VM space)
  *
  * under NOMMU conditions, we only permit changing a mapping's size, and only
- * as long as it stays within the hole allocated by the kmalloc() call in
- * do_mmap_pgoff() and the block is not shareable
+ * as long as it stays within the region allocated by do_mmap_private() and the
+ * block is not shareable
  *
  * MREMAP_FIXED is not supported under NOMMU conditions
  */
@@ -1231,13 +1596,16 @@ unsigned long do_mremap(unsigned long addr,
 	struct vm_area_struct *vma;
 
 	/* insanity checks first */
-	if (new_len == 0)
+	if (old_len == 0 || new_len == 0)
 		return (unsigned long) -EINVAL;
 
+	if (addr & ~PAGE_MASK)
+		return -EINVAL;
+
 	if (flags & MREMAP_FIXED && new_addr != addr)
 		return (unsigned long) -EINVAL;
 
-	vma = find_vma_exact(current->mm, addr);
+	vma = find_vma_exact(current->mm, addr, old_len);
 	if (!vma)
 		return (unsigned long) -EINVAL;
 
@@ -1247,19 +1615,19 @@ unsigned long do_mremap(unsigned long addr,
 	if (vma->vm_flags & VM_MAYSHARE)
 		return (unsigned long) -EPERM;
 
-	if (new_len > kobjsize((void *) addr))
+	if (new_len > vma->vm_region->vm_end - vma->vm_region->vm_start)
 		return (unsigned long) -ENOMEM;
 
 	/* all checks complete - do it */
 	vma->vm_end = vma->vm_start + new_len;
-
 	return vma->vm_start;
 }
 EXPORT_SYMBOL(do_mremap);
 
-asmlinkage unsigned long sys_mremap(unsigned long addr,
-	unsigned long old_len, unsigned long new_len,
-	unsigned long flags, unsigned long new_addr)
+asmlinkage
+unsigned long sys_mremap(unsigned long addr,
+			 unsigned long old_len, unsigned long new_len,
+			 unsigned long flags, unsigned long new_addr)
 {
 	unsigned long ret;
 
-- 
cgit v1.2.3


From dd8632a12e500a684478fea0951f380478d56fed Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 8 Jan 2009 12:04:47 +0000
Subject: NOMMU: Make mmap allocation page trimming behaviour configurable.

NOMMU mmap allocates a piece of memory for an mmap that's rounded up in size to
the nearest power-of-2 number of pages.  Currently it then discards the excess
pages back to the page allocator, making that memory available for use by other
things.  This can, however, cause greater amount of fragmentation.

To counter this, a sysctl is added in order to fine-tune the trimming
behaviour.  The default behaviour remains to trim pages aggressively, while
this can either be disabled completely or set to a higher page-granular
watermark in order to have finer-grained control.

vm region vm_top bits taken from an earlier patch by David Howells.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: David Howells <dhowells@redhat.com>
Tested-by: Mike Frysinger <vapier.adi@gmail.com>
---
 Documentation/nommu-mmap.txt | 15 ++++++++++
 Documentation/sysctl/vm.txt  | 18 ++++++++++++
 include/linux/mm_types.h     |  1 +
 kernel/sysctl.c              | 14 ++++++++++
 mm/nommu.c                   | 65 ++++++++++++++++++++++++++++----------------
 5 files changed, 90 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/nommu-mmap.txt b/Documentation/nommu-mmap.txt
index 02b89dcf38ac..b565e8279d13 100644
--- a/Documentation/nommu-mmap.txt
+++ b/Documentation/nommu-mmap.txt
@@ -248,3 +248,18 @@ PROVIDING SHAREABLE BLOCK DEVICE SUPPORT
 Provision of shared mappings on block device files is exactly the same as for
 character devices. If there isn't a real device underneath, then the driver
 should allocate sufficient contiguous memory to honour any supported mapping.
+
+
+=================================
+ADJUSTING PAGE TRIMMING BEHAVIOUR
+=================================
+
+NOMMU mmap automatically rounds up to the nearest power-of-2 number of pages
+when performing an allocation.  This can have adverse effects on memory
+fragmentation, and as such, is left configurable.  The default behaviour is to
+aggressively trim allocations and discard any excess pages back in to the page
+allocator.  In order to retain finer-grained control over fragmentation, this
+behaviour can either be disabled completely, or bumped up to a higher page
+watermark where trimming begins.
+
+Page trimming behaviour is configurable via the sysctl `vm.nr_trim_pages'.
diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
index cd05994a49e6..a3415070bcac 100644
--- a/Documentation/sysctl/vm.txt
+++ b/Documentation/sysctl/vm.txt
@@ -38,6 +38,7 @@ Currently, these files are in /proc/sys/vm:
 - numa_zonelist_order
 - nr_hugepages
 - nr_overcommit_hugepages
+- nr_trim_pages		(only if CONFIG_MMU=n)
 
 ==============================================================
 
@@ -348,3 +349,20 @@ Change the maximum size of the hugepage pool. The maximum is
 nr_hugepages + nr_overcommit_hugepages.
 
 See Documentation/vm/hugetlbpage.txt
+
+==============================================================
+
+nr_trim_pages
+
+This is available only on NOMMU kernels.
+
+This value adjusts the excess page trimming behaviour of power-of-2 aligned
+NOMMU mmap allocations.
+
+A value of 0 disables trimming of allocations entirely, while a value of 1
+trims excess pages aggressively. Any value >= 1 acts as the watermark where
+trimming of allocations is initiated.
+
+The default value is 1.
+
+See Documentation/nommu-mmap.txt for more information.
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 1c1e0d3a1714..92915e81443f 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -106,6 +106,7 @@ struct vm_region {
 	unsigned long	vm_flags;	/* VMA vm_flags */
 	unsigned long	vm_start;	/* start address of region */
 	unsigned long	vm_end;		/* region initialised to here */
+	unsigned long	vm_top;		/* region allocated to here */
 	unsigned long	vm_pgoff;	/* the offset in vm_file corresponding to vm_start */
 	struct file	*vm_file;	/* the backing file or NULL */
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 92f6e5bc3c24..89d74436318c 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -82,6 +82,9 @@ extern int percpu_pagelist_fraction;
 extern int compat_log;
 extern int latencytop_enabled;
 extern int sysctl_nr_open_min, sysctl_nr_open_max;
+#ifndef CONFIG_MMU
+extern int sysctl_nr_trim_pages;
+#endif
 #ifdef CONFIG_RCU_TORTURE_TEST
 extern int rcutorture_runnable;
 #endif /* #ifdef CONFIG_RCU_TORTURE_TEST */
@@ -1102,6 +1105,17 @@ static struct ctl_table vm_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec
 	},
+#else
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "nr_trim_pages",
+		.data		= &sysctl_nr_trim_pages,
+		.maxlen		= sizeof(sysctl_nr_trim_pages),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &zero,
+	},
 #endif
 	{
 		.ctl_name	= VM_LAPTOP_MODE,
diff --git a/mm/nommu.c b/mm/nommu.c
index 0d363dfcf10e..a6e8ccfbd400 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -10,7 +10,7 @@
  *  Copyright (c) 2000-2003 David McCullough <davidm@snapgear.com>
  *  Copyright (c) 2000-2001 D Jeff Dionne <jeff@uClinux.org>
  *  Copyright (c) 2002      Greg Ungerer <gerg@snapgear.com>
- *  Copyright (c) 2007      Paul Mundt <lethal@linux-sh.org>
+ *  Copyright (c) 2007-2008 Paul Mundt <lethal@linux-sh.org>
  */
 
 #include <linux/module.h>
@@ -66,6 +66,7 @@ atomic_long_t vm_committed_space = ATOMIC_LONG_INIT(0);
 int sysctl_overcommit_memory = OVERCOMMIT_GUESS; /* heuristic overcommit */
 int sysctl_overcommit_ratio = 50; /* default is 50% */
 int sysctl_max_map_count = DEFAULT_MAX_MAP_COUNT;
+int sysctl_nr_trim_pages = 1; /* page trimming behaviour */
 int heap_stack_gap = 0;
 
 atomic_t mmap_pages_allocated;
@@ -455,6 +456,8 @@ static noinline void validate_nommu_regions(void)
 	last = rb_entry(lastp, struct vm_region, vm_rb);
 	if (unlikely(last->vm_end <= last->vm_start))
 		BUG();
+	if (unlikely(last->vm_top < last->vm_end))
+		BUG();
 
 	while ((p = rb_next(lastp))) {
 		region = rb_entry(p, struct vm_region, vm_rb);
@@ -462,7 +465,9 @@ static noinline void validate_nommu_regions(void)
 
 		if (unlikely(region->vm_end <= region->vm_start))
 			BUG();
-		if (unlikely(region->vm_start < last->vm_end))
+		if (unlikely(region->vm_top < region->vm_end))
+			BUG();
+		if (unlikely(region->vm_start < last->vm_top))
 			BUG();
 
 		lastp = p;
@@ -536,7 +541,7 @@ static void free_page_series(unsigned long from, unsigned long to)
 /*
  * release a reference to a region
  * - the caller must hold the region semaphore, which this releases
- * - the region may not have been added to the tree yet, in which case vm_end
+ * - the region may not have been added to the tree yet, in which case vm_top
  *   will equal vm_start
  */
 static void __put_nommu_region(struct vm_region *region)
@@ -547,7 +552,7 @@ static void __put_nommu_region(struct vm_region *region)
 	BUG_ON(!nommu_region_tree.rb_node);
 
 	if (atomic_dec_and_test(&region->vm_usage)) {
-		if (region->vm_end > region->vm_start)
+		if (region->vm_top > region->vm_start)
 			delete_nommu_region(region);
 		up_write(&nommu_region_sem);
 
@@ -558,7 +563,7 @@ static void __put_nommu_region(struct vm_region *region)
 		 * from ramfs/tmpfs mustn't be released here */
 		if (region->vm_flags & VM_MAPPED_COPY) {
 			kdebug("free series");
-			free_page_series(region->vm_start, region->vm_end);
+			free_page_series(region->vm_start, region->vm_top);
 		}
 		kmem_cache_free(vm_region_jar, region);
 	} else {
@@ -999,6 +1004,10 @@ static int do_mmap_shared_file(struct vm_area_struct *vma)
 	int ret;
 
 	ret = vma->vm_file->f_op->mmap(vma->vm_file, vma);
+	if (ret == 0) {
+		vma->vm_region->vm_top = vma->vm_region->vm_end;
+		return ret;
+	}
 	if (ret != -ENOSYS)
 		return ret;
 
@@ -1027,11 +1036,14 @@ static int do_mmap_private(struct vm_area_struct *vma,
 	 */
 	if (vma->vm_file) {
 		ret = vma->vm_file->f_op->mmap(vma->vm_file, vma);
-		if (ret != -ENOSYS) {
+		if (ret == 0) {
 			/* shouldn't return success if we're not sharing */
-			BUG_ON(ret == 0 && !(vma->vm_flags & VM_MAYSHARE));
-			return ret; /* success or a real error */
+			BUG_ON(!(vma->vm_flags & VM_MAYSHARE));
+			vma->vm_region->vm_top = vma->vm_region->vm_end;
+			return ret;
 		}
+		if (ret != -ENOSYS)
+			return ret;
 
 		/* getting an ENOSYS error indicates that direct mmap isn't
 		 * possible (as opposed to tried but failed) so we'll try to
@@ -1051,23 +1063,25 @@ static int do_mmap_private(struct vm_area_struct *vma,
 	if (!pages)
 		goto enomem;
 
-	/* we allocated a power-of-2 sized page set, so we need to trim off the
-	 * excess */
 	total = 1 << order;
 	atomic_add(total, &mmap_pages_allocated);
 
 	point = rlen >> PAGE_SHIFT;
-	while (total > point) {
-		order = ilog2(total - point);
-		n = 1 << order;
-		kdebug("shave %lu/%lu @%lu", n, total - point, total);
-		atomic_sub(n, &mmap_pages_allocated);
-		total -= n;
-		set_page_refcounted(pages + total);
-		__free_pages(pages + total, order);
+
+	/* we allocated a power-of-2 sized page set, so we may want to trim off
+	 * the excess */
+	if (sysctl_nr_trim_pages && total - point >= sysctl_nr_trim_pages) {
+		while (total > point) {
+			order = ilog2(total - point);
+			n = 1 << order;
+			kdebug("shave %lu/%lu @%lu", n, total - point, total);
+			atomic_sub(n, &mmap_pages_allocated);
+			total -= n;
+			set_page_refcounted(pages + total);
+			__free_pages(pages + total, order);
+		}
 	}
 
-	total = rlen >> PAGE_SHIFT;
 	for (point = 1; point < total; point++)
 		set_page_refcounted(&pages[point]);
 
@@ -1075,6 +1089,7 @@ static int do_mmap_private(struct vm_area_struct *vma,
 	region->vm_flags = vma->vm_flags |= VM_MAPPED_COPY;
 	region->vm_start = (unsigned long) base;
 	region->vm_end   = region->vm_start + rlen;
+	region->vm_top   = region->vm_start + (total << PAGE_SHIFT);
 
 	vma->vm_start = region->vm_start;
 	vma->vm_end   = region->vm_start + len;
@@ -1110,6 +1125,7 @@ error_free:
 	free_page_series(region->vm_start, region->vm_end);
 	region->vm_start = vma->vm_start = 0;
 	region->vm_end   = vma->vm_end = 0;
+	region->vm_top   = 0;
 	return ret;
 
 enomem:
@@ -1401,7 +1417,7 @@ int split_vma(struct mm_struct *mm, struct vm_area_struct *vma,
 	npages = (addr - vma->vm_start) >> PAGE_SHIFT;
 
 	if (new_below) {
-		region->vm_end = new->vm_end = addr;
+		region->vm_top = region->vm_end = new->vm_end = addr;
 	} else {
 		region->vm_start = new->vm_start = addr;
 		region->vm_pgoff = new->vm_pgoff += npages;
@@ -1418,6 +1434,7 @@ int split_vma(struct mm_struct *mm, struct vm_area_struct *vma,
 		vma->vm_region->vm_pgoff = vma->vm_pgoff += npages;
 	} else {
 		vma->vm_region->vm_end = vma->vm_end = addr;
+		vma->vm_region->vm_top = addr;
 	}
 	add_nommu_region(vma->vm_region);
 	add_nommu_region(new->vm_region);
@@ -1454,10 +1471,12 @@ static int shrink_vma(struct mm_struct *mm,
 
 	down_write(&nommu_region_sem);
 	delete_nommu_region(region);
-	if (from > region->vm_start)
-		region->vm_end = from;
-	else
+	if (from > region->vm_start) {
+		to = region->vm_top;
+		region->vm_top = region->vm_end = from;
+	} else {
 		region->vm_start = to;
+	}
 	add_nommu_region(region);
 	up_write(&nommu_region_sem);
 
-- 
cgit v1.2.3


From e2387d6c20752ccdb2895ba5de664fa39652f4cc Mon Sep 17 00:00:00 2001
From: Wolfram Sang <w.sang@pengutronix.de>
Date: Mon, 17 Nov 2008 14:35:44 +0000
Subject: leds: Make header variable naming consistent

There is one place where the struct led_classdev as the function
argument is named differently. Fix it.

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 include/linux/leds.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/leds.h b/include/linux/leds.h
index d3a73f5a48c3..3c1a8ce6a5ea 100644
--- a/include/linux/leds.h
+++ b/include/linux/leds.h
@@ -62,7 +62,7 @@ struct led_classdev {
 
 extern int led_classdev_register(struct device *parent,
 				 struct led_classdev *led_cdev);
-extern void led_classdev_unregister(struct led_classdev *lcd);
+extern void led_classdev_unregister(struct led_classdev *led_cdev);
 extern void led_classdev_suspend(struct led_classdev *led_cdev);
 extern void led_classdev_resume(struct led_classdev *led_cdev);
 
-- 
cgit v1.2.3


From 934cd3f979a1daacbd403398f2c7a8f6720c33aa Mon Sep 17 00:00:00 2001
From: Riku Voipio <riku.voipio@iki.fi>
Date: Wed, 3 Dec 2008 08:21:36 +0000
Subject: leds: leds-pcs9532 - Move i2c work to a workqueque

Apparently these might be called under atomic context,
and i2c operations may sleep. BUG found by
Ross Burton <ross@burtonini.com>

Signed-off-by: Riku Voipio <riku.voipio@iki.fi>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 drivers/leds/leds-pca9532.c  | 51 +++++++++++++++++++++++++++++++++++++-------
 include/linux/leds-pca9532.h |  2 ++
 2 files changed, 45 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/leds/leds-pca9532.c b/drivers/leds/leds-pca9532.c
index 62b60a038e2c..76ec7498e2d5 100644
--- a/drivers/leds/leds-pca9532.c
+++ b/drivers/leds/leds-pca9532.c
@@ -16,6 +16,7 @@
 #include <linux/leds.h>
 #include <linux/input.h>
 #include <linux/mutex.h>
+#include <linux/workqueue.h>
 #include <linux/leds-pca9532.h>
 
 static const unsigned short normal_i2c[] = { /*0x60,*/ I2C_CLIENT_END};
@@ -34,6 +35,7 @@ struct pca9532_data {
 	struct pca9532_led leds[16];
 	struct mutex update_lock;
 	struct input_dev    *idev;
+       struct work_struct work;
 	u8 pwm[2];
 	u8 psc[2];
 };
@@ -63,7 +65,7 @@ static struct i2c_driver pca9532_driver = {
  * as a compromise we average one pwm to the values requested by all
  * leds that are not ON/OFF.
  * */
-static int pca9532_setpwm(struct i2c_client *client, int pwm, int blink,
+static int pca9532_calcpwm(struct i2c_client *client, int pwm, int blink,
 	enum led_brightness value)
 {
 	int a = 0, b = 0, i = 0;
@@ -84,11 +86,17 @@ static int pca9532_setpwm(struct i2c_client *client, int pwm, int blink,
 	b = b/a;
 	if (b > 0xFF)
 		return -EINVAL;
-	mutex_lock(&data->update_lock);
 	data->pwm[pwm] = b;
+       data->psc[pwm] = blink;
+       return 0;
+}
+
+static int pca9532_setpwm(struct i2c_client *client, int pwm)
+{
+       struct pca9532_data *data = i2c_get_clientdata(client);
+       mutex_lock(&data->update_lock);
 	i2c_smbus_write_byte_data(client, PCA9532_REG_PWM(pwm),
 		data->pwm[pwm]);
-	data->psc[pwm] = blink;
 	i2c_smbus_write_byte_data(client, PCA9532_REG_PSC(pwm),
 		data->psc[pwm]);
 	mutex_unlock(&data->update_lock);
@@ -124,11 +132,11 @@ static void pca9532_set_brightness(struct led_classdev *led_cdev,
 		led->state = PCA9532_ON;
 	else {
 		led->state = PCA9532_PWM0; /* Thecus: hardcode one pwm */
-		err = pca9532_setpwm(led->client, 0, 0, value);
+               err = pca9532_calcpwm(led->client, 0, 0, value);
 		if (err)
 			return; /* XXX: led api doesn't allow error code? */
 	}
-	pca9532_setled(led);
+       schedule_work(&led->work);
 }
 
 static int pca9532_set_blink(struct led_classdev *led_cdev,
@@ -137,6 +145,7 @@ static int pca9532_set_blink(struct led_classdev *led_cdev,
 	struct pca9532_led *led = ldev_to_led(led_cdev);
 	struct i2c_client *client = led->client;
 	int psc;
+       int err = 0;
 
 	if (*delay_on == 0 && *delay_off == 0) {
 	/* led subsystem ask us for a blink rate */
@@ -148,7 +157,11 @@ static int pca9532_set_blink(struct led_classdev *led_cdev,
 
 	/* Thecus specific: only use PSC/PWM 0 */
 	psc = (*delay_on * 152-1)/1000;
-	return pca9532_setpwm(client, 0, psc, led_cdev->brightness);
+       err = pca9532_calcpwm(client, 0, psc, led_cdev->brightness);
+       if (err)
+               return err;
+       schedule_work(&led->work);
+       return 0;
 }
 
 static int pca9532_event(struct input_dev *dev, unsigned int type,
@@ -165,13 +178,28 @@ static int pca9532_event(struct input_dev *dev, unsigned int type,
 	else
 		data->pwm[1] = 0;
 
-	dev_info(&dev->dev, "setting beep to %d \n", data->pwm[1]);
+       schedule_work(&data->work);
+
+       return 0;
+}
+
+static void pca9532_input_work(struct work_struct *work)
+{
+       struct pca9532_data *data;
+       data = container_of(work, struct pca9532_data, work);
 	mutex_lock(&data->update_lock);
 	i2c_smbus_write_byte_data(data->client, PCA9532_REG_PWM(1),
 		data->pwm[1]);
 	mutex_unlock(&data->update_lock);
+}
 
-	return 0;
+static void pca9532_led_work(struct work_struct *work)
+{
+       struct pca9532_led *led;
+       led = container_of(work, struct pca9532_led, work);
+       if (led->state == PCA9532_PWM0)
+               pca9532_setpwm(led->client, 0);
+       pca9532_setled(led);
 }
 
 static int pca9532_configure(struct i2c_client *client,
@@ -204,6 +232,7 @@ static int pca9532_configure(struct i2c_client *client,
 			led->ldev.brightness = LED_OFF;
 			led->ldev.brightness_set = pca9532_set_brightness;
 			led->ldev.blink_set = pca9532_set_blink;
+                       INIT_WORK(&led->work, pca9532_led_work);
 			err = led_classdev_register(&client->dev, &led->ldev);
 			if (err < 0) {
 				dev_err(&client->dev,
@@ -233,9 +262,11 @@ static int pca9532_configure(struct i2c_client *client,
 						BIT_MASK(SND_TONE);
 			data->idev->event = pca9532_event;
 			input_set_drvdata(data->idev, data);
+                       INIT_WORK(&data->work, pca9532_input_work);
 			err = input_register_device(data->idev);
 			if (err) {
 				input_free_device(data->idev);
+                               cancel_work_sync(&data->work);
 				data->idev = NULL;
 				goto exit;
 			}
@@ -252,11 +283,13 @@ exit:
 				break;
 			case PCA9532_TYPE_LED:
 				led_classdev_unregister(&data->leds[i].ldev);
+                               cancel_work_sync(&data->leds[i].work);
 				break;
 			case PCA9532_TYPE_N2100_BEEP:
 				if (data->idev != NULL) {
 					input_unregister_device(data->idev);
 					input_free_device(data->idev);
+                                       cancel_work_sync(&data->work);
 					data->idev = NULL;
 				}
 				break;
@@ -307,11 +340,13 @@ static int pca9532_remove(struct i2c_client *client)
 			break;
 		case PCA9532_TYPE_LED:
 			led_classdev_unregister(&data->leds[i].ldev);
+                       cancel_work_sync(&data->leds[i].work);
 			break;
 		case PCA9532_TYPE_N2100_BEEP:
 			if (data->idev != NULL) {
 				input_unregister_device(data->idev);
 				input_free_device(data->idev);
+                               cancel_work_sync(&data->work);
 				data->idev = NULL;
 			}
 			break;
diff --git a/include/linux/leds-pca9532.h b/include/linux/leds-pca9532.h
index 81b4207deb95..96eea90f01a8 100644
--- a/include/linux/leds-pca9532.h
+++ b/include/linux/leds-pca9532.h
@@ -15,6 +15,7 @@
 #define __LINUX_PCA9532_H
 
 #include <linux/leds.h>
+#include <linux/workqueue.h>
 
 enum pca9532_state {
 	PCA9532_OFF  = 0x0,
@@ -31,6 +32,7 @@ struct pca9532_led {
 	struct i2c_client *client;
 	char *name;
 	struct led_classdev ldev;
+       struct work_struct work;
 	enum pca9532_type type;
 	enum pca9532_state state;
 };
-- 
cgit v1.2.3


From 0081e8020ebd814a99e45720a10e869a54ee08a6 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Thu, 4 Dec 2008 16:52:33 +0000
Subject: leds: Add WM8350 LED driver

The voltage and current regulators on the WM8350 AudioPlus PMIC can be
used in concert to provide a power efficient LED driver.  This driver
implements support for this within the standard LED class.

Platform initialisation code should configure the LED hardware in the
init callback provided by the WM8350 core driver.  The callback should
use wm8350_isink_set_flash(), wm8350_dcdc25_set_mode() and
wm8350_dcdc_set_slot() to configure the operating parameters of the
regulators for their hardware and then then use wm8350_register_led() to
instantiate the LED driver.

This driver was originally written by Liam Girdwood, though it has been
extensively modified since then.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 drivers/leds/Kconfig                 |   7 +
 drivers/leds/Makefile                |   1 +
 drivers/leds/leds-wm8350.c           | 333 +++++++++++++++++++++++++++++++++++
 drivers/mfd/wm8350-core.c            |   3 +
 drivers/regulator/wm8350-regulator.c |  91 ++++++++++
 include/linux/mfd/wm8350/pmic.h      |  36 ++++
 6 files changed, 471 insertions(+)
 create mode 100644 drivers/leds/leds-wm8350.c

(limited to 'include/linux')

diff --git a/drivers/leds/Kconfig b/drivers/leds/Kconfig
index 3d0d2625f6ab..a4a1ae214630 100644
--- a/drivers/leds/Kconfig
+++ b/drivers/leds/Kconfig
@@ -164,6 +164,13 @@ config LEDS_PCA955X
 	  LED driver chips accessed via the I2C bus.  Supported
 	  devices include PCA9550, PCA9551, PCA9552, and PCA9553.
 
+config LEDS_WM8350
+	tristate "LED Support for WM8350 AudioPlus PMIC"
+	depends on LEDS_CLASS && MFD_WM8350
+	help
+	  This option enables support for LEDs driven by the Wolfson
+	  Microelectronics WM8350 AudioPlus PMIC.
+
 config LEDS_DA903X
 	tristate "LED Support for DA9030/DA9034 PMIC"
 	depends on LEDS_CLASS && PMIC_DA903X
diff --git a/drivers/leds/Makefile b/drivers/leds/Makefile
index ff4616a49043..bc247cb02e82 100644
--- a/drivers/leds/Makefile
+++ b/drivers/leds/Makefile
@@ -24,6 +24,7 @@ obj-$(CONFIG_LEDS_FSG)			+= leds-fsg.o
 obj-$(CONFIG_LEDS_PCA955X)		+= leds-pca955x.o
 obj-$(CONFIG_LEDS_DA903X)		+= leds-da903x.o
 obj-$(CONFIG_LEDS_HP_DISK)		+= leds-hp-disk.o
+obj-$(CONFIG_LEDS_WM8350)		+= leds-wm8350.o
 
 # LED Triggers
 obj-$(CONFIG_LEDS_TRIGGER_TIMER)	+= ledtrig-timer.o
diff --git a/drivers/leds/leds-wm8350.c b/drivers/leds/leds-wm8350.c
new file mode 100644
index 000000000000..846ed978103f
--- /dev/null
+++ b/drivers/leds/leds-wm8350.c
@@ -0,0 +1,333 @@
+/*
+ * LED driver for WM8350 driven LEDS.
+ *
+ * Copyright(C) 2007, 2008 Wolfson Microelectronics PLC.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/leds.h>
+#include <linux/err.h>
+#include <linux/mfd/wm8350/pmic.h>
+#include <linux/regulator/consumer.h>
+
+/* Microamps */
+static const int isink_cur[] = {
+	4,
+	5,
+	6,
+	7,
+	8,
+	10,
+	11,
+	14,
+	16,
+	19,
+	23,
+	27,
+	32,
+	39,
+	46,
+	54,
+	65,
+	77,
+	92,
+	109,
+	130,
+	154,
+	183,
+	218,
+	259,
+	308,
+	367,
+	436,
+	518,
+	616,
+	733,
+	872,
+	1037,
+	1233,
+	1466,
+	1744,
+	2073,
+	2466,
+	2933,
+	3487,
+	4147,
+	4932,
+	5865,
+	6975,
+	8294,
+	9864,
+	11730,
+	13949,
+	16589,
+	19728,
+	23460,
+	27899,
+	33178,
+	39455,
+	46920,
+	55798,
+	66355,
+	78910,
+	93840,
+	111596,
+	132710,
+	157820,
+	187681,
+	223191
+};
+
+#define to_wm8350_led(led_cdev) \
+	container_of(led_cdev, struct wm8350_led, cdev)
+
+static void wm8350_led_enable(struct wm8350_led *led)
+{
+	int ret;
+
+	if (led->enabled)
+		return;
+
+	ret = regulator_enable(led->isink);
+	if (ret != 0) {
+		dev_err(led->cdev.dev, "Failed to enable ISINK: %d\n", ret);
+		return;
+	}
+
+	ret = regulator_enable(led->dcdc);
+	if (ret != 0) {
+		dev_err(led->cdev.dev, "Failed to enable DCDC: %d\n", ret);
+		regulator_disable(led->isink);
+		return;
+	}
+
+	led->enabled = 1;
+}
+
+static void wm8350_led_disable(struct wm8350_led *led)
+{
+	int ret;
+
+	if (!led->enabled)
+		return;
+
+	ret = regulator_disable(led->dcdc);
+	if (ret != 0) {
+		dev_err(led->cdev.dev, "Failed to disable DCDC: %d\n", ret);
+		return;
+	}
+
+	ret = regulator_disable(led->isink);
+	if (ret != 0) {
+		dev_err(led->cdev.dev, "Failed to disable ISINK: %d\n", ret);
+		regulator_enable(led->dcdc);
+		return;
+	}
+
+	led->enabled = 0;
+}
+
+static void led_work(struct work_struct *work)
+{
+	struct wm8350_led *led = container_of(work, struct wm8350_led, work);
+	int ret;
+	int uA;
+	unsigned long flags;
+
+	mutex_lock(&led->mutex);
+
+	spin_lock_irqsave(&led->value_lock, flags);
+
+	if (led->value == LED_OFF) {
+		spin_unlock_irqrestore(&led->value_lock, flags);
+		wm8350_led_disable(led);
+		goto out;
+	}
+
+	/* This scales linearly into the index of valid current
+	 * settings which results in a linear scaling of perceived
+	 * brightness due to the non-linear current settings provided
+	 * by the hardware.
+	 */
+	uA = (led->max_uA_index * led->value) / LED_FULL;
+	spin_unlock_irqrestore(&led->value_lock, flags);
+	BUG_ON(uA >= ARRAY_SIZE(isink_cur));
+
+	ret = regulator_set_current_limit(led->isink, isink_cur[uA],
+					  isink_cur[uA]);
+	if (ret != 0)
+		dev_err(led->cdev.dev, "Failed to set %duA: %d\n",
+			isink_cur[uA], ret);
+
+	wm8350_led_enable(led);
+
+out:
+	mutex_unlock(&led->mutex);
+}
+
+static void wm8350_led_set(struct led_classdev *led_cdev,
+			   enum led_brightness value)
+{
+	struct wm8350_led *led = to_wm8350_led(led_cdev);
+	unsigned long flags;
+
+	spin_lock_irqsave(&led->value_lock, flags);
+	led->value = value;
+	schedule_work(&led->work);
+	spin_unlock_irqrestore(&led->value_lock, flags);
+}
+
+#ifdef CONFIG_PM
+static int wm8350_led_suspend(struct platform_device *pdev, pm_message_t state)
+{
+	struct wm8350_led *led = platform_get_drvdata(pdev);
+
+	led_classdev_suspend(&led->cdev);
+	return 0;
+}
+
+static int wm8350_led_resume(struct platform_device *pdev)
+{
+	struct wm8350_led *led = platform_get_drvdata(pdev);
+
+	led_classdev_resume(&led->cdev);
+	return 0;
+}
+#else
+#define wm8350_led_suspend NULL
+#define wm8350_led_resume NULL
+#endif
+
+static void wm8350_led_shutdown(struct platform_device *pdev)
+{
+	struct wm8350_led *led = platform_get_drvdata(pdev);
+
+	mutex_lock(&led->mutex);
+	led->value = LED_OFF;
+	wm8350_led_disable(led);
+	mutex_unlock(&led->mutex);
+}
+
+static int wm8350_led_probe(struct platform_device *pdev)
+{
+	struct regulator *isink, *dcdc;
+	struct wm8350_led *led;
+	struct wm8350_led_platform_data *pdata = pdev->dev.platform_data;
+	int ret, i;
+
+	if (pdata == NULL) {
+		dev_err(&pdev->dev, "no platform data\n");
+		return -ENODEV;
+	}
+
+	if (pdata->max_uA < isink_cur[0]) {
+		dev_err(&pdev->dev, "Invalid maximum current %duA\n",
+			pdata->max_uA);
+		return -EINVAL;
+	}
+
+	isink = regulator_get(&pdev->dev, "led_isink");
+	if (IS_ERR(isink)) {
+		printk(KERN_ERR "%s: cant get ISINK\n", __func__);
+		return PTR_ERR(isink);
+	}
+
+	dcdc = regulator_get(&pdev->dev, "led_vcc");
+	if (IS_ERR(dcdc)) {
+		printk(KERN_ERR "%s: cant get DCDC\n", __func__);
+		ret = PTR_ERR(dcdc);
+		goto err_isink;
+	}
+
+	led = kzalloc(sizeof(*led), GFP_KERNEL);
+	if (led == NULL) {
+		ret = -ENOMEM;
+		goto err_dcdc;
+	}
+
+	led->cdev.brightness_set = wm8350_led_set;
+	led->cdev.default_trigger = pdata->default_trigger;
+	led->cdev.name = pdata->name;
+	led->enabled = regulator_is_enabled(isink);
+	led->isink = isink;
+	led->dcdc = dcdc;
+
+	for (i = 0; i < ARRAY_SIZE(isink_cur) - 1; i++)
+		if (isink_cur[i] >= pdata->max_uA)
+			break;
+	led->max_uA_index = i;
+	if (pdata->max_uA != isink_cur[i])
+		dev_warn(&pdev->dev,
+			 "Maximum current %duA is not directly supported,"
+			 " check platform data\n",
+			 pdata->max_uA);
+
+	spin_lock_init(&led->value_lock);
+	mutex_init(&led->mutex);
+	INIT_WORK(&led->work, led_work);
+	led->value = LED_OFF;
+	platform_set_drvdata(pdev, led);
+
+	ret = led_classdev_register(&pdev->dev, &led->cdev);
+	if (ret < 0)
+		goto err_led;
+
+	return 0;
+
+ err_led:
+	kfree(led);
+ err_dcdc:
+	regulator_put(dcdc);
+ err_isink:
+	regulator_put(isink);
+	return ret;
+}
+
+static int wm8350_led_remove(struct platform_device *pdev)
+{
+	struct wm8350_led *led = platform_get_drvdata(pdev);
+
+	led_classdev_unregister(&led->cdev);
+	flush_scheduled_work();
+	wm8350_led_disable(led);
+	regulator_put(led->dcdc);
+	regulator_put(led->isink);
+	kfree(led);
+	return 0;
+}
+
+static struct platform_driver wm8350_led_driver = {
+	.driver = {
+		   .name = "wm8350-led",
+		   .owner = THIS_MODULE,
+		   },
+	.probe = wm8350_led_probe,
+	.remove = wm8350_led_remove,
+	.shutdown = wm8350_led_shutdown,
+	.suspend = wm8350_led_suspend,
+	.resume = wm8350_led_resume,
+};
+
+static int __devinit wm8350_led_init(void)
+{
+	return platform_driver_register(&wm8350_led_driver);
+}
+module_init(wm8350_led_init);
+
+static void wm8350_led_exit(void)
+{
+	platform_driver_unregister(&wm8350_led_driver);
+}
+module_exit(wm8350_led_exit);
+
+MODULE_AUTHOR("Mark Brown");
+MODULE_DESCRIPTION("WM8350 LED driver");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:wm8350-led");
diff --git a/drivers/mfd/wm8350-core.c b/drivers/mfd/wm8350-core.c
index 3a273ccef3f2..f92595c8f165 100644
--- a/drivers/mfd/wm8350-core.c
+++ b/drivers/mfd/wm8350-core.c
@@ -1453,6 +1453,9 @@ void wm8350_device_exit(struct wm8350 *wm8350)
 {
 	int i;
 
+	for (i = 0; i < ARRAY_SIZE(wm8350->pmic.led); i++)
+		platform_device_unregister(wm8350->pmic.led[i].pdev);
+
 	for (i = 0; i < ARRAY_SIZE(wm8350->pmic.pdev); i++)
 		platform_device_unregister(wm8350->pmic.pdev[i]);
 
diff --git a/drivers/regulator/wm8350-regulator.c b/drivers/regulator/wm8350-regulator.c
index c68c496b2c49..7aa35248181b 100644
--- a/drivers/regulator/wm8350-regulator.c
+++ b/drivers/regulator/wm8350-regulator.c
@@ -1412,6 +1412,97 @@ int wm8350_register_regulator(struct wm8350 *wm8350, int reg,
 }
 EXPORT_SYMBOL_GPL(wm8350_register_regulator);
 
+/**
+ * wm8350_register_led - Register a WM8350 LED output
+ *
+ * @param wm8350 The WM8350 device to configure.
+ * @param lednum LED device index to create.
+ * @param dcdc The DCDC to use for the LED.
+ * @param isink The ISINK to use for the LED.
+ * @param pdata Configuration for the LED.
+ *
+ * The WM8350 supports the use of an ISINK together with a DCDC to
+ * provide a power-efficient LED driver.  This function registers the
+ * regulators and instantiates the platform device for a LED.  The
+ * operating modes for the LED regulators must be configured using
+ * wm8350_isink_set_flash(), wm8350_dcdc25_set_mode() and
+ * wm8350_dcdc_set_slot() prior to calling this function.
+ */
+int wm8350_register_led(struct wm8350 *wm8350, int lednum, int dcdc, int isink,
+			struct wm8350_led_platform_data *pdata)
+{
+	struct wm8350_led *led;
+	struct platform_device *pdev;
+	int ret;
+
+	if (lednum > ARRAY_SIZE(wm8350->pmic.led) || lednum < 0) {
+		dev_err(wm8350->dev, "Invalid LED index %d\n", lednum);
+		return -ENODEV;
+	}
+
+	led = &wm8350->pmic.led[lednum];
+
+	if (led->pdev) {
+		dev_err(wm8350->dev, "LED %d already allocated\n", lednum);
+		return -EINVAL;
+	}
+
+	pdev = platform_device_alloc("wm8350-led", lednum);
+	if (pdev == NULL) {
+		dev_err(wm8350->dev, "Failed to allocate LED %d\n", lednum);
+		return -ENOMEM;
+	}
+
+	led->isink_consumer.dev = &pdev->dev;
+	led->isink_consumer.supply = "led_isink";
+	led->isink_init.num_consumer_supplies = 1;
+	led->isink_init.consumer_supplies = &led->isink_consumer;
+	led->isink_init.constraints.min_uA = 0;
+	led->isink_init.constraints.max_uA = pdata->max_uA;
+	led->isink_init.constraints.valid_ops_mask = REGULATOR_CHANGE_CURRENT;
+	led->isink_init.constraints.valid_modes_mask = REGULATOR_MODE_NORMAL;
+	ret = wm8350_register_regulator(wm8350, isink, &led->isink_init);
+	if (ret != 0) {
+		platform_device_put(pdev);
+		return ret;
+	}
+
+	led->dcdc_consumer.dev = &pdev->dev;
+	led->dcdc_consumer.supply = "led_vcc";
+	led->dcdc_init.num_consumer_supplies = 1;
+	led->dcdc_init.consumer_supplies = &led->dcdc_consumer;
+	led->dcdc_init.constraints.valid_modes_mask = REGULATOR_MODE_NORMAL;
+	ret = wm8350_register_regulator(wm8350, dcdc, &led->dcdc_init);
+	if (ret != 0) {
+		platform_device_put(pdev);
+		return ret;
+	}
+
+	switch (isink) {
+	case WM8350_ISINK_A:
+		wm8350->pmic.isink_A_dcdc = dcdc;
+		break;
+	case WM8350_ISINK_B:
+		wm8350->pmic.isink_B_dcdc = dcdc;
+		break;
+	}
+
+	pdev->dev.platform_data = pdata;
+	pdev->dev.parent = wm8350->dev;
+	ret = platform_device_add(pdev);
+	if (ret != 0) {
+		dev_err(wm8350->dev, "Failed to register LED %d: %d\n",
+			lednum, ret);
+		platform_device_put(pdev);
+		return ret;
+	}
+
+	led->pdev = pdev;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(wm8350_register_led);
+
 static struct platform_driver wm8350_regulator_driver = {
 	.probe = wm8350_regulator_probe,
 	.remove = wm8350_regulator_remove,
diff --git a/include/linux/mfd/wm8350/pmic.h b/include/linux/mfd/wm8350/pmic.h
index 96acbfc8aa12..be3264e286e0 100644
--- a/include/linux/mfd/wm8350/pmic.h
+++ b/include/linux/mfd/wm8350/pmic.h
@@ -13,6 +13,10 @@
 #ifndef __LINUX_MFD_WM8350_PMIC_H
 #define __LINUX_MFD_WM8350_PMIC_H
 
+#include <linux/platform_device.h>
+#include <linux/leds.h>
+#include <linux/regulator/machine.h>
+
 /*
  * Register values.
  */
@@ -700,6 +704,33 @@ struct wm8350;
 struct platform_device;
 struct regulator_init_data;
 
+/*
+ * WM8350 LED platform data
+ */
+struct wm8350_led_platform_data {
+	const char *name;
+	const char *default_trigger;
+	int max_uA;
+};
+
+struct wm8350_led {
+	struct platform_device *pdev;
+	struct mutex mutex;
+	struct work_struct work;
+	spinlock_t value_lock;
+	enum led_brightness value;
+	struct led_classdev cdev;
+	int max_uA_index;
+	int enabled;
+
+	struct regulator *isink;
+	struct regulator_consumer_supply isink_consumer;
+	struct regulator_init_data isink_init;
+	struct regulator *dcdc;
+	struct regulator_consumer_supply dcdc_consumer;
+	struct regulator_init_data dcdc_init;
+};
+
 struct wm8350_pmic {
 	/* Number of regulators of each type on this device */
 	int max_dcdc;
@@ -717,10 +748,15 @@ struct wm8350_pmic {
 
 	/* regulator devices */
 	struct platform_device *pdev[NUM_WM8350_REGULATORS];
+
+	/* LED devices */
+	struct wm8350_led led[2];
 };
 
 int wm8350_register_regulator(struct wm8350 *wm8350, int reg,
 			      struct regulator_init_data *initdata);
+int wm8350_register_led(struct wm8350 *wm8350, int lednum, int dcdc, int isink,
+			struct wm8350_led_platform_data *pdata);
 
 /*
  * Additional DCDC control not supported via regulator API
-- 
cgit v1.2.3


From c835ee7f4154992e6cf0674d7ee136f5d36247a4 Mon Sep 17 00:00:00 2001
From: Richard Purdie <rpurdie@linux.intel.com>
Date: Tue, 6 Jan 2009 21:00:19 +0000
Subject: backlight: Add suspend/resume support to the backlight core

Add suspend/resume support to the backlight core and enable use of it
by appropriate drivers.

Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 drivers/video/backlight/backlight.c     | 34 +++++++++++++++++++++++++
 drivers/video/backlight/corgi_bl.c      | 45 +++++++++------------------------
 drivers/video/backlight/mbp_nvidia_bl.c |  1 +
 include/linux/backlight.h               | 16 ++++++++++++
 4 files changed, 63 insertions(+), 33 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/video/backlight/backlight.c b/drivers/video/backlight/backlight.c
index a9c013bb9f20..157057c79ca3 100644
--- a/drivers/video/backlight/backlight.c
+++ b/drivers/video/backlight/backlight.c
@@ -40,6 +40,10 @@ static int fb_notifier_callback(struct notifier_block *self,
 		if (!bd->ops->check_fb ||
 		    bd->ops->check_fb(evdata->info)) {
 			bd->props.fb_blank = *(int *)evdata->data;
+			if (bd->props.fb_blank == FB_BLANK_UNBLANK)
+				bd->props.state &= ~BL_CORE_FBBLANK;
+			else
+				bd->props.state |= BL_CORE_FBBLANK;
 			backlight_update_status(bd);
 		}
 	mutex_unlock(&bd->ops_lock);
@@ -165,6 +169,34 @@ static ssize_t backlight_show_actual_brightness(struct device *dev,
 
 static struct class *backlight_class;
 
+static int backlight_suspend(struct device *dev, pm_message_t state)
+{
+	struct backlight_device *bd = to_backlight_device(dev);
+
+	if (bd->ops->options & BL_CORE_SUSPENDRESUME) {
+		mutex_lock(&bd->ops_lock);
+		bd->props.state |= BL_CORE_SUSPENDED;
+		backlight_update_status(bd);
+		mutex_unlock(&bd->ops_lock);
+	}
+
+	return 0;
+}
+
+static int backlight_resume(struct device *dev)
+{
+	struct backlight_device *bd = to_backlight_device(dev);
+
+	if (bd->ops->options & BL_CORE_SUSPENDRESUME) {
+		mutex_lock(&bd->ops_lock);
+		bd->props.state &= ~BL_CORE_SUSPENDED;
+		backlight_update_status(bd);
+		mutex_unlock(&bd->ops_lock);
+	}
+
+	return 0;
+}
+
 static void bl_device_release(struct device *dev)
 {
 	struct backlight_device *bd = to_backlight_device(dev);
@@ -281,6 +313,8 @@ static int __init backlight_class_init(void)
 	}
 
 	backlight_class->dev_attrs = bl_device_attributes;
+	backlight_class->suspend = backlight_suspend;
+	backlight_class->resume = backlight_resume;
 	return 0;
 }
 
diff --git a/drivers/video/backlight/corgi_bl.c b/drivers/video/backlight/corgi_bl.c
index 4d4d037e3ec9..c6abf51ee3e0 100644
--- a/drivers/video/backlight/corgi_bl.c
+++ b/drivers/video/backlight/corgi_bl.c
@@ -24,9 +24,8 @@ static struct backlight_properties corgibl_data;
 static struct backlight_device *corgi_backlight_device;
 static struct generic_bl_info *bl_machinfo;
 
-static unsigned long corgibl_flags;
-#define CORGIBL_SUSPENDED     0x01
-#define CORGIBL_BATTLOW       0x02
+/* Flag to signal when the battery is low */
+#define CORGIBL_BATTLOW       BL_CORE_DRIVER1
 
 static int corgibl_send_intensity(struct backlight_device *bd)
 {
@@ -34,11 +33,11 @@ static int corgibl_send_intensity(struct backlight_device *bd)
 
 	if (bd->props.power != FB_BLANK_UNBLANK)
 		intensity = 0;
-	if (bd->props.fb_blank != FB_BLANK_UNBLANK)
+	if (bd->props.state & BL_CORE_FBBLANK)
 		intensity = 0;
-	if (corgibl_flags & CORGIBL_SUSPENDED)
+	if (bd->props.state & BL_CORE_SUSPENDED)
 		intensity = 0;
-	if (corgibl_flags & CORGIBL_BATTLOW)
+	if (bd->props.state & CORGIBL_BATTLOW)
 		intensity &= bl_machinfo->limit_mask;
 
 	bl_machinfo->set_bl_intensity(intensity);
@@ -51,29 +50,6 @@ static int corgibl_send_intensity(struct backlight_device *bd)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int corgibl_suspend(struct platform_device *pdev, pm_message_t state)
-{
-	struct backlight_device *bd = platform_get_drvdata(pdev);
-
-	corgibl_flags |= CORGIBL_SUSPENDED;
-	backlight_update_status(bd);
-	return 0;
-}
-
-static int corgibl_resume(struct platform_device *pdev)
-{
-	struct backlight_device *bd = platform_get_drvdata(pdev);
-
-	corgibl_flags &= ~CORGIBL_SUSPENDED;
-	backlight_update_status(bd);
-	return 0;
-}
-#else
-#define corgibl_suspend	NULL
-#define corgibl_resume	NULL
-#endif
-
 static int corgibl_get_intensity(struct backlight_device *bd)
 {
 	return corgibl_intensity;
@@ -85,16 +61,21 @@ static int corgibl_get_intensity(struct backlight_device *bd)
  */
 void corgibl_limit_intensity(int limit)
 {
+	struct backlight_device *bd = corgi_backlight_device;
+
+	mutex_lock(&bd->ops_lock);
 	if (limit)
-		corgibl_flags |= CORGIBL_BATTLOW;
+		bd->props.state |= CORGIBL_BATTLOW;
 	else
-		corgibl_flags &= ~CORGIBL_BATTLOW;
+		bd->props.state &= ~CORGIBL_BATTLOW;
 	backlight_update_status(corgi_backlight_device);
+	mutex_unlock(&bd->ops_lock);
 }
 EXPORT_SYMBOL(corgibl_limit_intensity);
 
 
 static struct backlight_ops corgibl_ops = {
+	.options = BL_CORE_SUSPENDRESUME,
 	.get_brightness = corgibl_get_intensity,
 	.update_status  = corgibl_send_intensity,
 };
@@ -144,8 +125,6 @@ static int corgibl_remove(struct platform_device *pdev)
 static struct platform_driver corgibl_driver = {
 	.probe		= corgibl_probe,
 	.remove		= corgibl_remove,
-	.suspend	= corgibl_suspend,
-	.resume		= corgibl_resume,
 	.driver		= {
 		.name	= "generic-bl",
 	},
diff --git a/drivers/video/backlight/mbp_nvidia_bl.c b/drivers/video/backlight/mbp_nvidia_bl.c
index 06964af761c6..65864c500455 100644
--- a/drivers/video/backlight/mbp_nvidia_bl.c
+++ b/drivers/video/backlight/mbp_nvidia_bl.c
@@ -70,6 +70,7 @@ static int mbp_get_intensity(struct backlight_device *bd)
 }
 
 static struct backlight_ops mbp_ops = {
+	.options = BL_CORE_SUSPENDRESUME,
 	.get_brightness = mbp_get_intensity,
 	.update_status  = mbp_send_intensity,
 };
diff --git a/include/linux/backlight.h b/include/linux/backlight.h
index 1ee9488ca2e4..79ca2da81c87 100644
--- a/include/linux/backlight.h
+++ b/include/linux/backlight.h
@@ -31,6 +31,10 @@ struct backlight_device;
 struct fb_info;
 
 struct backlight_ops {
+	unsigned int options;
+
+#define BL_CORE_SUSPENDRESUME	(1 << 0)
+
 	/* Notify the backlight driver some property has changed */
 	int (*update_status)(struct backlight_device *);
 	/* Return the current backlight brightness (accounting for power,
@@ -51,7 +55,19 @@ struct backlight_properties {
 	   modes; 4: full off), see FB_BLANK_XXX */
 	int power;
 	/* FB Blanking active? (values as for power) */
+	/* Due to be removed, please use (state & BL_CORE_FBBLANK) */
 	int fb_blank;
+	/* Flags used to signal drivers of state changes */
+	/* Upper 4 bits are reserved for driver internal use */
+	unsigned int state;
+
+#define BL_CORE_SUSPENDED	(1 << 0)	/* backlight is suspended */
+#define BL_CORE_FBBLANK		(1 << 1)	/* backlight is under an fb blank event */
+#define BL_CORE_DRIVER4		(1 << 28)	/* reserved for driver specific use */
+#define BL_CORE_DRIVER3		(1 << 29)	/* reserved for driver specific use */
+#define BL_CORE_DRIVER2		(1 << 30)	/* reserved for driver specific use */
+#define BL_CORE_DRIVER1		(1 << 31)	/* reserved for driver specific use */
+
 };
 
 struct backlight_device {
-- 
cgit v1.2.3


From 1107ba885e46964316c083d441d5dd185b6c9e49 Mon Sep 17 00:00:00 2001
From: Alex Zeffertt <alex.zeffertt@eu.citrix.com>
Date: Wed, 7 Jan 2009 18:07:11 -0800
Subject: xen: add xenfs to allow usermode <-> Xen interaction

The xenfs filesystem exports various interfaces to usermode.  Initially
this exports a file to allow usermode to interact with xenbus/xenstore.

Traditionally this appeared in /proc/xen.  Rather than extending procfs,
this patch adds a backward-compat mountpoint on /proc/xen, and provides
a xenfs filesystem which can be mounted there.

Signed-off-by: Alex Zeffertt <alex.zeffertt@eu.citrix.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/xen/Kconfig               |  24 ++
 drivers/xen/Makefile              |   2 +
 drivers/xen/xenbus/xenbus_probe.c |  28 +-
 drivers/xen/xenbus/xenbus_xs.c    |   1 +
 drivers/xen/xenfs/Makefile        |   3 +
 drivers/xen/xenfs/super.c         |  64 ++++
 drivers/xen/xenfs/xenbus.c        | 593 ++++++++++++++++++++++++++++++++++++++
 drivers/xen/xenfs/xenfs.h         |   6 +
 include/linux/magic.h             |   1 +
 include/xen/xenbus.h              |   2 -
 10 files changed, 716 insertions(+), 8 deletions(-)
 create mode 100644 drivers/xen/xenfs/Makefile
 create mode 100644 drivers/xen/xenfs/super.c
 create mode 100644 drivers/xen/xenfs/xenbus.c
 create mode 100644 drivers/xen/xenfs/xenfs.h

(limited to 'include/linux')

diff --git a/drivers/xen/Kconfig b/drivers/xen/Kconfig
index 4b75a16de009..526187c8a12d 100644
--- a/drivers/xen/Kconfig
+++ b/drivers/xen/Kconfig
@@ -17,3 +17,27 @@ config XEN_SCRUB_PAGES
 	  is not accidentally visible to other domains.  Is it more
 	  secure, but slightly less efficient.
 	  If in doubt, say yes.
+
+config XENFS
+	tristate "Xen filesystem"
+	depends on XEN
+	default y
+	help
+	  The xen filesystem provides a way for domains to share
+	  information with each other and with the hypervisor.
+	  For example, by reading and writing the "xenbus" file, guests
+	  may pass arbitrary information to the initial domain.
+	  If in doubt, say yes.
+
+config XEN_COMPAT_XENFS
+       bool "Create compatibility mount point /proc/xen"
+       depends on XENFS
+       default y
+       help
+         The old xenstore userspace tools expect to find "xenbus"
+         under /proc/xen, but "xenbus" is now found at the root of the
+         xenfs filesystem.  Selecting this causes the kernel to create
+         the compatibilty mount point /proc/xen if it is running on
+         a xen platform.
+         If in doubt, say yes.
+
diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile
index d2a8fdf0e191..ff8accc9e103 100644
--- a/drivers/xen/Makefile
+++ b/drivers/xen/Makefile
@@ -1,5 +1,7 @@
 obj-y	+= grant-table.o features.o events.o manage.o
 obj-y	+= xenbus/
+
 obj-$(CONFIG_HOTPLUG_CPU)	+= cpu_hotplug.o
 obj-$(CONFIG_XEN_XENCOMM)	+= xencomm.o
 obj-$(CONFIG_XEN_BALLOON)	+= balloon.o
+obj-$(CONFIG_XENFS)		+= xenfs/
\ No newline at end of file
diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c
index b2a03184a246..773d1cf23283 100644
--- a/drivers/xen/xenbus/xenbus_probe.c
+++ b/drivers/xen/xenbus/xenbus_probe.c
@@ -40,6 +40,7 @@
 #include <linux/ctype.h>
 #include <linux/fcntl.h>
 #include <linux/mm.h>
+#include <linux/proc_fs.h>
 #include <linux/notifier.h>
 #include <linux/kthread.h>
 #include <linux/mutex.h>
@@ -55,7 +56,10 @@
 #include "xenbus_comms.h"
 #include "xenbus_probe.h"
 
+
 int xen_store_evtchn;
+EXPORT_SYMBOL(xen_store_evtchn);
+
 struct xenstore_domain_interface *xen_store_interface;
 static unsigned long xen_store_mfn;
 
@@ -166,6 +170,9 @@ static int read_backend_details(struct xenbus_device *xendev)
 	return read_otherend_details(xendev, "backend-id", "backend");
 }
 
+static struct device_attribute xenbus_dev_attrs[] = {
+	__ATTR_NULL
+};
 
 /* Bus type for frontend drivers. */
 static struct xen_bus_type xenbus_frontend = {
@@ -174,12 +181,13 @@ static struct xen_bus_type xenbus_frontend = {
 	.get_bus_id = frontend_bus_id,
 	.probe = xenbus_probe_frontend,
 	.bus = {
-		.name     = "xen",
-		.match    = xenbus_match,
-		.uevent   = xenbus_uevent,
-		.probe    = xenbus_dev_probe,
-		.remove   = xenbus_dev_remove,
-		.shutdown = xenbus_dev_shutdown,
+		.name      = "xen",
+		.match     = xenbus_match,
+		.uevent    = xenbus_uevent,
+		.probe     = xenbus_dev_probe,
+		.remove    = xenbus_dev_remove,
+		.shutdown  = xenbus_dev_shutdown,
+		.dev_attrs = xenbus_dev_attrs,
 	},
 };
 
@@ -852,6 +860,14 @@ static int __init xenbus_probe_init(void)
 	if (!xen_initial_domain())
 		xenbus_probe(NULL);
 
+#ifdef CONFIG_XEN_COMPAT_XENFS
+	/*
+	 * Create xenfs mountpoint in /proc for compatibility with
+	 * utilities that expect to find "xenbus" under "/proc/xen".
+	 */
+	proc_mkdir("xen", NULL);
+#endif
+
 	return 0;
 
   out_unreg_back:
diff --git a/drivers/xen/xenbus/xenbus_xs.c b/drivers/xen/xenbus/xenbus_xs.c
index 7f2f91c0e11d..e325eab4724d 100644
--- a/drivers/xen/xenbus/xenbus_xs.c
+++ b/drivers/xen/xenbus/xenbus_xs.c
@@ -184,6 +184,7 @@ void *xenbus_dev_request_and_reply(struct xsd_sockmsg *msg)
 
 	return ret;
 }
+EXPORT_SYMBOL(xenbus_dev_request_and_reply);
 
 /* Send message to xs, get kmalloc'ed reply.  ERR_PTR() on error. */
 static void *xs_talkv(struct xenbus_transaction t,
diff --git a/drivers/xen/xenfs/Makefile b/drivers/xen/xenfs/Makefile
new file mode 100644
index 000000000000..25275c3bbdff
--- /dev/null
+++ b/drivers/xen/xenfs/Makefile
@@ -0,0 +1,3 @@
+obj-$(CONFIG_XENFS) += xenfs.o
+
+xenfs-objs = super.o xenbus.o
\ No newline at end of file
diff --git a/drivers/xen/xenfs/super.c b/drivers/xen/xenfs/super.c
new file mode 100644
index 000000000000..515741a8e6b8
--- /dev/null
+++ b/drivers/xen/xenfs/super.c
@@ -0,0 +1,64 @@
+/*
+ *  xenfs.c - a filesystem for passing info between the a domain and
+ *  the hypervisor.
+ *
+ * 2008-10-07  Alex Zeffertt    Replaced /proc/xen/xenbus with xenfs filesystem
+ *                              and /proc/xen compatibility mount point.
+ *                              Turned xenfs into a loadable module.
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/magic.h>
+
+#include "xenfs.h"
+
+#include <asm/xen/hypervisor.h>
+
+MODULE_DESCRIPTION("Xen filesystem");
+MODULE_LICENSE("GPL");
+
+static int xenfs_fill_super(struct super_block *sb, void *data, int silent)
+{
+	static struct tree_descr xenfs_files[] = {
+		[2] = {"xenbus", &xenbus_file_ops, S_IRUSR|S_IWUSR},
+		{""},
+	};
+
+	return simple_fill_super(sb, XENFS_SUPER_MAGIC, xenfs_files);
+}
+
+static int xenfs_get_sb(struct file_system_type *fs_type,
+			int flags, const char *dev_name,
+			void *data, struct vfsmount *mnt)
+{
+	return get_sb_single(fs_type, flags, data, xenfs_fill_super, mnt);
+}
+
+static struct file_system_type xenfs_type = {
+	.owner =	THIS_MODULE,
+	.name =		"xenfs",
+	.get_sb =	xenfs_get_sb,
+	.kill_sb =	kill_litter_super,
+};
+
+static int __init xenfs_init(void)
+{
+	if (xen_pv_domain())
+		return register_filesystem(&xenfs_type);
+
+	printk(KERN_INFO "XENFS: not registering filesystem on non-xen platform\n");
+	return 0;
+}
+
+static void __exit xenfs_exit(void)
+{
+	if (xen_pv_domain())
+		unregister_filesystem(&xenfs_type);
+}
+
+module_init(xenfs_init);
+module_exit(xenfs_exit);
+
diff --git a/drivers/xen/xenfs/xenbus.c b/drivers/xen/xenfs/xenbus.c
new file mode 100644
index 000000000000..875a4c59c594
--- /dev/null
+++ b/drivers/xen/xenfs/xenbus.c
@@ -0,0 +1,593 @@
+/*
+ * Driver giving user-space access to the kernel's xenbus connection
+ * to xenstore.
+ *
+ * Copyright (c) 2005, Christian Limpach
+ * Copyright (c) 2005, Rusty Russell, IBM Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Changes:
+ * 2008-10-07  Alex Zeffertt    Replaced /proc/xen/xenbus with xenfs filesystem
+ *                              and /proc/xen compatibility mount point.
+ *                              Turned xenfs into a loadable module.
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/uio.h>
+#include <linux/notifier.h>
+#include <linux/wait.h>
+#include <linux/fs.h>
+#include <linux/poll.h>
+#include <linux/mutex.h>
+#include <linux/spinlock.h>
+#include <linux/mount.h>
+#include <linux/pagemap.h>
+#include <linux/uaccess.h>
+#include <linux/init.h>
+#include <linux/namei.h>
+#include <linux/string.h>
+
+#include "xenfs.h"
+#include "../xenbus/xenbus_comms.h"
+
+#include <xen/xenbus.h>
+#include <asm/xen/hypervisor.h>
+
+/*
+ * An element of a list of outstanding transactions, for which we're
+ * still waiting a reply.
+ */
+struct xenbus_transaction_holder {
+	struct list_head list;
+	struct xenbus_transaction handle;
+};
+
+/*
+ * A buffer of data on the queue.
+ */
+struct read_buffer {
+	struct list_head list;
+	unsigned int cons;
+	unsigned int len;
+	char msg[];
+};
+
+struct xenbus_file_priv {
+	/*
+	 * msgbuffer_mutex is held while partial requests are built up
+	 * and complete requests are acted on.  It therefore protects
+	 * the "transactions" and "watches" lists, and the partial
+	 * request length and buffer.
+	 *
+	 * reply_mutex protects the reply being built up to return to
+	 * usermode.  It nests inside msgbuffer_mutex but may be held
+	 * alone during a watch callback.
+	 */
+	struct mutex msgbuffer_mutex;
+
+	/* In-progress transactions */
+	struct list_head transactions;
+
+	/* Active watches. */
+	struct list_head watches;
+
+	/* Partial request. */
+	unsigned int len;
+	union {
+		struct xsd_sockmsg msg;
+		char buffer[PAGE_SIZE];
+	} u;
+
+	/* Response queue. */
+	struct mutex reply_mutex;
+	struct list_head read_buffers;
+	wait_queue_head_t read_waitq;
+
+};
+
+/* Read out any raw xenbus messages queued up. */
+static ssize_t xenbus_file_read(struct file *filp,
+			       char __user *ubuf,
+			       size_t len, loff_t *ppos)
+{
+	struct xenbus_file_priv *u = filp->private_data;
+	struct read_buffer *rb;
+	unsigned i;
+	int ret;
+
+	mutex_lock(&u->reply_mutex);
+	while (list_empty(&u->read_buffers)) {
+		mutex_unlock(&u->reply_mutex);
+		ret = wait_event_interruptible(u->read_waitq,
+					       !list_empty(&u->read_buffers));
+		if (ret)
+			return ret;
+		mutex_lock(&u->reply_mutex);
+	}
+
+	rb = list_entry(u->read_buffers.next, struct read_buffer, list);
+	i = 0;
+	while (i < len) {
+		unsigned sz = min((unsigned)len - i, rb->len - rb->cons);
+
+		ret = copy_to_user(ubuf + i, &rb->msg[rb->cons], sz);
+
+		i += sz - ret;
+		rb->cons += sz - ret;
+
+		if (ret != sz) {
+			if (i == 0)
+				i = -EFAULT;
+			goto out;
+		}
+
+		/* Clear out buffer if it has been consumed */
+		if (rb->cons == rb->len) {
+			list_del(&rb->list);
+			kfree(rb);
+			if (list_empty(&u->read_buffers))
+				break;
+			rb = list_entry(u->read_buffers.next,
+					struct read_buffer, list);
+		}
+	}
+
+out:
+	mutex_unlock(&u->reply_mutex);
+	return i;
+}
+
+/*
+ * Add a buffer to the queue.  Caller must hold the appropriate lock
+ * if the queue is not local.  (Commonly the caller will build up
+ * multiple queued buffers on a temporary local list, and then add it
+ * to the appropriate list under lock once all the buffers have een
+ * successfully allocated.)
+ */
+static int queue_reply(struct list_head *queue, const void *data, size_t len)
+{
+	struct read_buffer *rb;
+
+	if (len == 0)
+		return 0;
+
+	rb = kmalloc(sizeof(*rb) + len, GFP_KERNEL);
+	if (rb == NULL)
+		return -ENOMEM;
+
+	rb->cons = 0;
+	rb->len = len;
+
+	memcpy(rb->msg, data, len);
+
+	list_add_tail(&rb->list, queue);
+	return 0;
+}
+
+/*
+ * Free all the read_buffer s on a list.
+ * Caller must have sole reference to list.
+ */
+static void queue_cleanup(struct list_head *list)
+{
+	struct read_buffer *rb;
+
+	while (!list_empty(list)) {
+		rb = list_entry(list->next, struct read_buffer, list);
+		list_del(list->next);
+		kfree(rb);
+	}
+}
+
+struct watch_adapter {
+	struct list_head list;
+	struct xenbus_watch watch;
+	struct xenbus_file_priv *dev_data;
+	char *token;
+};
+
+static void free_watch_adapter(struct watch_adapter *watch)
+{
+	kfree(watch->watch.node);
+	kfree(watch->token);
+	kfree(watch);
+}
+
+static struct watch_adapter *alloc_watch_adapter(const char *path,
+						 const char *token)
+{
+	struct watch_adapter *watch;
+
+	watch = kzalloc(sizeof(*watch), GFP_KERNEL);
+	if (watch == NULL)
+		goto out_fail;
+
+	watch->watch.node = kstrdup(path, GFP_KERNEL);
+	if (watch->watch.node == NULL)
+		goto out_free;
+
+	watch->token = kstrdup(token, GFP_KERNEL);
+	if (watch->token == NULL)
+		goto out_free;
+
+	return watch;
+
+out_free:
+	free_watch_adapter(watch);
+
+out_fail:
+	return NULL;
+}
+
+static void watch_fired(struct xenbus_watch *watch,
+			const char **vec,
+			unsigned int len)
+{
+	struct watch_adapter *adap;
+	struct xsd_sockmsg hdr;
+	const char *path, *token;
+	int path_len, tok_len, body_len, data_len = 0;
+	int ret;
+	LIST_HEAD(staging_q);
+
+	adap = container_of(watch, struct watch_adapter, watch);
+
+	path = vec[XS_WATCH_PATH];
+	token = adap->token;
+
+	path_len = strlen(path) + 1;
+	tok_len = strlen(token) + 1;
+	if (len > 2)
+		data_len = vec[len] - vec[2] + 1;
+	body_len = path_len + tok_len + data_len;
+
+	hdr.type = XS_WATCH_EVENT;
+	hdr.len = body_len;
+
+	mutex_lock(&adap->dev_data->reply_mutex);
+
+	ret = queue_reply(&staging_q, &hdr, sizeof(hdr));
+	if (!ret)
+		ret = queue_reply(&staging_q, path, path_len);
+	if (!ret)
+		ret = queue_reply(&staging_q, token, tok_len);
+	if (!ret && len > 2)
+		ret = queue_reply(&staging_q, vec[2], data_len);
+
+	if (!ret) {
+		/* success: pass reply list onto watcher */
+		list_splice_tail(&staging_q, &adap->dev_data->read_buffers);
+		wake_up(&adap->dev_data->read_waitq);
+	} else
+		queue_cleanup(&staging_q);
+
+	mutex_unlock(&adap->dev_data->reply_mutex);
+}
+
+static int xenbus_write_transaction(unsigned msg_type,
+				    struct xenbus_file_priv *u)
+{
+	int rc, ret;
+	void *reply;
+	struct xenbus_transaction_holder *trans = NULL;
+	LIST_HEAD(staging_q);
+
+	if (msg_type == XS_TRANSACTION_START) {
+		trans = kmalloc(sizeof(*trans), GFP_KERNEL);
+		if (!trans) {
+			rc = -ENOMEM;
+			goto out;
+		}
+	}
+
+	reply = xenbus_dev_request_and_reply(&u->u.msg);
+	if (IS_ERR(reply)) {
+		kfree(trans);
+		rc = PTR_ERR(reply);
+		goto out;
+	}
+
+	if (msg_type == XS_TRANSACTION_START) {
+		trans->handle.id = simple_strtoul(reply, NULL, 0);
+
+		list_add(&trans->list, &u->transactions);
+	} else if (msg_type == XS_TRANSACTION_END) {
+		list_for_each_entry(trans, &u->transactions, list)
+			if (trans->handle.id == u->u.msg.tx_id)
+				break;
+		BUG_ON(&trans->list == &u->transactions);
+		list_del(&trans->list);
+
+		kfree(trans);
+	}
+
+	mutex_lock(&u->reply_mutex);
+	ret = queue_reply(&staging_q, &u->u.msg, sizeof(u->u.msg));
+	if (!ret)
+		ret = queue_reply(&staging_q, reply, u->u.msg.len);
+	if (!ret) {
+		list_splice_tail(&staging_q, &u->read_buffers);
+		wake_up(&u->read_waitq);
+	} else {
+		queue_cleanup(&staging_q);
+		rc = ret;
+	}
+	mutex_unlock(&u->reply_mutex);
+
+	kfree(reply);
+
+out:
+	return rc;
+}
+
+static int xenbus_write_watch(unsigned msg_type, struct xenbus_file_priv *u)
+{
+	struct watch_adapter *watch, *tmp_watch;
+	char *path, *token;
+	int err, rc;
+	LIST_HEAD(staging_q);
+
+	path = u->u.buffer + sizeof(u->u.msg);
+	token = memchr(path, 0, u->u.msg.len);
+	if (token == NULL) {
+		rc = -EILSEQ;
+		goto out;
+	}
+	token++;
+
+	if (msg_type == XS_WATCH) {
+		watch = alloc_watch_adapter(path, token);
+		if (watch == NULL) {
+			rc = -ENOMEM;
+			goto out;
+		}
+
+		watch->watch.callback = watch_fired;
+		watch->dev_data = u;
+
+		err = register_xenbus_watch(&watch->watch);
+		if (err) {
+			free_watch_adapter(watch);
+			rc = err;
+			goto out;
+		}
+		list_add(&watch->list, &u->watches);
+	} else {
+		list_for_each_entry_safe(watch, tmp_watch, &u->watches, list) {
+			if (!strcmp(watch->token, token) &&
+			    !strcmp(watch->watch.node, path)) {
+				unregister_xenbus_watch(&watch->watch);
+				list_del(&watch->list);
+				free_watch_adapter(watch);
+				break;
+			}
+		}
+	}
+
+	/* Success.  Synthesize a reply to say all is OK. */
+	{
+		struct {
+			struct xsd_sockmsg hdr;
+			char body[3];
+		} __packed reply = {
+			{
+				.type = msg_type,
+				.len = sizeof(reply.body)
+			},
+			"OK"
+		};
+
+		mutex_lock(&u->reply_mutex);
+		rc = queue_reply(&u->read_buffers, &reply, sizeof(reply));
+		mutex_unlock(&u->reply_mutex);
+	}
+
+out:
+	return rc;
+}
+
+static ssize_t xenbus_file_write(struct file *filp,
+				const char __user *ubuf,
+				size_t len, loff_t *ppos)
+{
+	struct xenbus_file_priv *u = filp->private_data;
+	uint32_t msg_type;
+	int rc = len;
+	int ret;
+	LIST_HEAD(staging_q);
+
+	/*
+	 * We're expecting usermode to be writing properly formed
+	 * xenbus messages.  If they write an incomplete message we
+	 * buffer it up.  Once it is complete, we act on it.
+	 */
+
+	/*
+	 * Make sure concurrent writers can't stomp all over each
+	 * other's messages and make a mess of our partial message
+	 * buffer.  We don't make any attemppt to stop multiple
+	 * writers from making a mess of each other's incomplete
+	 * messages; we're just trying to guarantee our own internal
+	 * consistency and make sure that single writes are handled
+	 * atomically.
+	 */
+	mutex_lock(&u->msgbuffer_mutex);
+
+	/* Get this out of the way early to avoid confusion */
+	if (len == 0)
+		goto out;
+
+	/* Can't write a xenbus message larger we can buffer */
+	if ((len + u->len) > sizeof(u->u.buffer)) {
+		/* On error, dump existing buffer */
+		u->len = 0;
+		rc = -EINVAL;
+		goto out;
+	}
+
+	ret = copy_from_user(u->u.buffer + u->len, ubuf, len);
+
+	if (ret == len) {
+		rc = -EFAULT;
+		goto out;
+	}
+
+	/* Deal with a partial copy. */
+	len -= ret;
+	rc = len;
+
+	u->len += len;
+
+	/* Return if we haven't got a full message yet */
+	if (u->len < sizeof(u->u.msg))
+		goto out;	/* not even the header yet */
+
+	/* If we're expecting a message that's larger than we can
+	   possibly send, dump what we have and return an error. */
+	if ((sizeof(u->u.msg) + u->u.msg.len) > sizeof(u->u.buffer)) {
+		rc = -E2BIG;
+		u->len = 0;
+		goto out;
+	}
+
+	if (u->len < (sizeof(u->u.msg) + u->u.msg.len))
+		goto out;	/* incomplete data portion */
+
+	/*
+	 * OK, now we have a complete message.  Do something with it.
+	 */
+
+	msg_type = u->u.msg.type;
+
+	switch (msg_type) {
+	case XS_TRANSACTION_START:
+	case XS_TRANSACTION_END:
+	case XS_DIRECTORY:
+	case XS_READ:
+	case XS_GET_PERMS:
+	case XS_RELEASE:
+	case XS_GET_DOMAIN_PATH:
+	case XS_WRITE:
+	case XS_MKDIR:
+	case XS_RM:
+	case XS_SET_PERMS:
+		/* Send out a transaction */
+		ret = xenbus_write_transaction(msg_type, u);
+		break;
+
+	case XS_WATCH:
+	case XS_UNWATCH:
+		/* (Un)Ask for some path to be watched for changes */
+		ret = xenbus_write_watch(msg_type, u);
+		break;
+
+	default:
+		ret = -EINVAL;
+		break;
+	}
+	if (ret != 0)
+		rc = ret;
+
+	/* Buffered message consumed */
+	u->len = 0;
+
+ out:
+	mutex_unlock(&u->msgbuffer_mutex);
+	return rc;
+}
+
+static int xenbus_file_open(struct inode *inode, struct file *filp)
+{
+	struct xenbus_file_priv *u;
+
+	if (xen_store_evtchn == 0)
+		return -ENOENT;
+
+	nonseekable_open(inode, filp);
+
+	u = kzalloc(sizeof(*u), GFP_KERNEL);
+	if (u == NULL)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&u->transactions);
+	INIT_LIST_HEAD(&u->watches);
+	INIT_LIST_HEAD(&u->read_buffers);
+	init_waitqueue_head(&u->read_waitq);
+
+	mutex_init(&u->reply_mutex);
+	mutex_init(&u->msgbuffer_mutex);
+
+	filp->private_data = u;
+
+	return 0;
+}
+
+static int xenbus_file_release(struct inode *inode, struct file *filp)
+{
+	struct xenbus_file_priv *u = filp->private_data;
+	struct xenbus_transaction_holder *trans, *tmp;
+	struct watch_adapter *watch, *tmp_watch;
+
+	/*
+	 * No need for locking here because there are no other users,
+	 * by definition.
+	 */
+
+	list_for_each_entry_safe(trans, tmp, &u->transactions, list) {
+		xenbus_transaction_end(trans->handle, 1);
+		list_del(&trans->list);
+		kfree(trans);
+	}
+
+	list_for_each_entry_safe(watch, tmp_watch, &u->watches, list) {
+		unregister_xenbus_watch(&watch->watch);
+		list_del(&watch->list);
+		free_watch_adapter(watch);
+	}
+
+	kfree(u);
+
+	return 0;
+}
+
+static unsigned int xenbus_file_poll(struct file *file, poll_table *wait)
+{
+	struct xenbus_file_priv *u = file->private_data;
+
+	poll_wait(file, &u->read_waitq, wait);
+	if (!list_empty(&u->read_buffers))
+		return POLLIN | POLLRDNORM;
+	return 0;
+}
+
+const struct file_operations xenbus_file_ops = {
+	.read = xenbus_file_read,
+	.write = xenbus_file_write,
+	.open = xenbus_file_open,
+	.release = xenbus_file_release,
+	.poll = xenbus_file_poll,
+};
diff --git a/drivers/xen/xenfs/xenfs.h b/drivers/xen/xenfs/xenfs.h
new file mode 100644
index 000000000000..51f08b2d0bf1
--- /dev/null
+++ b/drivers/xen/xenfs/xenfs.h
@@ -0,0 +1,6 @@
+#ifndef _XENFS_XENBUS_H
+#define _XENFS_XENBUS_H
+
+extern const struct file_operations xenbus_file_ops;
+
+#endif	/* _XENFS_XENBUS_H */
diff --git a/include/linux/magic.h b/include/linux/magic.h
index f7f3fdddbef0..439f6f3cb0c4 100644
--- a/include/linux/magic.h
+++ b/include/linux/magic.h
@@ -13,6 +13,7 @@
 #define EFS_SUPER_MAGIC		0x414A53
 #define EXT2_SUPER_MAGIC	0xEF53
 #define EXT3_SUPER_MAGIC	0xEF53
+#define XENFS_SUPER_MAGIC	0xabba1974
 #define EXT4_SUPER_MAGIC	0xEF53
 #define HPFS_SUPER_MAGIC	0xf995e849
 #define ISOFS_SUPER_MAGIC	0x9660
diff --git a/include/xen/xenbus.h b/include/xen/xenbus.h
index 6369d89c25d5..f87f9614844d 100644
--- a/include/xen/xenbus.h
+++ b/include/xen/xenbus.h
@@ -136,8 +136,6 @@ struct xenbus_transaction
 /* Nil transaction ID. */
 #define XBT_NIL ((struct xenbus_transaction) { 0 })
 
-int __init xenbus_dev_init(void);
-
 char **xenbus_directory(struct xenbus_transaction t,
 			const char *dir, const char *node, unsigned int *num);
 void *xenbus_read(struct xenbus_transaction t,
-- 
cgit v1.2.3


From 18a82eb9f980b5e02cea651e4ecda26265d98933 Mon Sep 17 00:00:00 2001
From: Pekka J Enberg <penberg@cs.helsinki.fi>
Date: Wed, 7 Jan 2009 18:07:19 -0800
Subject: ext2: allocate ->s_blockgroup_lock separately

As spotted by kmemtrace, struct ext2_sb_info is 17024 bytes on 64-bit
which makes it a very bad fit for SLAB allocators.  The culprit of the
wasted memory is ->s_blockgroup_lock which can be as big as 16 KB when
NR_CPUS >= 32.

To fix that, allocate ->s_blockgroup_lock, which fits nicely in a order 2
page in the worst case, separately.  This shinks down struct ext2_sb_info
enough to fit a 1 KB slab cache so now we allocate 16 KB + 1 KB instead of
32 KB saving 15 KB of memory.

Acked-by: Andreas Dilger <adilger@sun.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext2/super.c            | 10 +++++++++-
 include/linux/ext2_fs_sb.h |  4 ++--
 2 files changed, 11 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 647cd888ac87..da8bdeaa2e6d 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -132,6 +132,7 @@ static void ext2_put_super (struct super_block * sb)
 	percpu_counter_destroy(&sbi->s_dirs_counter);
 	brelse (sbi->s_sbh);
 	sb->s_fs_info = NULL;
+	kfree(sbi->s_blockgroup_lock);
 	kfree(sbi);
 
 	return;
@@ -756,6 +757,13 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
+
+	sbi->s_blockgroup_lock =
+		kzalloc(sizeof(struct blockgroup_lock), GFP_KERNEL);
+	if (!sbi->s_blockgroup_lock) {
+		kfree(sbi);
+		return -ENOMEM;
+	}
 	sb->s_fs_info = sbi;
 	sbi->s_sb_block = sb_block;
 
@@ -983,7 +991,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 		printk ("EXT2-fs: not enough memory\n");
 		goto failed_mount;
 	}
-	bgl_lock_init(&sbi->s_blockgroup_lock);
+	bgl_lock_init(sbi->s_blockgroup_lock);
 	sbi->s_debts = kcalloc(sbi->s_groups_count, sizeof(*sbi->s_debts), GFP_KERNEL);
 	if (!sbi->s_debts) {
 		printk ("EXT2-fs: not enough memory\n");
diff --git a/include/linux/ext2_fs_sb.h b/include/linux/ext2_fs_sb.h
index dc541f3653d1..1cdb66367c98 100644
--- a/include/linux/ext2_fs_sb.h
+++ b/include/linux/ext2_fs_sb.h
@@ -101,7 +101,7 @@ struct ext2_sb_info {
 	struct percpu_counter s_freeblocks_counter;
 	struct percpu_counter s_freeinodes_counter;
 	struct percpu_counter s_dirs_counter;
-	struct blockgroup_lock s_blockgroup_lock;
+	struct blockgroup_lock *s_blockgroup_lock;
 	/* root of the per fs reservation window tree */
 	spinlock_t s_rsv_window_lock;
 	struct rb_root s_rsv_window_root;
@@ -111,7 +111,7 @@ struct ext2_sb_info {
 static inline spinlock_t *
 sb_bgl_lock(struct ext2_sb_info *sbi, unsigned int block_group)
 {
-	return bgl_lock_ptr(&sbi->s_blockgroup_lock, block_group);
+	return bgl_lock_ptr(sbi->s_blockgroup_lock, block_group);
 }
 
 #endif	/* _LINUX_EXT2_FS_SB */
-- 
cgit v1.2.3


From 0e090f1e05a563cc9acdda442767176bf1616001 Mon Sep 17 00:00:00 2001
From: Duane Griffin <duaneg@dghda.com>
Date: Wed, 7 Jan 2009 18:07:20 -0800
Subject: ext2: don't inherit inappropriate inode flags from parent

At present BTREE/INDEX is the only flag that new ext2 inodes do NOT
inherit from their parent.  In addition prevent the flags DIRTY, ECOMPR,
INDEX, IMAGIC and TOPDIR from being inherited.  List inheritable flags
explicitly to prevent future flags from accidentally being inherited.

This fixes the TOPDIR flag inheritance bug reported at
http://bugzilla.kernel.org/show_bug.cgi?id=9866.

Signed-off-by: Duane Griffin <duaneg@dghda.com>
Acked-by: Andreas Dilger <adilger@sun.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext2/ialloc.c        | 2 +-
 include/linux/ext2_fs.h | 7 +++++++
 2 files changed, 8 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c
index c454d5db28a5..b5598e1393d9 100644
--- a/fs/ext2/ialloc.c
+++ b/fs/ext2/ialloc.c
@@ -565,7 +565,7 @@ got:
 	inode->i_blocks = 0;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
 	memset(ei->i_data, 0, sizeof(ei->i_data));
-	ei->i_flags = EXT2_I(dir)->i_flags & ~EXT2_BTREE_FL;
+	ei->i_flags = EXT2_I(dir)->i_flags & EXT2_FL_INHERITED;
 	if (S_ISLNK(mode))
 		ei->i_flags &= ~(EXT2_IMMUTABLE_FL|EXT2_APPEND_FL);
 	/* dirsync is only applied to directories */
diff --git a/include/linux/ext2_fs.h b/include/linux/ext2_fs.h
index 78c775a83f7c..c3a051819363 100644
--- a/include/linux/ext2_fs.h
+++ b/include/linux/ext2_fs.h
@@ -194,6 +194,13 @@ struct ext2_group_desc
 #define EXT2_FL_USER_VISIBLE		FS_FL_USER_VISIBLE	/* User visible flags */
 #define EXT2_FL_USER_MODIFIABLE		FS_FL_USER_MODIFIABLE	/* User modifiable flags */
 
+/* Flags that should be inherited by new inodes from their parent. */
+#define EXT2_FL_INHERITED (EXT2_SECRM_FL | EXT2_UNRM_FL | EXT2_COMPR_FL |\
+			   EXT2_SYNC_FL | EXT2_IMMUTABLE_FL | EXT2_APPEND_FL |\
+			   EXT2_NODUMP_FL | EXT2_NOATIME_FL | EXT2_COMPRBLK_FL|\
+			   EXT2_NOCOMP_FL | EXT2_JOURNAL_DATA_FL |\
+			   EXT2_NOTAIL_FL | EXT2_DIRSYNC_FL)
+
 /*
  * ioctl commands
  */
-- 
cgit v1.2.3


From ef8b646183868b2d042fa6cde0eef2a31263ff85 Mon Sep 17 00:00:00 2001
From: Duane Griffin <duaneg@dghda.com>
Date: Wed, 7 Jan 2009 18:07:21 -0800
Subject: ext2: tighten restrictions on inode flags

At the moment there are few restrictions on which flags may be set on
which inodes.  Specifically DIRSYNC may only be set on directories and
IMMUTABLE and APPEND may not be set on links.  Tighten that to disallow
TOPDIR being set on non-directories and only NODUMP and NOATIME to be set
on non-regular file, non-directories.

Introduces a flags masking function which masks flags based on mode and
use it during inode creation and when flags are set via the ioctl to
facilitate future consistency.

Signed-off-by: Duane Griffin <duaneg@dghda.com>
Acked-by: Andreas Dilger <adilger@sun.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext2/ialloc.c        |  8 ++------
 fs/ext2/ioctl.c         |  3 +--
 include/linux/ext2_fs.h | 17 +++++++++++++++++
 3 files changed, 20 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c
index b5598e1393d9..66321a877e74 100644
--- a/fs/ext2/ialloc.c
+++ b/fs/ext2/ialloc.c
@@ -565,12 +565,8 @@ got:
 	inode->i_blocks = 0;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
 	memset(ei->i_data, 0, sizeof(ei->i_data));
-	ei->i_flags = EXT2_I(dir)->i_flags & EXT2_FL_INHERITED;
-	if (S_ISLNK(mode))
-		ei->i_flags &= ~(EXT2_IMMUTABLE_FL|EXT2_APPEND_FL);
-	/* dirsync is only applied to directories */
-	if (!S_ISDIR(mode))
-		ei->i_flags &= ~EXT2_DIRSYNC_FL;
+	ei->i_flags =
+		ext2_mask_flags(mode, EXT2_I(dir)->i_flags & EXT2_FL_INHERITED);
 	ei->i_faddr = 0;
 	ei->i_frag_no = 0;
 	ei->i_frag_size = 0;
diff --git a/fs/ext2/ioctl.c b/fs/ext2/ioctl.c
index de876fa793e1..7cb4badef927 100644
--- a/fs/ext2/ioctl.c
+++ b/fs/ext2/ioctl.c
@@ -50,8 +50,7 @@ long ext2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 			goto setflags_out;
 		}
 
-		if (!S_ISDIR(inode->i_mode))
-			flags &= ~EXT2_DIRSYNC_FL;
+		flags = ext2_mask_flags(inode->i_mode, flags);
 
 		mutex_lock(&inode->i_mutex);
 		/* Is it quota file? Do not allow user to mess with it */
diff --git a/include/linux/ext2_fs.h b/include/linux/ext2_fs.h
index c3a051819363..121720d74e15 100644
--- a/include/linux/ext2_fs.h
+++ b/include/linux/ext2_fs.h
@@ -201,6 +201,23 @@ struct ext2_group_desc
 			   EXT2_NOCOMP_FL | EXT2_JOURNAL_DATA_FL |\
 			   EXT2_NOTAIL_FL | EXT2_DIRSYNC_FL)
 
+/* Flags that are appropriate for regular files (all but dir-specific ones). */
+#define EXT2_REG_FLMASK (~(EXT2_DIRSYNC_FL | EXT2_TOPDIR_FL))
+
+/* Flags that are appropriate for non-directories/regular files. */
+#define EXT2_OTHER_FLMASK (EXT2_NODUMP_FL | EXT2_NOATIME_FL)
+
+/* Mask out flags that are inappropriate for the given type of inode. */
+static inline __u32 ext2_mask_flags(umode_t mode, __u32 flags)
+{
+	if (S_ISDIR(mode))
+		return flags;
+	else if (S_ISREG(mode))
+		return flags & EXT2_REG_FLMASK;
+	else
+		return flags & EXT2_OTHER_FLMASK;
+}
+
 /*
  * ioctl commands
  */
-- 
cgit v1.2.3


From f420d4dc4272fd223986762df2ad06056ddebada Mon Sep 17 00:00:00 2001
From: Josef Bacik <jbacik@redhat.com>
Date: Wed, 7 Jan 2009 18:07:24 -0800
Subject: jbd: improve fsync batching

There is a flaw with the way jbd handles fsync batching.  If we fsync() a
file and we were not the last person to run fsync() on this fs then we
automatically sleep for 1 jiffie in order to wait for new writers to join
into the transaction before forcing the commit.  The problem with this is
that with really fast storage (ie a Clariion) the time it takes to commit
a transaction to disk is way faster than 1 jiffie in most cases, so
sleeping means waiting longer with nothing to do than if we just committed
the transaction and kept going.  Ric Wheeler noticed this when using
fs_mark with more than 1 thread, the throughput would plummet as he added
more threads.

This patch attempts to fix this problem by recording the average time in
nanoseconds that it takes to commit a transaction to disk, and what time
we started the transaction.  If we run an fsync() and we have been running
for less time than it takes to commit the transaction to disk, we sleep
for the delta amount of time and then commit to disk.  We acheive
sub-jiffie sleeping using schedule_hrtimeout.  This means that the wait
time is auto-tuned to the speed of the underlying disk, instead of having
this static timeout.  I weighted the average according to somebody's
comments (Andreas Dilger I think) in order to help normalize random
outliers where we take way longer or way less time to commit than the
average.  I also have a min() check in there to make sure we don't sleep
longer than a jiffie in case our storage is super slow, this was requested
by Andrew.

I unfortunately do not have access to a Clariion, so I had to use a
ramdisk to represent a super fast array.  I tested with a SATA drive with
barrier=1 to make sure there was no regression with local disks, I tested
with a 4 way multipathed Apple Xserve RAID array and of course the
ramdisk.  I ran the following command

fs_mark -d /mnt/ext3-test -s 4096 -n 2000 -D 64 -t $i

where $i was 2, 4, 8, 16 and 32.  I mkfs'ed the fs each time.  Here are my
results

type	threads		with patch	without patch
sata	2		24.6		26.3
sata	4		49.2		48.1
sata	8		70.1		67.0
sata	16		104.0		94.1
sata	32		153.6		142.7

xserve	2		246.4		222.0
xserve	4		480.0		440.8
xserve	8		829.5		730.8
xserve	16		1172.7		1026.9
xserve	32		1816.3		1650.5

ramdisk	2		2538.3		1745.6
ramdisk	4		2942.3		661.9
ramdisk	8		2882.5		999.8
ramdisk	16		2738.7		1801.9
ramdisk	32		2541.9		2394.0

Signed-off-by: Josef Bacik <jbacik@redhat.com>
Cc: Andreas Dilger <adilger@sun.com>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Ric Wheeler <rwheeler@redhat.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/jbd/commit.c      | 15 +++++++++++++++
 fs/jbd/transaction.c | 38 +++++++++++++++++++++++++++++++++-----
 include/linux/jbd.h  | 15 +++++++++++++++
 3 files changed, 63 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c
index 25719d902c51..3fbffb1ea714 100644
--- a/fs/jbd/commit.c
+++ b/fs/jbd/commit.c
@@ -306,6 +306,8 @@ void journal_commit_transaction(journal_t *journal)
 	int flags;
 	int err;
 	unsigned long blocknr;
+	ktime_t start_time;
+	u64 commit_time;
 	char *tagp = NULL;
 	journal_header_t *header;
 	journal_block_tag_t *tag = NULL;
@@ -418,6 +420,7 @@ void journal_commit_transaction(journal_t *journal)
 	commit_transaction->t_state = T_FLUSH;
 	journal->j_committing_transaction = commit_transaction;
 	journal->j_running_transaction = NULL;
+	start_time = ktime_get();
 	commit_transaction->t_log_start = journal->j_head;
 	wake_up(&journal->j_wait_transaction_locked);
 	spin_unlock(&journal->j_state_lock);
@@ -913,6 +916,18 @@ restart_loop:
 	J_ASSERT(commit_transaction == journal->j_committing_transaction);
 	journal->j_commit_sequence = commit_transaction->t_tid;
 	journal->j_committing_transaction = NULL;
+	commit_time = ktime_to_ns(ktime_sub(ktime_get(), start_time));
+
+	/*
+	 * weight the commit time higher than the average time so we don't
+	 * react too strongly to vast changes in commit time
+	 */
+	if (likely(journal->j_average_commit_time))
+		journal->j_average_commit_time = (commit_time*3 +
+				journal->j_average_commit_time) / 4;
+	else
+		journal->j_average_commit_time = commit_time;
+
 	spin_unlock(&journal->j_state_lock);
 
 	if (commit_transaction->t_checkpoint_list == NULL &&
diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c
index 60d4c32c8808..b51fbd4b2913 100644
--- a/fs/jbd/transaction.c
+++ b/fs/jbd/transaction.c
@@ -25,6 +25,7 @@
 #include <linux/timer.h>
 #include <linux/mm.h>
 #include <linux/highmem.h>
+#include <linux/hrtimer.h>
 
 static void __journal_temp_unlink_buffer(struct journal_head *jh);
 
@@ -49,6 +50,7 @@ get_transaction(journal_t *journal, transaction_t *transaction)
 {
 	transaction->t_journal = journal;
 	transaction->t_state = T_RUNNING;
+	transaction->t_start_time = ktime_get();
 	transaction->t_tid = journal->j_transaction_sequence++;
 	transaction->t_expires = jiffies + journal->j_commit_interval;
 	spin_lock_init(&transaction->t_handle_lock);
@@ -1370,7 +1372,7 @@ int journal_stop(handle_t *handle)
 {
 	transaction_t *transaction = handle->h_transaction;
 	journal_t *journal = transaction->t_journal;
-	int old_handle_count, err;
+	int err;
 	pid_t pid;
 
 	J_ASSERT(journal_current_handle() == handle);
@@ -1399,6 +1401,17 @@ int journal_stop(handle_t *handle)
 	 * on IO anyway.  Speeds up many-threaded, many-dir operations
 	 * by 30x or more...
 	 *
+	 * We try and optimize the sleep time against what the underlying disk
+	 * can do, instead of having a static sleep time.  This is usefull for
+	 * the case where our storage is so fast that it is more optimal to go
+	 * ahead and force a flush and wait for the transaction to be committed
+	 * than it is to wait for an arbitrary amount of time for new writers to
+	 * join the transaction.  We acheive this by measuring how long it takes
+	 * to commit a transaction, and compare it with how long this
+	 * transaction has been running, and if run time < commit time then we
+	 * sleep for the delta and commit.  This greatly helps super fast disks
+	 * that would see slowdowns as more threads started doing fsyncs.
+	 *
 	 * But don't do this if this process was the most recent one to
 	 * perform a synchronous write.  We do this to detect the case where a
 	 * single process is doing a stream of sync writes.  No point in waiting
@@ -1406,11 +1419,26 @@ int journal_stop(handle_t *handle)
 	 */
 	pid = current->pid;
 	if (handle->h_sync && journal->j_last_sync_writer != pid) {
+		u64 commit_time, trans_time;
+
 		journal->j_last_sync_writer = pid;
-		do {
-			old_handle_count = transaction->t_handle_count;
-			schedule_timeout_uninterruptible(1);
-		} while (old_handle_count != transaction->t_handle_count);
+
+		spin_lock(&journal->j_state_lock);
+		commit_time = journal->j_average_commit_time;
+		spin_unlock(&journal->j_state_lock);
+
+		trans_time = ktime_to_ns(ktime_sub(ktime_get(),
+						   transaction->t_start_time));
+
+		commit_time = min_t(u64, commit_time,
+				    1000*jiffies_to_usecs(1));
+
+		if (trans_time < commit_time) {
+			ktime_t expires = ktime_add_ns(ktime_get(),
+						       commit_time);
+			set_current_state(TASK_UNINTERRUPTIBLE);
+			schedule_hrtimeout(&expires, HRTIMER_MODE_ABS);
+		}
 	}
 
 	current->journal_info = NULL;
diff --git a/include/linux/jbd.h b/include/linux/jbd.h
index 346e2b80be7d..6384b19efe64 100644
--- a/include/linux/jbd.h
+++ b/include/linux/jbd.h
@@ -542,6 +542,11 @@ struct transaction_s
 	 */
 	unsigned long		t_expires;
 
+	/*
+	 * When this transaction started, in nanoseconds [no locking]
+	 */
+	ktime_t			t_start_time;
+
 	/*
 	 * How many handles used this transaction? [t_handle_lock]
 	 */
@@ -798,8 +803,18 @@ struct journal_s
 	struct buffer_head	**j_wbuf;
 	int			j_wbufsize;
 
+	/*
+	 * this is the pid of the last person to run a synchronous operation
+	 * through the journal.
+	 */
 	pid_t			j_last_sync_writer;
 
+	/*
+	 * the average amount of time in nanoseconds it takes to commit a
+	 * transaction to the disk.  [j_state_lock]
+	 */
+	u64			j_average_commit_time;
+
 	/*
 	 * An opaque pointer to fs-private information.  ext3 puts its
 	 * superblock pointer here
-- 
cgit v1.2.3


From 5df096d67ec2b6578518caed7d57317a4b807aa1 Mon Sep 17 00:00:00 2001
From: Pekka Enberg <penberg@cs.helsinki.fi>
Date: Wed, 7 Jan 2009 18:07:25 -0800
Subject: ext3: allocate ->s_blockgroup_lock separately

As spotted by kmemtrace, struct ext3_sb_info is 17152 bytes on 64-bit
which makes it a very bad fit for SLAB allocators.  The culprit of the
wasted memory is ->s_blockgroup_lock which can be as big as 16 KB when
NR_CPUS >= 32.

To fix that, allocate ->s_blockgroup_lock, which fits nicely in a order 2
page in the worst case, separately.  This shinks down struct ext3_sb_info
enough to fit a 1 KB slab cache so now we allocate 16 KB + 1 KB instead of
32 KB saving 15 KB of memory.

Acked-by: Andreas Dilger <adilger@sun.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext3/super.c            | 10 +++++++++-
 include/linux/ext3_fs_sb.h |  4 ++--
 2 files changed, 11 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index c22d01467bd1..01c235bc2054 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -439,6 +439,7 @@ static void ext3_put_super (struct super_block * sb)
 		ext3_blkdev_remove(sbi);
 	}
 	sb->s_fs_info = NULL;
+	kfree(sbi->s_blockgroup_lock);
 	kfree(sbi);
 	return;
 }
@@ -1546,6 +1547,13 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
+
+	sbi->s_blockgroup_lock =
+		kzalloc(sizeof(struct blockgroup_lock), GFP_KERNEL);
+	if (!sbi->s_blockgroup_lock) {
+		kfree(sbi);
+		return -ENOMEM;
+	}
 	sb->s_fs_info = sbi;
 	sbi->s_mount_opt = 0;
 	sbi->s_resuid = EXT3_DEF_RESUID;
@@ -1786,7 +1794,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 		goto failed_mount;
 	}
 
-	bgl_lock_init(&sbi->s_blockgroup_lock);
+	bgl_lock_init(sbi->s_blockgroup_lock);
 
 	for (i = 0; i < db_count; i++) {
 		block = descriptor_loc(sb, logic_sb_block, i);
diff --git a/include/linux/ext3_fs_sb.h b/include/linux/ext3_fs_sb.h
index e024e38248ff..76fdc0f4b028 100644
--- a/include/linux/ext3_fs_sb.h
+++ b/include/linux/ext3_fs_sb.h
@@ -60,7 +60,7 @@ struct ext3_sb_info {
 	struct percpu_counter s_freeblocks_counter;
 	struct percpu_counter s_freeinodes_counter;
 	struct percpu_counter s_dirs_counter;
-	struct blockgroup_lock s_blockgroup_lock;
+	struct blockgroup_lock *s_blockgroup_lock;
 
 	/* root of the per fs reservation window tree */
 	spinlock_t s_rsv_window_lock;
@@ -86,7 +86,7 @@ struct ext3_sb_info {
 static inline spinlock_t *
 sb_bgl_lock(struct ext3_sb_info *sbi, unsigned int block_group)
 {
-	return bgl_lock_ptr(&sbi->s_blockgroup_lock, block_group);
+	return bgl_lock_ptr(sbi->s_blockgroup_lock, block_group);
 }
 
 #endif	/* _LINUX_EXT3_FS_SB */
-- 
cgit v1.2.3


From 2e8671cb566da993425d324fc355af31edc6e7f1 Mon Sep 17 00:00:00 2001
From: Duane Griffin <duaneg@dghda.com>
Date: Wed, 7 Jan 2009 18:07:26 -0800
Subject: ext3: don't inherit inappropriate inode flags from parent

At present INDEX is the only flag that new ext3 inodes do NOT inherit from
their parent.  In addition prevent the flags DIRTY, ECOMPR, IMAGIC and
TOPDIR from being inherited.  List inheritable flags explicitly to prevent
future flags from accidentally being inherited.

This fixes the TOPDIR flag inheritance bug reported at
http://bugzilla.kernel.org/show_bug.cgi?id=9866.

Signed-off-by: Duane Griffin <duaneg@dghda.com>
Acked-by: Andreas Dilger <adilger@sun.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext3/ialloc.c        | 2 +-
 include/linux/ext3_fs.h | 7 +++++++
 2 files changed, 8 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c
index 5655fbcbd11f..ba9186a21d1e 100644
--- a/fs/ext3/ialloc.c
+++ b/fs/ext3/ialloc.c
@@ -559,7 +559,7 @@ got:
 	ei->i_dir_start_lookup = 0;
 	ei->i_disksize = 0;
 
-	ei->i_flags = EXT3_I(dir)->i_flags & ~EXT3_INDEX_FL;
+	ei->i_flags = EXT3_I(dir)->i_flags & EXT3_FL_INHERITED;
 	if (S_ISLNK(mode))
 		ei->i_flags &= ~(EXT3_IMMUTABLE_FL|EXT3_APPEND_FL);
 	/* dirsync only applies to directories */
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index d14f02918483..b745619a9b8e 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -178,6 +178,13 @@ struct ext3_group_desc
 #define EXT3_FL_USER_VISIBLE		0x0003DFFF /* User visible flags */
 #define EXT3_FL_USER_MODIFIABLE		0x000380FF /* User modifiable flags */
 
+/* Flags that should be inherited by new inodes from their parent. */
+#define EXT3_FL_INHERITED (EXT3_SECRM_FL | EXT3_UNRM_FL | EXT3_COMPR_FL |\
+			   EXT3_SYNC_FL | EXT3_IMMUTABLE_FL | EXT3_APPEND_FL |\
+			   EXT3_NODUMP_FL | EXT3_NOATIME_FL | EXT3_COMPRBLK_FL|\
+			   EXT3_NOCOMPR_FL | EXT3_JOURNAL_DATA_FL |\
+			   EXT3_NOTAIL_FL | EXT3_DIRSYNC_FL)
+
 /*
  * Inode dynamic state flags
  */
-- 
cgit v1.2.3


From 04143e2fb9d512c21e1dcfb561dbb0445dcfdc8c Mon Sep 17 00:00:00 2001
From: Duane Griffin <duaneg@dghda.com>
Date: Wed, 7 Jan 2009 18:07:26 -0800
Subject: ext3: tighten restrictions on inode flags

At the moment there are few restrictions on which flags may be set on
which inodes.  Specifically DIRSYNC may only be set on directories and
IMMUTABLE and APPEND may not be set on links.  Tighten that to disallow
TOPDIR being set on non-directories and only NODUMP and NOATIME to be set
on non-regular file, non-directories.

Introduces a flags masking function which masks flags based on mode and
use it during inode creation and when flags are set via the ioctl to
facilitate future consistency.

Signed-off-by: Duane Griffin <duaneg@dghda.com>
Acked-by: Andreas Dilger <adilger@sun.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext3/ialloc.c        |  8 ++------
 fs/ext3/ioctl.c         |  3 +--
 include/linux/ext3_fs.h | 17 +++++++++++++++++
 3 files changed, 20 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c
index ba9186a21d1e..8de6c720e510 100644
--- a/fs/ext3/ialloc.c
+++ b/fs/ext3/ialloc.c
@@ -559,12 +559,8 @@ got:
 	ei->i_dir_start_lookup = 0;
 	ei->i_disksize = 0;
 
-	ei->i_flags = EXT3_I(dir)->i_flags & EXT3_FL_INHERITED;
-	if (S_ISLNK(mode))
-		ei->i_flags &= ~(EXT3_IMMUTABLE_FL|EXT3_APPEND_FL);
-	/* dirsync only applies to directories */
-	if (!S_ISDIR(mode))
-		ei->i_flags &= ~EXT3_DIRSYNC_FL;
+	ei->i_flags =
+		ext3_mask_flags(mode, EXT3_I(dir)->i_flags & EXT3_FL_INHERITED);
 #ifdef EXT3_FRAGMENTS
 	ei->i_faddr = 0;
 	ei->i_frag_no = 0;
diff --git a/fs/ext3/ioctl.c b/fs/ext3/ioctl.c
index b7394d05ee8e..5e86ce9a86e0 100644
--- a/fs/ext3/ioctl.c
+++ b/fs/ext3/ioctl.c
@@ -53,8 +53,7 @@ int ext3_ioctl (struct inode * inode, struct file * filp, unsigned int cmd,
 			goto flags_out;
 		}
 
-		if (!S_ISDIR(inode->i_mode))
-			flags &= ~EXT3_DIRSYNC_FL;
+		flags = ext3_mask_flags(inode->i_mode, flags);
 
 		mutex_lock(&inode->i_mutex);
 		/* Is it quota file? Do not allow user to mess with it */
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index b745619a9b8e..d76800f6ecf0 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -185,6 +185,23 @@ struct ext3_group_desc
 			   EXT3_NOCOMPR_FL | EXT3_JOURNAL_DATA_FL |\
 			   EXT3_NOTAIL_FL | EXT3_DIRSYNC_FL)
 
+/* Flags that are appropriate for regular files (all but dir-specific ones). */
+#define EXT3_REG_FLMASK (~(EXT3_DIRSYNC_FL | EXT3_TOPDIR_FL))
+
+/* Flags that are appropriate for non-directories/regular files. */
+#define EXT3_OTHER_FLMASK (EXT3_NODUMP_FL | EXT3_NOATIME_FL)
+
+/* Mask out flags that are inappropriate for the given type of inode. */
+static inline __u32 ext3_mask_flags(umode_t mode, __u32 flags)
+{
+	if (S_ISDIR(mode))
+		return flags;
+	else if (S_ISREG(mode))
+		return flags & EXT3_REG_FLMASK;
+	else
+		return flags & EXT3_OTHER_FLMASK;
+}
+
 /*
  * Inode dynamic state flags
  */
-- 
cgit v1.2.3


From b2aa30f7bb381e04c93eed106089ba55553955f1 Mon Sep 17 00:00:00 2001
From: Lai Jiangshan <laijs@cn.fujitsu.com>
Date: Wed, 7 Jan 2009 18:07:37 -0800
Subject: cgroups: don't put struct cgroupfs_root protected by RCU

We don't access struct cgroupfs_root in fast path, so we should not put
struct cgroupfs_root protected by RCU

But the comment in struct cgroup_subsys.root confuse us.

struct cgroup_subsys.root is used in these places:

1 find_css_set(): if (ss->root->subsys_list.next == &ss->sibling)
2 rebind_subsystems(): if (ss->root != &rootnode)
                       rcu_assign_pointer(ss->root, root);
                       rcu_assign_pointer(subsys[i]->root, &rootnode);
3 cgroup_has_css_refs(): if (ss->root != cgrp->root)
4 cgroup_init_subsys(): ss->root = &rootnode;
5 proc_cgroupstats_show(): ss->name, ss->root->subsys_bits,
                           ss->root->number_of_cgroups, !ss->disabled);
6 cgroup_clone(): root = subsys->root;
                  if ((root != subsys->root) ||

All these place we have held cgroup_lock() or we don't dereference to
struct cgroupfs_root.  It's means wo don't need RCU when use struct
cgroup_subsys.root, and we should not put struct cgroupfs_root protected
by RCU.

Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Reviewed-by: Paul Menage <menage@google.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cgroup.h | 1 -
 kernel/cgroup.c        | 4 ++--
 2 files changed, 2 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index 08b78c09b09a..f68dfd8dd53a 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -337,7 +337,6 @@ struct cgroup_subsys {
 #define MAX_CGROUP_TYPE_NAMELEN 32
 	const char *name;
 
-	/* Protected by RCU */
 	struct cgroupfs_root *root;
 
 	struct list_head sibling;
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index a391ab3bdfc6..a288da176e46 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -713,7 +713,7 @@ static int rebind_subsystems(struct cgroupfs_root *root,
 			cgrp->subsys[i] = dummytop->subsys[i];
 			cgrp->subsys[i]->cgroup = cgrp;
 			list_add(&ss->sibling, &root->subsys_list);
-			rcu_assign_pointer(ss->root, root);
+			ss->root = root;
 			if (ss->bind)
 				ss->bind(ss, cgrp);
 
@@ -725,7 +725,7 @@ static int rebind_subsystems(struct cgroupfs_root *root,
 				ss->bind(ss, dummytop);
 			dummytop->subsys[i]->cgroup = dummytop;
 			cgrp->subsys[i] = NULL;
-			rcu_assign_pointer(subsys[i]->root, &rootnode);
+			subsys[i]->root = &rootnode;
 			list_del(&ss->sibling);
 		} else if (bit & final_bits) {
 			/* Subsystem state should already exist */
-- 
cgit v1.2.3


From a47295e6bc42ad35f9c15ac66f598aa24debd4e2 Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Wed, 7 Jan 2009 18:07:44 -0800
Subject: cgroups: make cgroup_path() RCU-safe

Fix races between /proc/sched_debug by freeing cgroup objects via an RCU
callback.  Thus any cgroup reference obtained from an RCU-safe source will
remain valid during the RCU section.  Since dentries are also RCU-safe,
this allows us to traverse up the tree safely.

Additionally, make cgroup_path() check for a NULL cgrp->dentry to avoid
trying to report a path for a partially-created cgroup.

[lizf@cn.fujitsu.com: call deactive_super() in cgroup_diput()]
Signed-off-by: Paul Menage <menage@google.com>
Reviewed-by: Li Zefan <lizf@cn.fujitsu.com>
Tested-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cgroup.h |  5 ++++-
 kernel/cgroup.c        | 30 +++++++++++++++++++++---------
 2 files changed, 25 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index f68dfd8dd53a..73d1c730c3c4 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -116,7 +116,7 @@ struct cgroup {
 	struct list_head children;	/* my children */
 
 	struct cgroup *parent;	/* my parent */
-	struct dentry *dentry;	  	/* cgroup fs entry */
+	struct dentry *dentry;	  	/* cgroup fs entry, RCU protected */
 
 	/* Private pointers for each registered subsystem */
 	struct cgroup_subsys_state *subsys[CGROUP_SUBSYS_COUNT];
@@ -145,6 +145,9 @@ struct cgroup {
 	int pids_use_count;
 	/* Length of the current tasks_pids array */
 	int pids_length;
+
+	/* For RCU-protected deletion */
+	struct rcu_head rcu_head;
 };
 
 /* A css_set is a structure holding pointers to a set of
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index cb7c72b91f46..83ea4f524be5 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -271,7 +271,7 @@ static void __put_css_set(struct css_set *cg, int taskexit)
 
 	rcu_read_lock();
 	for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) {
-		struct cgroup *cgrp = cg->subsys[i]->cgroup;
+		struct cgroup *cgrp = rcu_dereference(cg->subsys[i]->cgroup);
 		if (atomic_dec_and_test(&cgrp->count) &&
 		    notify_on_release(cgrp)) {
 			if (taskexit)
@@ -594,6 +594,13 @@ static void cgroup_call_pre_destroy(struct cgroup *cgrp)
 	return;
 }
 
+static void free_cgroup_rcu(struct rcu_head *obj)
+{
+	struct cgroup *cgrp = container_of(obj, struct cgroup, rcu_head);
+
+	kfree(cgrp);
+}
+
 static void cgroup_diput(struct dentry *dentry, struct inode *inode)
 {
 	/* is dentry a directory ? if so, kfree() associated cgroup */
@@ -619,11 +626,13 @@ static void cgroup_diput(struct dentry *dentry, struct inode *inode)
 		cgrp->root->number_of_cgroups--;
 		mutex_unlock(&cgroup_mutex);
 
-		/* Drop the active superblock reference that we took when we
-		 * created the cgroup */
+		/*
+		 * Drop the active superblock reference that we took when we
+		 * created the cgroup
+		 */
 		deactivate_super(cgrp->root->sb);
 
-		kfree(cgrp);
+		call_rcu(&cgrp->rcu_head, free_cgroup_rcu);
 	}
 	iput(inode);
 }
@@ -1134,14 +1143,16 @@ static inline struct cftype *__d_cft(struct dentry *dentry)
  * @buf: the buffer to write the path into
  * @buflen: the length of the buffer
  *
- * Called with cgroup_mutex held. Writes path of cgroup into buf.
- * Returns 0 on success, -errno on error.
+ * Called with cgroup_mutex held or else with an RCU-protected cgroup
+ * reference.  Writes path of cgroup into buf.  Returns 0 on success,
+ * -errno on error.
  */
 int cgroup_path(const struct cgroup *cgrp, char *buf, int buflen)
 {
 	char *start;
+	struct dentry *dentry = rcu_dereference(cgrp->dentry);
 
-	if (cgrp == dummytop) {
+	if (!dentry || cgrp == dummytop) {
 		/*
 		 * Inactive subsystems have no dentry for their root
 		 * cgroup
@@ -1154,13 +1165,14 @@ int cgroup_path(const struct cgroup *cgrp, char *buf, int buflen)
 
 	*--start = '\0';
 	for (;;) {
-		int len = cgrp->dentry->d_name.len;
+		int len = dentry->d_name.len;
 		if ((start -= len) < buf)
 			return -ENAMETOOLONG;
 		memcpy(start, cgrp->dentry->d_name.name, len);
 		cgrp = cgrp->parent;
 		if (!cgrp)
 			break;
+		dentry = rcu_dereference(cgrp->dentry);
 		if (!cgrp->parent)
 			continue;
 		if (--start < buf)
@@ -1663,7 +1675,7 @@ static int cgroup_create_dir(struct cgroup *cgrp, struct dentry *dentry,
 	if (!error) {
 		dentry->d_fsdata = cgrp;
 		inc_nlink(parent->d_inode);
-		cgrp->dentry = dentry;
+		rcu_assign_pointer(cgrp->dentry, dentry);
 		dget(dentry);
 	}
 	dput(dentry);
-- 
cgit v1.2.3


From 7a81b88cb53e335ff7d019e6398c95792c817d93 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:07:48 -0800
Subject: memcg: introduce charge-commit-cancel style of functions

There is a small race in do_swap_page().  When the page swapped-in is
charged, the mapcount can be greater than 0.  But, at the same time some
process (shares it ) call unmap and make mapcount 1->0 and the page is
uncharged.

      CPUA 			CPUB
       mapcount == 1.
   (1) charge if mapcount==0     zap_pte_range()
                                (2) mapcount 1 => 0.
			        (3) uncharge(). (success)
   (4) set page's rmap()
       mapcount 0=>1

Then, this swap page's account is leaked.

For fixing this, I added a new interface.
  - charge
   account to res_counter by PAGE_SIZE and try to free pages if necessary.
  - commit
   register page_cgroup and add to LRU if necessary.
  - cancel
   uncharge PAGE_SIZE because of do_swap_page failure.

     CPUA
  (1) charge (always)
  (2) set page's rmap (mapcount > 0)
  (3) commit charge was necessary or not after set_pte().

This protocol uses PCG_USED bit on page_cgroup for avoiding over accounting.
Usual mem_cgroup_charge_common() does charge -> commit at a time.

And this patch also adds following function to clarify all charges.

  - mem_cgroup_newpage_charge() ....replacement for mem_cgroup_charge()
	called against newly allocated anon pages.

  - mem_cgroup_charge_migrate_fixup()
        called only from remove_migration_ptes().
	we'll have to rewrite this later.(this patch just keeps old behavior)
	This function will be removed by additional patch to make migration
	clearer.

Good for clarifying "what we do"

Then, we have 4 following charge points.
  - newpage
  - swap-in
  - add-to-cache.
  - migration.

[akpm@linux-foundation.org: add missing inline directives to stubs]
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h |  36 ++++++++++-
 mm/memcontrol.c            | 155 ++++++++++++++++++++++++++++++++++++---------
 mm/memory.c                |  12 ++--
 mm/migrate.c               |   2 +-
 mm/swapfile.c              |   6 +-
 5 files changed, 170 insertions(+), 41 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 1fbe14d39521..c592f315cd02 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -27,8 +27,17 @@ struct mm_struct;
 
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR
 
-extern int mem_cgroup_charge(struct page *page, struct mm_struct *mm,
+extern int mem_cgroup_newpage_charge(struct page *page, struct mm_struct *mm,
 				gfp_t gfp_mask);
+extern int mem_cgroup_charge_migrate_fixup(struct page *page,
+				struct mm_struct *mm, gfp_t gfp_mask);
+/* for swap handling */
+extern int mem_cgroup_try_charge(struct mm_struct *mm,
+		gfp_t gfp_mask, struct mem_cgroup **ptr);
+extern void mem_cgroup_commit_charge_swapin(struct page *page,
+					struct mem_cgroup *ptr);
+extern void mem_cgroup_cancel_charge_swapin(struct mem_cgroup *ptr);
+
 extern int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 					gfp_t gfp_mask);
 extern void mem_cgroup_move_lists(struct page *page, enum lru_list lru);
@@ -71,7 +80,9 @@ extern long mem_cgroup_calc_reclaim(struct mem_cgroup *mem, struct zone *zone,
 
 
 #else /* CONFIG_CGROUP_MEM_RES_CTLR */
-static inline int mem_cgroup_charge(struct page *page,
+struct mem_cgroup;
+
+static inline int mem_cgroup_newpage_charge(struct page *page,
 					struct mm_struct *mm, gfp_t gfp_mask)
 {
 	return 0;
@@ -83,6 +94,27 @@ static inline int mem_cgroup_cache_charge(struct page *page,
 	return 0;
 }
 
+static inline int mem_cgroup_charge_migrate_fixup(struct page *page,
+					struct mm_struct *mm, gfp_t gfp_mask)
+{
+	return 0;
+}
+
+static inline int mem_cgroup_try_charge(struct mm_struct *mm,
+				gfp_t gfp_mask, struct mem_cgroup **ptr)
+{
+	return 0;
+}
+
+static inline void mem_cgroup_commit_charge_swapin(struct page *page,
+					  struct mem_cgroup *ptr)
+{
+}
+
+static inline void mem_cgroup_cancel_charge_swapin(struct mem_cgroup *ptr)
+{
+}
+
 static inline void mem_cgroup_uncharge_page(struct page *page)
 {
 }
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 51ee96545579..f568b1964551 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -467,35 +467,31 @@ unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
 	return nr_taken;
 }
 
-/*
- * Charge the memory controller for page usage.
- * Return
- * 0 if the charge was successful
- * < 0 if the cgroup is over its limit
+
+/**
+ * mem_cgroup_try_charge - get charge of PAGE_SIZE.
+ * @mm: an mm_struct which is charged against. (when *memcg is NULL)
+ * @gfp_mask: gfp_mask for reclaim.
+ * @memcg: a pointer to memory cgroup which is charged against.
+ *
+ * charge against memory cgroup pointed by *memcg. if *memcg == NULL, estimated
+ * memory cgroup from @mm is got and stored in *memcg.
+ *
+ * Returns 0 if success. -ENOMEM at failure.
  */
-static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm,
-				gfp_t gfp_mask, enum charge_type ctype,
-				struct mem_cgroup *memcg)
+
+int mem_cgroup_try_charge(struct mm_struct *mm,
+			gfp_t gfp_mask, struct mem_cgroup **memcg)
 {
 	struct mem_cgroup *mem;
-	struct page_cgroup *pc;
-	unsigned long nr_retries = MEM_CGROUP_RECLAIM_RETRIES;
-	struct mem_cgroup_per_zone *mz;
-	unsigned long flags;
-
-	pc = lookup_page_cgroup(page);
-	/* can happen at boot */
-	if (unlikely(!pc))
-		return 0;
-	prefetchw(pc);
+	int nr_retries = MEM_CGROUP_RECLAIM_RETRIES;
 	/*
 	 * We always charge the cgroup the mm_struct belongs to.
 	 * The mm_struct's mem_cgroup changes on task migration if the
 	 * thread group leader migrates. It's possible that mm is not
 	 * set, if so charge the init_mm (happens for pagecache usage).
 	 */
-
-	if (likely(!memcg)) {
+	if (likely(!*memcg)) {
 		rcu_read_lock();
 		mem = mem_cgroup_from_task(rcu_dereference(mm->owner));
 		if (unlikely(!mem)) {
@@ -506,15 +502,17 @@ static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm,
 		 * For every charge from the cgroup, increment reference count
 		 */
 		css_get(&mem->css);
+		*memcg = mem;
 		rcu_read_unlock();
 	} else {
-		mem = memcg;
-		css_get(&memcg->css);
+		mem = *memcg;
+		css_get(&mem->css);
 	}
 
+
 	while (unlikely(res_counter_charge(&mem->res, PAGE_SIZE))) {
 		if (!(gfp_mask & __GFP_WAIT))
-			goto out;
+			goto nomem;
 
 		if (try_to_free_mem_cgroup_pages(mem, gfp_mask))
 			continue;
@@ -531,18 +529,37 @@ static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm,
 
 		if (!nr_retries--) {
 			mem_cgroup_out_of_memory(mem, gfp_mask);
-			goto out;
+			goto nomem;
 		}
 	}
+	return 0;
+nomem:
+	css_put(&mem->css);
+	return -ENOMEM;
+}
+
+/*
+ * commit a charge got by mem_cgroup_try_charge() and makes page_cgroup to be
+ * USED state. If already USED, uncharge and return.
+ */
+
+static void __mem_cgroup_commit_charge(struct mem_cgroup *mem,
+				     struct page_cgroup *pc,
+				     enum charge_type ctype)
+{
+	struct mem_cgroup_per_zone *mz;
+	unsigned long flags;
 
+	/* try_charge() can return NULL to *memcg, taking care of it. */
+	if (!mem)
+		return;
 
 	lock_page_cgroup(pc);
 	if (unlikely(PageCgroupUsed(pc))) {
 		unlock_page_cgroup(pc);
 		res_counter_uncharge(&mem->res, PAGE_SIZE);
 		css_put(&mem->css);
-
-		goto done;
+		return;
 	}
 	pc->mem_cgroup = mem;
 	/*
@@ -557,15 +574,39 @@ static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm,
 	__mem_cgroup_add_list(mz, pc);
 	spin_unlock_irqrestore(&mz->lru_lock, flags);
 	unlock_page_cgroup(pc);
+}
 
-done:
+/*
+ * Charge the memory controller for page usage.
+ * Return
+ * 0 if the charge was successful
+ * < 0 if the cgroup is over its limit
+ */
+static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm,
+				gfp_t gfp_mask, enum charge_type ctype,
+				struct mem_cgroup *memcg)
+{
+	struct mem_cgroup *mem;
+	struct page_cgroup *pc;
+	int ret;
+
+	pc = lookup_page_cgroup(page);
+	/* can happen at boot */
+	if (unlikely(!pc))
+		return 0;
+	prefetchw(pc);
+
+	mem = memcg;
+	ret = mem_cgroup_try_charge(mm, gfp_mask, &mem);
+	if (ret)
+		return ret;
+
+	__mem_cgroup_commit_charge(mem, pc, ctype);
 	return 0;
-out:
-	css_put(&mem->css);
-	return -ENOMEM;
 }
 
-int mem_cgroup_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask)
+int mem_cgroup_newpage_charge(struct page *page,
+			      struct mm_struct *mm, gfp_t gfp_mask)
 {
 	if (mem_cgroup_subsys.disabled)
 		return 0;
@@ -586,6 +627,34 @@ int mem_cgroup_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask)
 				MEM_CGROUP_CHARGE_TYPE_MAPPED, NULL);
 }
 
+/*
+ * same as mem_cgroup_newpage_charge(), now.
+ * But what we assume is different from newpage, and this is special case.
+ * treat this in special function. easy for maintenance.
+ */
+
+int mem_cgroup_charge_migrate_fixup(struct page *page,
+				struct mm_struct *mm, gfp_t gfp_mask)
+{
+	if (mem_cgroup_subsys.disabled)
+		return 0;
+
+	if (PageCompound(page))
+		return 0;
+
+	if (page_mapped(page) || (page->mapping && !PageAnon(page)))
+		return 0;
+
+	if (unlikely(!mm))
+		mm = &init_mm;
+
+	return mem_cgroup_charge_common(page, mm, gfp_mask,
+				MEM_CGROUP_CHARGE_TYPE_MAPPED, NULL);
+}
+
+
+
+
 int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 				gfp_t gfp_mask)
 {
@@ -628,6 +697,30 @@ int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 				MEM_CGROUP_CHARGE_TYPE_SHMEM, NULL);
 }
 
+
+void mem_cgroup_commit_charge_swapin(struct page *page, struct mem_cgroup *ptr)
+{
+	struct page_cgroup *pc;
+
+	if (mem_cgroup_subsys.disabled)
+		return;
+	if (!ptr)
+		return;
+	pc = lookup_page_cgroup(page);
+	__mem_cgroup_commit_charge(ptr, pc, MEM_CGROUP_CHARGE_TYPE_MAPPED);
+}
+
+void mem_cgroup_cancel_charge_swapin(struct mem_cgroup *mem)
+{
+	if (mem_cgroup_subsys.disabled)
+		return;
+	if (!mem)
+		return;
+	res_counter_uncharge(&mem->res, PAGE_SIZE);
+	css_put(&mem->css);
+}
+
+
 /*
  * uncharge if !page_mapped(page)
  */
diff --git a/mm/memory.c b/mm/memory.c
index 3f8fa06b963b..7f210f160990 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2000,7 +2000,7 @@ gotten:
 	cow_user_page(new_page, old_page, address, vma);
 	__SetPageUptodate(new_page);
 
-	if (mem_cgroup_charge(new_page, mm, GFP_KERNEL))
+	if (mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL))
 		goto oom_free_new;
 
 	/*
@@ -2392,6 +2392,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	struct page *page;
 	swp_entry_t entry;
 	pte_t pte;
+	struct mem_cgroup *ptr = NULL;
 	int ret = 0;
 
 	if (!pte_unmap_same(mm, pmd, page_table, orig_pte))
@@ -2430,7 +2431,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	lock_page(page);
 	delayacct_clear_flag(DELAYACCT_PF_SWAPIN);
 
-	if (mem_cgroup_charge(page, mm, GFP_KERNEL)) {
+	if (mem_cgroup_try_charge(mm, GFP_KERNEL, &ptr) == -ENOMEM) {
 		ret = VM_FAULT_OOM;
 		unlock_page(page);
 		goto out;
@@ -2460,6 +2461,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	flush_icache_page(vma, page);
 	set_pte_at(mm, address, page_table, pte);
 	page_add_anon_rmap(page, vma, address);
+	mem_cgroup_commit_charge_swapin(page, ptr);
 
 	swap_free(entry);
 	if (vm_swap_full() || (vma->vm_flags & VM_LOCKED) || PageMlocked(page))
@@ -2480,7 +2482,7 @@ unlock:
 out:
 	return ret;
 out_nomap:
-	mem_cgroup_uncharge_page(page);
+	mem_cgroup_cancel_charge_swapin(ptr);
 	pte_unmap_unlock(page_table, ptl);
 	unlock_page(page);
 	page_cache_release(page);
@@ -2510,7 +2512,7 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		goto oom;
 	__SetPageUptodate(page);
 
-	if (mem_cgroup_charge(page, mm, GFP_KERNEL))
+	if (mem_cgroup_newpage_charge(page, mm, GFP_KERNEL))
 		goto oom_free_page;
 
 	entry = mk_pte(page, vma->vm_page_prot);
@@ -2601,7 +2603,7 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 				ret = VM_FAULT_OOM;
 				goto out;
 			}
-			if (mem_cgroup_charge(page, mm, GFP_KERNEL)) {
+			if (mem_cgroup_newpage_charge(page, mm, GFP_KERNEL)) {
 				ret = VM_FAULT_OOM;
 				page_cache_release(page);
 				goto out;
diff --git a/mm/migrate.c b/mm/migrate.c
index 55373983c9c6..246dcb973ae7 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -133,7 +133,7 @@ static void remove_migration_pte(struct vm_area_struct *vma,
 	 * be reliable, and this charge can actually fail: oh well, we don't
 	 * make the situation any worse by proceeding as if it had succeeded.
 	 */
-	mem_cgroup_charge(new, mm, GFP_ATOMIC);
+	mem_cgroup_charge_migrate_fixup(new, mm, GFP_ATOMIC);
 
 	get_page(new);
 	pte = pte_mkold(mk_pte(new, vma->vm_page_prot));
diff --git a/mm/swapfile.c b/mm/swapfile.c
index eec5ca758a23..fb926efb5167 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -690,17 +690,18 @@ unsigned int count_swap_pages(int type, int free)
 static int unuse_pte(struct vm_area_struct *vma, pmd_t *pmd,
 		unsigned long addr, swp_entry_t entry, struct page *page)
 {
+	struct mem_cgroup *ptr = NULL;
 	spinlock_t *ptl;
 	pte_t *pte;
 	int ret = 1;
 
-	if (mem_cgroup_charge(page, vma->vm_mm, GFP_KERNEL))
+	if (mem_cgroup_try_charge(vma->vm_mm, GFP_KERNEL, &ptr))
 		ret = -ENOMEM;
 
 	pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
 	if (unlikely(!pte_same(*pte, swp_entry_to_pte(entry)))) {
 		if (ret > 0)
-			mem_cgroup_uncharge_page(page);
+			mem_cgroup_cancel_charge_swapin(ptr);
 		ret = 0;
 		goto out;
 	}
@@ -710,6 +711,7 @@ static int unuse_pte(struct vm_area_struct *vma, pmd_t *pmd,
 	set_pte_at(vma->vm_mm, addr, pte,
 		   pte_mkold(mk_pte(page, vma->vm_page_prot)));
 	page_add_anon_rmap(page, vma, addr);
+	mem_cgroup_commit_charge_swapin(page, ptr);
 	swap_free(entry);
 	/*
 	 * Move the page to the active list so it is not
-- 
cgit v1.2.3


From 01b1ae63c2270cbacfd43fea94578c17950eb548 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:07:50 -0800
Subject: memcg: simple migration handling

Now, management of "charge" under page migration is done under following
manner. (Assume migrate page contents from oldpage to newpage)

 before
  - "newpage" is charged before migration.
 at success.
  - "oldpage" is uncharged at somewhere(unmap, radix-tree-replace)
 at failure
  - "newpage" is uncharged.
  - "oldpage" is charged if necessary (*1)

But (*1) is not reliable....because of GFP_ATOMIC.

This patch tries to change behavior as following by charge/commit/cancel ops.

 before
  - charge PAGE_SIZE (no target page)
 success
  - commit charge against "newpage".
 failure
  - commit charge against "oldpage".
    (PCG_USED bit works effectively to avoid double-counting)
  - if "oldpage" is obsolete, cancel charge of PAGE_SIZE.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h |  19 +++-----
 mm/memcontrol.c            | 108 ++++++++++++++++++++++-----------------------
 mm/migrate.c               |  42 ++++++------------
 3 files changed, 73 insertions(+), 96 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index c592f315cd02..b095f5f6ecf7 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -29,8 +29,6 @@ struct mm_struct;
 
 extern int mem_cgroup_newpage_charge(struct page *page, struct mm_struct *mm,
 				gfp_t gfp_mask);
-extern int mem_cgroup_charge_migrate_fixup(struct page *page,
-				struct mm_struct *mm, gfp_t gfp_mask);
 /* for swap handling */
 extern int mem_cgroup_try_charge(struct mm_struct *mm,
 		gfp_t gfp_mask, struct mem_cgroup **ptr);
@@ -60,8 +58,9 @@ extern struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p);
 	((cgroup) == mem_cgroup_from_task((mm)->owner))
 
 extern int
-mem_cgroup_prepare_migration(struct page *page, struct page *newpage);
-extern void mem_cgroup_end_migration(struct page *page);
+mem_cgroup_prepare_migration(struct page *page, struct mem_cgroup **ptr);
+extern void mem_cgroup_end_migration(struct mem_cgroup *mem,
+	struct page *oldpage, struct page *newpage);
 
 /*
  * For memory reclaim.
@@ -94,12 +93,6 @@ static inline int mem_cgroup_cache_charge(struct page *page,
 	return 0;
 }
 
-static inline int mem_cgroup_charge_migrate_fixup(struct page *page,
-					struct mm_struct *mm, gfp_t gfp_mask)
-{
-	return 0;
-}
-
 static inline int mem_cgroup_try_charge(struct mm_struct *mm,
 				gfp_t gfp_mask, struct mem_cgroup **ptr)
 {
@@ -144,12 +137,14 @@ static inline int task_in_mem_cgroup(struct task_struct *task,
 }
 
 static inline int
-mem_cgroup_prepare_migration(struct page *page, struct page *newpage)
+mem_cgroup_prepare_migration(struct page *page, struct mem_cgroup **ptr)
 {
 	return 0;
 }
 
-static inline void mem_cgroup_end_migration(struct page *page)
+static inline void mem_cgroup_end_migration(struct mem_cgroup *mem,
+					struct page *oldpage,
+					struct page *newpage)
 {
 }
 
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index c34eb52bdc3f..b71195e8198b 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -627,34 +627,6 @@ int mem_cgroup_newpage_charge(struct page *page,
 				MEM_CGROUP_CHARGE_TYPE_MAPPED, NULL);
 }
 
-/*
- * same as mem_cgroup_newpage_charge(), now.
- * But what we assume is different from newpage, and this is special case.
- * treat this in special function. easy for maintenance.
- */
-
-int mem_cgroup_charge_migrate_fixup(struct page *page,
-				struct mm_struct *mm, gfp_t gfp_mask)
-{
-	if (mem_cgroup_subsys.disabled)
-		return 0;
-
-	if (PageCompound(page))
-		return 0;
-
-	if (page_mapped(page) || (page->mapping && !PageAnon(page)))
-		return 0;
-
-	if (unlikely(!mm))
-		mm = &init_mm;
-
-	return mem_cgroup_charge_common(page, mm, gfp_mask,
-				MEM_CGROUP_CHARGE_TYPE_MAPPED, NULL);
-}
-
-
-
-
 int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 				gfp_t gfp_mask)
 {
@@ -697,7 +669,6 @@ int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 				MEM_CGROUP_CHARGE_TYPE_SHMEM, NULL);
 }
 
-
 void mem_cgroup_commit_charge_swapin(struct page *page, struct mem_cgroup *ptr)
 {
 	struct page_cgroup *pc;
@@ -782,13 +753,13 @@ void mem_cgroup_uncharge_cache_page(struct page *page)
 }
 
 /*
- * Before starting migration, account against new page.
+ * Before starting migration, account PAGE_SIZE to mem_cgroup that the old
+ * page belongs to.
  */
-int mem_cgroup_prepare_migration(struct page *page, struct page *newpage)
+int mem_cgroup_prepare_migration(struct page *page, struct mem_cgroup **ptr)
 {
 	struct page_cgroup *pc;
 	struct mem_cgroup *mem = NULL;
-	enum charge_type ctype = MEM_CGROUP_CHARGE_TYPE_MAPPED;
 	int ret = 0;
 
 	if (mem_cgroup_subsys.disabled)
@@ -799,42 +770,67 @@ int mem_cgroup_prepare_migration(struct page *page, struct page *newpage)
 	if (PageCgroupUsed(pc)) {
 		mem = pc->mem_cgroup;
 		css_get(&mem->css);
-		if (PageCgroupCache(pc)) {
-			if (page_is_file_cache(page))
-				ctype = MEM_CGROUP_CHARGE_TYPE_CACHE;
-			else
-				ctype = MEM_CGROUP_CHARGE_TYPE_SHMEM;
-		}
 	}
 	unlock_page_cgroup(pc);
+
 	if (mem) {
-		ret = mem_cgroup_charge_common(newpage, NULL,
-					GFP_HIGHUSER_MOVABLE,
-					ctype, mem);
+		ret = mem_cgroup_try_charge(NULL, GFP_HIGHUSER_MOVABLE, &mem);
 		css_put(&mem->css);
 	}
+	*ptr = mem;
 	return ret;
 }
 
 /* remove redundant charge if migration failed*/
-void mem_cgroup_end_migration(struct page *newpage)
+void mem_cgroup_end_migration(struct mem_cgroup *mem,
+		struct page *oldpage, struct page *newpage)
 {
+	struct page *target, *unused;
+	struct page_cgroup *pc;
+	enum charge_type ctype;
+
+	if (!mem)
+		return;
+
+	/* at migration success, oldpage->mapping is NULL. */
+	if (oldpage->mapping) {
+		target = oldpage;
+		unused = NULL;
+	} else {
+		target = newpage;
+		unused = oldpage;
+	}
+
+	if (PageAnon(target))
+		ctype = MEM_CGROUP_CHARGE_TYPE_MAPPED;
+	else if (page_is_file_cache(target))
+		ctype = MEM_CGROUP_CHARGE_TYPE_CACHE;
+	else
+		ctype = MEM_CGROUP_CHARGE_TYPE_SHMEM;
+
+	/* unused page is not on radix-tree now. */
+	if (unused && ctype != MEM_CGROUP_CHARGE_TYPE_MAPPED)
+		__mem_cgroup_uncharge_common(unused, ctype);
+
+	pc = lookup_page_cgroup(target);
 	/*
-	 * At success, page->mapping is not NULL.
-	 * special rollback care is necessary when
-	 * 1. at migration failure. (newpage->mapping is cleared in this case)
-	 * 2. the newpage was moved but not remapped again because the task
-	 *    exits and the newpage is obsolete. In this case, the new page
-	 *    may be a swapcache. So, we just call mem_cgroup_uncharge_page()
-	 *    always for avoiding mess. The  page_cgroup will be removed if
-	 *    unnecessary. File cache pages is still on radix-tree. Don't
-	 *    care it.
+	 * __mem_cgroup_commit_charge() check PCG_USED bit of page_cgroup.
+	 * So, double-counting is effectively avoided.
+	 */
+	__mem_cgroup_commit_charge(mem, pc, ctype);
+
+	/*
+	 * Both of oldpage and newpage are still under lock_page().
+	 * Then, we don't have to care about race in radix-tree.
+	 * But we have to be careful that this page is unmapped or not.
+	 *
+	 * There is a case for !page_mapped(). At the start of
+	 * migration, oldpage was mapped. But now, it's zapped.
+	 * But we know *target* page is not freed/reused under us.
+	 * mem_cgroup_uncharge_page() does all necessary checks.
 	 */
-	if (!newpage->mapping)
-		__mem_cgroup_uncharge_common(newpage,
-				MEM_CGROUP_CHARGE_TYPE_FORCE);
-	else if (PageAnon(newpage))
-		mem_cgroup_uncharge_page(newpage);
+	if (ctype == MEM_CGROUP_CHARGE_TYPE_MAPPED)
+		mem_cgroup_uncharge_page(target);
 }
 
 /*
diff --git a/mm/migrate.c b/mm/migrate.c
index 246dcb973ae7..a30ea5fcf9f1 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -121,20 +121,6 @@ static void remove_migration_pte(struct vm_area_struct *vma,
 	if (!is_migration_entry(entry) || migration_entry_to_page(entry) != old)
 		goto out;
 
-	/*
-	 * Yes, ignore the return value from a GFP_ATOMIC mem_cgroup_charge.
-	 * Failure is not an option here: we're now expected to remove every
-	 * migration pte, and will cause crashes otherwise.  Normally this
-	 * is not an issue: mem_cgroup_prepare_migration bumped up the old
-	 * page_cgroup count for safety, that's now attached to the new page,
-	 * so this charge should just be another incrementation of the count,
-	 * to keep in balance with rmap.c's mem_cgroup_uncharging.  But if
-	 * there's been a force_empty, those reference counts may no longer
-	 * be reliable, and this charge can actually fail: oh well, we don't
-	 * make the situation any worse by proceeding as if it had succeeded.
-	 */
-	mem_cgroup_charge_migrate_fixup(new, mm, GFP_ATOMIC);
-
 	get_page(new);
 	pte = pte_mkold(mk_pte(new, vma->vm_page_prot));
 	if (is_write_migration_entry(entry))
@@ -378,9 +364,6 @@ static void migrate_page_copy(struct page *newpage, struct page *page)
 	anon = PageAnon(page);
 	page->mapping = NULL;
 
-	if (!anon) /* This page was removed from radix-tree. */
-		mem_cgroup_uncharge_cache_page(page);
-
 	/*
 	 * If any waiters have accumulated on the new page then
 	 * wake them up.
@@ -614,6 +597,7 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
 	struct page *newpage = get_new_page(page, private, &result);
 	int rcu_locked = 0;
 	int charge = 0;
+	struct mem_cgroup *mem;
 
 	if (!newpage)
 		return -ENOMEM;
@@ -623,24 +607,26 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
 		goto move_newpage;
 	}
 
-	charge = mem_cgroup_prepare_migration(page, newpage);
-	if (charge == -ENOMEM) {
-		rc = -ENOMEM;
-		goto move_newpage;
-	}
 	/* prepare cgroup just returns 0 or -ENOMEM */
-	BUG_ON(charge);
-
 	rc = -EAGAIN;
+
 	if (!trylock_page(page)) {
 		if (!force)
 			goto move_newpage;
 		lock_page(page);
 	}
 
+	/* charge against new page */
+	charge = mem_cgroup_prepare_migration(page, &mem);
+	if (charge == -ENOMEM) {
+		rc = -ENOMEM;
+		goto unlock;
+	}
+	BUG_ON(charge);
+
 	if (PageWriteback(page)) {
 		if (!force)
-			goto unlock;
+			goto uncharge;
 		wait_on_page_writeback(page);
 	}
 	/*
@@ -693,7 +679,9 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
 rcu_unlock:
 	if (rcu_locked)
 		rcu_read_unlock();
-
+uncharge:
+	if (!charge)
+		mem_cgroup_end_migration(mem, page, newpage);
 unlock:
 	unlock_page(page);
 
@@ -709,8 +697,6 @@ unlock:
 	}
 
 move_newpage:
-	if (!charge)
-		mem_cgroup_end_migration(newpage);
 
 	/*
 	 * Move the new page to the LRU. If migration was not successful
-- 
cgit v1.2.3


From d13d144309d2e5a3e6ad978b16c1d0226ddc9231 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:07:56 -0800
Subject: memcg: handle swap caches

SwapCache support for memory resource controller (memcg)

Before mem+swap controller, memcg itself should handle SwapCache in proper
way.  This is cut-out from it.

In current memcg, SwapCache is just leaked and the user can create tons of
SwapCache.  This is a leak of account and should be handled.

SwapCache accounting is done as following.

  charge (anon)
	- charged when it's mapped.
	  (because of readahead, charge at add_to_swap_cache() is not sane)
  uncharge (anon)
	- uncharged when it's dropped from swapcache and fully unmapped.
	  means it's not uncharged at unmap.
	  Note: delete from swap cache at swap-in is done after rmap information
	        is established.
  charge (shmem)
	- charged at swap-in. this prevents charge at add_to_page_cache().

  uncharge (shmem)
	- uncharged when it's dropped from swapcache and not on shmem's
	  radix-tree.

  at migration, check against 'old page' is modified to handle shmem.

Comparing to the old version discussed (and caused troubles), we have
advantages of
  - PCG_USED bit.
  - simple migrating handling.

So, situation is much easier than several months ago, maybe.

[hugh@veritas.com: memcg: handle swap caches build fix]
Reviewed-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Tested-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/controllers/memory.txt |  5 +++
 include/linux/swap.h                 | 22 ++++++++++++
 mm/memcontrol.c                      | 67 ++++++++++++++++++++++++++++++++----
 mm/shmem.c                           | 18 ++++++++--
 mm/swap_state.c                      |  1 +
 5 files changed, 105 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/controllers/memory.txt b/Documentation/controllers/memory.txt
index 54253b7a8db2..9fe2d0eabe05 100644
--- a/Documentation/controllers/memory.txt
+++ b/Documentation/controllers/memory.txt
@@ -137,6 +137,11 @@ behind this approach is that a cgroup that aggressively uses a shared
 page will eventually get charged for it (once it is uncharged from
 the cgroup that brought it in -- this will happen on memory pressure).
 
+Exception: When you do swapoff and make swapped-out pages of shmem(tmpfs) to
+be backed into memory in force, charges for pages are accounted against the
+caller of swapoff rather than the users of shmem.
+
+
 2.4 Reclaim
 
 Each cgroup maintains a per cgroup LRU that consists of an active
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 91dee50fe260..f8f3907533f0 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -333,6 +333,22 @@ static inline void disable_swap_token(void)
 	put_swap_token(swap_token_mm);
 }
 
+#ifdef CONFIG_CGROUP_MEM_RES_CTLR
+extern int mem_cgroup_cache_charge_swapin(struct page *page,
+				struct mm_struct *mm, gfp_t mask, bool locked);
+extern void mem_cgroup_uncharge_swapcache(struct page *page);
+#else
+static inline
+int mem_cgroup_cache_charge_swapin(struct page *page,
+				struct mm_struct *mm, gfp_t mask, bool locked)
+{
+	return 0;
+}
+static inline void mem_cgroup_uncharge_swapcache(struct page *page)
+{
+}
+#endif
+
 #else /* CONFIG_SWAP */
 
 #define nr_swap_pages				0L
@@ -409,6 +425,12 @@ static inline swp_entry_t get_swap_page(void)
 #define has_swap_token(x) 0
 #define disable_swap_token() do { } while(0)
 
+static inline int mem_cgroup_cache_charge_swapin(struct page *page,
+			struct mm_struct *mm, gfp_t mask, bool locked)
+{
+	return 0;
+}
+
 #endif /* CONFIG_SWAP */
 #endif /* __KERNEL__*/
 #endif /* _LINUX_SWAP_H */
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index decace3bb57e..7288e9d85ca7 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -21,6 +21,7 @@
 #include <linux/memcontrol.h>
 #include <linux/cgroup.h>
 #include <linux/mm.h>
+#include <linux/pagemap.h>
 #include <linux/smp.h>
 #include <linux/page-flags.h>
 #include <linux/backing-dev.h>
@@ -139,6 +140,7 @@ enum charge_type {
 	MEM_CGROUP_CHARGE_TYPE_MAPPED,
 	MEM_CGROUP_CHARGE_TYPE_SHMEM,	/* used by page migration of shmem */
 	MEM_CGROUP_CHARGE_TYPE_FORCE,	/* used by force_empty */
+	MEM_CGROUP_CHARGE_TYPE_SWAPOUT,	/* for accounting swapcache */
 	NR_CHARGE_TYPE,
 };
 
@@ -780,6 +782,33 @@ int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 				MEM_CGROUP_CHARGE_TYPE_SHMEM, NULL);
 }
 
+#ifdef CONFIG_SWAP
+int mem_cgroup_cache_charge_swapin(struct page *page,
+			struct mm_struct *mm, gfp_t mask, bool locked)
+{
+	int ret = 0;
+
+	if (mem_cgroup_subsys.disabled)
+		return 0;
+	if (unlikely(!mm))
+		mm = &init_mm;
+	if (!locked)
+		lock_page(page);
+	/*
+	 * If not locked, the page can be dropped from SwapCache until
+	 * we reach here.
+	 */
+	if (PageSwapCache(page)) {
+		ret = mem_cgroup_charge_common(page, mm, mask,
+				MEM_CGROUP_CHARGE_TYPE_SHMEM, NULL);
+	}
+	if (!locked)
+		unlock_page(page);
+
+	return ret;
+}
+#endif
+
 void mem_cgroup_commit_charge_swapin(struct page *page, struct mem_cgroup *ptr)
 {
 	struct page_cgroup *pc;
@@ -817,6 +846,9 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype)
 	if (mem_cgroup_subsys.disabled)
 		return;
 
+	if (PageSwapCache(page))
+		return;
+
 	/*
 	 * Check if our page_cgroup is valid
 	 */
@@ -825,12 +857,26 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype)
 		return;
 
 	lock_page_cgroup(pc);
-	if ((ctype == MEM_CGROUP_CHARGE_TYPE_MAPPED && page_mapped(page))
-	     || !PageCgroupUsed(pc)) {
-		/* This happens at race in zap_pte_range() and do_swap_page()*/
-		unlock_page_cgroup(pc);
-		return;
+
+	if (!PageCgroupUsed(pc))
+		goto unlock_out;
+
+	switch (ctype) {
+	case MEM_CGROUP_CHARGE_TYPE_MAPPED:
+		if (page_mapped(page))
+			goto unlock_out;
+		break;
+	case MEM_CGROUP_CHARGE_TYPE_SWAPOUT:
+		if (!PageAnon(page)) {	/* Shared memory */
+			if (page->mapping && !page_is_file_cache(page))
+				goto unlock_out;
+		} else if (page_mapped(page)) /* Anon */
+				goto unlock_out;
+		break;
+	default:
+		break;
 	}
+
 	ClearPageCgroupUsed(pc);
 	mem = pc->mem_cgroup;
 
@@ -844,6 +890,10 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype)
 	css_put(&mem->css);
 
 	return;
+
+unlock_out:
+	unlock_page_cgroup(pc);
+	return;
 }
 
 void mem_cgroup_uncharge_page(struct page *page)
@@ -863,6 +913,11 @@ void mem_cgroup_uncharge_cache_page(struct page *page)
 	__mem_cgroup_uncharge_common(page, MEM_CGROUP_CHARGE_TYPE_CACHE);
 }
 
+void mem_cgroup_uncharge_swapcache(struct page *page)
+{
+	__mem_cgroup_uncharge_common(page, MEM_CGROUP_CHARGE_TYPE_SWAPOUT);
+}
+
 /*
  * Before starting migration, account PAGE_SIZE to mem_cgroup that the old
  * page belongs to.
@@ -920,7 +975,7 @@ void mem_cgroup_end_migration(struct mem_cgroup *mem,
 		ctype = MEM_CGROUP_CHARGE_TYPE_SHMEM;
 
 	/* unused page is not on radix-tree now. */
-	if (unused && ctype != MEM_CGROUP_CHARGE_TYPE_MAPPED)
+	if (unused)
 		__mem_cgroup_uncharge_common(unused, ctype);
 
 	pc = lookup_page_cgroup(target);
diff --git a/mm/shmem.c b/mm/shmem.c
index bd9b4ea307b2..adf5c3eedbc9 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -928,8 +928,12 @@ found:
 	error = 1;
 	if (!inode)
 		goto out;
-	/* Charge page using GFP_HIGHUSER_MOVABLE while we can wait */
-	error = mem_cgroup_cache_charge(page, current->mm, GFP_HIGHUSER_MOVABLE);
+	/*
+	 * Charge page using GFP_HIGHUSER_MOVABLE while we can wait.
+	 * charged back to the user(not to caller) when swap account is used.
+	 */
+	error = mem_cgroup_cache_charge_swapin(page,
+			current->mm, GFP_HIGHUSER_MOVABLE, true);
 	if (error)
 		goto out;
 	error = radix_tree_preload(GFP_KERNEL);
@@ -1266,6 +1270,16 @@ repeat:
 				goto repeat;
 			}
 			wait_on_page_locked(swappage);
+			/*
+			 * We want to avoid charge at add_to_page_cache().
+			 * charge against this swap cache here.
+			 */
+			if (mem_cgroup_cache_charge_swapin(swappage,
+						current->mm, gfp, false)) {
+				page_cache_release(swappage);
+				error = -ENOMEM;
+				goto failed;
+			}
 			page_cache_release(swappage);
 			goto repeat;
 		}
diff --git a/mm/swap_state.c b/mm/swap_state.c
index 81c825f67a7f..09291ca11f5f 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -118,6 +118,7 @@ void __delete_from_swap_cache(struct page *page)
 	total_swapcache_pages--;
 	__dec_zone_page_state(page, NR_FILE_PAGES);
 	INC_CACHE_INFO(del_total);
+	mem_cgroup_uncharge_swapcache(page);
 }
 
 /**
-- 
cgit v1.2.3


From c077719be8e9e6b55702117513d1b5f41d80404a Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:07:57 -0800
Subject: memcg: mem+swap controller Kconfig

Config and control variable for mem+swap controller.

This patch adds CONFIG_CGROUP_MEM_RES_CTLR_SWAP
(memory resource controller swap extension.)

For accounting swap, it's obvious that we have to use additional memory to
remember "who uses swap".  This adds more overhead.  So, it's better to
offer "choice" to users.  This patch adds 2 choices.

This patch adds 2 parameters to enable swap extension or not.
  - CONFIG
  - boot option

Reviewed-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/kernel-parameters.txt |  3 +++
 include/linux/memcontrol.h          |  3 +++
 init/Kconfig                        | 17 +++++++++++++++++
 mm/memcontrol.c                     | 34 ++++++++++++++++++++++++++++++++++
 4 files changed, 57 insertions(+)

(limited to 'include/linux')

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 532eacbbed62..fb849020aea9 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1562,6 +1562,9 @@ and is between 256 and 4096 characters. It is defined in the file
 
 	nosoftlockup	[KNL] Disable the soft-lockup detector.
 
+	noswapaccount	[KNL] Disable accounting of swap in memory resource
+			controller. (See Documentation/controllers/memory.txt)
+
 	nosync		[HW,M68K] Disables sync negotiation for all devices.
 
 	notsc		[BUGS=X86-32] Disable Time Stamp Counter
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index b095f5f6ecf7..41b46cc9d1f1 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -77,6 +77,9 @@ extern void mem_cgroup_record_reclaim_priority(struct mem_cgroup *mem,
 extern long mem_cgroup_calc_reclaim(struct mem_cgroup *mem, struct zone *zone,
 					int priority, enum lru_list lru);
 
+#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
+extern int do_swap_account;
+#endif
 
 #else /* CONFIG_CGROUP_MEM_RES_CTLR */
 struct mem_cgroup;
diff --git a/init/Kconfig b/init/Kconfig
index 7cbe1f43ca22..a724a149bf3f 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -428,6 +428,23 @@ config CGROUP_MEM_RES_CTLR
 config MM_OWNER
 	bool
 
+config CGROUP_MEM_RES_CTLR_SWAP
+	bool "Memory Resource Controller Swap Extension(EXPERIMENTAL)"
+	depends on CGROUP_MEM_RES_CTLR && SWAP && EXPERIMENTAL
+	help
+	  Add swap management feature to memory resource controller. When you
+	  enable this, you can limit mem+swap usage per cgroup. In other words,
+	  when you disable this, memory resource controller has no cares to
+	  usage of swap...a process can exhaust all of the swap. This extension
+	  is useful when you want to avoid exhaustion swap but this itself
+	  adds more overheads and consumes memory for remembering information.
+	  Especially if you use 32bit system or small memory system, please
+	  be careful about enabling this. When memory resource controller
+	  is disabled by boot option, this will be automatically disabled and
+	  there will be no overhead from this. Even when you set this config=y,
+	  if boot option "noswapaccount" is set, swap will not be accounted.
+
+
 endmenu
 
 config SYSFS_DEPRECATED
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 7288e9d85ca7..59dd8c116372 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -41,6 +41,15 @@
 struct cgroup_subsys mem_cgroup_subsys __read_mostly;
 #define MEM_CGROUP_RECLAIM_RETRIES	5
 
+#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
+/* Turned on only when memory cgroup is enabled && really_do_swap_account = 0 */
+int do_swap_account __read_mostly;
+static int really_do_swap_account __initdata = 1; /* for remember boot option*/
+#else
+#define do_swap_account		(0)
+#endif
+
+
 /*
  * Statistics for memory cgroup.
  */
@@ -1404,6 +1413,18 @@ static void mem_cgroup_free(struct mem_cgroup *mem)
 }
 
 
+#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
+static void __init enable_swap_cgroup(void)
+{
+	if (!mem_cgroup_subsys.disabled && really_do_swap_account)
+		do_swap_account = 1;
+}
+#else
+static void __init enable_swap_cgroup(void)
+{
+}
+#endif
+
 static struct cgroup_subsys_state *
 mem_cgroup_create(struct cgroup_subsys *ss, struct cgroup *cont)
 {
@@ -1419,6 +1440,9 @@ mem_cgroup_create(struct cgroup_subsys *ss, struct cgroup *cont)
 	for_each_node_state(node, N_POSSIBLE)
 		if (alloc_mem_cgroup_per_zone_info(mem, node))
 			goto free_out;
+	/* root ? */
+	if (cont->parent == NULL)
+		enable_swap_cgroup();
 
 	return &mem->css;
 free_out:
@@ -1490,3 +1514,13 @@ struct cgroup_subsys mem_cgroup_subsys = {
 	.attach = mem_cgroup_move_task,
 	.early_init = 0,
 };
+
+#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
+
+static int __init disable_swap_account(char *s)
+{
+	really_do_swap_account = 0;
+	return 1;
+}
+__setup("noswapaccount", disable_swap_account);
+#endif
-- 
cgit v1.2.3


From 27a7faa0779dd13729196c1a818c294f44bbd1ee Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:07:58 -0800
Subject: memcg: swap cgroup for remembering usage

For accounting swap, we need a record per swap entry, at least.

This patch adds following function.
  - swap_cgroup_swapon() .... called from swapon
  - swap_cgroup_swapoff() ... called at the end of swapoff.

  - swap_cgroup_record() .... record information of swap entry.
  - swap_cgroup_lookup() .... lookup information of swap entry.

This patch just implements "how to record information".  No actual method
for limit the usage of swap.  These routine uses flat table to record and
lookup.  "wise" lookup system like radix-tree requires requires memory
allocation at new records but swap-out is usually called under memory
shortage (or memcg hits limit.) So, I used static allocation.  (maybe
dynamic allocation is not very hard but it adds additional memory
allocation in memory shortage path.)

Note1: In this, we use pointer to record information and this means
      8bytes per swap entry. I think we can reduce this when we
      create "id of cgroup" in the range of 0-65535 or 0-255.

Reported-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Reviewed-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Tested-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Reported-by: Hugh Dickins <hugh@veritas.com>
Reported-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Reported-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Pavel Emelianov <xemul@openvz.org>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/page_cgroup.h |  35 ++++++++
 mm/page_cgroup.c            | 197 ++++++++++++++++++++++++++++++++++++++++++++
 mm/swapfile.c               |  10 +++
 3 files changed, 242 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/page_cgroup.h b/include/linux/page_cgroup.h
index 1e6d34bfa094..d754b2dfbf2d 100644
--- a/include/linux/page_cgroup.h
+++ b/include/linux/page_cgroup.h
@@ -104,5 +104,40 @@ static inline void page_cgroup_init(void)
 {
 }
 
+#endif
+
+#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
+#include <linux/swap.h>
+extern struct mem_cgroup *
+swap_cgroup_record(swp_entry_t ent, struct mem_cgroup *mem);
+extern struct mem_cgroup *lookup_swap_cgroup(swp_entry_t ent);
+extern int swap_cgroup_swapon(int type, unsigned long max_pages);
+extern void swap_cgroup_swapoff(int type);
+#else
+#include <linux/swap.h>
+
+static inline
+struct mem_cgroup *swap_cgroup_record(swp_entry_t ent, struct mem_cgroup *mem)
+{
+	return NULL;
+}
+
+static inline
+struct mem_cgroup *lookup_swap_cgroup(swp_entry_t ent)
+{
+	return NULL;
+}
+
+static inline int
+swap_cgroup_swapon(int type, unsigned long max_pages)
+{
+	return 0;
+}
+
+static inline void swap_cgroup_swapoff(int type)
+{
+	return;
+}
+
 #endif
 #endif
diff --git a/mm/page_cgroup.c b/mm/page_cgroup.c
index df1e54a5ed19..685e7c8e1fd6 100644
--- a/mm/page_cgroup.c
+++ b/mm/page_cgroup.c
@@ -8,6 +8,7 @@
 #include <linux/memory.h>
 #include <linux/vmalloc.h>
 #include <linux/cgroup.h>
+#include <linux/swapops.h>
 
 static void __meminit
 __init_page_cgroup(struct page_cgroup *pc, unsigned long pfn)
@@ -270,3 +271,199 @@ void __meminit pgdat_page_cgroup_init(struct pglist_data *pgdat)
 }
 
 #endif
+
+
+#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
+
+static DEFINE_MUTEX(swap_cgroup_mutex);
+struct swap_cgroup_ctrl {
+	struct page **map;
+	unsigned long length;
+};
+
+struct swap_cgroup_ctrl swap_cgroup_ctrl[MAX_SWAPFILES];
+
+/*
+ * This 8bytes seems big..maybe we can reduce this when we can use "id" for
+ * cgroup rather than pointer.
+ */
+struct swap_cgroup {
+	struct mem_cgroup	*val;
+};
+#define SC_PER_PAGE	(PAGE_SIZE/sizeof(struct swap_cgroup))
+#define SC_POS_MASK	(SC_PER_PAGE - 1)
+
+/*
+ * SwapCgroup implements "lookup" and "exchange" operations.
+ * In typical usage, this swap_cgroup is accessed via memcg's charge/uncharge
+ * against SwapCache. At swap_free(), this is accessed directly from swap.
+ *
+ * This means,
+ *  - we have no race in "exchange" when we're accessed via SwapCache because
+ *    SwapCache(and its swp_entry) is under lock.
+ *  - When called via swap_free(), there is no user of this entry and no race.
+ * Then, we don't need lock around "exchange".
+ *
+ * TODO: we can push these buffers out to HIGHMEM.
+ */
+
+/*
+ * allocate buffer for swap_cgroup.
+ */
+static int swap_cgroup_prepare(int type)
+{
+	struct page *page;
+	struct swap_cgroup_ctrl *ctrl;
+	unsigned long idx, max;
+
+	if (!do_swap_account)
+		return 0;
+	ctrl = &swap_cgroup_ctrl[type];
+
+	for (idx = 0; idx < ctrl->length; idx++) {
+		page = alloc_page(GFP_KERNEL | __GFP_ZERO);
+		if (!page)
+			goto not_enough_page;
+		ctrl->map[idx] = page;
+	}
+	return 0;
+not_enough_page:
+	max = idx;
+	for (idx = 0; idx < max; idx++)
+		__free_page(ctrl->map[idx]);
+
+	return -ENOMEM;
+}
+
+/**
+ * swap_cgroup_record - record mem_cgroup for this swp_entry.
+ * @ent: swap entry to be recorded into
+ * @mem: mem_cgroup to be recorded
+ *
+ * Returns old value at success, NULL at failure.
+ * (Of course, old value can be NULL.)
+ */
+struct mem_cgroup *swap_cgroup_record(swp_entry_t ent, struct mem_cgroup *mem)
+{
+	int type = swp_type(ent);
+	unsigned long offset = swp_offset(ent);
+	unsigned long idx = offset / SC_PER_PAGE;
+	unsigned long pos = offset & SC_POS_MASK;
+	struct swap_cgroup_ctrl *ctrl;
+	struct page *mappage;
+	struct swap_cgroup *sc;
+	struct mem_cgroup *old;
+
+	if (!do_swap_account)
+		return NULL;
+
+	ctrl = &swap_cgroup_ctrl[type];
+
+	mappage = ctrl->map[idx];
+	sc = page_address(mappage);
+	sc += pos;
+	old = sc->val;
+	sc->val = mem;
+
+	return old;
+}
+
+/**
+ * lookup_swap_cgroup - lookup mem_cgroup tied to swap entry
+ * @ent: swap entry to be looked up.
+ *
+ * Returns pointer to mem_cgroup at success. NULL at failure.
+ */
+struct mem_cgroup *lookup_swap_cgroup(swp_entry_t ent)
+{
+	int type = swp_type(ent);
+	unsigned long offset = swp_offset(ent);
+	unsigned long idx = offset / SC_PER_PAGE;
+	unsigned long pos = offset & SC_POS_MASK;
+	struct swap_cgroup_ctrl *ctrl;
+	struct page *mappage;
+	struct swap_cgroup *sc;
+	struct mem_cgroup *ret;
+
+	if (!do_swap_account)
+		return NULL;
+
+	ctrl = &swap_cgroup_ctrl[type];
+	mappage = ctrl->map[idx];
+	sc = page_address(mappage);
+	sc += pos;
+	ret = sc->val;
+	return ret;
+}
+
+int swap_cgroup_swapon(int type, unsigned long max_pages)
+{
+	void *array;
+	unsigned long array_size;
+	unsigned long length;
+	struct swap_cgroup_ctrl *ctrl;
+
+	if (!do_swap_account)
+		return 0;
+
+	length = ((max_pages/SC_PER_PAGE) + 1);
+	array_size = length * sizeof(void *);
+
+	array = vmalloc(array_size);
+	if (!array)
+		goto nomem;
+
+	memset(array, 0, array_size);
+	ctrl = &swap_cgroup_ctrl[type];
+	mutex_lock(&swap_cgroup_mutex);
+	ctrl->length = length;
+	ctrl->map = array;
+	if (swap_cgroup_prepare(type)) {
+		/* memory shortage */
+		ctrl->map = NULL;
+		ctrl->length = 0;
+		vfree(array);
+		mutex_unlock(&swap_cgroup_mutex);
+		goto nomem;
+	}
+	mutex_unlock(&swap_cgroup_mutex);
+
+	printk(KERN_INFO
+		"swap_cgroup: uses %ld bytes of vmalloc for pointer array space"
+		" and %ld bytes to hold mem_cgroup pointers on swap\n",
+		array_size, length * PAGE_SIZE);
+	printk(KERN_INFO
+	"swap_cgroup can be disabled by noswapaccount boot option.\n");
+
+	return 0;
+nomem:
+	printk(KERN_INFO "couldn't allocate enough memory for swap_cgroup.\n");
+	printk(KERN_INFO
+		"swap_cgroup can be disabled by noswapaccount boot option\n");
+	return -ENOMEM;
+}
+
+void swap_cgroup_swapoff(int type)
+{
+	int i;
+	struct swap_cgroup_ctrl *ctrl;
+
+	if (!do_swap_account)
+		return;
+
+	mutex_lock(&swap_cgroup_mutex);
+	ctrl = &swap_cgroup_ctrl[type];
+	if (ctrl->map) {
+		for (i = 0; i < ctrl->length; i++) {
+			struct page *page = ctrl->map[i];
+			if (page)
+				__free_page(page);
+		}
+		vfree(ctrl->map);
+		ctrl->map = NULL;
+		ctrl->length = 0;
+	}
+	mutex_unlock(&swap_cgroup_mutex);
+}
+
+#endif
diff --git a/mm/swapfile.c b/mm/swapfile.c
index ddc6d92be2cb..1e7a715a3866 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -33,6 +33,7 @@
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
 #include <linux/swapops.h>
+#include <linux/page_cgroup.h>
 
 static DEFINE_SPINLOCK(swap_lock);
 static unsigned int nr_swapfiles;
@@ -1494,6 +1495,9 @@ asmlinkage long sys_swapoff(const char __user * specialfile)
 	spin_unlock(&swap_lock);
 	mutex_unlock(&swapon_mutex);
 	vfree(swap_map);
+	/* Destroy swap account informatin */
+	swap_cgroup_swapoff(type);
+
 	inode = mapping->host;
 	if (S_ISBLK(inode->i_mode)) {
 		struct block_device *bdev = I_BDEV(inode);
@@ -1811,6 +1815,11 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags)
 		}
 		swap_map[page_nr] = SWAP_MAP_BAD;
 	}
+
+	error = swap_cgroup_swapon(type, maxpages);
+	if (error)
+		goto bad_swap;
+
 	nr_good_pages = swap_header->info.last_page -
 			swap_header->info.nr_badpages -
 			1 /* header page */;
@@ -1882,6 +1891,7 @@ bad_swap:
 		bd_release(bdev);
 	}
 	destroy_swap_extents(p);
+	swap_cgroup_swapoff(type);
 bad_swap_2:
 	spin_lock(&swap_lock);
 	p->swap_file = NULL;
-- 
cgit v1.2.3


From 8c7c6e34a1256a5082d38c8e9bd1474476912715 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:00 -0800
Subject: memcg: mem+swap controller core

This patch implements per cgroup limit for usage of memory+swap.  However
there are SwapCache, double counting of swap-cache and swap-entry is
avoided.

Mem+Swap controller works as following.
  - memory usage is limited by memory.limit_in_bytes.
  - memory + swap usage is limited by memory.memsw_limit_in_bytes.

This has following benefits.
  - A user can limit total resource usage of mem+swap.

    Without this, because memory resource controller doesn't take care of
    usage of swap, a process can exhaust all the swap (by memory leak.)
    We can avoid this case.

    And Swap is shared resource but it cannot be reclaimed (goes back to memory)
    until it's used. This characteristic can be trouble when the memory
    is divided into some parts by cpuset or memcg.
    Assume group A and group B.
    After some application executes, the system can be..

    Group A -- very large free memory space but occupy 99% of swap.
    Group B -- under memory shortage but cannot use swap...it's nearly full.

    Ability to set appropriate swap limit for each group is required.

Maybe someone wonder "why not swap but mem+swap ?"

  - The global LRU(kswapd) can swap out arbitrary pages. Swap-out means
    to move account from memory to swap...there is no change in usage of
    mem+swap.

    In other words, when we want to limit the usage of swap without affecting
    global LRU, mem+swap limit is better than just limiting swap.

Accounting target information is stored in swap_cgroup which is
per swap entry record.

Charge is done as following.
  map
    - charge  page and memsw.

  unmap
    - uncharge page/memsw if not SwapCache.

  swap-out (__delete_from_swap_cache)
    - uncharge page
    - record mem_cgroup information to swap_cgroup.

  swap-in (do_swap_page)
    - charged as page and memsw.
      record in swap_cgroup is cleared.
      memsw accounting is decremented.

  swap-free (swap_free())
    - if swap entry is freed, memsw is uncharged by PAGE_SIZE.

There are people work under never-swap environments and consider swap as
something bad. For such people, this mem+swap controller extension is just an
overhead.  This overhead is avoided by config or boot option.
(see Kconfig. detail is not in this patch.)

TODO:
 - maybe more optimization can be don in swap-in path. (but not very safe.)
   But we just do simple accounting at this stage.

[nishimura@mxp.nes.nec.co.jp: make resize limit hold mutex]
[hugh@veritas.com: memswap controller core swapcache fixes]
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/controllers/memory.txt |  29 ++-
 include/linux/memcontrol.h           |  11 +-
 include/linux/swap.h                 |  14 +-
 mm/memcontrol.c                      | 400 +++++++++++++++++++++++++++++++----
 mm/memory.c                          |  18 +-
 mm/swap_state.c                      |   5 +-
 mm/swapfile.c                        |  11 +-
 mm/vmscan.c                          |   6 +-
 8 files changed, 440 insertions(+), 54 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/controllers/memory.txt b/Documentation/controllers/memory.txt
index 9fe2d0eabe05..05fe29ab1e58 100644
--- a/Documentation/controllers/memory.txt
+++ b/Documentation/controllers/memory.txt
@@ -137,12 +137,32 @@ behind this approach is that a cgroup that aggressively uses a shared
 page will eventually get charged for it (once it is uncharged from
 the cgroup that brought it in -- this will happen on memory pressure).
 
-Exception: When you do swapoff and make swapped-out pages of shmem(tmpfs) to
+Exception: If CONFIG_CGROUP_CGROUP_MEM_RES_CTLR_SWAP is not used..
+When you do swapoff and make swapped-out pages of shmem(tmpfs) to
 be backed into memory in force, charges for pages are accounted against the
 caller of swapoff rather than the users of shmem.
 
 
-2.4 Reclaim
+2.4 Swap Extension (CONFIG_CGROUP_MEM_RES_CTLR_SWAP)
+Swap Extension allows you to record charge for swap. A swapped-in page is
+charged back to original page allocator if possible.
+
+When swap is accounted, following files are added.
+ - memory.memsw.usage_in_bytes.
+ - memory.memsw.limit_in_bytes.
+
+usage of mem+swap is limited by memsw.limit_in_bytes.
+
+Note: why 'mem+swap' rather than swap.
+The global LRU(kswapd) can swap out arbitrary pages. Swap-out means
+to move account from memory to swap...there is no change in usage of
+mem+swap.
+
+In other words, when we want to limit the usage of swap without affecting
+global LRU, mem+swap limit is better than just limiting swap from OS point
+of view.
+
+2.5 Reclaim
 
 Each cgroup maintains a per cgroup LRU that consists of an active
 and inactive list. When a cgroup goes over its limit, we first try
@@ -246,6 +266,11 @@ Such charges are freed(at default) or moved to its parent. When moved,
 both of RSS and CACHES are moved to parent.
 If both of them are busy, rmdir() returns -EBUSY. See 5.1 Also.
 
+Charges recorded in swap information is not updated at removal of cgroup.
+Recorded information is discarded and a cgroup which uses swap (swapcache)
+will be charged as a new owner of it.
+
+
 5. Misc. interfaces.
 
 5.1 force_empty
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 41b46cc9d1f1..ca51ac72d6c0 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -32,6 +32,8 @@ extern int mem_cgroup_newpage_charge(struct page *page, struct mm_struct *mm,
 /* for swap handling */
 extern int mem_cgroup_try_charge(struct mm_struct *mm,
 		gfp_t gfp_mask, struct mem_cgroup **ptr);
+extern int mem_cgroup_try_charge_swapin(struct mm_struct *mm,
+		struct page *page, gfp_t mask, struct mem_cgroup **ptr);
 extern void mem_cgroup_commit_charge_swapin(struct page *page,
 					struct mem_cgroup *ptr);
 extern void mem_cgroup_cancel_charge_swapin(struct mem_cgroup *ptr);
@@ -80,7 +82,6 @@ extern long mem_cgroup_calc_reclaim(struct mem_cgroup *mem, struct zone *zone,
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
 extern int do_swap_account;
 #endif
-
 #else /* CONFIG_CGROUP_MEM_RES_CTLR */
 struct mem_cgroup;
 
@@ -97,7 +98,13 @@ static inline int mem_cgroup_cache_charge(struct page *page,
 }
 
 static inline int mem_cgroup_try_charge(struct mm_struct *mm,
-				gfp_t gfp_mask, struct mem_cgroup **ptr)
+			gfp_t gfp_mask, struct mem_cgroup **ptr)
+{
+	return 0;
+}
+
+static inline int mem_cgroup_try_charge_swapin(struct mm_struct *mm,
+		struct page *page, gfp_t gfp_mask, struct mem_cgroup **ptr)
 {
 	return 0;
 }
diff --git a/include/linux/swap.h b/include/linux/swap.h
index f8f3907533f0..be938ce4895a 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -214,7 +214,7 @@ static inline void lru_cache_add_active_file(struct page *page)
 extern unsigned long try_to_free_pages(struct zonelist *zonelist, int order,
 					gfp_t gfp_mask);
 extern unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem,
-							gfp_t gfp_mask);
+						gfp_t gfp_mask, bool noswap);
 extern int __isolate_lru_page(struct page *page, int mode, int file);
 extern unsigned long shrink_all_memory(unsigned long nr_pages);
 extern int vm_swappiness;
@@ -336,7 +336,7 @@ static inline void disable_swap_token(void)
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR
 extern int mem_cgroup_cache_charge_swapin(struct page *page,
 				struct mm_struct *mm, gfp_t mask, bool locked);
-extern void mem_cgroup_uncharge_swapcache(struct page *page);
+extern void mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent);
 #else
 static inline
 int mem_cgroup_cache_charge_swapin(struct page *page,
@@ -344,7 +344,15 @@ int mem_cgroup_cache_charge_swapin(struct page *page,
 {
 	return 0;
 }
-static inline void mem_cgroup_uncharge_swapcache(struct page *page)
+static inline void
+mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent)
+{
+}
+#endif
+#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
+extern void mem_cgroup_uncharge_swap(swp_entry_t ent);
+#else
+static inline void mem_cgroup_uncharge_swap(swp_entry_t ent)
 {
 }
 #endif
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 59dd8c116372..2efcf38f3b73 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -27,6 +27,7 @@
 #include <linux/backing-dev.h>
 #include <linux/bit_spinlock.h>
 #include <linux/rcupdate.h>
+#include <linux/mutex.h>
 #include <linux/slab.h>
 #include <linux/swap.h>
 #include <linux/spinlock.h>
@@ -131,6 +132,10 @@ struct mem_cgroup {
 	 * the counter to account for memory usage
 	 */
 	struct res_counter res;
+	/*
+	 * the counter to account for mem+swap usage.
+	 */
+	struct res_counter memsw;
 	/*
 	 * Per cgroup active and inactive list, similar to the
 	 * per zone LRU lists.
@@ -138,6 +143,8 @@ struct mem_cgroup {
 	struct mem_cgroup_lru_info info;
 
 	int	prev_priority;	/* for recording reclaim priority */
+	int		obsolete;
+	atomic_t	refcnt;
 	/*
 	 * statistics. This must be placed at the end of memcg.
 	 */
@@ -167,6 +174,17 @@ pcg_default_flags[NR_CHARGE_TYPE] = {
 	0, /* FORCE */
 };
 
+
+/* for encoding cft->private value on file */
+#define _MEM			(0)
+#define _MEMSWAP		(1)
+#define MEMFILE_PRIVATE(x, val)	(((x) << 16) | (val))
+#define MEMFILE_TYPE(val)	(((val) >> 16) & 0xffff)
+#define MEMFILE_ATTR(val)	((val) & 0xffff)
+
+static void mem_cgroup_get(struct mem_cgroup *mem);
+static void mem_cgroup_put(struct mem_cgroup *mem);
+
 /*
  * Always modified under lru lock. Then, not necessary to preempt_disable()
  */
@@ -485,7 +503,8 @@ unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
  * oom-killer can be invoked.
  */
 static int __mem_cgroup_try_charge(struct mm_struct *mm,
-			gfp_t gfp_mask, struct mem_cgroup **memcg, bool oom)
+			gfp_t gfp_mask, struct mem_cgroup **memcg,
+			bool oom)
 {
 	struct mem_cgroup *mem;
 	int nr_retries = MEM_CGROUP_RECLAIM_RETRIES;
@@ -513,12 +532,25 @@ static int __mem_cgroup_try_charge(struct mm_struct *mm,
 		css_get(&mem->css);
 	}
 
+	while (1) {
+		int ret;
+		bool noswap = false;
 
-	while (unlikely(res_counter_charge(&mem->res, PAGE_SIZE))) {
+		ret = res_counter_charge(&mem->res, PAGE_SIZE);
+		if (likely(!ret)) {
+			if (!do_swap_account)
+				break;
+			ret = res_counter_charge(&mem->memsw, PAGE_SIZE);
+			if (likely(!ret))
+				break;
+			/* mem+swap counter fails */
+			res_counter_uncharge(&mem->res, PAGE_SIZE);
+			noswap = true;
+		}
 		if (!(gfp_mask & __GFP_WAIT))
 			goto nomem;
 
-		if (try_to_free_mem_cgroup_pages(mem, gfp_mask))
+		if (try_to_free_mem_cgroup_pages(mem, gfp_mask, noswap))
 			continue;
 
 		/*
@@ -527,8 +559,13 @@ static int __mem_cgroup_try_charge(struct mm_struct *mm,
 		 * moved to swap cache or just unmapped from the cgroup.
 		 * Check the limit again to see if the reclaim reduced the
 		 * current usage of the cgroup before giving up
+		 *
 		 */
-		if (res_counter_check_under_limit(&mem->res))
+		if (!do_swap_account &&
+			res_counter_check_under_limit(&mem->res))
+			continue;
+		if (do_swap_account &&
+			res_counter_check_under_limit(&mem->memsw))
 			continue;
 
 		if (!nr_retries--) {
@@ -582,6 +619,8 @@ static void __mem_cgroup_commit_charge(struct mem_cgroup *mem,
 	if (unlikely(PageCgroupUsed(pc))) {
 		unlock_page_cgroup(pc);
 		res_counter_uncharge(&mem->res, PAGE_SIZE);
+		if (do_swap_account)
+			res_counter_uncharge(&mem->memsw, PAGE_SIZE);
 		css_put(&mem->css);
 		return;
 	}
@@ -646,6 +685,8 @@ static int mem_cgroup_move_account(struct page_cgroup *pc,
 		__mem_cgroup_remove_list(from_mz, pc);
 		css_put(&from->css);
 		res_counter_uncharge(&from->res, PAGE_SIZE);
+		if (do_swap_account)
+			res_counter_uncharge(&from->memsw, PAGE_SIZE);
 		pc->mem_cgroup = to;
 		css_get(&to->css);
 		__mem_cgroup_add_list(to_mz, pc, false);
@@ -692,8 +733,11 @@ static int mem_cgroup_move_parent(struct page_cgroup *pc,
 	/* drop extra refcnt */
 	css_put(&parent->css);
 	/* uncharge if move fails */
-	if (ret)
+	if (ret) {
 		res_counter_uncharge(&parent->res, PAGE_SIZE);
+		if (do_swap_account)
+			res_counter_uncharge(&parent->memsw, PAGE_SIZE);
+	}
 
 	return ret;
 }
@@ -791,7 +835,42 @@ int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 				MEM_CGROUP_CHARGE_TYPE_SHMEM, NULL);
 }
 
+int mem_cgroup_try_charge_swapin(struct mm_struct *mm,
+				 struct page *page,
+				 gfp_t mask, struct mem_cgroup **ptr)
+{
+	struct mem_cgroup *mem;
+	swp_entry_t     ent;
+
+	if (mem_cgroup_subsys.disabled)
+		return 0;
+
+	if (!do_swap_account)
+		goto charge_cur_mm;
+
+	/*
+	 * A racing thread's fault, or swapoff, may have already updated
+	 * the pte, and even removed page from swap cache: return success
+	 * to go on to do_swap_page()'s pte_same() test, which should fail.
+	 */
+	if (!PageSwapCache(page))
+		return 0;
+
+	ent.val = page_private(page);
+
+	mem = lookup_swap_cgroup(ent);
+	if (!mem || mem->obsolete)
+		goto charge_cur_mm;
+	*ptr = mem;
+	return __mem_cgroup_try_charge(NULL, mask, ptr, true);
+charge_cur_mm:
+	if (unlikely(!mm))
+		mm = &init_mm;
+	return __mem_cgroup_try_charge(mm, mask, ptr, true);
+}
+
 #ifdef CONFIG_SWAP
+
 int mem_cgroup_cache_charge_swapin(struct page *page,
 			struct mm_struct *mm, gfp_t mask, bool locked)
 {
@@ -808,8 +887,28 @@ int mem_cgroup_cache_charge_swapin(struct page *page,
 	 * we reach here.
 	 */
 	if (PageSwapCache(page)) {
+		struct mem_cgroup *mem = NULL;
+		swp_entry_t ent;
+
+		ent.val = page_private(page);
+		if (do_swap_account) {
+			mem = lookup_swap_cgroup(ent);
+			if (mem && mem->obsolete)
+				mem = NULL;
+			if (mem)
+				mm = NULL;
+		}
 		ret = mem_cgroup_charge_common(page, mm, mask,
-				MEM_CGROUP_CHARGE_TYPE_SHMEM, NULL);
+				MEM_CGROUP_CHARGE_TYPE_SHMEM, mem);
+
+		if (!ret && do_swap_account) {
+			/* avoid double counting */
+			mem = swap_cgroup_record(ent, NULL);
+			if (mem) {
+				res_counter_uncharge(&mem->memsw, PAGE_SIZE);
+				mem_cgroup_put(mem);
+			}
+		}
 	}
 	if (!locked)
 		unlock_page(page);
@@ -828,6 +927,23 @@ void mem_cgroup_commit_charge_swapin(struct page *page, struct mem_cgroup *ptr)
 		return;
 	pc = lookup_page_cgroup(page);
 	__mem_cgroup_commit_charge(ptr, pc, MEM_CGROUP_CHARGE_TYPE_MAPPED);
+	/*
+	 * Now swap is on-memory. This means this page may be
+	 * counted both as mem and swap....double count.
+	 * Fix it by uncharging from memsw. This SwapCache is stable
+	 * because we're still under lock_page().
+	 */
+	if (do_swap_account) {
+		swp_entry_t ent = {.val = page_private(page)};
+		struct mem_cgroup *memcg;
+		memcg = swap_cgroup_record(ent, NULL);
+		if (memcg) {
+			/* If memcg is obsolete, memcg can be != ptr */
+			res_counter_uncharge(&memcg->memsw, PAGE_SIZE);
+			mem_cgroup_put(memcg);
+		}
+
+	}
 }
 
 void mem_cgroup_cancel_charge_swapin(struct mem_cgroup *mem)
@@ -837,6 +953,8 @@ void mem_cgroup_cancel_charge_swapin(struct mem_cgroup *mem)
 	if (!mem)
 		return;
 	res_counter_uncharge(&mem->res, PAGE_SIZE);
+	if (do_swap_account)
+		res_counter_uncharge(&mem->memsw, PAGE_SIZE);
 	css_put(&mem->css);
 }
 
@@ -844,29 +962,31 @@ void mem_cgroup_cancel_charge_swapin(struct mem_cgroup *mem)
 /*
  * uncharge if !page_mapped(page)
  */
-static void
+static struct mem_cgroup *
 __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype)
 {
 	struct page_cgroup *pc;
-	struct mem_cgroup *mem;
+	struct mem_cgroup *mem = NULL;
 	struct mem_cgroup_per_zone *mz;
 	unsigned long flags;
 
 	if (mem_cgroup_subsys.disabled)
-		return;
+		return NULL;
 
 	if (PageSwapCache(page))
-		return;
+		return NULL;
 
 	/*
 	 * Check if our page_cgroup is valid
 	 */
 	pc = lookup_page_cgroup(page);
 	if (unlikely(!pc || !PageCgroupUsed(pc)))
-		return;
+		return NULL;
 
 	lock_page_cgroup(pc);
 
+	mem = pc->mem_cgroup;
+
 	if (!PageCgroupUsed(pc))
 		goto unlock_out;
 
@@ -886,8 +1006,11 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype)
 		break;
 	}
 
+	res_counter_uncharge(&mem->res, PAGE_SIZE);
+	if (do_swap_account && (ctype != MEM_CGROUP_CHARGE_TYPE_SWAPOUT))
+		res_counter_uncharge(&mem->memsw, PAGE_SIZE);
+
 	ClearPageCgroupUsed(pc);
-	mem = pc->mem_cgroup;
 
 	mz = page_cgroup_zoneinfo(pc);
 	spin_lock_irqsave(&mz->lru_lock, flags);
@@ -895,14 +1018,13 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype)
 	spin_unlock_irqrestore(&mz->lru_lock, flags);
 	unlock_page_cgroup(pc);
 
-	res_counter_uncharge(&mem->res, PAGE_SIZE);
 	css_put(&mem->css);
 
-	return;
+	return mem;
 
 unlock_out:
 	unlock_page_cgroup(pc);
-	return;
+	return NULL;
 }
 
 void mem_cgroup_uncharge_page(struct page *page)
@@ -922,10 +1044,42 @@ void mem_cgroup_uncharge_cache_page(struct page *page)
 	__mem_cgroup_uncharge_common(page, MEM_CGROUP_CHARGE_TYPE_CACHE);
 }
 
-void mem_cgroup_uncharge_swapcache(struct page *page)
+/*
+ * called from __delete_from_swap_cache() and drop "page" account.
+ * memcg information is recorded to swap_cgroup of "ent"
+ */
+void mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent)
+{
+	struct mem_cgroup *memcg;
+
+	memcg = __mem_cgroup_uncharge_common(page,
+					MEM_CGROUP_CHARGE_TYPE_SWAPOUT);
+	/* record memcg information */
+	if (do_swap_account && memcg) {
+		swap_cgroup_record(ent, memcg);
+		mem_cgroup_get(memcg);
+	}
+}
+
+#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
+/*
+ * called from swap_entry_free(). remove record in swap_cgroup and
+ * uncharge "memsw" account.
+ */
+void mem_cgroup_uncharge_swap(swp_entry_t ent)
 {
-	__mem_cgroup_uncharge_common(page, MEM_CGROUP_CHARGE_TYPE_SWAPOUT);
+	struct mem_cgroup *memcg;
+
+	if (!do_swap_account)
+		return;
+
+	memcg = swap_cgroup_record(ent, NULL);
+	if (memcg) {
+		res_counter_uncharge(&memcg->memsw, PAGE_SIZE);
+		mem_cgroup_put(memcg);
+	}
 }
+#endif
 
 /*
  * Before starting migration, account PAGE_SIZE to mem_cgroup that the old
@@ -1034,7 +1188,7 @@ int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask)
 	rcu_read_unlock();
 
 	do {
-		progress = try_to_free_mem_cgroup_pages(mem, gfp_mask);
+		progress = try_to_free_mem_cgroup_pages(mem, gfp_mask, true);
 		progress += res_counter_check_under_limit(&mem->res);
 	} while (!progress && --retry);
 
@@ -1044,26 +1198,84 @@ int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask)
 	return 0;
 }
 
+static DEFINE_MUTEX(set_limit_mutex);
+
 static int mem_cgroup_resize_limit(struct mem_cgroup *memcg,
-				   unsigned long long val)
+				unsigned long long val)
 {
 
 	int retry_count = MEM_CGROUP_RECLAIM_RETRIES;
 	int progress;
+	u64 memswlimit;
 	int ret = 0;
 
-	while (res_counter_set_limit(&memcg->res, val)) {
+	while (retry_count) {
 		if (signal_pending(current)) {
 			ret = -EINTR;
 			break;
 		}
-		if (!retry_count) {
-			ret = -EBUSY;
+		/*
+		 * Rather than hide all in some function, I do this in
+		 * open coded manner. You see what this really does.
+		 * We have to guarantee mem->res.limit < mem->memsw.limit.
+		 */
+		mutex_lock(&set_limit_mutex);
+		memswlimit = res_counter_read_u64(&memcg->memsw, RES_LIMIT);
+		if (memswlimit < val) {
+			ret = -EINVAL;
+			mutex_unlock(&set_limit_mutex);
 			break;
 		}
+		ret = res_counter_set_limit(&memcg->res, val);
+		mutex_unlock(&set_limit_mutex);
+
+		if (!ret)
+			break;
+
 		progress = try_to_free_mem_cgroup_pages(memcg,
-				GFP_HIGHUSER_MOVABLE);
-		if (!progress)
+				GFP_HIGHUSER_MOVABLE, false);
+  		if (!progress)			retry_count--;
+	}
+	return ret;
+}
+
+int mem_cgroup_resize_memsw_limit(struct mem_cgroup *memcg,
+				unsigned long long val)
+{
+	int retry_count = MEM_CGROUP_RECLAIM_RETRIES;
+	u64 memlimit, oldusage, curusage;
+	int ret;
+
+	if (!do_swap_account)
+		return -EINVAL;
+
+	while (retry_count) {
+		if (signal_pending(current)) {
+			ret = -EINTR;
+			break;
+		}
+		/*
+		 * Rather than hide all in some function, I do this in
+		 * open coded manner. You see what this really does.
+		 * We have to guarantee mem->res.limit < mem->memsw.limit.
+		 */
+		mutex_lock(&set_limit_mutex);
+		memlimit = res_counter_read_u64(&memcg->res, RES_LIMIT);
+		if (memlimit > val) {
+			ret = -EINVAL;
+			mutex_unlock(&set_limit_mutex);
+			break;
+		}
+		ret = res_counter_set_limit(&memcg->memsw, val);
+		mutex_unlock(&set_limit_mutex);
+
+		if (!ret)
+			break;
+
+		oldusage = res_counter_read_u64(&memcg->memsw, RES_USAGE);
+		try_to_free_mem_cgroup_pages(memcg, GFP_HIGHUSER_MOVABLE, true);
+		curusage = res_counter_read_u64(&memcg->memsw, RES_USAGE);
+		if (curusage >= oldusage)
 			retry_count--;
 	}
 	return ret;
@@ -1193,7 +1405,7 @@ try_to_free:
 			goto out;
 		}
 		progress = try_to_free_mem_cgroup_pages(mem,
-						  GFP_HIGHUSER_MOVABLE);
+						  GFP_HIGHUSER_MOVABLE, false);
 		if (!progress) {
 			nr_retries--;
 			/* maybe some writeback is necessary */
@@ -1216,8 +1428,25 @@ int mem_cgroup_force_empty_write(struct cgroup *cont, unsigned int event)
 
 static u64 mem_cgroup_read(struct cgroup *cont, struct cftype *cft)
 {
-	return res_counter_read_u64(&mem_cgroup_from_cont(cont)->res,
-				    cft->private);
+	struct mem_cgroup *mem = mem_cgroup_from_cont(cont);
+	u64 val = 0;
+	int type, name;
+
+	type = MEMFILE_TYPE(cft->private);
+	name = MEMFILE_ATTR(cft->private);
+	switch (type) {
+	case _MEM:
+		val = res_counter_read_u64(&mem->res, name);
+		break;
+	case _MEMSWAP:
+		if (do_swap_account)
+			val = res_counter_read_u64(&mem->memsw, name);
+		break;
+	default:
+		BUG();
+		break;
+	}
+	return val;
 }
 /*
  * The user of this function is...
@@ -1227,15 +1456,22 @@ static int mem_cgroup_write(struct cgroup *cont, struct cftype *cft,
 			    const char *buffer)
 {
 	struct mem_cgroup *memcg = mem_cgroup_from_cont(cont);
+	int type, name;
 	unsigned long long val;
 	int ret;
 
-	switch (cft->private) {
+	type = MEMFILE_TYPE(cft->private);
+	name = MEMFILE_ATTR(cft->private);
+	switch (name) {
 	case RES_LIMIT:
 		/* This function does all necessary parse...reuse it */
 		ret = res_counter_memparse_write_strategy(buffer, &val);
-		if (!ret)
+		if (ret)
+			break;
+		if (type == _MEM)
 			ret = mem_cgroup_resize_limit(memcg, val);
+		else
+			ret = mem_cgroup_resize_memsw_limit(memcg, val);
 		break;
 	default:
 		ret = -EINVAL; /* should be BUG() ? */
@@ -1247,14 +1483,23 @@ static int mem_cgroup_write(struct cgroup *cont, struct cftype *cft,
 static int mem_cgroup_reset(struct cgroup *cont, unsigned int event)
 {
 	struct mem_cgroup *mem;
+	int type, name;
 
 	mem = mem_cgroup_from_cont(cont);
-	switch (event) {
+	type = MEMFILE_TYPE(event);
+	name = MEMFILE_ATTR(event);
+	switch (name) {
 	case RES_MAX_USAGE:
-		res_counter_reset_max(&mem->res);
+		if (type == _MEM)
+			res_counter_reset_max(&mem->res);
+		else
+			res_counter_reset_max(&mem->memsw);
 		break;
 	case RES_FAILCNT:
-		res_counter_reset_failcnt(&mem->res);
+		if (type == _MEM)
+			res_counter_reset_failcnt(&mem->res);
+		else
+			res_counter_reset_failcnt(&mem->memsw);
 		break;
 	}
 	return 0;
@@ -1315,24 +1560,24 @@ static int mem_control_stat_show(struct cgroup *cont, struct cftype *cft,
 static struct cftype mem_cgroup_files[] = {
 	{
 		.name = "usage_in_bytes",
-		.private = RES_USAGE,
+		.private = MEMFILE_PRIVATE(_MEM, RES_USAGE),
 		.read_u64 = mem_cgroup_read,
 	},
 	{
 		.name = "max_usage_in_bytes",
-		.private = RES_MAX_USAGE,
+		.private = MEMFILE_PRIVATE(_MEM, RES_MAX_USAGE),
 		.trigger = mem_cgroup_reset,
 		.read_u64 = mem_cgroup_read,
 	},
 	{
 		.name = "limit_in_bytes",
-		.private = RES_LIMIT,
+		.private = MEMFILE_PRIVATE(_MEM, RES_LIMIT),
 		.write_string = mem_cgroup_write,
 		.read_u64 = mem_cgroup_read,
 	},
 	{
 		.name = "failcnt",
-		.private = RES_FAILCNT,
+		.private = MEMFILE_PRIVATE(_MEM, RES_FAILCNT),
 		.trigger = mem_cgroup_reset,
 		.read_u64 = mem_cgroup_read,
 	},
@@ -1346,6 +1591,47 @@ static struct cftype mem_cgroup_files[] = {
 	},
 };
 
+#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
+static struct cftype memsw_cgroup_files[] = {
+	{
+		.name = "memsw.usage_in_bytes",
+		.private = MEMFILE_PRIVATE(_MEMSWAP, RES_USAGE),
+		.read_u64 = mem_cgroup_read,
+	},
+	{
+		.name = "memsw.max_usage_in_bytes",
+		.private = MEMFILE_PRIVATE(_MEMSWAP, RES_MAX_USAGE),
+		.trigger = mem_cgroup_reset,
+		.read_u64 = mem_cgroup_read,
+	},
+	{
+		.name = "memsw.limit_in_bytes",
+		.private = MEMFILE_PRIVATE(_MEMSWAP, RES_LIMIT),
+		.write_string = mem_cgroup_write,
+		.read_u64 = mem_cgroup_read,
+	},
+	{
+		.name = "memsw.failcnt",
+		.private = MEMFILE_PRIVATE(_MEMSWAP, RES_FAILCNT),
+		.trigger = mem_cgroup_reset,
+		.read_u64 = mem_cgroup_read,
+	},
+};
+
+static int register_memsw_files(struct cgroup *cont, struct cgroup_subsys *ss)
+{
+	if (!do_swap_account)
+		return 0;
+	return cgroup_add_files(cont, ss, memsw_cgroup_files,
+				ARRAY_SIZE(memsw_cgroup_files));
+};
+#else
+static int register_memsw_files(struct cgroup *cont, struct cgroup_subsys *ss)
+{
+	return 0;
+}
+#endif
+
 static int alloc_mem_cgroup_per_zone_info(struct mem_cgroup *mem, int node)
 {
 	struct mem_cgroup_per_node *pn;
@@ -1404,14 +1690,44 @@ static struct mem_cgroup *mem_cgroup_alloc(void)
 	return mem;
 }
 
+/*
+ * At destroying mem_cgroup, references from swap_cgroup can remain.
+ * (scanning all at force_empty is too costly...)
+ *
+ * Instead of clearing all references at force_empty, we remember
+ * the number of reference from swap_cgroup and free mem_cgroup when
+ * it goes down to 0.
+ *
+ * When mem_cgroup is destroyed, mem->obsolete will be set to 0 and
+ * entry which points to this memcg will be ignore at swapin.
+ *
+ * Removal of cgroup itself succeeds regardless of refs from swap.
+ */
+
 static void mem_cgroup_free(struct mem_cgroup *mem)
 {
+	if (atomic_read(&mem->refcnt) > 0)
+		return;
 	if (mem_cgroup_size() < PAGE_SIZE)
 		kfree(mem);
 	else
 		vfree(mem);
 }
 
+static void mem_cgroup_get(struct mem_cgroup *mem)
+{
+	atomic_inc(&mem->refcnt);
+}
+
+static void mem_cgroup_put(struct mem_cgroup *mem)
+{
+	if (atomic_dec_and_test(&mem->refcnt)) {
+		if (!mem->obsolete)
+			return;
+		mem_cgroup_free(mem);
+	}
+}
+
 
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
 static void __init enable_swap_cgroup(void)
@@ -1436,6 +1752,7 @@ mem_cgroup_create(struct cgroup_subsys *ss, struct cgroup *cont)
 		return ERR_PTR(-ENOMEM);
 
 	res_counter_init(&mem->res);
+	res_counter_init(&mem->memsw);
 
 	for_each_node_state(node, N_POSSIBLE)
 		if (alloc_mem_cgroup_per_zone_info(mem, node))
@@ -1456,6 +1773,7 @@ static void mem_cgroup_pre_destroy(struct cgroup_subsys *ss,
 					struct cgroup *cont)
 {
 	struct mem_cgroup *mem = mem_cgroup_from_cont(cont);
+	mem->obsolete = 1;
 	mem_cgroup_force_empty(mem, false);
 }
 
@@ -1474,8 +1792,14 @@ static void mem_cgroup_destroy(struct cgroup_subsys *ss,
 static int mem_cgroup_populate(struct cgroup_subsys *ss,
 				struct cgroup *cont)
 {
-	return cgroup_add_files(cont, ss, mem_cgroup_files,
-					ARRAY_SIZE(mem_cgroup_files));
+	int ret;
+
+	ret = cgroup_add_files(cont, ss, mem_cgroup_files,
+				ARRAY_SIZE(mem_cgroup_files));
+
+	if (!ret)
+		ret = register_memsw_files(cont, ss);
+	return ret;
 }
 
 static void mem_cgroup_move_task(struct cgroup_subsys *ss,
diff --git a/mm/memory.c b/mm/memory.c
index ba5189e322e6..1358012ffa73 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2431,7 +2431,8 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	lock_page(page);
 	delayacct_clear_flag(DELAYACCT_PF_SWAPIN);
 
-	if (mem_cgroup_try_charge(mm, GFP_HIGHUSER_MOVABLE, &ptr) == -ENOMEM) {
+	if (mem_cgroup_try_charge_swapin(mm, page,
+				GFP_HIGHUSER_MOVABLE, &ptr) == -ENOMEM) {
 		ret = VM_FAULT_OOM;
 		unlock_page(page);
 		goto out;
@@ -2449,8 +2450,20 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		goto out_nomap;
 	}
 
-	/* The page isn't present yet, go ahead with the fault. */
+	/*
+	 * The page isn't present yet, go ahead with the fault.
+	 *
+	 * Be careful about the sequence of operations here.
+	 * To get its accounting right, reuse_swap_page() must be called
+	 * while the page is counted on swap but not yet in mapcount i.e.
+	 * before page_add_anon_rmap() and swap_free(); try_to_free_swap()
+	 * must be called after the swap_free(), or it will never succeed.
+	 * And mem_cgroup_commit_charge_swapin(), which uses the swp_entry
+	 * in page->private, must be called before reuse_swap_page(),
+	 * which may delete_from_swap_cache().
+	 */
 
+	mem_cgroup_commit_charge_swapin(page, ptr);
 	inc_mm_counter(mm, anon_rss);
 	pte = mk_pte(page, vma->vm_page_prot);
 	if (write_access && reuse_swap_page(page)) {
@@ -2461,7 +2474,6 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	flush_icache_page(vma, page);
 	set_pte_at(mm, address, page_table, pte);
 	page_add_anon_rmap(page, vma, address);
-	mem_cgroup_commit_charge_swapin(page, ptr);
 
 	swap_free(entry);
 	if (vm_swap_full() || (vma->vm_flags & VM_LOCKED) || PageMlocked(page))
diff --git a/mm/swap_state.c b/mm/swap_state.c
index 09291ca11f5f..3ecea98ecb45 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -17,6 +17,7 @@
 #include <linux/backing-dev.h>
 #include <linux/pagevec.h>
 #include <linux/migrate.h>
+#include <linux/page_cgroup.h>
 
 #include <asm/pgtable.h>
 
@@ -108,6 +109,8 @@ int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask)
  */
 void __delete_from_swap_cache(struct page *page)
 {
+	swp_entry_t ent = {.val = page_private(page)};
+
 	VM_BUG_ON(!PageLocked(page));
 	VM_BUG_ON(!PageSwapCache(page));
 	VM_BUG_ON(PageWriteback(page));
@@ -118,7 +121,7 @@ void __delete_from_swap_cache(struct page *page)
 	total_swapcache_pages--;
 	__dec_zone_page_state(page, NR_FILE_PAGES);
 	INC_CACHE_INFO(del_total);
-	mem_cgroup_uncharge_swapcache(page);
+	mem_cgroup_uncharge_swapcache(page, ent);
 }
 
 /**
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 1e7a715a3866..0579d9069b61 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -471,8 +471,9 @@ out:
 	return NULL;
 }
 
-static int swap_entry_free(struct swap_info_struct *p, unsigned long offset)
+static int swap_entry_free(struct swap_info_struct *p, swp_entry_t ent)
 {
+	unsigned long offset = swp_offset(ent);
 	int count = p->swap_map[offset];
 
 	if (count < SWAP_MAP_MAX) {
@@ -487,6 +488,7 @@ static int swap_entry_free(struct swap_info_struct *p, unsigned long offset)
 				swap_list.next = p - swap_info;
 			nr_swap_pages++;
 			p->inuse_pages--;
+			mem_cgroup_uncharge_swap(ent);
 		}
 	}
 	return count;
@@ -502,7 +504,7 @@ void swap_free(swp_entry_t entry)
 
 	p = swap_info_get(entry);
 	if (p) {
-		swap_entry_free(p, swp_offset(entry));
+		swap_entry_free(p, entry);
 		spin_unlock(&swap_lock);
 	}
 }
@@ -582,7 +584,7 @@ int free_swap_and_cache(swp_entry_t entry)
 
 	p = swap_info_get(entry);
 	if (p) {
-		if (swap_entry_free(p, swp_offset(entry)) == 1) {
+		if (swap_entry_free(p, entry) == 1) {
 			page = find_get_page(&swapper_space, entry.val);
 			if (page && !trylock_page(page)) {
 				page_cache_release(page);
@@ -696,7 +698,8 @@ static int unuse_pte(struct vm_area_struct *vma, pmd_t *pmd,
 	pte_t *pte;
 	int ret = 1;
 
-	if (mem_cgroup_try_charge(vma->vm_mm, GFP_HIGHUSER_MOVABLE, &ptr))
+	if (mem_cgroup_try_charge_swapin(vma->vm_mm, page,
+					GFP_HIGHUSER_MOVABLE, &ptr))
 		ret = -ENOMEM;
 
 	pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
diff --git a/mm/vmscan.c b/mm/vmscan.c
index b07c48b09a93..f63b20dd7714 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1661,7 +1661,8 @@ unsigned long try_to_free_pages(struct zonelist *zonelist, int order,
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR
 
 unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem_cont,
-						gfp_t gfp_mask)
+						gfp_t gfp_mask,
+					   bool noswap)
 {
 	struct scan_control sc = {
 		.may_writepage = !laptop_mode,
@@ -1674,6 +1675,9 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem_cont,
 	};
 	struct zonelist *zonelist;
 
+	if (noswap)
+		sc.may_swap = 0;
+
 	sc.gfp_mask = (gfp_mask & GFP_RECLAIM_MASK) |
 			(GFP_HIGHUSER_MOVABLE & ~GFP_RECLAIM_MASK);
 	zonelist = NODE_DATA(numa_node_id())->node_zonelists;
-- 
cgit v1.2.3


From 08e552c69c6930d64722de3ec18c51844d06ee28 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:01 -0800
Subject: memcg: synchronized LRU

A big patch for changing memcg's LRU semantics.

Now,
  - page_cgroup is linked to mem_cgroup's its own LRU (per zone).

  - LRU of page_cgroup is not synchronous with global LRU.

  - page and page_cgroup is one-to-one and statically allocated.

  - To find page_cgroup is on what LRU, you have to check pc->mem_cgroup as
    - lru = page_cgroup_zoneinfo(pc, nid_of_pc, zid_of_pc);

  - SwapCache is handled.

And, when we handle LRU list of page_cgroup, we do following.

	pc = lookup_page_cgroup(page);
	lock_page_cgroup(pc); .....................(1)
	mz = page_cgroup_zoneinfo(pc);
	spin_lock(&mz->lru_lock);
	.....add to LRU
	spin_unlock(&mz->lru_lock);
	unlock_page_cgroup(pc);

But (1) is spin_lock and we have to be afraid of dead-lock with zone->lru_lock.
So, trylock() is used at (1), now. Without (1), we can't trust "mz" is correct.

This is a trial to remove this dirty nesting of locks.
This patch changes mz->lru_lock to be zone->lru_lock.
Then, above sequence will be written as

        spin_lock(&zone->lru_lock); # in vmscan.c or swap.c via global LRU
	mem_cgroup_add/remove/etc_lru() {
		pc = lookup_page_cgroup(page);
		mz = page_cgroup_zoneinfo(pc);
		if (PageCgroupUsed(pc)) {
			....add to LRU
		}
        spin_lock(&zone->lru_lock); # in vmscan.c or swap.c via global LRU

This is much simpler.
(*) We're safe even if we don't take lock_page_cgroup(pc). Because..
    1. When pc->mem_cgroup can be modified.
       - at charge.
       - at account_move().
    2. at charge
       the PCG_USED bit is not set before pc->mem_cgroup is fixed.
    3. at account_move()
       the page is isolated and not on LRU.

Pros.
  - easy for maintenance.
  - memcg can make use of laziness of pagevec.
  - we don't have to duplicated LRU/Active/Unevictable bit in page_cgroup.
  - LRU status of memcg will be synchronized with global LRU's one.
  - # of locks are reduced.
  - account_move() is simplified very much.
Cons.
  - may increase cost of LRU rotation.
    (no impact if memcg is not configured.)

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/splice.c                 |   1 +
 include/linux/memcontrol.h  |  29 +++-
 include/linux/mm_inline.h   |   3 +
 include/linux/page_cgroup.h |  17 ---
 mm/memcontrol.c             | 323 +++++++++++++++++++-------------------------
 mm/page_cgroup.c            |   1 +
 mm/swap.c                   |   1 -
 mm/vmscan.c                 |   9 +-
 8 files changed, 178 insertions(+), 206 deletions(-)

(limited to 'include/linux')

diff --git a/fs/splice.c b/fs/splice.c
index 1abab5cee4ba..a54b3e3f10a7 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -21,6 +21,7 @@
 #include <linux/file.h>
 #include <linux/pagemap.h>
 #include <linux/splice.h>
+#include <linux/memcontrol.h>
 #include <linux/mm_inline.h>
 #include <linux/swap.h>
 #include <linux/writeback.h>
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index ca51ac72d6c0..32c07b1852d6 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -40,7 +40,12 @@ extern void mem_cgroup_cancel_charge_swapin(struct mem_cgroup *ptr);
 
 extern int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 					gfp_t gfp_mask);
-extern void mem_cgroup_move_lists(struct page *page, enum lru_list lru);
+extern void mem_cgroup_add_lru_list(struct page *page, enum lru_list lru);
+extern void mem_cgroup_del_lru_list(struct page *page, enum lru_list lru);
+extern void mem_cgroup_rotate_lru_list(struct page *page, enum lru_list lru);
+extern void mem_cgroup_del_lru(struct page *page);
+extern void mem_cgroup_move_lists(struct page *page,
+				  enum lru_list from, enum lru_list to);
 extern void mem_cgroup_uncharge_page(struct page *page);
 extern void mem_cgroup_uncharge_cache_page(struct page *page);
 extern int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask);
@@ -131,7 +136,27 @@ static inline int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask)
 	return 0;
 }
 
-static inline void mem_cgroup_move_lists(struct page *page, bool active)
+static inline void mem_cgroup_add_lru_list(struct page *page, int lru)
+{
+}
+
+static inline void mem_cgroup_del_lru_list(struct page *page, int lru)
+{
+	return ;
+}
+
+static inline void mem_cgroup_rotate_lru_list(struct page *page, int lru)
+{
+	return ;
+}
+
+static inline void mem_cgroup_del_lru(struct page *page)
+{
+	return ;
+}
+
+static inline void
+mem_cgroup_move_lists(struct page *page, enum lru_list from, enum lru_list to)
 {
 }
 
diff --git a/include/linux/mm_inline.h b/include/linux/mm_inline.h
index c948350c378e..37ef13d0f01e 100644
--- a/include/linux/mm_inline.h
+++ b/include/linux/mm_inline.h
@@ -28,6 +28,7 @@ add_page_to_lru_list(struct zone *zone, struct page *page, enum lru_list l)
 {
 	list_add(&page->lru, &zone->lru[l].list);
 	__inc_zone_state(zone, NR_LRU_BASE + l);
+	mem_cgroup_add_lru_list(page, l);
 }
 
 static inline void
@@ -35,6 +36,7 @@ del_page_from_lru_list(struct zone *zone, struct page *page, enum lru_list l)
 {
 	list_del(&page->lru);
 	__dec_zone_state(zone, NR_LRU_BASE + l);
+	mem_cgroup_del_lru_list(page, l);
 }
 
 static inline void
@@ -54,6 +56,7 @@ del_page_from_lru(struct zone *zone, struct page *page)
 		l += page_is_file_cache(page);
 	}
 	__dec_zone_state(zone, NR_LRU_BASE + l);
+	mem_cgroup_del_lru_list(page, l);
 }
 
 /**
diff --git a/include/linux/page_cgroup.h b/include/linux/page_cgroup.h
index d754b2dfbf2d..602cc1fdee90 100644
--- a/include/linux/page_cgroup.h
+++ b/include/linux/page_cgroup.h
@@ -26,10 +26,6 @@ enum {
 	PCG_LOCK,  /* page cgroup is locked */
 	PCG_CACHE, /* charged as cache */
 	PCG_USED, /* this object is in use. */
-	/* flags for LRU placement */
-	PCG_ACTIVE, /* page is active in this cgroup */
-	PCG_FILE, /* page is file system backed */
-	PCG_UNEVICTABLE, /* page is unevictableable */
 };
 
 #define TESTPCGFLAG(uname, lname)			\
@@ -50,19 +46,6 @@ TESTPCGFLAG(Cache, CACHE)
 TESTPCGFLAG(Used, USED)
 CLEARPCGFLAG(Used, USED)
 
-/* LRU management flags (from global-lru definition) */
-TESTPCGFLAG(File, FILE)
-SETPCGFLAG(File, FILE)
-CLEARPCGFLAG(File, FILE)
-
-TESTPCGFLAG(Active, ACTIVE)
-SETPCGFLAG(Active, ACTIVE)
-CLEARPCGFLAG(Active, ACTIVE)
-
-TESTPCGFLAG(Unevictable, UNEVICTABLE)
-SETPCGFLAG(Unevictable, UNEVICTABLE)
-CLEARPCGFLAG(Unevictable, UNEVICTABLE)
-
 static inline int page_cgroup_nid(struct page_cgroup *pc)
 {
 	return page_to_nid(pc->page);
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 2efcf38f3b73..8ce4e9e47959 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -36,6 +36,7 @@
 #include <linux/vmalloc.h>
 #include <linux/mm_inline.h>
 #include <linux/page_cgroup.h>
+#include "internal.h"
 
 #include <asm/uaccess.h>
 
@@ -100,7 +101,6 @@ struct mem_cgroup_per_zone {
 	/*
 	 * spin_lock to protect the per cgroup LRU
 	 */
-	spinlock_t		lru_lock;
 	struct list_head	lists[NR_LRU_LISTS];
 	unsigned long		count[NR_LRU_LISTS];
 };
@@ -163,14 +163,12 @@ enum charge_type {
 /* only for here (for easy reading.) */
 #define PCGF_CACHE	(1UL << PCG_CACHE)
 #define PCGF_USED	(1UL << PCG_USED)
-#define PCGF_ACTIVE	(1UL << PCG_ACTIVE)
 #define PCGF_LOCK	(1UL << PCG_LOCK)
-#define PCGF_FILE	(1UL << PCG_FILE)
 static const unsigned long
 pcg_default_flags[NR_CHARGE_TYPE] = {
-	PCGF_CACHE | PCGF_FILE | PCGF_USED | PCGF_LOCK, /* File Cache */
-	PCGF_ACTIVE | PCGF_USED | PCGF_LOCK, /* Anon */
-	PCGF_ACTIVE | PCGF_CACHE | PCGF_USED | PCGF_LOCK, /* Shmem */
+	PCGF_CACHE | PCGF_USED | PCGF_LOCK, /* File Cache */
+	PCGF_USED | PCGF_LOCK, /* Anon */
+	PCGF_CACHE | PCGF_USED | PCGF_LOCK, /* Shmem */
 	0, /* FORCE */
 };
 
@@ -185,9 +183,6 @@ pcg_default_flags[NR_CHARGE_TYPE] = {
 static void mem_cgroup_get(struct mem_cgroup *mem);
 static void mem_cgroup_put(struct mem_cgroup *mem);
 
-/*
- * Always modified under lru lock. Then, not necessary to preempt_disable()
- */
 static void mem_cgroup_charge_statistics(struct mem_cgroup *mem,
 					 struct page_cgroup *pc,
 					 bool charge)
@@ -195,10 +190,9 @@ static void mem_cgroup_charge_statistics(struct mem_cgroup *mem,
 	int val = (charge)? 1 : -1;
 	struct mem_cgroup_stat *stat = &mem->stat;
 	struct mem_cgroup_stat_cpu *cpustat;
+	int cpu = get_cpu();
 
-	VM_BUG_ON(!irqs_disabled());
-
-	cpustat = &stat->cpustat[smp_processor_id()];
+	cpustat = &stat->cpustat[cpu];
 	if (PageCgroupCache(pc))
 		__mem_cgroup_stat_add_safe(cpustat, MEM_CGROUP_STAT_CACHE, val);
 	else
@@ -210,6 +204,7 @@ static void mem_cgroup_charge_statistics(struct mem_cgroup *mem,
 	else
 		__mem_cgroup_stat_add_safe(cpustat,
 				MEM_CGROUP_STAT_PGPGOUT_COUNT, 1);
+	put_cpu();
 }
 
 static struct mem_cgroup_per_zone *
@@ -264,80 +259,95 @@ struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p)
 				struct mem_cgroup, css);
 }
 
-static void __mem_cgroup_remove_list(struct mem_cgroup_per_zone *mz,
-			struct page_cgroup *pc)
-{
-	int lru = LRU_BASE;
+/*
+ * Following LRU functions are allowed to be used without PCG_LOCK.
+ * Operations are called by routine of global LRU independently from memcg.
+ * What we have to take care of here is validness of pc->mem_cgroup.
+ *
+ * Changes to pc->mem_cgroup happens when
+ * 1. charge
+ * 2. moving account
+ * In typical case, "charge" is done before add-to-lru. Exception is SwapCache.
+ * It is added to LRU before charge.
+ * If PCG_USED bit is not set, page_cgroup is not added to this private LRU.
+ * When moving account, the page is not on LRU. It's isolated.
+ */
 
-	if (PageCgroupUnevictable(pc))
-		lru = LRU_UNEVICTABLE;
-	else {
-		if (PageCgroupActive(pc))
-			lru += LRU_ACTIVE;
-		if (PageCgroupFile(pc))
-			lru += LRU_FILE;
-	}
+void mem_cgroup_del_lru_list(struct page *page, enum lru_list lru)
+{
+	struct page_cgroup *pc;
+	struct mem_cgroup *mem;
+	struct mem_cgroup_per_zone *mz;
 
+	if (mem_cgroup_subsys.disabled)
+		return;
+	pc = lookup_page_cgroup(page);
+	/* can happen while we handle swapcache. */
+	if (list_empty(&pc->lru))
+		return;
+	mz = page_cgroup_zoneinfo(pc);
+	mem = pc->mem_cgroup;
 	MEM_CGROUP_ZSTAT(mz, lru) -= 1;
-
-	mem_cgroup_charge_statistics(pc->mem_cgroup, pc, false);
-	list_del(&pc->lru);
+	list_del_init(&pc->lru);
+	return;
 }
 
-static void __mem_cgroup_add_list(struct mem_cgroup_per_zone *mz,
-				struct page_cgroup *pc, bool hot)
+void mem_cgroup_del_lru(struct page *page)
 {
-	int lru = LRU_BASE;
+	mem_cgroup_del_lru_list(page, page_lru(page));
+}
 
-	if (PageCgroupUnevictable(pc))
-		lru = LRU_UNEVICTABLE;
-	else {
-		if (PageCgroupActive(pc))
-			lru += LRU_ACTIVE;
-		if (PageCgroupFile(pc))
-			lru += LRU_FILE;
-	}
+void mem_cgroup_rotate_lru_list(struct page *page, enum lru_list lru)
+{
+	struct mem_cgroup_per_zone *mz;
+	struct page_cgroup *pc;
 
-	MEM_CGROUP_ZSTAT(mz, lru) += 1;
-	if (hot)
-		list_add(&pc->lru, &mz->lists[lru]);
-	else
-		list_add_tail(&pc->lru, &mz->lists[lru]);
+	if (mem_cgroup_subsys.disabled)
+		return;
 
-	mem_cgroup_charge_statistics(pc->mem_cgroup, pc, true);
+	pc = lookup_page_cgroup(page);
+	smp_rmb();
+	/* unused page is not rotated. */
+	if (!PageCgroupUsed(pc))
+		return;
+	mz = page_cgroup_zoneinfo(pc);
+	list_move(&pc->lru, &mz->lists[lru]);
 }
 
-static void __mem_cgroup_move_lists(struct page_cgroup *pc, enum lru_list lru)
+void mem_cgroup_add_lru_list(struct page *page, enum lru_list lru)
 {
-	struct mem_cgroup_per_zone *mz = page_cgroup_zoneinfo(pc);
-	int active    = PageCgroupActive(pc);
-	int file      = PageCgroupFile(pc);
-	int unevictable = PageCgroupUnevictable(pc);
-	enum lru_list from = unevictable ? LRU_UNEVICTABLE :
-				(LRU_FILE * !!file + !!active);
+	struct page_cgroup *pc;
+	struct mem_cgroup_per_zone *mz;
 
-	if (lru == from)
+	if (mem_cgroup_subsys.disabled)
+		return;
+	pc = lookup_page_cgroup(page);
+	/* barrier to sync with "charge" */
+	smp_rmb();
+	if (!PageCgroupUsed(pc))
 		return;
 
-	MEM_CGROUP_ZSTAT(mz, from) -= 1;
-	/*
-	 * However this is done under mz->lru_lock, another flags, which
-	 * are not related to LRU, will be modified from out-of-lock.
-	 * We have to use atomic set/clear flags.
-	 */
-	if (is_unevictable_lru(lru)) {
-		ClearPageCgroupActive(pc);
-		SetPageCgroupUnevictable(pc);
-	} else {
-		if (is_active_lru(lru))
-			SetPageCgroupActive(pc);
-		else
-			ClearPageCgroupActive(pc);
-		ClearPageCgroupUnevictable(pc);
-	}
-
+	mz = page_cgroup_zoneinfo(pc);
 	MEM_CGROUP_ZSTAT(mz, lru) += 1;
-	list_move(&pc->lru, &mz->lists[lru]);
+	list_add(&pc->lru, &mz->lists[lru]);
+}
+/*
+ * To add swapcache into LRU. Be careful to all this function.
+ * zone->lru_lock shouldn't be held and irq must not be disabled.
+ */
+static void mem_cgroup_lru_fixup(struct page *page)
+{
+	if (!isolate_lru_page(page))
+		putback_lru_page(page);
+}
+
+void mem_cgroup_move_lists(struct page *page,
+			   enum lru_list from, enum lru_list to)
+{
+	if (mem_cgroup_subsys.disabled)
+		return;
+	mem_cgroup_del_lru_list(page, from);
+	mem_cgroup_add_lru_list(page, to);
 }
 
 int task_in_mem_cgroup(struct task_struct *task, const struct mem_cgroup *mem)
@@ -350,37 +360,6 @@ int task_in_mem_cgroup(struct task_struct *task, const struct mem_cgroup *mem)
 	return ret;
 }
 
-/*
- * This routine assumes that the appropriate zone's lru lock is already held
- */
-void mem_cgroup_move_lists(struct page *page, enum lru_list lru)
-{
-	struct page_cgroup *pc;
-	struct mem_cgroup_per_zone *mz;
-	unsigned long flags;
-
-	if (mem_cgroup_subsys.disabled)
-		return;
-
-	/*
-	 * We cannot lock_page_cgroup while holding zone's lru_lock,
-	 * because other holders of lock_page_cgroup can be interrupted
-	 * with an attempt to rotate_reclaimable_page.  But we cannot
-	 * safely get to page_cgroup without it, so just try_lock it:
-	 * mem_cgroup_isolate_pages allows for page left on wrong list.
-	 */
-	pc = lookup_page_cgroup(page);
-	if (!trylock_page_cgroup(pc))
-		return;
-	if (pc && PageCgroupUsed(pc)) {
-		mz = page_cgroup_zoneinfo(pc);
-		spin_lock_irqsave(&mz->lru_lock, flags);
-		__mem_cgroup_move_lists(pc, lru);
-		spin_unlock_irqrestore(&mz->lru_lock, flags);
-	}
-	unlock_page_cgroup(pc);
-}
-
 /*
  * Calculate mapped_ratio under memory controller. This will be used in
  * vmscan.c for deteremining we have to reclaim mapped pages.
@@ -460,40 +439,24 @@ unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
 	mz = mem_cgroup_zoneinfo(mem_cont, nid, zid);
 	src = &mz->lists[lru];
 
-	spin_lock(&mz->lru_lock);
 	scan = 0;
 	list_for_each_entry_safe_reverse(pc, tmp, src, lru) {
 		if (scan >= nr_to_scan)
 			break;
+
+		page = pc->page;
 		if (unlikely(!PageCgroupUsed(pc)))
 			continue;
-		page = pc->page;
-
 		if (unlikely(!PageLRU(page)))
 			continue;
 
-		/*
-		 * TODO: play better with lumpy reclaim, grabbing anything.
-		 */
-		if (PageUnevictable(page) ||
-		    (PageActive(page) && !active) ||
-		    (!PageActive(page) && active)) {
-			__mem_cgroup_move_lists(pc, page_lru(page));
-			continue;
-		}
-
 		scan++;
-		list_move(&pc->lru, &pc_list);
-
 		if (__isolate_lru_page(page, mode, file) == 0) {
 			list_move(&page->lru, dst);
 			nr_taken++;
 		}
 	}
 
-	list_splice(&pc_list, src);
-	spin_unlock(&mz->lru_lock);
-
 	*scanned = scan;
 	return nr_taken;
 }
@@ -608,9 +571,6 @@ static void __mem_cgroup_commit_charge(struct mem_cgroup *mem,
 				     struct page_cgroup *pc,
 				     enum charge_type ctype)
 {
-	struct mem_cgroup_per_zone *mz;
-	unsigned long flags;
-
 	/* try_charge() can return NULL to *memcg, taking care of it. */
 	if (!mem)
 		return;
@@ -625,17 +585,11 @@ static void __mem_cgroup_commit_charge(struct mem_cgroup *mem,
 		return;
 	}
 	pc->mem_cgroup = mem;
-	/*
-	 * If a page is accounted as a page cache, insert to inactive list.
-	 * If anon, insert to active list.
-	 */
+	smp_wmb();
 	pc->flags = pcg_default_flags[ctype];
 
-	mz = page_cgroup_zoneinfo(pc);
+	mem_cgroup_charge_statistics(mem, pc, true);
 
-	spin_lock_irqsave(&mz->lru_lock, flags);
-	__mem_cgroup_add_list(mz, pc, true);
-	spin_unlock_irqrestore(&mz->lru_lock, flags);
 	unlock_page_cgroup(pc);
 }
 
@@ -646,8 +600,7 @@ static void __mem_cgroup_commit_charge(struct mem_cgroup *mem,
  * @to:	mem_cgroup which the page is moved to. @from != @to.
  *
  * The caller must confirm following.
- * 1. disable irq.
- * 2. lru_lock of old mem_cgroup(@from) should be held.
+ * - page is not on LRU (isolate_page() is useful.)
  *
  * returns 0 at success,
  * returns -EBUSY when lock is busy or "pc" is unstable.
@@ -663,15 +616,14 @@ static int mem_cgroup_move_account(struct page_cgroup *pc,
 	int nid, zid;
 	int ret = -EBUSY;
 
-	VM_BUG_ON(!irqs_disabled());
 	VM_BUG_ON(from == to);
+	VM_BUG_ON(PageLRU(pc->page));
 
 	nid = page_cgroup_nid(pc);
 	zid = page_cgroup_zid(pc);
 	from_mz =  mem_cgroup_zoneinfo(from, nid, zid);
 	to_mz =  mem_cgroup_zoneinfo(to, nid, zid);
 
-
 	if (!trylock_page_cgroup(pc))
 		return ret;
 
@@ -681,18 +633,15 @@ static int mem_cgroup_move_account(struct page_cgroup *pc,
 	if (pc->mem_cgroup != from)
 		goto out;
 
-	if (spin_trylock(&to_mz->lru_lock)) {
-		__mem_cgroup_remove_list(from_mz, pc);
-		css_put(&from->css);
-		res_counter_uncharge(&from->res, PAGE_SIZE);
-		if (do_swap_account)
-			res_counter_uncharge(&from->memsw, PAGE_SIZE);
-		pc->mem_cgroup = to;
-		css_get(&to->css);
-		__mem_cgroup_add_list(to_mz, pc, false);
-		ret = 0;
-		spin_unlock(&to_mz->lru_lock);
-	}
+	css_put(&from->css);
+	res_counter_uncharge(&from->res, PAGE_SIZE);
+	mem_cgroup_charge_statistics(from, pc, false);
+	if (do_swap_account)
+		res_counter_uncharge(&from->memsw, PAGE_SIZE);
+	pc->mem_cgroup = to;
+	mem_cgroup_charge_statistics(to, pc, true);
+	css_get(&to->css);
+	ret = 0;
 out:
 	unlock_page_cgroup(pc);
 	return ret;
@@ -706,39 +655,47 @@ static int mem_cgroup_move_parent(struct page_cgroup *pc,
 				  struct mem_cgroup *child,
 				  gfp_t gfp_mask)
 {
+	struct page *page = pc->page;
 	struct cgroup *cg = child->css.cgroup;
 	struct cgroup *pcg = cg->parent;
 	struct mem_cgroup *parent;
-	struct mem_cgroup_per_zone *mz;
-	unsigned long flags;
 	int ret;
 
 	/* Is ROOT ? */
 	if (!pcg)
 		return -EINVAL;
 
+
 	parent = mem_cgroup_from_cont(pcg);
 
+
 	ret = __mem_cgroup_try_charge(NULL, gfp_mask, &parent, false);
 	if (ret)
 		return ret;
 
-	mz = mem_cgroup_zoneinfo(child,
-			page_cgroup_nid(pc), page_cgroup_zid(pc));
+	if (!get_page_unless_zero(page))
+		return -EBUSY;
+
+	ret = isolate_lru_page(page);
+
+	if (ret)
+		goto cancel;
 
-	spin_lock_irqsave(&mz->lru_lock, flags);
 	ret = mem_cgroup_move_account(pc, child, parent);
-	spin_unlock_irqrestore(&mz->lru_lock, flags);
 
-	/* drop extra refcnt */
+	/* drop extra refcnt by try_charge() (move_account increment one) */
 	css_put(&parent->css);
-	/* uncharge if move fails */
-	if (ret) {
-		res_counter_uncharge(&parent->res, PAGE_SIZE);
-		if (do_swap_account)
-			res_counter_uncharge(&parent->memsw, PAGE_SIZE);
+	putback_lru_page(page);
+	if (!ret) {
+		put_page(page);
+		return 0;
 	}
-
+	/* uncharge if move fails */
+cancel:
+	res_counter_uncharge(&parent->res, PAGE_SIZE);
+	if (do_swap_account)
+		res_counter_uncharge(&parent->memsw, PAGE_SIZE);
+	put_page(page);
 	return ret;
 }
 
@@ -912,6 +869,8 @@ int mem_cgroup_cache_charge_swapin(struct page *page,
 	}
 	if (!locked)
 		unlock_page(page);
+	/* add this page(page_cgroup) to the LRU we want. */
+	mem_cgroup_lru_fixup(page);
 
 	return ret;
 }
@@ -944,6 +903,8 @@ void mem_cgroup_commit_charge_swapin(struct page *page, struct mem_cgroup *ptr)
 		}
 
 	}
+	/* add this page(page_cgroup) to the LRU we want. */
+	mem_cgroup_lru_fixup(page);
 }
 
 void mem_cgroup_cancel_charge_swapin(struct mem_cgroup *mem)
@@ -968,7 +929,6 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype)
 	struct page_cgroup *pc;
 	struct mem_cgroup *mem = NULL;
 	struct mem_cgroup_per_zone *mz;
-	unsigned long flags;
 
 	if (mem_cgroup_subsys.disabled)
 		return NULL;
@@ -1010,12 +970,10 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype)
 	if (do_swap_account && (ctype != MEM_CGROUP_CHARGE_TYPE_SWAPOUT))
 		res_counter_uncharge(&mem->memsw, PAGE_SIZE);
 
+	mem_cgroup_charge_statistics(mem, pc, false);
 	ClearPageCgroupUsed(pc);
 
 	mz = page_cgroup_zoneinfo(pc);
-	spin_lock_irqsave(&mz->lru_lock, flags);
-	__mem_cgroup_remove_list(mz, pc);
-	spin_unlock_irqrestore(&mz->lru_lock, flags);
 	unlock_page_cgroup(pc);
 
 	css_put(&mem->css);
@@ -1281,21 +1239,22 @@ int mem_cgroup_resize_memsw_limit(struct mem_cgroup *memcg,
 	return ret;
 }
 
-
 /*
  * This routine traverse page_cgroup in given list and drop them all.
  * *And* this routine doesn't reclaim page itself, just removes page_cgroup.
  */
 static int mem_cgroup_force_empty_list(struct mem_cgroup *mem,
-			    struct mem_cgroup_per_zone *mz,
-			    enum lru_list lru)
+				int node, int zid, enum lru_list lru)
 {
+	struct zone *zone;
+	struct mem_cgroup_per_zone *mz;
 	struct page_cgroup *pc, *busy;
-	unsigned long flags;
-	unsigned long loop;
+	unsigned long flags, loop;
 	struct list_head *list;
 	int ret = 0;
 
+	zone = &NODE_DATA(node)->node_zones[zid];
+	mz = mem_cgroup_zoneinfo(mem, node, zid);
 	list = &mz->lists[lru];
 
 	loop = MEM_CGROUP_ZSTAT(mz, lru);
@@ -1304,19 +1263,19 @@ static int mem_cgroup_force_empty_list(struct mem_cgroup *mem,
 	busy = NULL;
 	while (loop--) {
 		ret = 0;
-		spin_lock_irqsave(&mz->lru_lock, flags);
+		spin_lock_irqsave(&zone->lru_lock, flags);
 		if (list_empty(list)) {
-			spin_unlock_irqrestore(&mz->lru_lock, flags);
+			spin_unlock_irqrestore(&zone->lru_lock, flags);
 			break;
 		}
 		pc = list_entry(list->prev, struct page_cgroup, lru);
 		if (busy == pc) {
 			list_move(&pc->lru, list);
 			busy = 0;
-			spin_unlock_irqrestore(&mz->lru_lock, flags);
+			spin_unlock_irqrestore(&zone->lru_lock, flags);
 			continue;
 		}
-		spin_unlock_irqrestore(&mz->lru_lock, flags);
+		spin_unlock_irqrestore(&zone->lru_lock, flags);
 
 		ret = mem_cgroup_move_parent(pc, mem, GFP_HIGHUSER_MOVABLE);
 		if (ret == -ENOMEM)
@@ -1329,6 +1288,7 @@ static int mem_cgroup_force_empty_list(struct mem_cgroup *mem,
 		} else
 			busy = NULL;
 	}
+
 	if (!ret && !list_empty(list))
 		return -EBUSY;
 	return ret;
@@ -1364,12 +1324,10 @@ move_account:
 		ret = 0;
 		for_each_node_state(node, N_POSSIBLE) {
 			for (zid = 0; !ret && zid < MAX_NR_ZONES; zid++) {
-				struct mem_cgroup_per_zone *mz;
 				enum lru_list l;
-				mz = mem_cgroup_zoneinfo(mem, node, zid);
 				for_each_lru(l) {
 					ret = mem_cgroup_force_empty_list(mem,
-								  mz, l);
+							node, zid, l);
 					if (ret)
 						break;
 				}
@@ -1413,6 +1371,7 @@ try_to_free:
 		}
 
 	}
+	lru_add_drain();
 	/* try move_account...there may be some *locked* pages. */
 	if (mem->res.usage)
 		goto move_account;
@@ -1657,7 +1616,6 @@ static int alloc_mem_cgroup_per_zone_info(struct mem_cgroup *mem, int node)
 
 	for (zone = 0; zone < MAX_NR_ZONES; zone++) {
 		mz = &pn->zoneinfo[zone];
-		spin_lock_init(&mz->lru_lock);
 		for_each_lru(l)
 			INIT_LIST_HEAD(&mz->lists[l]);
 	}
@@ -1706,8 +1664,15 @@ static struct mem_cgroup *mem_cgroup_alloc(void)
 
 static void mem_cgroup_free(struct mem_cgroup *mem)
 {
+	int node;
+
 	if (atomic_read(&mem->refcnt) > 0)
 		return;
+
+
+	for_each_node_state(node, N_POSSIBLE)
+		free_mem_cgroup_per_zone_info(mem, node);
+
 	if (mem_cgroup_size() < PAGE_SIZE)
 		kfree(mem);
 	else
@@ -1780,12 +1745,6 @@ static void mem_cgroup_pre_destroy(struct cgroup_subsys *ss,
 static void mem_cgroup_destroy(struct cgroup_subsys *ss,
 				struct cgroup *cont)
 {
-	int node;
-	struct mem_cgroup *mem = mem_cgroup_from_cont(cont);
-
-	for_each_node_state(node, N_POSSIBLE)
-		free_mem_cgroup_per_zone_info(mem, node);
-
 	mem_cgroup_free(mem_cgroup_from_cont(cont));
 }
 
diff --git a/mm/page_cgroup.c b/mm/page_cgroup.c
index 685e7c8e1fd6..74ae8e01d071 100644
--- a/mm/page_cgroup.c
+++ b/mm/page_cgroup.c
@@ -16,6 +16,7 @@ __init_page_cgroup(struct page_cgroup *pc, unsigned long pfn)
 	pc->flags = 0;
 	pc->mem_cgroup = NULL;
 	pc->page = pfn_to_page(pfn);
+	INIT_LIST_HEAD(&pc->lru);
 }
 static unsigned long total_usage;
 
diff --git a/mm/swap.c b/mm/swap.c
index ba2c0e8b8b54..8a98a9c90704 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -168,7 +168,6 @@ void activate_page(struct page *page)
 		lru += LRU_ACTIVE;
 		add_page_to_lru_list(zone, page, lru);
 		__count_vm_event(PGACTIVATE);
-		mem_cgroup_move_lists(page, lru);
 
 		zone->recent_rotated[!!file]++;
 		zone->recent_scanned[!!file]++;
diff --git a/mm/vmscan.c b/mm/vmscan.c
index f63b20dd7714..45983af1de3d 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -512,7 +512,6 @@ redo:
 		lru = LRU_UNEVICTABLE;
 		add_page_to_unevictable_list(page);
 	}
-	mem_cgroup_move_lists(page, lru);
 
 	/*
 	 * page's status can change while we move it among lru. If an evictable
@@ -547,7 +546,6 @@ void putback_lru_page(struct page *page)
 
 	lru = !!TestClearPageActive(page) + page_is_file_cache(page);
 	lru_cache_add_lru(page, lru);
-	mem_cgroup_move_lists(page, lru);
 	put_page(page);
 }
 #endif /* CONFIG_UNEVICTABLE_LRU */
@@ -813,6 +811,7 @@ int __isolate_lru_page(struct page *page, int mode, int file)
 		return ret;
 
 	ret = -EBUSY;
+
 	if (likely(get_page_unless_zero(page))) {
 		/*
 		 * Be careful not to clear PageLRU until after we're
@@ -821,6 +820,7 @@ int __isolate_lru_page(struct page *page, int mode, int file)
 		 */
 		ClearPageLRU(page);
 		ret = 0;
+		mem_cgroup_del_lru(page);
 	}
 
 	return ret;
@@ -1134,7 +1134,6 @@ static unsigned long shrink_inactive_list(unsigned long max_scan,
 			SetPageLRU(page);
 			lru = page_lru(page);
 			add_page_to_lru_list(zone, page, lru);
-			mem_cgroup_move_lists(page, lru);
 			if (PageActive(page) && scan_global_lru(sc)) {
 				int file = !!page_is_file_cache(page);
 				zone->recent_rotated[file]++;
@@ -1263,7 +1262,7 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 		ClearPageActive(page);
 
 		list_move(&page->lru, &zone->lru[lru].list);
-		mem_cgroup_move_lists(page, lru);
+		mem_cgroup_add_lru_list(page, lru);
 		pgmoved++;
 		if (!pagevec_add(&pvec, page)) {
 			__mod_zone_page_state(zone, NR_LRU_BASE + lru, pgmoved);
@@ -2408,6 +2407,7 @@ retry:
 
 		__dec_zone_state(zone, NR_UNEVICTABLE);
 		list_move(&page->lru, &zone->lru[l].list);
+		mem_cgroup_move_lists(page, LRU_UNEVICTABLE, l);
 		__inc_zone_state(zone, NR_INACTIVE_ANON + l);
 		__count_vm_event(UNEVICTABLE_PGRESCUED);
 	} else {
@@ -2416,6 +2416,7 @@ retry:
 		 */
 		SetPageUnevictable(page);
 		list_move(&page->lru, &zone->lru[LRU_UNEVICTABLE].list);
+		mem_cgroup_rotate_lru_list(page, LRU_UNEVICTABLE);
 		if (page_evictable(page, NULL))
 			goto retry;
 	}
-- 
cgit v1.2.3


From f8d665422603ee1b8ed04dcad4242f14d623c941 Mon Sep 17 00:00:00 2001
From: Hirokazu Takahashi <taka@valinux.co.jp>
Date: Wed, 7 Jan 2009 18:08:02 -0800
Subject: memcg: add mem_cgroup_disabled()

We check mem_cgroup is disabled or not by checking
mem_cgroup_subsys.disabled.  I think it has more references than expected,
now.

replacing
   if (mem_cgroup_subsys.disabled)
with
   if (mem_cgroup_disabled())

give us good look, I think.

[kamezawa.hiroyu@jp.fujitsu.com: fix typo]
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h | 15 ++++++++++++++-
 mm/memcontrol.c            | 28 ++++++++++++++--------------
 mm/page_cgroup.c           |  4 ++--
 3 files changed, 30 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 32c07b1852d6..472efd09118c 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -19,7 +19,7 @@
 
 #ifndef _LINUX_MEMCONTROL_H
 #define _LINUX_MEMCONTROL_H
-
+#include <linux/cgroup.h>
 struct mem_cgroup;
 struct page_cgroup;
 struct page;
@@ -87,6 +87,14 @@ extern long mem_cgroup_calc_reclaim(struct mem_cgroup *mem, struct zone *zone,
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
 extern int do_swap_account;
 #endif
+
+static inline bool mem_cgroup_disabled(void)
+{
+	if (mem_cgroup_subsys.disabled)
+		return true;
+	return false;
+}
+
 #else /* CONFIG_CGROUP_MEM_RES_CTLR */
 struct mem_cgroup;
 
@@ -214,6 +222,11 @@ static inline long mem_cgroup_calc_reclaim(struct mem_cgroup *mem,
 {
 	return 0;
 }
+
+static inline bool mem_cgroup_disabled(void)
+{
+	return true;
+}
 #endif /* CONFIG_CGROUP_MEM_CONT */
 
 #endif /* _LINUX_MEMCONTROL_H */
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 8ce4e9e47959..9846f617115d 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -279,7 +279,7 @@ void mem_cgroup_del_lru_list(struct page *page, enum lru_list lru)
 	struct mem_cgroup *mem;
 	struct mem_cgroup_per_zone *mz;
 
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return;
 	pc = lookup_page_cgroup(page);
 	/* can happen while we handle swapcache. */
@@ -302,7 +302,7 @@ void mem_cgroup_rotate_lru_list(struct page *page, enum lru_list lru)
 	struct mem_cgroup_per_zone *mz;
 	struct page_cgroup *pc;
 
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return;
 
 	pc = lookup_page_cgroup(page);
@@ -319,7 +319,7 @@ void mem_cgroup_add_lru_list(struct page *page, enum lru_list lru)
 	struct page_cgroup *pc;
 	struct mem_cgroup_per_zone *mz;
 
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return;
 	pc = lookup_page_cgroup(page);
 	/* barrier to sync with "charge" */
@@ -344,7 +344,7 @@ static void mem_cgroup_lru_fixup(struct page *page)
 void mem_cgroup_move_lists(struct page *page,
 			   enum lru_list from, enum lru_list to)
 {
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return;
 	mem_cgroup_del_lru_list(page, from);
 	mem_cgroup_add_lru_list(page, to);
@@ -731,7 +731,7 @@ static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm,
 int mem_cgroup_newpage_charge(struct page *page,
 			      struct mm_struct *mm, gfp_t gfp_mask)
 {
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return 0;
 	if (PageCompound(page))
 		return 0;
@@ -753,7 +753,7 @@ int mem_cgroup_newpage_charge(struct page *page,
 int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 				gfp_t gfp_mask)
 {
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return 0;
 	if (PageCompound(page))
 		return 0;
@@ -799,7 +799,7 @@ int mem_cgroup_try_charge_swapin(struct mm_struct *mm,
 	struct mem_cgroup *mem;
 	swp_entry_t     ent;
 
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return 0;
 
 	if (!do_swap_account)
@@ -833,7 +833,7 @@ int mem_cgroup_cache_charge_swapin(struct page *page,
 {
 	int ret = 0;
 
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return 0;
 	if (unlikely(!mm))
 		mm = &init_mm;
@@ -880,7 +880,7 @@ void mem_cgroup_commit_charge_swapin(struct page *page, struct mem_cgroup *ptr)
 {
 	struct page_cgroup *pc;
 
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return;
 	if (!ptr)
 		return;
@@ -909,7 +909,7 @@ void mem_cgroup_commit_charge_swapin(struct page *page, struct mem_cgroup *ptr)
 
 void mem_cgroup_cancel_charge_swapin(struct mem_cgroup *mem)
 {
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return;
 	if (!mem)
 		return;
@@ -930,7 +930,7 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype)
 	struct mem_cgroup *mem = NULL;
 	struct mem_cgroup_per_zone *mz;
 
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return NULL;
 
 	if (PageSwapCache(page))
@@ -1049,7 +1049,7 @@ int mem_cgroup_prepare_migration(struct page *page, struct mem_cgroup **ptr)
 	struct mem_cgroup *mem = NULL;
 	int ret = 0;
 
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return 0;
 
 	pc = lookup_page_cgroup(page);
@@ -1131,7 +1131,7 @@ int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask)
 	int progress = 0;
 	int retry = MEM_CGROUP_RECLAIM_RETRIES;
 
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return 0;
 	if (!mm)
 		return 0;
@@ -1697,7 +1697,7 @@ static void mem_cgroup_put(struct mem_cgroup *mem)
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
 static void __init enable_swap_cgroup(void)
 {
-	if (!mem_cgroup_subsys.disabled && really_do_swap_account)
+	if (!mem_cgroup_disabled() && really_do_swap_account)
 		do_swap_account = 1;
 }
 #else
diff --git a/mm/page_cgroup.c b/mm/page_cgroup.c
index 74ae8e01d071..7006a11350c8 100644
--- a/mm/page_cgroup.c
+++ b/mm/page_cgroup.c
@@ -74,7 +74,7 @@ void __init page_cgroup_init(void)
 
 	int nid, fail;
 
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return;
 
 	for_each_online_node(nid)  {
@@ -247,7 +247,7 @@ void __init page_cgroup_init(void)
 	unsigned long pfn;
 	int fail = 0;
 
-	if (mem_cgroup_subsys.disabled)
+	if (mem_cgroup_disabled())
 		return;
 
 	for (pfn = 0; !fail && pfn < max_pfn; pfn += PAGES_PER_SECTION) {
-- 
cgit v1.2.3


From 28dbc4b6a01fb579a9441c7b81e3d3413dc452df Mon Sep 17 00:00:00 2001
From: Balbir Singh <balbir@linux.vnet.ibm.com>
Date: Wed, 7 Jan 2009 18:08:05 -0800
Subject: memcg: memory cgroup resource counters for hierarchy

Add support for building hierarchies in resource counters.  Cgroups allows
us to build a deep hierarchy, but we currently don't link the resource
counters belonging to the memory controller control groups, in the same
fashion as the corresponding cgroup entries in the cgroup hierarchy.  This
patch provides the infrastructure for resource counters that have the same
hiearchy as their cgroup counter parts.

These set of patches are based on the resource counter hiearchy patches
posted by Pavel Emelianov.

NOTE: Building hiearchies is expensive, deeper hierarchies imply charging
the all the way up to the root.  It is known that hiearchies are
expensive, so the user needs to be careful and aware of the trade-offs
before creating very deep ones.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Cc: Paul Menage <menage@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Pavel Emelianov <xemul@openvz.org>
Cc: Dhaval Giani <dhaval@linux.vnet.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/res_counter.h |  8 ++++++--
 kernel/res_counter.c        | 44 +++++++++++++++++++++++++++++++++++---------
 mm/memcontrol.c             | 20 +++++++++++++-------
 3 files changed, 54 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/res_counter.h b/include/linux/res_counter.h
index 271c1c2c9f6f..dede0a2cfc45 100644
--- a/include/linux/res_counter.h
+++ b/include/linux/res_counter.h
@@ -43,6 +43,10 @@ struct res_counter {
 	 * the routines below consider this to be IRQ-safe
 	 */
 	spinlock_t lock;
+	/*
+	 * Parent counter, used for hierarchial resource accounting
+	 */
+	struct res_counter *parent;
 };
 
 /**
@@ -87,7 +91,7 @@ enum {
  * helpers for accounting
  */
 
-void res_counter_init(struct res_counter *counter);
+void res_counter_init(struct res_counter *counter, struct res_counter *parent);
 
 /*
  * charge - try to consume more resource.
@@ -103,7 +107,7 @@ void res_counter_init(struct res_counter *counter);
 int __must_check res_counter_charge_locked(struct res_counter *counter,
 		unsigned long val);
 int __must_check res_counter_charge(struct res_counter *counter,
-		unsigned long val);
+		unsigned long val, struct res_counter **limit_fail_at);
 
 /*
  * uncharge - tell that some portion of the resource is released
diff --git a/kernel/res_counter.c b/kernel/res_counter.c
index f275c8eca772..bf8e7534c803 100644
--- a/kernel/res_counter.c
+++ b/kernel/res_counter.c
@@ -15,10 +15,11 @@
 #include <linux/uaccess.h>
 #include <linux/mm.h>
 
-void res_counter_init(struct res_counter *counter)
+void res_counter_init(struct res_counter *counter, struct res_counter *parent)
 {
 	spin_lock_init(&counter->lock);
 	counter->limit = (unsigned long long)LLONG_MAX;
+	counter->parent = parent;
 }
 
 int res_counter_charge_locked(struct res_counter *counter, unsigned long val)
@@ -34,14 +35,34 @@ int res_counter_charge_locked(struct res_counter *counter, unsigned long val)
 	return 0;
 }
 
-int res_counter_charge(struct res_counter *counter, unsigned long val)
+int res_counter_charge(struct res_counter *counter, unsigned long val,
+			struct res_counter **limit_fail_at)
 {
 	int ret;
 	unsigned long flags;
-
-	spin_lock_irqsave(&counter->lock, flags);
-	ret = res_counter_charge_locked(counter, val);
-	spin_unlock_irqrestore(&counter->lock, flags);
+	struct res_counter *c, *u;
+
+	*limit_fail_at = NULL;
+	local_irq_save(flags);
+	for (c = counter; c != NULL; c = c->parent) {
+		spin_lock(&c->lock);
+		ret = res_counter_charge_locked(c, val);
+		spin_unlock(&c->lock);
+		if (ret < 0) {
+			*limit_fail_at = c;
+			goto undo;
+		}
+	}
+	ret = 0;
+	goto done;
+undo:
+	for (u = counter; u != c; u = u->parent) {
+		spin_lock(&u->lock);
+		res_counter_uncharge_locked(u, val);
+		spin_unlock(&u->lock);
+	}
+done:
+	local_irq_restore(flags);
 	return ret;
 }
 
@@ -56,10 +77,15 @@ void res_counter_uncharge_locked(struct res_counter *counter, unsigned long val)
 void res_counter_uncharge(struct res_counter *counter, unsigned long val)
 {
 	unsigned long flags;
+	struct res_counter *c;
 
-	spin_lock_irqsave(&counter->lock, flags);
-	res_counter_uncharge_locked(counter, val);
-	spin_unlock_irqrestore(&counter->lock, flags);
+	local_irq_save(flags);
+	for (c = counter; c != NULL; c = c->parent) {
+		spin_lock(&c->lock);
+		res_counter_uncharge_locked(c, val);
+		spin_unlock(&c->lock);
+	}
+	local_irq_restore(flags);
 }
 
 
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 9846f617115d..e72fb2b4a7d8 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -471,6 +471,7 @@ static int __mem_cgroup_try_charge(struct mm_struct *mm,
 {
 	struct mem_cgroup *mem;
 	int nr_retries = MEM_CGROUP_RECLAIM_RETRIES;
+	struct res_counter *fail_res;
 	/*
 	 * We always charge the cgroup the mm_struct belongs to.
 	 * The mm_struct's mem_cgroup changes on task migration if the
@@ -499,11 +500,12 @@ static int __mem_cgroup_try_charge(struct mm_struct *mm,
 		int ret;
 		bool noswap = false;
 
-		ret = res_counter_charge(&mem->res, PAGE_SIZE);
+		ret = res_counter_charge(&mem->res, PAGE_SIZE, &fail_res);
 		if (likely(!ret)) {
 			if (!do_swap_account)
 				break;
-			ret = res_counter_charge(&mem->memsw, PAGE_SIZE);
+			ret = res_counter_charge(&mem->memsw, PAGE_SIZE,
+							&fail_res);
 			if (likely(!ret))
 				break;
 			/* mem+swap counter fails */
@@ -1709,22 +1711,26 @@ static void __init enable_swap_cgroup(void)
 static struct cgroup_subsys_state *
 mem_cgroup_create(struct cgroup_subsys *ss, struct cgroup *cont)
 {
-	struct mem_cgroup *mem;
+	struct mem_cgroup *mem, *parent;
 	int node;
 
 	mem = mem_cgroup_alloc();
 	if (!mem)
 		return ERR_PTR(-ENOMEM);
 
-	res_counter_init(&mem->res);
-	res_counter_init(&mem->memsw);
-
 	for_each_node_state(node, N_POSSIBLE)
 		if (alloc_mem_cgroup_per_zone_info(mem, node))
 			goto free_out;
 	/* root ? */
-	if (cont->parent == NULL)
+	if (cont->parent == NULL) {
 		enable_swap_cgroup();
+		parent = NULL;
+	} else
+		parent = mem_cgroup_from_cont(cont->parent);
+
+	res_counter_init(&mem->res, parent ? &parent->res : NULL);
+	res_counter_init(&mem->memsw, parent ? &parent->memsw : NULL);
+
 
 	return &mem->css;
 free_out:
-- 
cgit v1.2.3


From 2e4d40915fb85207fe48cfc31201824ec6d7426e Mon Sep 17 00:00:00 2001
From: Lai Jiangshan <laijs@cn.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:07 -0800
Subject: memcontrol: rcu_read_lock() to protect mm_match_cgroup()

mm_match_cgroup() calls cgroup_subsys_state().

We must use rcu_read_lock() to protect cgroup_subsys_state().

Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Paul Menage <menage@google.com>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 472efd09118c..2de6504e01fb 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -61,8 +61,15 @@ int task_in_mem_cgroup(struct task_struct *task, const struct mem_cgroup *mem);
 
 extern struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p);
 
-#define mm_match_cgroup(mm, cgroup)	\
-	((cgroup) == mem_cgroup_from_task((mm)->owner))
+static inline
+int mm_match_cgroup(const struct mm_struct *mm, const struct mem_cgroup *cgroup)
+{
+	struct mem_cgroup *mem;
+	rcu_read_lock();
+	mem = mem_cgroup_from_task((mm)->owner);
+	rcu_read_unlock();
+	return cgroup == mem;
+}
 
 extern int
 mem_cgroup_prepare_migration(struct page *page, struct mem_cgroup **ptr);
-- 
cgit v1.2.3


From a636b327f731143ccc544b966cfd8de6cb6d72c6 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:08 -0800
Subject: memcg: avoid unnecessary system-wide-oom-killer

Current mmtom has new oom function as pagefault_out_of_memory().  It's
added for select bad process rathar than killing current.

When memcg hit limit and calls OOM at page_fault, this handler called and
system-wide-oom handling happens.  (means kernel panics if panic_on_oom is
true....)

To avoid overkill, check memcg's recent behavior before starting
system-wide-oom.

And this patch also fixes to guarantee "don't accnout against process with
TIF_MEMDIE".  This is necessary for smooth OOM.

[akpm@linux-foundation.org: build fix]
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Cc: Jan Blunck <jblunck@suse.de>
Cc: Hirokazu Takahashi <taka@valinux.co.jp>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h |  7 +++++++
 mm/memcontrol.c            | 33 +++++++++++++++++++++++++++++----
 mm/oom_kill.c              |  8 ++++++++
 3 files changed, 44 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 2de6504e01fb..2fdd1380bf0a 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -102,6 +102,8 @@ static inline bool mem_cgroup_disabled(void)
 	return false;
 }
 
+extern bool mem_cgroup_oom_called(struct task_struct *task);
+
 #else /* CONFIG_CGROUP_MEM_RES_CTLR */
 struct mem_cgroup;
 
@@ -234,6 +236,11 @@ static inline bool mem_cgroup_disabled(void)
 {
 	return true;
 }
+
+static inline bool mem_cgroup_oom_called(struct task_struct *task)
+{
+	return false;
+}
 #endif /* CONFIG_CGROUP_MEM_CONT */
 
 #endif /* _LINUX_MEMCONTROL_H */
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 886e2224c5fd..659b0c58f13e 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -153,7 +153,7 @@ struct mem_cgroup {
 	 * Should the accounting and control be hierarchical, per subtree?
 	 */
 	bool use_hierarchy;
-
+	unsigned long	last_oom_jiffies;
 	int		obsolete;
 	atomic_t	refcnt;
 	/*
@@ -615,6 +615,22 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem,
 	return ret;
 }
 
+bool mem_cgroup_oom_called(struct task_struct *task)
+{
+	bool ret = false;
+	struct mem_cgroup *mem;
+	struct mm_struct *mm;
+
+	rcu_read_lock();
+	mm = task->mm;
+	if (!mm)
+		mm = &init_mm;
+	mem = mem_cgroup_from_task(rcu_dereference(mm->owner));
+	if (mem && time_before(jiffies, mem->last_oom_jiffies + HZ/10))
+		ret = true;
+	rcu_read_unlock();
+	return ret;
+}
 /*
  * Unlike exported interface, "oom" parameter is added. if oom==true,
  * oom-killer can be invoked.
@@ -626,6 +642,13 @@ static int __mem_cgroup_try_charge(struct mm_struct *mm,
 	struct mem_cgroup *mem, *mem_over_limit;
 	int nr_retries = MEM_CGROUP_RECLAIM_RETRIES;
 	struct res_counter *fail_res;
+
+	if (unlikely(test_thread_flag(TIF_MEMDIE))) {
+		/* Don't account this! */
+		*memcg = NULL;
+		return 0;
+	}
+
 	/*
 	 * We always charge the cgroup the mm_struct belongs to.
 	 * The mm_struct's mem_cgroup changes on task migration if the
@@ -694,8 +717,10 @@ static int __mem_cgroup_try_charge(struct mm_struct *mm,
 			continue;
 
 		if (!nr_retries--) {
-			if (oom)
+			if (oom) {
 				mem_cgroup_out_of_memory(mem, gfp_mask);
+				mem->last_oom_jiffies = jiffies;
+			}
 			goto nomem;
 		}
 	}
@@ -832,7 +857,7 @@ static int mem_cgroup_move_parent(struct page_cgroup *pc,
 
 
 	ret = __mem_cgroup_try_charge(NULL, gfp_mask, &parent, false);
-	if (ret)
+	if (ret || !parent)
 		return ret;
 
 	if (!get_page_unless_zero(page))
@@ -883,7 +908,7 @@ static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm,
 
 	mem = memcg;
 	ret = __mem_cgroup_try_charge(mm, gfp_mask, &mem, true);
-	if (ret)
+	if (ret || !mem)
 		return ret;
 
 	__mem_cgroup_commit_charge(mem, pc, ctype);
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 6b9e758c98a5..fd150e3a2567 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -560,6 +560,13 @@ void pagefault_out_of_memory(void)
 		/* Got some memory back in the last second. */
 		return;
 
+	/*
+	 * If this is from memcg, oom-killer is already invoked.
+	 * and not worth to go system-wide-oom.
+	 */
+	if (mem_cgroup_oom_called(current))
+		goto rest_and_return;
+
 	if (sysctl_panic_on_oom)
 		panic("out of memory from page fault. panic_on_oom is selected.\n");
 
@@ -571,6 +578,7 @@ void pagefault_out_of_memory(void)
 	 * Give "p" a good chance of killing itself before we
 	 * retry to allocate memory.
 	 */
+rest_and_return:
 	if (!test_thread_flag(TIF_MEMDIE))
 		schedule_timeout_uninterruptible(1);
 }
-- 
cgit v1.2.3


From 2c26fdd70c3094fa3e84caf9ef434911933d5477 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:10 -0800
Subject: memcg: revert gfp mask fix

My patch, memcg-fix-gfp_mask-of-callers-of-charge.patch changed gfp_mask
of callers of charge to be GFP_HIGHUSER_MOVABLE for showing what will
happen at memory reclaim.

But in recent discussion, it's NACKed because it sounds ugly.

This patch is for reverting it and add some clean up to gfp_mask of
callers of charge.  No behavior change but need review before generating
HUNK in deep queue.

This patch also adds explanation to meaning of gfp_mask passed to charge
functions in memcontrol.h.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h | 10 ++++++++++
 mm/filemap.c               |  2 +-
 mm/memcontrol.c            | 10 +++++-----
 mm/memory.c                | 10 ++++------
 mm/shmem.c                 |  8 ++++----
 mm/swapfile.c              |  3 +--
 6 files changed, 25 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 2fdd1380bf0a..59ac95a64508 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -26,6 +26,16 @@ struct page;
 struct mm_struct;
 
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR
+/*
+ * All "charge" functions with gfp_mask should use GFP_KERNEL or
+ * (gfp_mask & GFP_RECLAIM_MASK). In current implementatin, memcg doesn't
+ * alloc memory but reclaims memory from all available zones. So, "where I want
+ * memory from" bits of gfp_mask has no meaning. So any bits of that field is
+ * available but adding a rule is better. charge functions' gfp_mask should
+ * be set to GFP_KERNEL or gfp_mask & GFP_RECLAIM_MASK for avoiding ambiguous
+ * codes.
+ * (Of course, if memcg does memory allocation in future, GFP_KERNEL is sane.)
+ */
 
 extern int mem_cgroup_newpage_charge(struct page *page, struct mm_struct *mm,
 				gfp_t gfp_mask);
diff --git a/mm/filemap.c b/mm/filemap.c
index 2f55a1e2baf7..ceba0bd03662 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -460,7 +460,7 @@ int add_to_page_cache_locked(struct page *page, struct address_space *mapping,
 	VM_BUG_ON(!PageLocked(page));
 
 	error = mem_cgroup_cache_charge(page, current->mm,
-					gfp_mask & ~__GFP_HIGHMEM);
+					gfp_mask & GFP_RECLAIM_MASK);
 	if (error)
 		goto out;
 
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 9bf5d7c8ede7..b9cd57b667d6 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1248,7 +1248,7 @@ int mem_cgroup_prepare_migration(struct page *page, struct mem_cgroup **ptr)
 	unlock_page_cgroup(pc);
 
 	if (mem) {
-		ret = mem_cgroup_try_charge(NULL, GFP_HIGHUSER_MOVABLE, &mem);
+		ret = mem_cgroup_try_charge(NULL, GFP_KERNEL, &mem);
 		css_put(&mem->css);
 	}
 	*ptr = mem;
@@ -1378,7 +1378,7 @@ static int mem_cgroup_resize_limit(struct mem_cgroup *memcg,
 			break;
 
 		progress = try_to_free_mem_cgroup_pages(memcg,
-				GFP_HIGHUSER_MOVABLE, false);
+				GFP_KERNEL, false);
   		if (!progress)			retry_count--;
 	}
 	return ret;
@@ -1418,7 +1418,7 @@ int mem_cgroup_resize_memsw_limit(struct mem_cgroup *memcg,
 			break;
 
 		oldusage = res_counter_read_u64(&memcg->memsw, RES_USAGE);
-		try_to_free_mem_cgroup_pages(memcg, GFP_HIGHUSER_MOVABLE, true);
+		try_to_free_mem_cgroup_pages(memcg, GFP_KERNEL, true);
 		curusage = res_counter_read_u64(&memcg->memsw, RES_USAGE);
 		if (curusage >= oldusage)
 			retry_count--;
@@ -1464,7 +1464,7 @@ static int mem_cgroup_force_empty_list(struct mem_cgroup *mem,
 		}
 		spin_unlock_irqrestore(&zone->lru_lock, flags);
 
-		ret = mem_cgroup_move_parent(pc, mem, GFP_HIGHUSER_MOVABLE);
+		ret = mem_cgroup_move_parent(pc, mem, GFP_KERNEL);
 		if (ret == -ENOMEM)
 			break;
 
@@ -1550,7 +1550,7 @@ try_to_free:
 			goto out;
 		}
 		progress = try_to_free_mem_cgroup_pages(mem,
-						  GFP_HIGHUSER_MOVABLE, false);
+						  GFP_KERNEL, false);
 		if (!progress) {
 			nr_retries--;
 			/* maybe some writeback is necessary */
diff --git a/mm/memory.c b/mm/memory.c
index 1358012ffa73..e5bfbe6b594c 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2000,7 +2000,7 @@ gotten:
 	cow_user_page(new_page, old_page, address, vma);
 	__SetPageUptodate(new_page);
 
-	if (mem_cgroup_newpage_charge(new_page, mm, GFP_HIGHUSER_MOVABLE))
+	if (mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL))
 		goto oom_free_new;
 
 	/*
@@ -2431,8 +2431,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	lock_page(page);
 	delayacct_clear_flag(DELAYACCT_PF_SWAPIN);
 
-	if (mem_cgroup_try_charge_swapin(mm, page,
-				GFP_HIGHUSER_MOVABLE, &ptr) == -ENOMEM) {
+	if (mem_cgroup_try_charge_swapin(mm, page, GFP_KERNEL, &ptr)) {
 		ret = VM_FAULT_OOM;
 		unlock_page(page);
 		goto out;
@@ -2524,7 +2523,7 @@ static int do_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		goto oom;
 	__SetPageUptodate(page);
 
-	if (mem_cgroup_newpage_charge(page, mm, GFP_HIGHUSER_MOVABLE))
+	if (mem_cgroup_newpage_charge(page, mm, GFP_KERNEL))
 		goto oom_free_page;
 
 	entry = mk_pte(page, vma->vm_page_prot);
@@ -2615,8 +2614,7 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 				ret = VM_FAULT_OOM;
 				goto out;
 			}
-			if (mem_cgroup_newpage_charge(page,
-						mm, GFP_HIGHUSER_MOVABLE)) {
+			if (mem_cgroup_newpage_charge(page, mm, GFP_KERNEL)) {
 				ret = VM_FAULT_OOM;
 				page_cache_release(page);
 				goto out;
diff --git a/mm/shmem.c b/mm/shmem.c
index adf5c3eedbc9..bbb7b043c986 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -932,8 +932,8 @@ found:
 	 * Charge page using GFP_HIGHUSER_MOVABLE while we can wait.
 	 * charged back to the user(not to caller) when swap account is used.
 	 */
-	error = mem_cgroup_cache_charge_swapin(page,
-			current->mm, GFP_HIGHUSER_MOVABLE, true);
+	error = mem_cgroup_cache_charge_swapin(page, current->mm, GFP_KERNEL,
+					true);
 	if (error)
 		goto out;
 	error = radix_tree_preload(GFP_KERNEL);
@@ -1275,7 +1275,7 @@ repeat:
 			 * charge against this swap cache here.
 			 */
 			if (mem_cgroup_cache_charge_swapin(swappage,
-						current->mm, gfp, false)) {
+				current->mm, gfp & GFP_RECLAIM_MASK, false)) {
 				page_cache_release(swappage);
 				error = -ENOMEM;
 				goto failed;
@@ -1393,7 +1393,7 @@ repeat:
 
 			/* Precharge page while we can wait, compensate after */
 			error = mem_cgroup_cache_charge(filepage, current->mm,
-					GFP_HIGHUSER_MOVABLE);
+					GFP_KERNEL);
 			if (error) {
 				page_cache_release(filepage);
 				shmem_unacct_blocks(info->flags, 1);
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 0579d9069b61..da422c47e2ee 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -698,8 +698,7 @@ static int unuse_pte(struct vm_area_struct *vma, pmd_t *pmd,
 	pte_t *pte;
 	int ret = 1;
 
-	if (mem_cgroup_try_charge_swapin(vma->vm_mm, page,
-					GFP_HIGHUSER_MOVABLE, &ptr))
+	if (mem_cgroup_try_charge_swapin(vma->vm_mm, page, GFP_KERNEL, &ptr))
 		ret = -ENOMEM;
 
 	pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
-- 
cgit v1.2.3


From f89eb90e33fd4e4e0cc1a6d20afd63c5a561885a Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:14 -0800
Subject: inactive_anon_is_low: move to vmscan

The inactive_anon_is_low() is called only vmscan.  Then it can move to
vmscan.c

This patch doesn't have any functional change.

Reviewd-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/mm_inline.h | 19 -------------------
 mm/vmscan.c               | 20 ++++++++++++++++++++
 2 files changed, 20 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/mm_inline.h b/include/linux/mm_inline.h
index 37ef13d0f01e..7fbb97267556 100644
--- a/include/linux/mm_inline.h
+++ b/include/linux/mm_inline.h
@@ -81,23 +81,4 @@ static inline enum lru_list page_lru(struct page *page)
 	return lru;
 }
 
-/**
- * inactive_anon_is_low - check if anonymous pages need to be deactivated
- * @zone: zone to check
- *
- * Returns true if the zone does not have enough inactive anon pages,
- * meaning some active anon pages need to be deactivated.
- */
-static inline int inactive_anon_is_low(struct zone *zone)
-{
-	unsigned long active, inactive;
-
-	active = zone_page_state(zone, NR_ACTIVE_ANON);
-	inactive = zone_page_state(zone, NR_INACTIVE_ANON);
-
-	if (inactive * zone->inactive_ratio < active)
-		return 1;
-
-	return 0;
-}
 #endif
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 45983af1de3d..f75d924cb4f4 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1291,6 +1291,26 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 	pagevec_release(&pvec);
 }
 
+/**
+ * inactive_anon_is_low - check if anonymous pages need to be deactivated
+ * @zone: zone to check
+ *
+ * Returns true if the zone does not have enough inactive anon pages,
+ * meaning some active anon pages need to be deactivated.
+ */
+static int inactive_anon_is_low(struct zone *zone)
+{
+	unsigned long active, inactive;
+
+	active = zone_page_state(zone, NR_ACTIVE_ANON);
+	inactive = zone_page_state(zone, NR_INACTIVE_ANON);
+
+	if (inactive * zone->inactive_ratio < active)
+		return 1;
+
+	return 0;
+}
+
 static unsigned long shrink_list(enum lru_list lru, unsigned long nr_to_scan,
 	struct zone *zone, struct scan_control *sc, int priority)
 {
-- 
cgit v1.2.3


From 6e9015716ae9b59e9635d692fddfcfb9582c146c Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:15 -0800
Subject: mm: introduce zone_reclaim struct

Add zone_reclam_stat struct for later enhancement.

A later patch uses this.  This patch doesn't any behavior change (yet).

Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/mmzone.h | 24 ++++++++++++++----------
 mm/page_alloc.c        |  8 ++++----
 mm/swap.c              | 12 ++++++++----
 mm/vmscan.c            | 47 ++++++++++++++++++++++++++++++-----------------
 4 files changed, 56 insertions(+), 35 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 35a7b5e19465..09c14e213b63 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -263,6 +263,19 @@ enum zone_type {
 #error ZONES_SHIFT -- too many zones configured adjust calculation
 #endif
 
+struct zone_reclaim_stat {
+	/*
+	 * The pageout code in vmscan.c keeps track of how many of the
+	 * mem/swap backed and file backed pages are refeferenced.
+	 * The higher the rotated/scanned ratio, the more valuable
+	 * that cache is.
+	 *
+	 * The anon LRU stats live in [0], file LRU stats in [1]
+	 */
+	unsigned long		recent_rotated[2];
+	unsigned long		recent_scanned[2];
+};
+
 struct zone {
 	/* Fields commonly accessed by the page allocator */
 	unsigned long		pages_min, pages_low, pages_high;
@@ -315,16 +328,7 @@ struct zone {
 		unsigned long nr_scan;
 	} lru[NR_LRU_LISTS];
 
-	/*
-	 * The pageout code in vmscan.c keeps track of how many of the
-	 * mem/swap backed and file backed pages are refeferenced.
-	 * The higher the rotated/scanned ratio, the more valuable
-	 * that cache is.
-	 *
-	 * The anon LRU stats live in [0], file LRU stats in [1]
-	 */
-	unsigned long		recent_rotated[2];
-	unsigned long		recent_scanned[2];
+	struct zone_reclaim_stat reclaim_stat;
 
 	unsigned long		pages_scanned;	   /* since last reclaim */
 	unsigned long		flags;		   /* zone flags, see below */
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 7bf22e045318..5675b3073854 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3523,10 +3523,10 @@ static void __paginginit free_area_init_core(struct pglist_data *pgdat,
 			INIT_LIST_HEAD(&zone->lru[l].list);
 			zone->lru[l].nr_scan = 0;
 		}
-		zone->recent_rotated[0] = 0;
-		zone->recent_rotated[1] = 0;
-		zone->recent_scanned[0] = 0;
-		zone->recent_scanned[1] = 0;
+		zone->reclaim_stat.recent_rotated[0] = 0;
+		zone->reclaim_stat.recent_rotated[1] = 0;
+		zone->reclaim_stat.recent_scanned[0] = 0;
+		zone->reclaim_stat.recent_scanned[1] = 0;
 		zap_zone_vm_stats(zone);
 		zone->flags = 0;
 		if (!size)
diff --git a/mm/swap.c b/mm/swap.c
index 8a98a9c90704..26b07e7bc3d3 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -157,6 +157,7 @@ void  rotate_reclaimable_page(struct page *page)
 void activate_page(struct page *page)
 {
 	struct zone *zone = page_zone(page);
+	struct zone_reclaim_stat *reclaim_stat = &zone->reclaim_stat;
 
 	spin_lock_irq(&zone->lru_lock);
 	if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) {
@@ -169,8 +170,8 @@ void activate_page(struct page *page)
 		add_page_to_lru_list(zone, page, lru);
 		__count_vm_event(PGACTIVATE);
 
-		zone->recent_rotated[!!file]++;
-		zone->recent_scanned[!!file]++;
+		reclaim_stat->recent_rotated[!!file]++;
+		reclaim_stat->recent_scanned[!!file]++;
 	}
 	spin_unlock_irq(&zone->lru_lock);
 }
@@ -385,6 +386,8 @@ void ____pagevec_lru_add(struct pagevec *pvec, enum lru_list lru)
 {
 	int i;
 	struct zone *zone = NULL;
+	struct zone_reclaim_stat *reclaim_stat = NULL;
+
 	VM_BUG_ON(is_unevictable_lru(lru));
 
 	for (i = 0; i < pagevec_count(pvec); i++) {
@@ -396,6 +399,7 @@ void ____pagevec_lru_add(struct pagevec *pvec, enum lru_list lru)
 			if (zone)
 				spin_unlock_irq(&zone->lru_lock);
 			zone = pagezone;
+			reclaim_stat = &zone->reclaim_stat;
 			spin_lock_irq(&zone->lru_lock);
 		}
 		VM_BUG_ON(PageActive(page));
@@ -403,10 +407,10 @@ void ____pagevec_lru_add(struct pagevec *pvec, enum lru_list lru)
 		VM_BUG_ON(PageLRU(page));
 		SetPageLRU(page);
 		file = is_file_lru(lru);
-		zone->recent_scanned[file]++;
+		reclaim_stat->recent_scanned[file]++;
 		if (is_active_lru(lru)) {
 			SetPageActive(page);
-			zone->recent_rotated[file]++;
+			reclaim_stat->recent_rotated[file]++;
 		}
 		add_page_to_lru_list(zone, page, lru);
 	}
diff --git a/mm/vmscan.c b/mm/vmscan.c
index f75d924cb4f4..03ca923c6656 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -130,6 +130,12 @@ static DECLARE_RWSEM(shrinker_rwsem);
 #define scan_global_lru(sc)	(1)
 #endif
 
+static struct zone_reclaim_stat *get_reclaim_stat(struct zone *zone,
+						  struct scan_control *sc)
+{
+	return &zone->reclaim_stat;
+}
+
 /*
  * Add a shrinker callback to be called from the vm
  */
@@ -1029,6 +1035,7 @@ static unsigned long shrink_inactive_list(unsigned long max_scan,
 	struct pagevec pvec;
 	unsigned long nr_scanned = 0;
 	unsigned long nr_reclaimed = 0;
+	struct zone_reclaim_stat *reclaim_stat = get_reclaim_stat(zone, sc);
 
 	pagevec_init(&pvec, 1);
 
@@ -1072,10 +1079,14 @@ static unsigned long shrink_inactive_list(unsigned long max_scan,
 
 		if (scan_global_lru(sc)) {
 			zone->pages_scanned += nr_scan;
-			zone->recent_scanned[0] += count[LRU_INACTIVE_ANON];
-			zone->recent_scanned[0] += count[LRU_ACTIVE_ANON];
-			zone->recent_scanned[1] += count[LRU_INACTIVE_FILE];
-			zone->recent_scanned[1] += count[LRU_ACTIVE_FILE];
+			reclaim_stat->recent_scanned[0] +=
+						      count[LRU_INACTIVE_ANON];
+			reclaim_stat->recent_scanned[0] +=
+						      count[LRU_ACTIVE_ANON];
+			reclaim_stat->recent_scanned[1] +=
+						      count[LRU_INACTIVE_FILE];
+			reclaim_stat->recent_scanned[1] +=
+						      count[LRU_ACTIVE_FILE];
 		}
 		spin_unlock_irq(&zone->lru_lock);
 
@@ -1136,7 +1147,7 @@ static unsigned long shrink_inactive_list(unsigned long max_scan,
 			add_page_to_lru_list(zone, page, lru);
 			if (PageActive(page) && scan_global_lru(sc)) {
 				int file = !!page_is_file_cache(page);
-				zone->recent_rotated[file]++;
+				reclaim_stat->recent_rotated[file]++;
 			}
 			if (!pagevec_add(&pvec, page)) {
 				spin_unlock_irq(&zone->lru_lock);
@@ -1196,6 +1207,7 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 	struct page *page;
 	struct pagevec pvec;
 	enum lru_list lru;
+	struct zone_reclaim_stat *reclaim_stat = get_reclaim_stat(zone, sc);
 
 	lru_add_drain();
 	spin_lock_irq(&zone->lru_lock);
@@ -1208,7 +1220,7 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 	 */
 	if (scan_global_lru(sc)) {
 		zone->pages_scanned += pgscanned;
-		zone->recent_scanned[!!file] += pgmoved;
+		reclaim_stat->recent_scanned[!!file] += pgmoved;
 	}
 
 	if (file)
@@ -1251,7 +1263,7 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 	 * pages in get_scan_ratio.
 	 */
 	if (scan_global_lru(sc))
-		zone->recent_rotated[!!file] += pgmoved;
+		reclaim_stat->recent_rotated[!!file] += pgmoved;
 
 	while (!list_empty(&l_inactive)) {
 		page = lru_to_page(&l_inactive);
@@ -1344,6 +1356,7 @@ static void get_scan_ratio(struct zone *zone, struct scan_control *sc,
 	unsigned long anon, file, free;
 	unsigned long anon_prio, file_prio;
 	unsigned long ap, fp;
+	struct zone_reclaim_stat *reclaim_stat = get_reclaim_stat(zone, sc);
 
 	/* If we have no swap space, do not bother scanning anon pages. */
 	if (nr_swap_pages <= 0) {
@@ -1376,17 +1389,17 @@ static void get_scan_ratio(struct zone *zone, struct scan_control *sc,
 	 *
 	 * anon in [0], file in [1]
 	 */
-	if (unlikely(zone->recent_scanned[0] > anon / 4)) {
+	if (unlikely(reclaim_stat->recent_scanned[0] > anon / 4)) {
 		spin_lock_irq(&zone->lru_lock);
-		zone->recent_scanned[0] /= 2;
-		zone->recent_rotated[0] /= 2;
+		reclaim_stat->recent_scanned[0] /= 2;
+		reclaim_stat->recent_rotated[0] /= 2;
 		spin_unlock_irq(&zone->lru_lock);
 	}
 
-	if (unlikely(zone->recent_scanned[1] > file / 4)) {
+	if (unlikely(reclaim_stat->recent_scanned[1] > file / 4)) {
 		spin_lock_irq(&zone->lru_lock);
-		zone->recent_scanned[1] /= 2;
-		zone->recent_rotated[1] /= 2;
+		reclaim_stat->recent_scanned[1] /= 2;
+		reclaim_stat->recent_rotated[1] /= 2;
 		spin_unlock_irq(&zone->lru_lock);
 	}
 
@@ -1402,11 +1415,11 @@ static void get_scan_ratio(struct zone *zone, struct scan_control *sc,
 	 * proportional to the fraction of recently scanned pages on
 	 * each list that were recently referenced and in active use.
 	 */
-	ap = (anon_prio + 1) * (zone->recent_scanned[0] + 1);
-	ap /= zone->recent_rotated[0] + 1;
+	ap = (anon_prio + 1) * (reclaim_stat->recent_scanned[0] + 1);
+	ap /= reclaim_stat->recent_rotated[0] + 1;
 
-	fp = (file_prio + 1) * (zone->recent_scanned[1] + 1);
-	fp /= zone->recent_rotated[1] + 1;
+	fp = (file_prio + 1) * (reclaim_stat->recent_scanned[1] + 1);
+	fp /= reclaim_stat->recent_rotated[1] + 1;
 
 	/* Normalize to percentages */
 	percent[0] = 100 * ap / (ap + fp + 1);
-- 
cgit v1.2.3


From 14797e2363c2b2f1ce139fd1c5a215e4e05aa1d9 Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:18 -0800
Subject: memcg: add inactive_anon_is_low()

The inactive_anon_is_low() is key component of active/inactive anon
balancing on reclaim.  However current inactive_anon_is_low() function
only consider global reclaim.

Therefore, we need following ugly scan_global_lru() condition.

	if (lru == LRU_ACTIVE_ANON &&
	    (!scan_global_lru(sc) || inactive_anon_is_low(zone))) {
		shrink_active_list(nr_to_scan, zone, sc, priority, file);
		return 0;

it cause that memcg reclaim always deactivate pages when shrink_list() is
called.  To make mem_cgroup_inactive_anon_is_low() improve active/inactive
anon balancing of memcgroup.

Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: "Pekka Enberg" <penberg@cs.helsinki.fi>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h |  9 +++++++++
 mm/memcontrol.c            | 46 +++++++++++++++++++++++++++++++++++++++++++++-
 mm/vmscan.c                | 37 +++++++++++++++++++++++--------------
 3 files changed, 77 insertions(+), 15 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 59ac95a64508..aad9377c9828 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -100,6 +100,8 @@ extern void mem_cgroup_record_reclaim_priority(struct mem_cgroup *mem,
 
 extern long mem_cgroup_calc_reclaim(struct mem_cgroup *mem, struct zone *zone,
 					int priority, enum lru_list lru);
+int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg,
+				    struct zone *zone);
 
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
 extern int do_swap_account;
@@ -251,6 +253,13 @@ static inline bool mem_cgroup_oom_called(struct task_struct *task)
 {
 	return false;
 }
+
+static inline int
+mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg, struct zone *zone)
+{
+	return 1;
+}
+
 #endif /* CONFIG_CGROUP_MEM_CONT */
 
 #endif /* _LINUX_MEMCONTROL_H */
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 457d671029b8..6611328460e9 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -156,6 +156,9 @@ struct mem_cgroup {
 	unsigned long	last_oom_jiffies;
 	int		obsolete;
 	atomic_t	refcnt;
+
+	unsigned int inactive_ratio;
+
 	/*
 	 * statistics. This must be placed at the end of memcg.
 	 */
@@ -431,6 +434,20 @@ long mem_cgroup_calc_reclaim(struct mem_cgroup *mem, struct zone *zone,
 	return (nr_pages >> priority);
 }
 
+int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg, struct zone *zone)
+{
+	unsigned long active;
+	unsigned long inactive;
+
+	inactive = mem_cgroup_get_all_zonestat(memcg, LRU_INACTIVE_ANON);
+	active = mem_cgroup_get_all_zonestat(memcg, LRU_ACTIVE_ANON);
+
+	if (inactive * memcg->inactive_ratio < active)
+		return 1;
+
+	return 0;
+}
+
 unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
 					struct list_head *dst,
 					unsigned long *scanned, int order,
@@ -1360,6 +1377,29 @@ int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask)
 	return 0;
 }
 
+/*
+ * The inactive anon list should be small enough that the VM never has to
+ * do too much work, but large enough that each inactive page has a chance
+ * to be referenced again before it is swapped out.
+ *
+ * this calculation is straightforward porting from
+ * page_alloc.c::setup_per_zone_inactive_ratio().
+ * it describe more detail.
+ */
+static void mem_cgroup_set_inactive_ratio(struct mem_cgroup *memcg)
+{
+	unsigned int gb, ratio;
+
+	gb = res_counter_read_u64(&memcg->res, RES_LIMIT) >> 30;
+	if (gb)
+		ratio = int_sqrt(10 * gb);
+	else
+		ratio = 1;
+
+	memcg->inactive_ratio = ratio;
+
+}
+
 static DEFINE_MUTEX(set_limit_mutex);
 
 static int mem_cgroup_resize_limit(struct mem_cgroup *memcg,
@@ -1398,6 +1438,10 @@ static int mem_cgroup_resize_limit(struct mem_cgroup *memcg,
 				GFP_KERNEL, false);
   		if (!progress)			retry_count--;
 	}
+
+	if (!ret)
+		mem_cgroup_set_inactive_ratio(memcg);
+
 	return ret;
 }
 
@@ -1982,7 +2026,7 @@ mem_cgroup_create(struct cgroup_subsys *ss, struct cgroup *cont)
 		res_counter_init(&mem->res, NULL);
 		res_counter_init(&mem->memsw, NULL);
 	}
-
+	mem_cgroup_set_inactive_ratio(mem);
 	mem->last_scanned_child = NULL;
 
 	return &mem->css;
diff --git a/mm/vmscan.c b/mm/vmscan.c
index e2b31a522a66..b2bc06bffcfb 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1310,14 +1310,7 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 	pagevec_release(&pvec);
 }
 
-/**
- * inactive_anon_is_low - check if anonymous pages need to be deactivated
- * @zone: zone to check
- *
- * Returns true if the zone does not have enough inactive anon pages,
- * meaning some active anon pages need to be deactivated.
- */
-static int inactive_anon_is_low(struct zone *zone)
+static int inactive_anon_is_low_global(struct zone *zone)
 {
 	unsigned long active, inactive;
 
@@ -1330,6 +1323,25 @@ static int inactive_anon_is_low(struct zone *zone)
 	return 0;
 }
 
+/**
+ * inactive_anon_is_low - check if anonymous pages need to be deactivated
+ * @zone: zone to check
+ * @sc:   scan control of this context
+ *
+ * Returns true if the zone does not have enough inactive anon pages,
+ * meaning some active anon pages need to be deactivated.
+ */
+static int inactive_anon_is_low(struct zone *zone, struct scan_control *sc)
+{
+	int low;
+
+	if (scan_global_lru(sc))
+		low = inactive_anon_is_low_global(zone);
+	else
+		low = mem_cgroup_inactive_anon_is_low(sc->mem_cgroup, zone);
+	return low;
+}
+
 static unsigned long shrink_list(enum lru_list lru, unsigned long nr_to_scan,
 	struct zone *zone, struct scan_control *sc, int priority)
 {
@@ -1340,8 +1352,7 @@ static unsigned long shrink_list(enum lru_list lru, unsigned long nr_to_scan,
 		return 0;
 	}
 
-	if (lru == LRU_ACTIVE_ANON &&
-	    (!scan_global_lru(sc) || inactive_anon_is_low(zone))) {
+	if (lru == LRU_ACTIVE_ANON && inactive_anon_is_low(zone, sc)) {
 		shrink_active_list(nr_to_scan, zone, sc, priority, file);
 		return 0;
 	}
@@ -1509,9 +1520,7 @@ static void shrink_zone(int priority, struct zone *zone,
 	 * Even if we did not try to evict anon pages at all, we want to
 	 * rebalance the anon lru active/inactive ratio.
 	 */
-	if (!scan_global_lru(sc) || inactive_anon_is_low(zone))
-		shrink_active_list(SWAP_CLUSTER_MAX, zone, sc, priority, 0);
-	else if (!scan_global_lru(sc))
+	if (inactive_anon_is_low(zone, sc))
 		shrink_active_list(SWAP_CLUSTER_MAX, zone, sc, priority, 0);
 
 	throttle_vm_writeout(sc->gfp_mask);
@@ -1807,7 +1816,7 @@ loop_again:
 			 * Do some background aging of the anon list, to give
 			 * pages a chance to be referenced before reclaiming.
 			 */
-			if (inactive_anon_is_low(zone))
+			if (inactive_anon_is_low(zone, &sc))
 				shrink_active_list(SWAP_CLUSTER_MAX, zone,
 							&sc, priority, 0);
 
-- 
cgit v1.2.3


From a3d8e0549d913e30968fa02e505dfe02c0a23e0d Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:19 -0800
Subject: memcg: add mem_cgroup_zone_nr_pages()

Introduce mem_cgroup_zone_nr_pages().  It is called by zone_nr_pages()
helper function.

This patch doesn't have any behavior change.

Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: Balbir Singh <balbir@in.ibm.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h | 11 +++++++++++
 mm/memcontrol.c            | 12 +++++++++++-
 mm/vmscan.c                |  3 +++
 3 files changed, 25 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index aad9377c9828..b1defd6a2783 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -102,6 +102,9 @@ extern long mem_cgroup_calc_reclaim(struct mem_cgroup *mem, struct zone *zone,
 					int priority, enum lru_list lru);
 int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg,
 				    struct zone *zone);
+unsigned long mem_cgroup_zone_nr_pages(struct mem_cgroup *memcg,
+				       struct zone *zone,
+				       enum lru_list lru);
 
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
 extern int do_swap_account;
@@ -260,6 +263,14 @@ mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg, struct zone *zone)
 	return 1;
 }
 
+static inline unsigned long
+mem_cgroup_zone_nr_pages(struct mem_cgroup *memcg, struct zone *zone,
+			 enum lru_list lru)
+{
+	return 0;
+}
+
+
 #endif /* CONFIG_CGROUP_MEM_CONT */
 
 #endif /* _LINUX_MEMCONTROL_H */
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 6611328460e9..313247e6c503 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -186,7 +186,6 @@ pcg_default_flags[NR_CHARGE_TYPE] = {
 	0, /* FORCE */
 };
 
-
 /* for encoding cft->private value on file */
 #define _MEM			(0)
 #define _MEMSWAP		(1)
@@ -448,6 +447,17 @@ int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg, struct zone *zone)
 	return 0;
 }
 
+unsigned long mem_cgroup_zone_nr_pages(struct mem_cgroup *memcg,
+				       struct zone *zone,
+				       enum lru_list lru)
+{
+	int nid = zone->zone_pgdat->node_id;
+	int zid = zone_idx(zone);
+	struct mem_cgroup_per_zone *mz = mem_cgroup_zoneinfo(memcg, nid, zid);
+
+	return MEM_CGROUP_ZSTAT(mz, lru);
+}
+
 unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
 					struct list_head *dst,
 					unsigned long *scanned, int order,
diff --git a/mm/vmscan.c b/mm/vmscan.c
index b2bc06bffcfb..d958d624d3ae 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -139,6 +139,9 @@ static struct zone_reclaim_stat *get_reclaim_stat(struct zone *zone,
 static unsigned long zone_nr_pages(struct zone *zone, struct scan_control *sc,
 				   enum lru_list lru)
 {
+	if (!scan_global_lru(sc))
+		return mem_cgroup_zone_nr_pages(sc->mem_cgroup, zone, lru);
+
 	return zone_page_state(zone, NR_LRU_BASE + lru);
 }
 
-- 
cgit v1.2.3


From 3e2f41f1f64744f7942980d93cc93dd3e5924560 Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:20 -0800
Subject: memcg: add zone_reclaim_stat

Introduce mem_cgroup_per_zone::reclaim_stat member and its statics
collecting function.

Now, get_scan_ratio() can calculate correct value on memcg reclaim.

[hugh@veritas.com: avoid reclaim_stat oops when disabled]
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h | 16 ++++++++++++++++
 mm/memcontrol.c            | 29 +++++++++++++++++++++++++++++
 mm/swap.c                  | 34 +++++++++++++++++++++++++---------
 mm/vmscan.c                | 27 +++++++++++++--------------
 4 files changed, 83 insertions(+), 23 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index b1defd6a2783..36b8ebb39b82 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -105,6 +105,10 @@ int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg,
 unsigned long mem_cgroup_zone_nr_pages(struct mem_cgroup *memcg,
 				       struct zone *zone,
 				       enum lru_list lru);
+struct zone_reclaim_stat *mem_cgroup_get_reclaim_stat(struct mem_cgroup *memcg,
+						      struct zone *zone);
+struct zone_reclaim_stat*
+mem_cgroup_get_reclaim_stat_from_page(struct page *page);
 
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
 extern int do_swap_account;
@@ -271,6 +275,18 @@ mem_cgroup_zone_nr_pages(struct mem_cgroup *memcg, struct zone *zone,
 }
 
 
+static inline struct zone_reclaim_stat*
+mem_cgroup_get_reclaim_stat(struct mem_cgroup *memcg, struct zone *zone)
+{
+	return NULL;
+}
+
+static inline struct zone_reclaim_stat*
+mem_cgroup_get_reclaim_stat_from_page(struct page *page)
+{
+	return NULL;
+}
+
 #endif /* CONFIG_CGROUP_MEM_CONT */
 
 #endif /* _LINUX_MEMCONTROL_H */
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 313247e6c503..7b7f4dc05035 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -103,6 +103,8 @@ struct mem_cgroup_per_zone {
 	 */
 	struct list_head	lists[NR_LRU_LISTS];
 	unsigned long		count[NR_LRU_LISTS];
+
+	struct zone_reclaim_stat reclaim_stat;
 };
 /* Macro for accessing counter */
 #define MEM_CGROUP_ZSTAT(mz, idx)	((mz)->count[(idx)])
@@ -458,6 +460,33 @@ unsigned long mem_cgroup_zone_nr_pages(struct mem_cgroup *memcg,
 	return MEM_CGROUP_ZSTAT(mz, lru);
 }
 
+struct zone_reclaim_stat *mem_cgroup_get_reclaim_stat(struct mem_cgroup *memcg,
+						      struct zone *zone)
+{
+	int nid = zone->zone_pgdat->node_id;
+	int zid = zone_idx(zone);
+	struct mem_cgroup_per_zone *mz = mem_cgroup_zoneinfo(memcg, nid, zid);
+
+	return &mz->reclaim_stat;
+}
+
+struct zone_reclaim_stat *
+mem_cgroup_get_reclaim_stat_from_page(struct page *page)
+{
+	struct page_cgroup *pc;
+	struct mem_cgroup_per_zone *mz;
+
+	if (mem_cgroup_disabled())
+		return NULL;
+
+	pc = lookup_page_cgroup(page);
+	mz = page_cgroup_zoneinfo(pc);
+	if (!mz)
+		return NULL;
+
+	return &mz->reclaim_stat;
+}
+
 unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
 					struct list_head *dst,
 					unsigned long *scanned, int order,
diff --git a/mm/swap.c b/mm/swap.c
index 26b07e7bc3d3..8adb9feb61e1 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -151,13 +151,32 @@ void  rotate_reclaimable_page(struct page *page)
 	}
 }
 
+static void update_page_reclaim_stat(struct zone *zone, struct page *page,
+				     int file, int rotated)
+{
+	struct zone_reclaim_stat *reclaim_stat = &zone->reclaim_stat;
+	struct zone_reclaim_stat *memcg_reclaim_stat;
+
+	memcg_reclaim_stat = mem_cgroup_get_reclaim_stat_from_page(page);
+
+	reclaim_stat->recent_scanned[file]++;
+	if (rotated)
+		reclaim_stat->recent_rotated[file]++;
+
+	if (!memcg_reclaim_stat)
+		return;
+
+	memcg_reclaim_stat->recent_scanned[file]++;
+	if (rotated)
+		memcg_reclaim_stat->recent_rotated[file]++;
+}
+
 /*
  * FIXME: speed this up?
  */
 void activate_page(struct page *page)
 {
 	struct zone *zone = page_zone(page);
-	struct zone_reclaim_stat *reclaim_stat = &zone->reclaim_stat;
 
 	spin_lock_irq(&zone->lru_lock);
 	if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) {
@@ -170,8 +189,7 @@ void activate_page(struct page *page)
 		add_page_to_lru_list(zone, page, lru);
 		__count_vm_event(PGACTIVATE);
 
-		reclaim_stat->recent_rotated[!!file]++;
-		reclaim_stat->recent_scanned[!!file]++;
+		update_page_reclaim_stat(zone, page, !!file, 1);
 	}
 	spin_unlock_irq(&zone->lru_lock);
 }
@@ -386,7 +404,6 @@ void ____pagevec_lru_add(struct pagevec *pvec, enum lru_list lru)
 {
 	int i;
 	struct zone *zone = NULL;
-	struct zone_reclaim_stat *reclaim_stat = NULL;
 
 	VM_BUG_ON(is_unevictable_lru(lru));
 
@@ -394,24 +411,23 @@ void ____pagevec_lru_add(struct pagevec *pvec, enum lru_list lru)
 		struct page *page = pvec->pages[i];
 		struct zone *pagezone = page_zone(page);
 		int file;
+		int active;
 
 		if (pagezone != zone) {
 			if (zone)
 				spin_unlock_irq(&zone->lru_lock);
 			zone = pagezone;
-			reclaim_stat = &zone->reclaim_stat;
 			spin_lock_irq(&zone->lru_lock);
 		}
 		VM_BUG_ON(PageActive(page));
 		VM_BUG_ON(PageUnevictable(page));
 		VM_BUG_ON(PageLRU(page));
 		SetPageLRU(page);
+		active = is_active_lru(lru);
 		file = is_file_lru(lru);
-		reclaim_stat->recent_scanned[file]++;
-		if (is_active_lru(lru)) {
+		if (active)
 			SetPageActive(page);
-			reclaim_stat->recent_rotated[file]++;
-		}
+		update_page_reclaim_stat(zone, page, file, active);
 		add_page_to_lru_list(zone, page, lru);
 	}
 	if (zone)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index d958d624d3ae..56fc7abe4d23 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -133,6 +133,9 @@ static DECLARE_RWSEM(shrinker_rwsem);
 static struct zone_reclaim_stat *get_reclaim_stat(struct zone *zone,
 						  struct scan_control *sc)
 {
+	if (!scan_global_lru(sc))
+		return mem_cgroup_get_reclaim_stat(sc->mem_cgroup, zone);
+
 	return &zone->reclaim_stat;
 }
 
@@ -1087,17 +1090,14 @@ static unsigned long shrink_inactive_list(unsigned long max_scan,
 		__mod_zone_page_state(zone, NR_INACTIVE_ANON,
 						-count[LRU_INACTIVE_ANON]);
 
-		if (scan_global_lru(sc)) {
+		if (scan_global_lru(sc))
 			zone->pages_scanned += nr_scan;
-			reclaim_stat->recent_scanned[0] +=
-						      count[LRU_INACTIVE_ANON];
-			reclaim_stat->recent_scanned[0] +=
-						      count[LRU_ACTIVE_ANON];
-			reclaim_stat->recent_scanned[1] +=
-						      count[LRU_INACTIVE_FILE];
-			reclaim_stat->recent_scanned[1] +=
-						      count[LRU_ACTIVE_FILE];
-		}
+
+		reclaim_stat->recent_scanned[0] += count[LRU_INACTIVE_ANON];
+		reclaim_stat->recent_scanned[0] += count[LRU_ACTIVE_ANON];
+		reclaim_stat->recent_scanned[1] += count[LRU_INACTIVE_FILE];
+		reclaim_stat->recent_scanned[1] += count[LRU_ACTIVE_FILE];
+
 		spin_unlock_irq(&zone->lru_lock);
 
 		nr_scanned += nr_scan;
@@ -1155,7 +1155,7 @@ static unsigned long shrink_inactive_list(unsigned long max_scan,
 			SetPageLRU(page);
 			lru = page_lru(page);
 			add_page_to_lru_list(zone, page, lru);
-			if (PageActive(page) && scan_global_lru(sc)) {
+			if (PageActive(page)) {
 				int file = !!page_is_file_cache(page);
 				reclaim_stat->recent_rotated[file]++;
 			}
@@ -1230,8 +1230,8 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 	 */
 	if (scan_global_lru(sc)) {
 		zone->pages_scanned += pgscanned;
-		reclaim_stat->recent_scanned[!!file] += pgmoved;
 	}
+	reclaim_stat->recent_scanned[!!file] += pgmoved;
 
 	if (file)
 		__mod_zone_page_state(zone, NR_ACTIVE_FILE, -pgmoved);
@@ -1272,8 +1272,7 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 	 * This helps balance scan pressure between file and anonymous
 	 * pages in get_scan_ratio.
 	 */
-	if (scan_global_lru(sc))
-		reclaim_stat->recent_rotated[!!file] += pgmoved;
+	reclaim_stat->recent_rotated[!!file] += pgmoved;
 
 	while (!list_empty(&l_inactive)) {
 		page = lru_to_page(&l_inactive);
-- 
cgit v1.2.3


From 9439c1c95b5c25b8031b2a7eb7e1590eb84be7f5 Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:21 -0800
Subject: memcg: remove mem_cgroup_cal_reclaim()

Now, get_scan_ratio() return correct value although memcg reclaim.  Then,
mem_cgroup_calc_reclaim() can be removed.

So, memcg reclaim get the same capability of anon/file reclaim balancing
as global reclaim now.

Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@redhat.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h | 10 ----------
 mm/memcontrol.c            | 21 ---------------------
 mm/vmscan.c                | 27 ++++++++++-----------------
 3 files changed, 10 insertions(+), 48 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 36b8ebb39b82..8752052da8df 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -97,9 +97,6 @@ extern void mem_cgroup_note_reclaim_priority(struct mem_cgroup *mem,
 							int priority);
 extern void mem_cgroup_record_reclaim_priority(struct mem_cgroup *mem,
 							int priority);
-
-extern long mem_cgroup_calc_reclaim(struct mem_cgroup *mem, struct zone *zone,
-					int priority, enum lru_list lru);
 int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg,
 				    struct zone *zone);
 unsigned long mem_cgroup_zone_nr_pages(struct mem_cgroup *memcg,
@@ -244,13 +241,6 @@ static inline void mem_cgroup_record_reclaim_priority(struct mem_cgroup *mem,
 {
 }
 
-static inline long mem_cgroup_calc_reclaim(struct mem_cgroup *mem,
-					struct zone *zone, int priority,
-					enum lru_list lru)
-{
-	return 0;
-}
-
 static inline bool mem_cgroup_disabled(void)
 {
 	return true;
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 7b7f4dc05035..b8c1e5acc25a 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -414,27 +414,6 @@ void mem_cgroup_record_reclaim_priority(struct mem_cgroup *mem, int priority)
 	mem->prev_priority = priority;
 }
 
-/*
- * Calculate # of pages to be scanned in this priority/zone.
- * See also vmscan.c
- *
- * priority starts from "DEF_PRIORITY" and decremented in each loop.
- * (see include/linux/mmzone.h)
- */
-
-long mem_cgroup_calc_reclaim(struct mem_cgroup *mem, struct zone *zone,
-					int priority, enum lru_list lru)
-{
-	long nr_pages;
-	int nid = zone->zone_pgdat->node_id;
-	int zid = zone_idx(zone);
-	struct mem_cgroup_per_zone *mz = mem_cgroup_zoneinfo(mem, nid, zid);
-
-	nr_pages = MEM_CGROUP_ZSTAT(mz, lru);
-
-	return (nr_pages >> priority);
-}
-
 int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg, struct zone *zone)
 {
 	unsigned long active;
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 56fc7abe4d23..66bb6ef44b5f 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1466,30 +1466,23 @@ static void shrink_zone(int priority, struct zone *zone,
 	get_scan_ratio(zone, sc, percent);
 
 	for_each_evictable_lru(l) {
-		if (scan_global_lru(sc)) {
-			int file = is_file_lru(l);
-			int scan;
+		int file = is_file_lru(l);
+		int scan;
 
-			scan = zone_page_state(zone, NR_LRU_BASE + l);
-			if (priority) {
-				scan >>= priority;
-				scan = (scan * percent[file]) / 100;
-			}
+		scan = zone_page_state(zone, NR_LRU_BASE + l);
+		if (priority) {
+			scan >>= priority;
+			scan = (scan * percent[file]) / 100;
+		}
+		if (scan_global_lru(sc)) {
 			zone->lru[l].nr_scan += scan;
 			nr[l] = zone->lru[l].nr_scan;
 			if (nr[l] >= swap_cluster_max)
 				zone->lru[l].nr_scan = 0;
 			else
 				nr[l] = 0;
-		} else {
-			/*
-			 * This reclaim occurs not because zone memory shortage
-			 * but because memory controller hits its limit.
-			 * Don't modify zone reclaim related data.
-			 */
-			nr[l] = mem_cgroup_calc_reclaim(sc->mem_cgroup, zone,
-								priority, l);
-		}
+		} else
+			nr[l] = scan;
 	}
 
 	while (nr[LRU_INACTIVE_ANON] || nr[LRU_ACTIVE_FILE] ||
-- 
cgit v1.2.3


From a7885eb8ad465ec9db99ac5b5e6680f0ca8e11c8 Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:24 -0800
Subject: memcg: swappiness

Currently, /proc/sys/vm/swappiness can change swappiness ratio for global
reclaim.  However, memcg reclaim doesn't have tuning parameter for itself.

In general, the optimal swappiness depend on workload.  (e.g.  hpc
workload need to low swappiness than the others.)

Then, per cgroup swappiness improve administrator tunability.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/controllers/memory.txt |  9 +++++
 include/linux/swap.h                 |  3 +-
 mm/memcontrol.c                      | 78 ++++++++++++++++++++++++++++++++----
 mm/vmscan.c                          |  7 ++--
 4 files changed, 86 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/controllers/memory.txt b/Documentation/controllers/memory.txt
index d71745cc2f00..e1501964df1e 100644
--- a/Documentation/controllers/memory.txt
+++ b/Documentation/controllers/memory.txt
@@ -314,6 +314,15 @@ will be charged as a new owner of it.
 	showing for better debug please see the code for meanings.
 
 
+5.3 swappiness
+  Similar to /proc/sys/vm/swappiness, but affecting a hierarchy of groups only.
+
+  Following cgroup's swapiness can't be changed.
+  - root cgroup (uses /proc/sys/vm/swappiness).
+  - a cgroup which uses hierarchy and it has child cgroup.
+  - a cgroup which uses hierarchy and not the root of hierarchy.
+
+
 6. Hierarchy support
 
 The memory controller supports a deep hierarchy and hierarchical accounting.
diff --git a/include/linux/swap.h b/include/linux/swap.h
index be938ce4895a..4ccca25d0f05 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -214,7 +214,8 @@ static inline void lru_cache_add_active_file(struct page *page)
 extern unsigned long try_to_free_pages(struct zonelist *zonelist, int order,
 					gfp_t gfp_mask);
 extern unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem,
-						gfp_t gfp_mask, bool noswap);
+						  gfp_t gfp_mask, bool noswap,
+						  unsigned int swappiness);
 extern int __isolate_lru_page(struct page *page, int mode, int file);
 extern unsigned long shrink_all_memory(unsigned long nr_pages);
 extern int vm_swappiness;
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 027c0dd7a83e..ab2ecbb95b8d 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -164,6 +164,9 @@ struct mem_cgroup {
 	int		obsolete;
 	atomic_t	refcnt;
 
+	unsigned int	swappiness;
+
+
 	unsigned int inactive_ratio;
 
 	/*
@@ -636,6 +639,22 @@ static bool mem_cgroup_check_under_limit(struct mem_cgroup *mem)
 	return false;
 }
 
+static unsigned int get_swappiness(struct mem_cgroup *memcg)
+{
+	struct cgroup *cgrp = memcg->css.cgroup;
+	unsigned int swappiness;
+
+	/* root ? */
+	if (cgrp->parent == NULL)
+		return vm_swappiness;
+
+	spin_lock(&memcg->reclaim_param_lock);
+	swappiness = memcg->swappiness;
+	spin_unlock(&memcg->reclaim_param_lock);
+
+	return swappiness;
+}
+
 /*
  * Dance down the hierarchy if needed to reclaim memory. We remember the
  * last child we reclaimed from, so that we don't end up penalizing
@@ -656,7 +675,8 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem,
 	 * but there might be left over accounting, even after children
 	 * have left.
 	 */
-	ret = try_to_free_mem_cgroup_pages(root_mem, gfp_mask, noswap);
+	ret = try_to_free_mem_cgroup_pages(root_mem, gfp_mask, noswap,
+					   get_swappiness(root_mem));
 	if (mem_cgroup_check_under_limit(root_mem))
 		return 0;
 	if (!root_mem->use_hierarchy)
@@ -672,7 +692,8 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem,
 			cgroup_unlock();
 			continue;
 		}
-		ret = try_to_free_mem_cgroup_pages(next_mem, gfp_mask, noswap);
+		ret = try_to_free_mem_cgroup_pages(next_mem, gfp_mask, noswap,
+						   get_swappiness(next_mem));
 		if (mem_cgroup_check_under_limit(root_mem))
 			return 0;
 		cgroup_lock();
@@ -1400,7 +1421,8 @@ int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask)
 	rcu_read_unlock();
 
 	do {
-		progress = try_to_free_mem_cgroup_pages(mem, gfp_mask, true);
+		progress = try_to_free_mem_cgroup_pages(mem, gfp_mask, true,
+							get_swappiness(mem));
 		progress += mem_cgroup_check_under_limit(mem);
 	} while (!progress && --retry);
 
@@ -1468,7 +1490,9 @@ static int mem_cgroup_resize_limit(struct mem_cgroup *memcg,
 			break;
 
 		progress = try_to_free_mem_cgroup_pages(memcg,
-				GFP_KERNEL, false);
+							GFP_KERNEL,
+							false,
+							get_swappiness(memcg));
   		if (!progress)			retry_count--;
 	}
 
@@ -1512,7 +1536,8 @@ int mem_cgroup_resize_memsw_limit(struct mem_cgroup *memcg,
 			break;
 
 		oldusage = res_counter_read_u64(&memcg->memsw, RES_USAGE);
-		try_to_free_mem_cgroup_pages(memcg, GFP_KERNEL, true);
+		try_to_free_mem_cgroup_pages(memcg, GFP_KERNEL, true,
+					     get_swappiness(memcg));
 		curusage = res_counter_read_u64(&memcg->memsw, RES_USAGE);
 		if (curusage >= oldusage)
 			retry_count--;
@@ -1643,8 +1668,8 @@ try_to_free:
 			ret = -EINTR;
 			goto out;
 		}
-		progress = try_to_free_mem_cgroup_pages(mem,
-						  GFP_KERNEL, false);
+		progress = try_to_free_mem_cgroup_pages(mem, GFP_KERNEL,
+						false, get_swappiness(mem));
 		if (!progress) {
 			nr_retries--;
 			/* maybe some writeback is necessary */
@@ -1864,6 +1889,37 @@ static int mem_control_stat_show(struct cgroup *cont, struct cftype *cft,
 	return 0;
 }
 
+static u64 mem_cgroup_swappiness_read(struct cgroup *cgrp, struct cftype *cft)
+{
+	struct mem_cgroup *memcg = mem_cgroup_from_cont(cgrp);
+
+	return get_swappiness(memcg);
+}
+
+static int mem_cgroup_swappiness_write(struct cgroup *cgrp, struct cftype *cft,
+				       u64 val)
+{
+	struct mem_cgroup *memcg = mem_cgroup_from_cont(cgrp);
+	struct mem_cgroup *parent;
+	if (val > 100)
+		return -EINVAL;
+
+	if (cgrp->parent == NULL)
+		return -EINVAL;
+
+	parent = mem_cgroup_from_cont(cgrp->parent);
+	/* If under hierarchy, only empty-root can set this value */
+	if ((parent->use_hierarchy) ||
+	    (memcg->use_hierarchy && !list_empty(&cgrp->children)))
+		return -EINVAL;
+
+	spin_lock(&memcg->reclaim_param_lock);
+	memcg->swappiness = val;
+	spin_unlock(&memcg->reclaim_param_lock);
+
+	return 0;
+}
+
 
 static struct cftype mem_cgroup_files[] = {
 	{
@@ -1902,6 +1958,11 @@ static struct cftype mem_cgroup_files[] = {
 		.write_u64 = mem_cgroup_hierarchy_write,
 		.read_u64 = mem_cgroup_hierarchy_read,
 	},
+	{
+		.name = "swappiness",
+		.read_u64 = mem_cgroup_swappiness_read,
+		.write_u64 = mem_cgroup_swappiness_write,
+	},
 };
 
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
@@ -2093,6 +2154,9 @@ mem_cgroup_create(struct cgroup_subsys *ss, struct cgroup *cont)
 	mem->last_scanned_child = NULL;
 	spin_lock_init(&mem->reclaim_param_lock);
 
+	if (parent)
+		mem->swappiness = get_swappiness(parent);
+
 	return &mem->css;
 free_out:
 	for_each_node_state(node, N_POSSIBLE)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index f03c239440ad..ece2f405187f 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1707,14 +1707,15 @@ unsigned long try_to_free_pages(struct zonelist *zonelist, int order,
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR
 
 unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *mem_cont,
-						gfp_t gfp_mask,
-					   bool noswap)
+					   gfp_t gfp_mask,
+					   bool noswap,
+					   unsigned int swappiness)
 {
 	struct scan_control sc = {
 		.may_writepage = !laptop_mode,
 		.may_swap = 1,
 		.swap_cluster_max = SWAP_CLUSTER_MAX,
-		.swappiness = vm_swappiness,
+		.swappiness = swappiness,
 		.order = 0,
 		.mem_cgroup = mem_cont,
 		.isolate_pages = mem_cgroup_isolate_pages,
-- 
cgit v1.2.3


From c772be939e078afd2505ede7d596a30f8f61de95 Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:25 -0800
Subject: memcg: fix calculation of active_ratio

Currently, inactive_ratio of memcg is calculated at setting limit.
because page_alloc.c does so and current implementation is straightforward
porting.

However, memcg introduced hierarchy feature recently.  In hierarchy
restriction, memory limit is not only decided memory.limit_in_bytes of
current cgroup, but also parent limit and sibling memory usage.

Then, The optimal inactive_ratio is changed frequently.  So, everytime
calculation is better.

Tested-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h |  5 ++--
 mm/memcontrol.c            | 64 ++++++++++++++++++++++------------------------
 mm/vmscan.c                |  2 +-
 3 files changed, 34 insertions(+), 37 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 8752052da8df..056cf82c0e86 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -97,8 +97,7 @@ extern void mem_cgroup_note_reclaim_priority(struct mem_cgroup *mem,
 							int priority);
 extern void mem_cgroup_record_reclaim_priority(struct mem_cgroup *mem,
 							int priority);
-int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg,
-				    struct zone *zone);
+int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg);
 unsigned long mem_cgroup_zone_nr_pages(struct mem_cgroup *memcg,
 				       struct zone *zone,
 				       enum lru_list lru);
@@ -252,7 +251,7 @@ static inline bool mem_cgroup_oom_called(struct task_struct *task)
 }
 
 static inline int
-mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg, struct zone *zone)
+mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg)
 {
 	return 1;
 }
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index ab2ecbb95b8d..c7d78eca8363 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -166,9 +166,6 @@ struct mem_cgroup {
 
 	unsigned int	swappiness;
 
-
-	unsigned int inactive_ratio;
-
 	/*
 	 * statistics. This must be placed at the end of memcg.
 	 */
@@ -432,15 +429,43 @@ void mem_cgroup_record_reclaim_priority(struct mem_cgroup *mem, int priority)
 	spin_unlock(&mem->reclaim_param_lock);
 }
 
-int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg, struct zone *zone)
+static int calc_inactive_ratio(struct mem_cgroup *memcg, unsigned long *present_pages)
 {
 	unsigned long active;
 	unsigned long inactive;
+	unsigned long gb;
+	unsigned long inactive_ratio;
 
 	inactive = mem_cgroup_get_all_zonestat(memcg, LRU_INACTIVE_ANON);
 	active = mem_cgroup_get_all_zonestat(memcg, LRU_ACTIVE_ANON);
 
-	if (inactive * memcg->inactive_ratio < active)
+	gb = (inactive + active) >> (30 - PAGE_SHIFT);
+	if (gb)
+		inactive_ratio = int_sqrt(10 * gb);
+	else
+		inactive_ratio = 1;
+
+	if (present_pages) {
+		present_pages[0] = inactive;
+		present_pages[1] = active;
+	}
+
+	return inactive_ratio;
+}
+
+int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg)
+{
+	unsigned long active;
+	unsigned long inactive;
+	unsigned long present_pages[2];
+	unsigned long inactive_ratio;
+
+	inactive_ratio = calc_inactive_ratio(memcg, present_pages);
+
+	inactive = present_pages[0];
+	active = present_pages[1];
+
+	if (inactive * inactive_ratio < active)
 		return 1;
 
 	return 0;
@@ -1432,29 +1457,6 @@ int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask)
 	return 0;
 }
 
-/*
- * The inactive anon list should be small enough that the VM never has to
- * do too much work, but large enough that each inactive page has a chance
- * to be referenced again before it is swapped out.
- *
- * this calculation is straightforward porting from
- * page_alloc.c::setup_per_zone_inactive_ratio().
- * it describe more detail.
- */
-static void mem_cgroup_set_inactive_ratio(struct mem_cgroup *memcg)
-{
-	unsigned int gb, ratio;
-
-	gb = res_counter_read_u64(&memcg->res, RES_LIMIT) >> 30;
-	if (gb)
-		ratio = int_sqrt(10 * gb);
-	else
-		ratio = 1;
-
-	memcg->inactive_ratio = ratio;
-
-}
-
 static DEFINE_MUTEX(set_limit_mutex);
 
 static int mem_cgroup_resize_limit(struct mem_cgroup *memcg,
@@ -1496,9 +1498,6 @@ static int mem_cgroup_resize_limit(struct mem_cgroup *memcg,
   		if (!progress)			retry_count--;
 	}
 
-	if (!ret)
-		mem_cgroup_set_inactive_ratio(memcg);
-
 	return ret;
 }
 
@@ -1858,7 +1857,7 @@ static int mem_control_stat_show(struct cgroup *cont, struct cftype *cft,
 	}
 
 #ifdef CONFIG_DEBUG_VM
-	cb->fill(cb, "inactive_ratio", mem_cont->inactive_ratio);
+	cb->fill(cb, "inactive_ratio", calc_inactive_ratio(mem_cont, NULL));
 
 	{
 		int nid, zid;
@@ -2150,7 +2149,6 @@ mem_cgroup_create(struct cgroup_subsys *ss, struct cgroup *cont)
 		res_counter_init(&mem->res, NULL);
 		res_counter_init(&mem->memsw, NULL);
 	}
-	mem_cgroup_set_inactive_ratio(mem);
 	mem->last_scanned_child = NULL;
 	spin_lock_init(&mem->reclaim_param_lock);
 
diff --git a/mm/vmscan.c b/mm/vmscan.c
index ece2f405187f..9a27c44aa327 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1340,7 +1340,7 @@ static int inactive_anon_is_low(struct zone *zone, struct scan_control *sc)
 	if (scanning_global_lru(sc))
 		low = inactive_anon_is_low_global(zone);
 	else
-		low = mem_cgroup_inactive_anon_is_low(sc->mem_cgroup, zone);
+		low = mem_cgroup_inactive_anon_is_low(sc->mem_cgroup);
 	return low;
 }
 
-- 
cgit v1.2.3


From a5e924f5f8abf97944e625d74967cc9452cfbce8 Mon Sep 17 00:00:00 2001
From: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Date: Wed, 7 Jan 2009 18:08:28 -0800
Subject: memcg: remove mem_cgroup_try_charge

After previous patch, mem_cgroup_try_charge is not used by anyone, so we
can remove it.

Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h |  8 --------
 mm/memcontrol.c            | 21 +--------------------
 2 files changed, 1 insertion(+), 28 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 056cf82c0e86..8ae6ece8c962 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -40,8 +40,6 @@ struct mm_struct;
 extern int mem_cgroup_newpage_charge(struct page *page, struct mm_struct *mm,
 				gfp_t gfp_mask);
 /* for swap handling */
-extern int mem_cgroup_try_charge(struct mm_struct *mm,
-		gfp_t gfp_mask, struct mem_cgroup **ptr);
 extern int mem_cgroup_try_charge_swapin(struct mm_struct *mm,
 		struct page *page, gfp_t mask, struct mem_cgroup **ptr);
 extern void mem_cgroup_commit_charge_swapin(struct page *page,
@@ -134,12 +132,6 @@ static inline int mem_cgroup_cache_charge(struct page *page,
 	return 0;
 }
 
-static inline int mem_cgroup_try_charge(struct mm_struct *mm,
-			gfp_t gfp_mask, struct mem_cgroup **ptr)
-{
-	return 0;
-}
-
 static inline int mem_cgroup_try_charge_swapin(struct mm_struct *mm,
 		struct page *page, gfp_t gfp_mask, struct mem_cgroup **ptr)
 {
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 3ba72e6556cc..435f08dac8bf 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -839,27 +839,8 @@ nomem:
 	return -ENOMEM;
 }
 
-/**
- * mem_cgroup_try_charge - get charge of PAGE_SIZE.
- * @mm: an mm_struct which is charged against. (when *memcg is NULL)
- * @gfp_mask: gfp_mask for reclaim.
- * @memcg: a pointer to memory cgroup which is charged against.
- *
- * charge against memory cgroup pointed by *memcg. if *memcg == NULL, estimated
- * memory cgroup from @mm is got and stored in *memcg.
- *
- * Returns 0 if success. -ENOMEM at failure.
- * This call can invoke OOM-Killer.
- */
-
-int mem_cgroup_try_charge(struct mm_struct *mm,
-			  gfp_t mask, struct mem_cgroup **memcg)
-{
-	return __mem_cgroup_try_charge(mm, mask, memcg, true);
-}
-
 /*
- * commit a charge got by mem_cgroup_try_charge() and makes page_cgroup to be
+ * commit a charge got by __mem_cgroup_try_charge() and makes page_cgroup to be
  * USED state. If already USED, uncharge and return.
  */
 
-- 
cgit v1.2.3


From b5a84319a4343a0db753436fd8147e61eaafa7ea Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:35 -0800
Subject: memcg: fix shmem's swap accounting

Now, you can see following even when swap accounting is enabled.

 1. Create Group 01, and 02.
 2. allocate a "file" on tmpfs by a task under 01.
 3. swap out the "file" (by memory pressure)
 4. Read "file" from a task in group 02.
 5. the charge of "file" is moved to group 02.

This is not ideal behavior. This is because SwapCache which was loaded
by read-ahead is not taken into account..

This is a patch to fix shmem's swapcache behavior.
  - remove mem_cgroup_cache_charge_swapin().
  - Add SwapCache handler routine to mem_cgroup_cache_charge().
    By this, shmem's file cache is charged at add_to_page_cache()
    with GFP_NOWAIT.
  - pass the page of swapcache to shrink_mem_cgroup.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Paul Menage <menage@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h |   6 +-
 include/linux/swap.h       |   8 ---
 mm/memcontrol.c            | 134 ++++++++++++++++++++-------------------------
 mm/shmem.c                 |  30 ++++------
 4 files changed, 76 insertions(+), 102 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 8ae6ece8c962..326f45c86530 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -56,7 +56,8 @@ extern void mem_cgroup_move_lists(struct page *page,
 				  enum lru_list from, enum lru_list to);
 extern void mem_cgroup_uncharge_page(struct page *page);
 extern void mem_cgroup_uncharge_cache_page(struct page *page);
-extern int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask);
+extern int mem_cgroup_shrink_usage(struct page *page,
+			struct mm_struct *mm, gfp_t gfp_mask);
 
 extern unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
 					struct list_head *dst,
@@ -155,7 +156,8 @@ static inline void mem_cgroup_uncharge_cache_page(struct page *page)
 {
 }
 
-static inline int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask)
+static inline int mem_cgroup_shrink_usage(struct page *page,
+			struct mm_struct *mm, gfp_t gfp_mask)
 {
 	return 0;
 }
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 4ccca25d0f05..d30215578877 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -335,16 +335,8 @@ static inline void disable_swap_token(void)
 }
 
 #ifdef CONFIG_CGROUP_MEM_RES_CTLR
-extern int mem_cgroup_cache_charge_swapin(struct page *page,
-				struct mm_struct *mm, gfp_t mask, bool locked);
 extern void mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent);
 #else
-static inline
-int mem_cgroup_cache_charge_swapin(struct page *page,
-				struct mm_struct *mm, gfp_t mask, bool locked)
-{
-	return 0;
-}
 static inline void
 mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent)
 {
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index f50cb7b1efdb..93a792871804 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -893,6 +893,23 @@ nomem:
 	return -ENOMEM;
 }
 
+static struct mem_cgroup *try_get_mem_cgroup_from_swapcache(struct page *page)
+{
+	struct mem_cgroup *mem;
+	swp_entry_t ent;
+
+	if (!PageSwapCache(page))
+		return NULL;
+
+	ent.val = page_private(page);
+	mem = lookup_swap_cgroup(ent);
+	if (!mem)
+		return NULL;
+	if (!css_tryget(&mem->css))
+		return NULL;
+	return mem;
+}
+
 /*
  * commit a charge got by __mem_cgroup_try_charge() and makes page_cgroup to be
  * USED state. If already USED, uncharge and return.
@@ -1084,6 +1101,9 @@ int mem_cgroup_newpage_charge(struct page *page,
 int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 				gfp_t gfp_mask)
 {
+	struct mem_cgroup *mem = NULL;
+	int ret;
+
 	if (mem_cgroup_disabled())
 		return 0;
 	if (PageCompound(page))
@@ -1096,6 +1116,8 @@ int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 	 * For GFP_NOWAIT case, the page may be pre-charged before calling
 	 * add_to_page_cache(). (See shmem.c) check it here and avoid to call
 	 * charge twice. (It works but has to pay a bit larger cost.)
+	 * And when the page is SwapCache, it should take swap information
+	 * into account. This is under lock_page() now.
 	 */
 	if (!(gfp_mask & __GFP_WAIT)) {
 		struct page_cgroup *pc;
@@ -1112,15 +1134,40 @@ int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 		unlock_page_cgroup(pc);
 	}
 
-	if (unlikely(!mm))
+	if (do_swap_account && PageSwapCache(page)) {
+		mem = try_get_mem_cgroup_from_swapcache(page);
+		if (mem)
+			mm = NULL;
+		  else
+			mem = NULL;
+		/* SwapCache may be still linked to LRU now. */
+		mem_cgroup_lru_del_before_commit_swapcache(page);
+	}
+
+	if (unlikely(!mm && !mem))
 		mm = &init_mm;
 
 	if (page_is_file_cache(page))
 		return mem_cgroup_charge_common(page, mm, gfp_mask,
 				MEM_CGROUP_CHARGE_TYPE_CACHE, NULL);
-	else
-		return mem_cgroup_charge_common(page, mm, gfp_mask,
-				MEM_CGROUP_CHARGE_TYPE_SHMEM, NULL);
+
+	ret = mem_cgroup_charge_common(page, mm, gfp_mask,
+				MEM_CGROUP_CHARGE_TYPE_SHMEM, mem);
+	if (mem)
+		css_put(&mem->css);
+	if (PageSwapCache(page))
+		mem_cgroup_lru_add_after_commit_swapcache(page);
+
+	if (do_swap_account && !ret && PageSwapCache(page)) {
+		swp_entry_t ent = {.val = page_private(page)};
+		/* avoid double counting */
+		mem = swap_cgroup_record(ent, NULL);
+		if (mem) {
+			res_counter_uncharge(&mem->memsw, PAGE_SIZE);
+			mem_cgroup_put(mem);
+		}
+	}
+	return ret;
 }
 
 /*
@@ -1134,7 +1181,6 @@ int mem_cgroup_try_charge_swapin(struct mm_struct *mm,
 				 gfp_t mask, struct mem_cgroup **ptr)
 {
 	struct mem_cgroup *mem;
-	swp_entry_t     ent;
 	int ret;
 
 	if (mem_cgroup_disabled())
@@ -1142,7 +1188,6 @@ int mem_cgroup_try_charge_swapin(struct mm_struct *mm,
 
 	if (!do_swap_account)
 		goto charge_cur_mm;
-
 	/*
 	 * A racing thread's fault, or swapoff, may have already updated
 	 * the pte, and even removed page from swap cache: return success
@@ -1150,14 +1195,9 @@ int mem_cgroup_try_charge_swapin(struct mm_struct *mm,
 	 */
 	if (!PageSwapCache(page))
 		return 0;
-
-	ent.val = page_private(page);
-
-	mem = lookup_swap_cgroup(ent);
+	mem = try_get_mem_cgroup_from_swapcache(page);
 	if (!mem)
 		goto charge_cur_mm;
-	if (!css_tryget(&mem->css))
-		goto charge_cur_mm;
 	*ptr = mem;
 	ret = __mem_cgroup_try_charge(NULL, mask, ptr, true);
 	/* drop extra refcnt from tryget */
@@ -1169,62 +1209,6 @@ charge_cur_mm:
 	return __mem_cgroup_try_charge(mm, mask, ptr, true);
 }
 
-#ifdef CONFIG_SWAP
-
-int mem_cgroup_cache_charge_swapin(struct page *page,
-			struct mm_struct *mm, gfp_t mask, bool locked)
-{
-	int ret = 0;
-
-	if (mem_cgroup_disabled())
-		return 0;
-	if (unlikely(!mm))
-		mm = &init_mm;
-	if (!locked)
-		lock_page(page);
-	/*
-	 * If not locked, the page can be dropped from SwapCache until
-	 * we reach here.
-	 */
-	if (PageSwapCache(page)) {
-		struct mem_cgroup *mem = NULL;
-		swp_entry_t ent;
-
-		ent.val = page_private(page);
-		if (do_swap_account) {
-			mem = lookup_swap_cgroup(ent);
-			if (mem) {
-				if (css_tryget(&mem->css))
-					mm = NULL; /* charge to recorded */
-				else
-					mem = NULL; /* charge to current */
-			}
-		}
-		/* SwapCache may be still linked to LRU now. */
-		mem_cgroup_lru_del_before_commit_swapcache(page);
-		ret = mem_cgroup_charge_common(page, mm, mask,
-				MEM_CGROUP_CHARGE_TYPE_SHMEM, mem);
-		mem_cgroup_lru_add_after_commit_swapcache(page);
-		/* drop extra refcnt from tryget */
-		if (mem)
-			css_put(&mem->css);
-
-		if (!ret && do_swap_account) {
-			/* avoid double counting */
-			mem = swap_cgroup_record(ent, NULL);
-			if (mem) {
-				res_counter_uncharge(&mem->memsw, PAGE_SIZE);
-				mem_cgroup_put(mem);
-			}
-		}
-	}
-	if (!locked)
-		unlock_page(page);
-
-	return ret;
-}
-#endif
-
 void mem_cgroup_commit_charge_swapin(struct page *page, struct mem_cgroup *ptr)
 {
 	struct page_cgroup *pc;
@@ -1486,18 +1470,20 @@ void mem_cgroup_end_migration(struct mem_cgroup *mem,
  * This is typically used for page reclaiming for shmem for reducing side
  * effect of page allocation from shmem, which is used by some mem_cgroup.
  */
-int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask)
+int mem_cgroup_shrink_usage(struct page *page,
+			    struct mm_struct *mm,
+			    gfp_t gfp_mask)
 {
-	struct mem_cgroup *mem;
+	struct mem_cgroup *mem = NULL;
 	int progress = 0;
 	int retry = MEM_CGROUP_RECLAIM_RETRIES;
 
 	if (mem_cgroup_disabled())
 		return 0;
-	if (!mm)
-		return 0;
-
-	mem = try_get_mem_cgroup_from_mm(mm);
+	if (page)
+		mem = try_get_mem_cgroup_from_swapcache(page);
+	if (!mem && mm)
+		mem = try_get_mem_cgroup_from_mm(mm);
 	if (unlikely(!mem))
 		return 0;
 
diff --git a/mm/shmem.c b/mm/shmem.c
index bbb7b043c986..5d0de96c9789 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -929,11 +929,11 @@ found:
 	if (!inode)
 		goto out;
 	/*
-	 * Charge page using GFP_HIGHUSER_MOVABLE while we can wait.
-	 * charged back to the user(not to caller) when swap account is used.
+	 * Charge page using GFP_KERNEL while we can wait.
+	 * Charged back to the user(not to caller) when swap account is used.
+	 * add_to_page_cache() will be called with GFP_NOWAIT.
 	 */
-	error = mem_cgroup_cache_charge_swapin(page, current->mm, GFP_KERNEL,
-					true);
+	error = mem_cgroup_cache_charge(page, current->mm, GFP_KERNEL);
 	if (error)
 		goto out;
 	error = radix_tree_preload(GFP_KERNEL);
@@ -1270,16 +1270,6 @@ repeat:
 				goto repeat;
 			}
 			wait_on_page_locked(swappage);
-			/*
-			 * We want to avoid charge at add_to_page_cache().
-			 * charge against this swap cache here.
-			 */
-			if (mem_cgroup_cache_charge_swapin(swappage,
-				current->mm, gfp & GFP_RECLAIM_MASK, false)) {
-				page_cache_release(swappage);
-				error = -ENOMEM;
-				goto failed;
-			}
 			page_cache_release(swappage);
 			goto repeat;
 		}
@@ -1334,15 +1324,19 @@ repeat:
 		} else {
 			shmem_swp_unmap(entry);
 			spin_unlock(&info->lock);
-			unlock_page(swappage);
-			page_cache_release(swappage);
 			if (error == -ENOMEM) {
 				/* allow reclaim from this memory cgroup */
-				error = mem_cgroup_shrink_usage(current->mm,
+				error = mem_cgroup_shrink_usage(swappage,
+								current->mm,
 								gfp);
-				if (error)
+				if (error) {
+					unlock_page(swappage);
+					page_cache_release(swappage);
 					goto failed;
+				}
 			}
+			unlock_page(swappage);
+			page_cache_release(swappage);
 			goto repeat;
 		}
 	} else if (sgp == SGP_READ && !filepage) {
-- 
cgit v1.2.3


From 999cd8a450f8f93701669a61cac4d3b19eca07e8 Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Wed, 7 Jan 2009 18:08:36 -0800
Subject: cgroups: add a per-subsystem hierarchy_mutex

These patches introduce new locking/refcount support for cgroups to
reduce the need for subsystems to call cgroup_lock(). This will
ultimately allow the atomicity of cgroup_rmdir() (which was removed
recently) to be restored.

These three patches give:

1/3 - introduce a per-subsystem hierarchy_mutex which a subsystem can
     use to prevent changes to its own cgroup tree

2/3 - use hierarchy_mutex in place of calling cgroup_lock() in the
     memory controller

3/3 - introduce a css_tryget() function similar to the one recently
      proposed by Kamezawa, but avoiding spurious refcount failures in
      the event of a race between a css_tryget() and an unsuccessful
      cgroup_rmdir()

Future patches will likely involve:

- using hierarchy mutex in place of cgroup_lock() in more subsystems
 where appropriate

- restoring the atomicity of cgroup_rmdir() with respect to cgroup_create()

This patch:

Add a hierarchy_mutex to the cgroup_subsys object that protects changes to
the hierarchy observed by that subsystem.  It is taken by the cgroup
subsystem (in addition to cgroup_mutex) for the following operations:

- linking a cgroup into that subsystem's cgroup tree
- unlinking a cgroup from that subsystem's cgroup tree
- moving the subsystem to/from a hierarchy (including across the
  bind() callback)

Thus if the subsystem holds its own hierarchy_mutex, it can safely
traverse its own hierarchy.

Signed-off-by: Paul Menage <menage@google.com>
Tested-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/cgroups/cgroups.txt |  2 +-
 include/linux/cgroup.h            | 17 ++++++++++++++++-
 kernel/cgroup.c                   | 37 +++++++++++++++++++++++++++++++++++--
 3 files changed, 52 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/cgroups/cgroups.txt b/Documentation/cgroups/cgroups.txt
index 60287e9e9d27..e33ee74eee77 100644
--- a/Documentation/cgroups/cgroups.txt
+++ b/Documentation/cgroups/cgroups.txt
@@ -528,7 +528,7 @@ example in cpusets, no task may attach before 'cpus' and 'mems' are set
 up.
 
 void bind(struct cgroup_subsys *ss, struct cgroup *root)
-(cgroup_mutex held by caller)
+(cgroup_mutex and ss->hierarchy_mutex held by caller)
 
 Called when a cgroup subsystem is rebound to a different hierarchy
 and root cgroup. Currently this will only involve movement between
diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index 73d1c730c3c4..ce1c1f34c30c 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -340,8 +340,23 @@ struct cgroup_subsys {
 #define MAX_CGROUP_TYPE_NAMELEN 32
 	const char *name;
 
-	struct cgroupfs_root *root;
+	/*
+	 * Protects sibling/children links of cgroups in this
+	 * hierarchy, plus protects which hierarchy (or none) the
+	 * subsystem is a part of (i.e. root/sibling).  To avoid
+	 * potential deadlocks, the following operations should not be
+	 * undertaken while holding any hierarchy_mutex:
+	 *
+	 * - allocating memory
+	 * - initiating hotplug events
+	 */
+	struct mutex hierarchy_mutex;
 
+	/*
+	 * Link to parent, and list entry in parent's children.
+	 * Protected by this->hierarchy_mutex and cgroup_lock()
+	 */
+	struct cgroupfs_root *root;
 	struct list_head sibling;
 };
 
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 83ea4f524be5..8b6379cdf637 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -722,23 +722,26 @@ static int rebind_subsystems(struct cgroupfs_root *root,
 			BUG_ON(cgrp->subsys[i]);
 			BUG_ON(!dummytop->subsys[i]);
 			BUG_ON(dummytop->subsys[i]->cgroup != dummytop);
+			mutex_lock(&ss->hierarchy_mutex);
 			cgrp->subsys[i] = dummytop->subsys[i];
 			cgrp->subsys[i]->cgroup = cgrp;
 			list_move(&ss->sibling, &root->subsys_list);
 			ss->root = root;
 			if (ss->bind)
 				ss->bind(ss, cgrp);
-
+			mutex_unlock(&ss->hierarchy_mutex);
 		} else if (bit & removed_bits) {
 			/* We're removing this subsystem */
 			BUG_ON(cgrp->subsys[i] != dummytop->subsys[i]);
 			BUG_ON(cgrp->subsys[i]->cgroup != cgrp);
+			mutex_lock(&ss->hierarchy_mutex);
 			if (ss->bind)
 				ss->bind(ss, dummytop);
 			dummytop->subsys[i]->cgroup = dummytop;
 			cgrp->subsys[i] = NULL;
 			subsys[i]->root = &rootnode;
 			list_move(&ss->sibling, &rootnode.subsys_list);
+			mutex_unlock(&ss->hierarchy_mutex);
 		} else if (bit & final_bits) {
 			/* Subsystem state should already exist */
 			BUG_ON(!cgrp->subsys[i]);
@@ -2338,6 +2341,29 @@ static void init_cgroup_css(struct cgroup_subsys_state *css,
 	cgrp->subsys[ss->subsys_id] = css;
 }
 
+static void cgroup_lock_hierarchy(struct cgroupfs_root *root)
+{
+	/* We need to take each hierarchy_mutex in a consistent order */
+	int i;
+
+	for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) {
+		struct cgroup_subsys *ss = subsys[i];
+		if (ss->root == root)
+			mutex_lock_nested(&ss->hierarchy_mutex, i);
+	}
+}
+
+static void cgroup_unlock_hierarchy(struct cgroupfs_root *root)
+{
+	int i;
+
+	for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) {
+		struct cgroup_subsys *ss = subsys[i];
+		if (ss->root == root)
+			mutex_unlock(&ss->hierarchy_mutex);
+	}
+}
+
 /*
  * cgroup_create - create a cgroup
  * @parent: cgroup that will be parent of the new cgroup
@@ -2386,7 +2412,9 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry,
 		init_cgroup_css(css, ss, cgrp);
 	}
 
+	cgroup_lock_hierarchy(root);
 	list_add(&cgrp->sibling, &cgrp->parent->children);
+	cgroup_unlock_hierarchy(root);
 	root->number_of_cgroups++;
 
 	err = cgroup_create_dir(cgrp, dentry, mode);
@@ -2504,8 +2532,12 @@ static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry)
 	if (!list_empty(&cgrp->release_list))
 		list_del(&cgrp->release_list);
 	spin_unlock(&release_list_lock);
-	/* delete my sibling from parent->children */
+
+	cgroup_lock_hierarchy(cgrp->root);
+	/* delete this cgroup from parent->children */
 	list_del(&cgrp->sibling);
+	cgroup_unlock_hierarchy(cgrp->root);
+
 	spin_lock(&cgrp->dentry->d_lock);
 	d = dget(cgrp->dentry);
 	spin_unlock(&d->d_lock);
@@ -2547,6 +2579,7 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss)
 	 * need to invoke fork callbacks here. */
 	BUG_ON(!list_empty(&init_task.tasks));
 
+	mutex_init(&ss->hierarchy_mutex);
 	ss->active = 1;
 }
 
-- 
cgit v1.2.3


From e7c5ec9193d32b9559a3bb8893ceedbda85201ff Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Wed, 7 Jan 2009 18:08:38 -0800
Subject: cgroups: add css_tryget()

Add css_tryget(), that obtains a counted reference on a CSS.  It is used
in situations where the caller has a "weak" reference to the CSS, i.e.
one that does not protect the cgroup from removal via a reference count,
but would instead be cleaned up by a destroy() callback.

css_tryget() will return true on success, or false if the cgroup is being
removed.

This is similar to Kamezawa Hiroyuki's patch from a week or two ago, but
with the difference that in the event of css_tryget() racing with a
cgroup_rmdir(), css_tryget() will only return false if the cgroup really
does get removed.

This implementation is done by biasing css->refcnt, so that a refcnt of 1
means "releasable" and 0 means "released or releasing".  In the event of a
race, css_tryget() distinguishes between "released" and "releasing" by
checking for the CSS_REMOVED flag in css->flags.

Signed-off-by: Paul Menage <menage@google.com>
Tested-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cgroup.h | 38 ++++++++++++++++++++++++++-----
 kernel/cgroup.c        | 61 +++++++++++++++++++++++++++++++++++++++++++++-----
 2 files changed, 88 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index ce1c1f34c30c..e267e62827bb 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -52,9 +52,9 @@ struct cgroup_subsys_state {
 	 * hierarchy structure */
 	struct cgroup *cgroup;
 
-	/* State maintained by the cgroup system to allow
-	 * subsystems to be "busy". Should be accessed via css_get()
-	 * and css_put() */
+	/* State maintained by the cgroup system to allow subsystems
+	 * to be "busy". Should be accessed via css_get(),
+	 * css_tryget() and and css_put(). */
 
 	atomic_t refcnt;
 
@@ -64,11 +64,14 @@ struct cgroup_subsys_state {
 /* bits in struct cgroup_subsys_state flags field */
 enum {
 	CSS_ROOT, /* This CSS is the root of the subsystem */
+	CSS_REMOVED, /* This CSS is dead */
 };
 
 /*
- * Call css_get() to hold a reference on the cgroup;
- *
+ * Call css_get() to hold a reference on the css; it can be used
+ * for a reference obtained via:
+ * - an existing ref-counted reference to the css
+ * - task->cgroups for a locked task
  */
 
 static inline void css_get(struct cgroup_subsys_state *css)
@@ -77,9 +80,32 @@ static inline void css_get(struct cgroup_subsys_state *css)
 	if (!test_bit(CSS_ROOT, &css->flags))
 		atomic_inc(&css->refcnt);
 }
+
+static inline bool css_is_removed(struct cgroup_subsys_state *css)
+{
+	return test_bit(CSS_REMOVED, &css->flags);
+}
+
+/*
+ * Call css_tryget() to take a reference on a css if your existing
+ * (known-valid) reference isn't already ref-counted. Returns false if
+ * the css has been destroyed.
+ */
+
+static inline bool css_tryget(struct cgroup_subsys_state *css)
+{
+	if (test_bit(CSS_ROOT, &css->flags))
+		return true;
+	while (!atomic_inc_not_zero(&css->refcnt)) {
+		if (test_bit(CSS_REMOVED, &css->flags))
+			return false;
+	}
+	return true;
+}
+
 /*
  * css_put() should be called to release a reference taken by
- * css_get()
+ * css_get() or css_tryget()
  */
 
 extern void __css_put(struct cgroup_subsys_state *css);
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 8b6379cdf637..c29831076e7a 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -2333,7 +2333,7 @@ static void init_cgroup_css(struct cgroup_subsys_state *css,
 			       struct cgroup *cgrp)
 {
 	css->cgroup = cgrp;
-	atomic_set(&css->refcnt, 0);
+	atomic_set(&css->refcnt, 1);
 	css->flags = 0;
 	if (cgrp == dummytop)
 		set_bit(CSS_ROOT, &css->flags);
@@ -2465,7 +2465,7 @@ static int cgroup_has_css_refs(struct cgroup *cgrp)
 {
 	/* Check the reference count on each subsystem. Since we
 	 * already established that there are no tasks in the
-	 * cgroup, if the css refcount is also 0, then there should
+	 * cgroup, if the css refcount is also 1, then there should
 	 * be no outstanding references, so the subsystem is safe to
 	 * destroy. We scan across all subsystems rather than using
 	 * the per-hierarchy linked list of mounted subsystems since
@@ -2486,12 +2486,62 @@ static int cgroup_has_css_refs(struct cgroup *cgrp)
 		 * matter, since it can only happen if the cgroup
 		 * has been deleted and hence no longer needs the
 		 * release agent to be called anyway. */
-		if (css && atomic_read(&css->refcnt))
+		if (css && (atomic_read(&css->refcnt) > 1))
 			return 1;
 	}
 	return 0;
 }
 
+/*
+ * Atomically mark all (or else none) of the cgroup's CSS objects as
+ * CSS_REMOVED. Return true on success, or false if the cgroup has
+ * busy subsystems. Call with cgroup_mutex held
+ */
+
+static int cgroup_clear_css_refs(struct cgroup *cgrp)
+{
+	struct cgroup_subsys *ss;
+	unsigned long flags;
+	bool failed = false;
+	local_irq_save(flags);
+	for_each_subsys(cgrp->root, ss) {
+		struct cgroup_subsys_state *css = cgrp->subsys[ss->subsys_id];
+		int refcnt;
+		do {
+			/* We can only remove a CSS with a refcnt==1 */
+			refcnt = atomic_read(&css->refcnt);
+			if (refcnt > 1) {
+				failed = true;
+				goto done;
+			}
+			BUG_ON(!refcnt);
+			/*
+			 * Drop the refcnt to 0 while we check other
+			 * subsystems. This will cause any racing
+			 * css_tryget() to spin until we set the
+			 * CSS_REMOVED bits or abort
+			 */
+		} while (atomic_cmpxchg(&css->refcnt, refcnt, 0) != refcnt);
+	}
+ done:
+	for_each_subsys(cgrp->root, ss) {
+		struct cgroup_subsys_state *css = cgrp->subsys[ss->subsys_id];
+		if (failed) {
+			/*
+			 * Restore old refcnt if we previously managed
+			 * to clear it from 1 to 0
+			 */
+			if (!atomic_read(&css->refcnt))
+				atomic_set(&css->refcnt, 1);
+		} else {
+			/* Commit the fact that the CSS is removed */
+			set_bit(CSS_REMOVED, &css->flags);
+		}
+	}
+	local_irq_restore(flags);
+	return !failed;
+}
+
 static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry)
 {
 	struct cgroup *cgrp = dentry->d_fsdata;
@@ -2522,7 +2572,7 @@ static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry)
 
 	if (atomic_read(&cgrp->count)
 	    || !list_empty(&cgrp->children)
-	    || cgroup_has_css_refs(cgrp)) {
+	    || !cgroup_clear_css_refs(cgrp)) {
 		mutex_unlock(&cgroup_mutex);
 		return -EBUSY;
 	}
@@ -3078,7 +3128,8 @@ void __css_put(struct cgroup_subsys_state *css)
 {
 	struct cgroup *cgrp = css->cgroup;
 	rcu_read_lock();
-	if (atomic_dec_and_test(&css->refcnt) && notify_on_release(cgrp)) {
+	if ((atomic_dec_return(&css->refcnt) == 1) &&
+	    notify_on_release(cgrp)) {
 		set_bit(CGRP_RELEASABLE, &cgrp->flags);
 		check_for_release(cgrp);
 	}
-- 
cgit v1.2.3


From 6af866af34a96fed24a55979a78b6f73bd4e8e87 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Wed, 7 Jan 2009 18:08:45 -0800
Subject: cpuset: remove remaining pointers to cpumask_t

Impact: cleanups, use new cpumask API

Final trivial cleanups: mainly s/cpumask_t/struct cpumask

Note there is a FIXME in generate_sched_domains(). A future patch will
change struct cpumask *doms to struct cpumask *doms[].
(I suppose Rusty will do this.)

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Mike Travis <travis@sgi.com>
Cc: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cpuset.h | 10 ++++++----
 kernel/cpuset.c        | 28 +++++++++++++++-------------
 2 files changed, 21 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index 51ea2bdea0f9..90c6074a36ca 100644
--- a/include/linux/cpuset.h
+++ b/include/linux/cpuset.h
@@ -20,8 +20,9 @@ extern int number_of_cpusets;	/* How many cpusets are defined in system? */
 extern int cpuset_init_early(void);
 extern int cpuset_init(void);
 extern void cpuset_init_smp(void);
-extern void cpuset_cpus_allowed(struct task_struct *p, cpumask_t *mask);
-extern void cpuset_cpus_allowed_locked(struct task_struct *p, cpumask_t *mask);
+extern void cpuset_cpus_allowed(struct task_struct *p, struct cpumask *mask);
+extern void cpuset_cpus_allowed_locked(struct task_struct *p,
+				       struct cpumask *mask);
 extern nodemask_t cpuset_mems_allowed(struct task_struct *p);
 #define cpuset_current_mems_allowed (current->mems_allowed)
 void cpuset_init_current_mems_allowed(void);
@@ -86,12 +87,13 @@ static inline int cpuset_init_early(void) { return 0; }
 static inline int cpuset_init(void) { return 0; }
 static inline void cpuset_init_smp(void) {}
 
-static inline void cpuset_cpus_allowed(struct task_struct *p, cpumask_t *mask)
+static inline void cpuset_cpus_allowed(struct task_struct *p,
+				       struct cpumask *mask)
 {
 	*mask = cpu_possible_map;
 }
 static inline void cpuset_cpus_allowed_locked(struct task_struct *p,
-								cpumask_t *mask)
+					      struct cpumask *mask)
 {
 	*mask = cpu_possible_map;
 }
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index fc294aa9a97a..647c77a88fcb 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -289,7 +289,8 @@ static struct file_system_type cpuset_fs_type = {
  * Call with callback_mutex held.
  */
 
-static void guarantee_online_cpus(const struct cpuset *cs, cpumask_t *pmask)
+static void guarantee_online_cpus(const struct cpuset *cs,
+				  struct cpumask *pmask)
 {
 	while (cs && !cpumask_intersects(cs->cpus_allowed, cpu_online_mask))
 		cs = cs->parent;
@@ -610,7 +611,8 @@ update_domain_attr_tree(struct sched_domain_attr *dattr, struct cpuset *c)
  *	element of the partition (one sched domain) to be passed to
  *	partition_sched_domains().
  */
-static int generate_sched_domains(cpumask_t **domains,
+/* FIXME: see the FIXME in partition_sched_domains() */
+static int generate_sched_domains(struct cpumask **domains,
 			struct sched_domain_attr **attributes)
 {
 	LIST_HEAD(q);		/* queue of cpusets to be scanned */
@@ -618,10 +620,10 @@ static int generate_sched_domains(cpumask_t **domains,
 	struct cpuset **csa;	/* array of all cpuset ptrs */
 	int csn;		/* how many cpuset ptrs in csa so far */
 	int i, j, k;		/* indices for partition finding loops */
-	cpumask_t *doms;	/* resulting partition; i.e. sched domains */
+	struct cpumask *doms;	/* resulting partition; i.e. sched domains */
 	struct sched_domain_attr *dattr;  /* attributes for custom domains */
 	int ndoms = 0;		/* number of sched domains in result */
-	int nslot;		/* next empty doms[] cpumask_t slot */
+	int nslot;		/* next empty doms[] struct cpumask slot */
 
 	doms = NULL;
 	dattr = NULL;
@@ -629,7 +631,7 @@ static int generate_sched_domains(cpumask_t **domains,
 
 	/* Special case for the 99% of systems with one, full, sched domain */
 	if (is_sched_load_balance(&top_cpuset)) {
-		doms = kmalloc(sizeof(cpumask_t), GFP_KERNEL);
+		doms = kmalloc(cpumask_size(), GFP_KERNEL);
 		if (!doms)
 			goto done;
 
@@ -708,7 +710,7 @@ restart:
 	 * Now we know how many domains to create.
 	 * Convert <csn, csa> to <ndoms, doms> and populate cpu masks.
 	 */
-	doms = kmalloc(ndoms * sizeof(cpumask_t), GFP_KERNEL);
+	doms = kmalloc(ndoms * cpumask_size(), GFP_KERNEL);
 	if (!doms)
 		goto done;
 
@@ -720,7 +722,7 @@ restart:
 
 	for (nslot = 0, i = 0; i < csn; i++) {
 		struct cpuset *a = csa[i];
-		cpumask_t *dp;
+		struct cpumask *dp;
 		int apn = a->pn;
 
 		if (apn < 0) {
@@ -743,7 +745,7 @@ restart:
 			continue;
 		}
 
-		cpus_clear(*dp);
+		cpumask_clear(dp);
 		if (dattr)
 			*(dattr + nslot) = SD_ATTR_INIT;
 		for (j = i; j < csn; j++) {
@@ -790,7 +792,7 @@ done:
 static void do_rebuild_sched_domains(struct work_struct *unused)
 {
 	struct sched_domain_attr *attr;
-	cpumask_t *doms;
+	struct cpumask *doms;
 	int ndoms;
 
 	get_online_cpus();
@@ -2044,7 +2046,7 @@ static int cpuset_track_online_cpus(struct notifier_block *unused_nb,
 				unsigned long phase, void *unused_cpu)
 {
 	struct sched_domain_attr *attr;
-	cpumask_t *doms;
+	struct cpumask *doms;
 	int ndoms;
 
 	switch (phase) {
@@ -2114,7 +2116,7 @@ void __init cpuset_init_smp(void)
 /**
  * cpuset_cpus_allowed - return cpus_allowed mask from a tasks cpuset.
  * @tsk: pointer to task_struct from which to obtain cpuset->cpus_allowed.
- * @pmask: pointer to cpumask_t variable to receive cpus_allowed set.
+ * @pmask: pointer to struct cpumask variable to receive cpus_allowed set.
  *
  * Description: Returns the cpumask_var_t cpus_allowed of the cpuset
  * attached to the specified @tsk.  Guaranteed to return some non-empty
@@ -2122,7 +2124,7 @@ void __init cpuset_init_smp(void)
  * tasks cpuset.
  **/
 
-void cpuset_cpus_allowed(struct task_struct *tsk, cpumask_t *pmask)
+void cpuset_cpus_allowed(struct task_struct *tsk, struct cpumask *pmask)
 {
 	mutex_lock(&callback_mutex);
 	cpuset_cpus_allowed_locked(tsk, pmask);
@@ -2133,7 +2135,7 @@ void cpuset_cpus_allowed(struct task_struct *tsk, cpumask_t *pmask)
  * cpuset_cpus_allowed_locked - return cpus_allowed mask from a tasks cpuset.
  * Must be called with callback_mutex held.
  **/
-void cpuset_cpus_allowed_locked(struct task_struct *tsk, cpumask_t *pmask)
+void cpuset_cpus_allowed_locked(struct task_struct *tsk, struct cpumask *pmask)
 {
 	task_lock(tsk);
 	guarantee_online_cpus(task_cs(tsk), pmask);
-- 
cgit v1.2.3


From f9fb860f67b9542cd78d1558dec7058092b57d8e Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Wed, 7 Jan 2009 18:08:46 -0800
Subject: pid: implement ns_of_pid

A current problem with the pid namespace is that it is easy to do pid
related work after exit_task_namespaces which drops the nsproxy pointer.

However if we are doing pid namespace related work we are always operating
on some struct pid which retains the pid_namespace pointer of the pid
namespace it was allocated in.

So provide ns_of_pid which allows us to find the pid namespace a pid was
allocated in.

Using this we have the needed infrastructure to do pid namespace related
work at anytime we have a struct pid, removing the chance of accidentally
having a NULL pointer dereference when accessing current->nsproxy.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: Bastian Blank <bastian@waldi.eu.org>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Nadia Derbey <Nadia.Derbey@bull.net>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/pid.h | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/pid.h b/include/linux/pid.h
index bb206c56d1f0..49f1c2f66e95 100644
--- a/include/linux/pid.h
+++ b/include/linux/pid.h
@@ -122,6 +122,24 @@ int next_pidmap(struct pid_namespace *pid_ns, int last);
 extern struct pid *alloc_pid(struct pid_namespace *ns);
 extern void free_pid(struct pid *pid);
 
+/*
+ * ns_of_pid() returns the pid namespace in which the specified pid was
+ * allocated.
+ *
+ * NOTE:
+ * 	ns_of_pid() is expected to be called for a process (task) that has
+ * 	an attached 'struct pid' (see attach_pid(), detach_pid()) i.e @pid
+ * 	is expected to be non-NULL. If @pid is NULL, caller should handle
+ * 	the resulting NULL pid-ns.
+ */
+static inline struct pid_namespace *ns_of_pid(struct pid *pid)
+{
+	struct pid_namespace *ns = NULL;
+	if (pid)
+		ns = pid->numbers[pid->level].ns;
+	return ns;
+}
+
 /*
  * the helpers to get the pid's id seen from different namespaces
  *
-- 
cgit v1.2.3


From 61bce0f1371cfff497fe85594fd39d1a0b15ebe1 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Wed, 7 Jan 2009 18:08:49 -0800
Subject: pid: generalize task_active_pid_ns

Currently task_active_pid_ns is not safe to call after a task becomes a
zombie and exit_task_namespaces is called, as nsproxy becomes NULL.  By
reading the pid namespace from the pid of the task we can trivially solve
this problem at the cost of one extra memory read in what should be the
same cacheline as we read the namespace from.

When moving things around I have made task_active_pid_ns out of line
because keeping it in pid_namespace.h would require adding includes of
pid.h and sched.h that I don't think we want.

This change does make task_active_pid_ns unsafe to call during
copy_process until we attach a pid on the task_struct which seems to be a
reasonable trade off.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: Bastian Blank <bastian@waldi.eu.org>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Nadia Derbey <Nadia.Derbey@bull.net>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/pid_namespace.h | 6 +-----
 kernel/fork.c                 | 4 ++--
 kernel/pid.c                  | 6 ++++++
 3 files changed, 9 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/pid_namespace.h b/include/linux/pid_namespace.h
index d82fe825d62f..38d10326246a 100644
--- a/include/linux/pid_namespace.h
+++ b/include/linux/pid_namespace.h
@@ -79,11 +79,7 @@ static inline void zap_pid_ns_processes(struct pid_namespace *ns)
 }
 #endif /* CONFIG_PID_NS */
 
-static inline struct pid_namespace *task_active_pid_ns(struct task_struct *tsk)
-{
-	return tsk->nsproxy->pid_ns;
-}
-
+extern struct pid_namespace *task_active_pid_ns(struct task_struct *tsk);
 void pidhash_init(void);
 void pidmap_init(void);
 
diff --git a/kernel/fork.c b/kernel/fork.c
index 7b8f2a78be3d..4018308048cf 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1126,12 +1126,12 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 
 	if (pid != &init_struct_pid) {
 		retval = -ENOMEM;
-		pid = alloc_pid(task_active_pid_ns(p));
+		pid = alloc_pid(p->nsproxy->pid_ns);
 		if (!pid)
 			goto bad_fork_cleanup_io;
 
 		if (clone_flags & CLONE_NEWPID) {
-			retval = pid_ns_prepare_proc(task_active_pid_ns(p));
+			retval = pid_ns_prepare_proc(p->nsproxy->pid_ns);
 			if (retval < 0)
 				goto bad_fork_free_pid;
 		}
diff --git a/kernel/pid.c b/kernel/pid.c
index af9224cdd6c0..1b3586fe753a 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -474,6 +474,12 @@ pid_t task_session_nr_ns(struct task_struct *tsk, struct pid_namespace *ns)
 }
 EXPORT_SYMBOL(task_session_nr_ns);
 
+struct pid_namespace *task_active_pid_ns(struct task_struct *tsk)
+{
+	return ns_of_pid(task_pid(tsk));
+}
+EXPORT_SYMBOL_GPL(task_active_pid_ns);
+
 /*
  * Used by proc to find the first pid that is greater than or equal to nr.
  *
-- 
cgit v1.2.3


From f06295b44c296c8fb08823a3118468ae343b60f2 Mon Sep 17 00:00:00 2001
From: Kees Cook <kees.cook@canonical.com>
Date: Wed, 7 Jan 2009 18:08:52 -0800
Subject: ELF: implement AT_RANDOM for glibc PRNG seeding

While discussing[1] the need for glibc to have access to random bytes
during program load, it seems that an earlier attempt to implement
AT_RANDOM got stalled.  This implements a random 16 byte string, available
to every ELF program via a new auxv AT_RANDOM vector.

[1] http://sourceware.org/ml/libc-alpha/2008-10/msg00006.html

Ulrich said:

glibc needs right after startup a bit of random data for internal
protections (stack canary etc).  What is now in upstream glibc is that we
always unconditionally open /dev/urandom, read some data, and use it.  For
every process startup.  That's slow.

...

The solution is to provide a limited amount of random data to the
starting process in the aux vector.  I suggested 16 bytes and this is
what the patch implements.  If we need only 16 bytes or less we use the
data directly.  If we need more we'll use the 16 bytes to see a PRNG.
This avoids the costly /dev/urandom use and it allows the kernel to use
the most adequate source of random data for this purpose.  It might not
be the same pool as that for /dev/urandom.

Concerns were expressed about the depletion of the randomness pool.  But
this patch doesn't make the situation worse, it doesn't deplete entropy
more than happens now.

Signed-off-by: Kees Cook <kees.cook@canonical.com>
Cc: Jakub Jelinek <jakub@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/binfmt_elf.c        | 12 ++++++++++++
 include/linux/auxvec.h |  6 +++---
 2 files changed, 15 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index c41fa2af7677..e3ff2b9e602f 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -152,8 +152,10 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 	elf_addr_t __user *sp;
 	elf_addr_t __user *u_platform;
 	elf_addr_t __user *u_base_platform;
+	elf_addr_t __user *u_rand_bytes;
 	const char *k_platform = ELF_PLATFORM;
 	const char *k_base_platform = ELF_BASE_PLATFORM;
+	unsigned char k_rand_bytes[16];
 	int items;
 	elf_addr_t *elf_info;
 	int ei_index = 0;
@@ -196,6 +198,15 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 			return -EFAULT;
 	}
 
+	/*
+	 * Generate 16 random bytes for userspace PRNG seeding.
+	 */
+	get_random_bytes(k_rand_bytes, sizeof(k_rand_bytes));
+	u_rand_bytes = (elf_addr_t __user *)
+		       STACK_ALLOC(p, sizeof(k_rand_bytes));
+	if (__copy_to_user(u_rand_bytes, k_rand_bytes, sizeof(k_rand_bytes)))
+		return -EFAULT;
+
 	/* Create the ELF interpreter info */
 	elf_info = (elf_addr_t *)current->mm->saved_auxv;
 	/* update AT_VECTOR_SIZE_BASE if the number of NEW_AUX_ENT() changes */
@@ -228,6 +239,7 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 	NEW_AUX_ENT(AT_GID, cred->gid);
 	NEW_AUX_ENT(AT_EGID, cred->egid);
  	NEW_AUX_ENT(AT_SECURE, security_bprm_secureexec(bprm));
+	NEW_AUX_ENT(AT_RANDOM, (elf_addr_t)(unsigned long)u_rand_bytes);
 	NEW_AUX_ENT(AT_EXECFN, bprm->exec);
 	if (k_platform) {
 		NEW_AUX_ENT(AT_PLATFORM,
diff --git a/include/linux/auxvec.h b/include/linux/auxvec.h
index d7afa9dd6635..f3b5d4e3a2ac 100644
--- a/include/linux/auxvec.h
+++ b/include/linux/auxvec.h
@@ -23,16 +23,16 @@
 #define AT_PLATFORM 15  /* string identifying CPU for optimizations */
 #define AT_HWCAP  16    /* arch dependent hints at CPU capabilities */
 #define AT_CLKTCK 17	/* frequency at which times() increments */
-
+/* AT_* values 18 through 22 are reserved */
 #define AT_SECURE 23   /* secure mode boolean */
-
 #define AT_BASE_PLATFORM 24	/* string identifying real platform, may
 				 * differ from AT_PLATFORM. */
+#define AT_RANDOM 25	/* address of 16 random bytes */
 
 #define AT_EXECFN  31	/* filename of program */
 
 #ifdef __KERNEL__
-#define AT_VECTOR_SIZE_BASE 18 /* NEW_AUX_ENT entries in auxiliary table */
+#define AT_VECTOR_SIZE_BASE 19 /* NEW_AUX_ENT entries in auxiliary table */
   /* number of "#define AT_.*" above, minus {AT_NULL, AT_IGNORE, AT_NOTELF} */
 #endif
 
-- 
cgit v1.2.3


From 91f68b7359144aa40bb9668124543d15284750b4 Mon Sep 17 00:00:00 2001
From: Wu Fengguang <fengguang.wu@intel.com>
Date: Wed, 7 Jan 2009 18:09:12 -0800
Subject: generic swap(): introduce global macro swap(a, b)

There have been some local definitions of swap(), it's time to replace
them all with a uniform one.

Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/kernel.h | 6 ++++++
 1 file changed, 6 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 6b8e2027165e..343df9ef2412 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -476,6 +476,12 @@ static inline char *pack_hex_byte(char *buf, u8 byte)
 	__val = __val < __min ? __min: __val;	\
 	__val > __max ? __max: __val; })
 
+
+/*
+ * swap - swap value of @a and @b
+ */
+#define swap(a, b) ({ typeof(a) __tmp = (a); (a) = (b); (b) = __tmp; })
+
 /**
  * container_of - cast a member of a structure out to the containing structure
  * @ptr:	the pointer to the member.
-- 
cgit v1.2.3


From 859cb7f2a4244ea6da206d3fe9cc8a6810947a68 Mon Sep 17 00:00:00 2001
From: Richard Purdie <rpurdie@linux.intel.com>
Date: Thu, 8 Jan 2009 17:55:03 +0000
Subject: leds: Add suspend/resume to the core class

Add suspend/resume to the core class and remove all the now unneeded
code from various drivers. Originally the class code couldn't support
suspend/resume but since class_device can there is no reason for
each driver doing its own suspend/resume anymore.
---
 drivers/leds/led-class.c       | 22 ++++++++++++++++++++++
 drivers/leds/leds-alix2.c      | 30 +-----------------------------
 drivers/leds/leds-ams-delta.c  | 29 +----------------------------
 drivers/leds/leds-clevo-mail.c | 21 +--------------------
 drivers/leds/leds-fsg.c        | 37 ++++++-------------------------------
 drivers/leds/leds-gpio.c       | 36 +-----------------------------------
 drivers/leds/leds-hp-disk.c    | 20 +-------------------
 drivers/leds/leds-hp6xx.c      | 22 ++--------------------
 drivers/leds/leds-net48xx.c    | 21 +--------------------
 drivers/leds/leds-s3c24xx.c    | 25 +------------------------
 drivers/leds/leds-wm8350.c     | 24 +-----------------------
 drivers/leds/leds-wrap.c       | 27 +++------------------------
 include/linux/leds.h           |  3 +++
 13 files changed, 44 insertions(+), 273 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/leds/led-class.c b/drivers/leds/led-class.c
index 1553d93b8c65..52f82e3ea13a 100644
--- a/drivers/leds/led-class.c
+++ b/drivers/leds/led-class.c
@@ -91,6 +91,26 @@ void led_classdev_resume(struct led_classdev *led_cdev)
 }
 EXPORT_SYMBOL_GPL(led_classdev_resume);
 
+static int led_suspend(struct device *dev, pm_message_t state)
+{
+	struct led_classdev *led_cdev = dev_get_drvdata(dev);
+
+	if (led_cdev->flags & LED_CORE_SUSPENDRESUME)
+		led_classdev_suspend(led_cdev);
+
+	return 0;
+}
+
+static int led_resume(struct device *dev)
+{
+	struct led_classdev *led_cdev = dev_get_drvdata(dev);
+
+	if (led_cdev->flags & LED_CORE_SUSPENDRESUME)
+		led_classdev_resume(led_cdev);
+
+	return 0;
+}
+
 /**
  * led_classdev_register - register a new object of led_classdev class.
  * @parent: The device to register.
@@ -174,6 +194,8 @@ static int __init leds_init(void)
 	leds_class = class_create(THIS_MODULE, "leds");
 	if (IS_ERR(leds_class))
 		return PTR_ERR(leds_class);
+	leds_class->suspend = led_suspend;
+	leds_class->resume = led_resume;
 	return 0;
 }
 
diff --git a/drivers/leds/leds-alix2.c b/drivers/leds/leds-alix2.c
index d7666e6cb425..ddbd7730dfc8 100644
--- a/drivers/leds/leds-alix2.c
+++ b/drivers/leds/leds-alix2.c
@@ -65,39 +65,13 @@ static struct alix_led alix_leds[] = {
 	},
 };
 
-#ifdef CONFIG_PM
-
-static int alix_led_suspend(struct platform_device *dev, pm_message_t state)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(alix_leds); i++)
-		led_classdev_suspend(&alix_leds[i].cdev);
-	return 0;
-}
-
-static int alix_led_resume(struct platform_device *dev)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(alix_leds); i++)
-		led_classdev_resume(&alix_leds[i].cdev);
-	return 0;
-}
-
-#else
-
-#define alix_led_suspend NULL
-#define alix_led_resume NULL
-
-#endif
-
 static int __init alix_led_probe(struct platform_device *pdev)
 {
 	int i;
 	int ret;
 
 	for (i = 0; i < ARRAY_SIZE(alix_leds); i++) {
+		alix_leds[i].cdev.flags |= LED_CORE_SUSPENDRESUME;
 		ret = led_classdev_register(&pdev->dev, &alix_leds[i].cdev);
 		if (ret < 0)
 			goto fail;
@@ -121,8 +95,6 @@ static int alix_led_remove(struct platform_device *pdev)
 
 static struct platform_driver alix_led_driver = {
 	.remove = alix_led_remove,
-	.suspend = alix_led_suspend,
-	.resume = alix_led_resume,
 	.driver = {
 		.name = KBUILD_MODNAME,
 		.owner = THIS_MODULE,
diff --git a/drivers/leds/leds-ams-delta.c b/drivers/leds/leds-ams-delta.c
index 072157c3ba72..446050759b4d 100644
--- a/drivers/leds/leds-ams-delta.c
+++ b/drivers/leds/leds-ams-delta.c
@@ -79,37 +79,12 @@ static struct ams_delta_led ams_delta_leds[] = {
 	},
 };
 
-#ifdef CONFIG_PM
-static int ams_delta_led_suspend(struct platform_device *dev,
-		pm_message_t state)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(ams_delta_leds); i++)
-		led_classdev_suspend(&ams_delta_leds[i].cdev);
-
-	return 0;
-}
-
-static int ams_delta_led_resume(struct platform_device *dev)
-{
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(ams_delta_leds); i++)
-		led_classdev_resume(&ams_delta_leds[i].cdev);
-
-	return 0;
-}
-#else
-#define ams_delta_led_suspend NULL
-#define ams_delta_led_resume NULL
-#endif
-
 static int ams_delta_led_probe(struct platform_device *pdev)
 {
 	int i, ret;
 
 	for (i = 0; i < ARRAY_SIZE(ams_delta_leds); i++) {
+		ams_delta_leds[i].cdev.flags |= LED_CORE_SUSPENDRESUME;
 		ret = led_classdev_register(&pdev->dev,
 				&ams_delta_leds[i].cdev);
 		if (ret < 0)
@@ -136,8 +111,6 @@ static int ams_delta_led_remove(struct platform_device *pdev)
 static struct platform_driver ams_delta_led_driver = {
 	.probe		= ams_delta_led_probe,
 	.remove		= ams_delta_led_remove,
-	.suspend	= ams_delta_led_suspend,
-	.resume		= ams_delta_led_resume,
 	.driver		= {
 		.name = "ams-delta-led",
 		.owner = THIS_MODULE,
diff --git a/drivers/leds/leds-clevo-mail.c b/drivers/leds/leds-clevo-mail.c
index eb3415e88f43..1813c84ea5fc 100644
--- a/drivers/leds/leds-clevo-mail.c
+++ b/drivers/leds/leds-clevo-mail.c
@@ -142,6 +142,7 @@ static struct led_classdev clevo_mail_led = {
 	.name			= "clevo::mail",
 	.brightness_set		= clevo_mail_led_set,
 	.blink_set		= clevo_mail_led_blink,
+	.flags			= LED_CORE_SUSPENDRESUME,
 };
 
 static int __init clevo_mail_led_probe(struct platform_device *pdev)
@@ -155,29 +156,9 @@ static int clevo_mail_led_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int clevo_mail_led_suspend(struct platform_device *dev,
-				  pm_message_t state)
-{
-	led_classdev_suspend(&clevo_mail_led);
-	return 0;
-}
-
-static int clevo_mail_led_resume(struct platform_device *dev)
-{
-	led_classdev_resume(&clevo_mail_led);
-	return 0;
-}
-#else
-#define clevo_mail_led_suspend    NULL
-#define clevo_mail_led_resume     NULL
-#endif
-
 static struct platform_driver clevo_mail_led_driver = {
 	.probe		= clevo_mail_led_probe,
 	.remove		= clevo_mail_led_remove,
-	.suspend	= clevo_mail_led_suspend,
-	.resume		= clevo_mail_led_resume,
 	.driver		= {
 		.name		= KBUILD_MODNAME,
 		.owner		= THIS_MODULE,
diff --git a/drivers/leds/leds-fsg.c b/drivers/leds/leds-fsg.c
index 34935155c1c0..5f7c9c5c09b1 100644
--- a/drivers/leds/leds-fsg.c
+++ b/drivers/leds/leds-fsg.c
@@ -99,64 +99,43 @@ static void fsg_led_ring_set(struct led_classdev *led_cdev,
 }
 
 
-
 static struct led_classdev fsg_wlan_led = {
 	.name			= "fsg:blue:wlan",
 	.brightness_set		= fsg_led_wlan_set,
+	.flags			= LED_CORE_SUSPENDRESUME,
 };
 
 static struct led_classdev fsg_wan_led = {
 	.name			= "fsg:blue:wan",
 	.brightness_set		= fsg_led_wan_set,
+	.flags			= LED_CORE_SUSPENDRESUME,
 };
 
 static struct led_classdev fsg_sata_led = {
 	.name			= "fsg:blue:sata",
 	.brightness_set		= fsg_led_sata_set,
+	.flags			= LED_CORE_SUSPENDRESUME,
 };
 
 static struct led_classdev fsg_usb_led = {
 	.name			= "fsg:blue:usb",
 	.brightness_set		= fsg_led_usb_set,
+	.flags			= LED_CORE_SUSPENDRESUME,
 };
 
 static struct led_classdev fsg_sync_led = {
 	.name			= "fsg:blue:sync",
 	.brightness_set		= fsg_led_sync_set,
+	.flags			= LED_CORE_SUSPENDRESUME,
 };
 
 static struct led_classdev fsg_ring_led = {
 	.name			= "fsg:blue:ring",
 	.brightness_set		= fsg_led_ring_set,
+	.flags			= LED_CORE_SUSPENDRESUME,
 };
 
 
-
-#ifdef CONFIG_PM
-static int fsg_led_suspend(struct platform_device *dev, pm_message_t state)
-{
-	led_classdev_suspend(&fsg_wlan_led);
-	led_classdev_suspend(&fsg_wan_led);
-	led_classdev_suspend(&fsg_sata_led);
-	led_classdev_suspend(&fsg_usb_led);
-	led_classdev_suspend(&fsg_sync_led);
-	led_classdev_suspend(&fsg_ring_led);
-	return 0;
-}
-
-static int fsg_led_resume(struct platform_device *dev)
-{
-	led_classdev_resume(&fsg_wlan_led);
-	led_classdev_resume(&fsg_wan_led);
-	led_classdev_resume(&fsg_sata_led);
-	led_classdev_resume(&fsg_usb_led);
-	led_classdev_resume(&fsg_sync_led);
-	led_classdev_resume(&fsg_ring_led);
-	return 0;
-}
-#endif
-
-
 static int fsg_led_probe(struct platform_device *pdev)
 {
 	int ret;
@@ -232,10 +211,6 @@ static int fsg_led_remove(struct platform_device *pdev)
 static struct platform_driver fsg_led_driver = {
 	.probe		= fsg_led_probe,
 	.remove		= fsg_led_remove,
-#ifdef CONFIG_PM
-	.suspend	= fsg_led_suspend,
-	.resume		= fsg_led_resume,
-#endif
 	.driver		= {
 		.name		= "fsg-led",
 	},
diff --git a/drivers/leds/leds-gpio.c b/drivers/leds/leds-gpio.c
index b13bd2950e95..2e3df08b649b 100644
--- a/drivers/leds/leds-gpio.c
+++ b/drivers/leds/leds-gpio.c
@@ -105,6 +105,7 @@ static int gpio_led_probe(struct platform_device *pdev)
 		}
 		led_dat->cdev.brightness_set = gpio_led_set;
 		led_dat->cdev.brightness = LED_OFF;
+		led_dat->cdev.flags |= LED_CORE_SUSPENDRESUME;
 
 		gpio_direction_output(led_dat->gpio, led_dat->active_low);
 
@@ -154,44 +155,9 @@ static int __devexit gpio_led_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int gpio_led_suspend(struct platform_device *pdev, pm_message_t state)
-{
-	struct gpio_led_platform_data *pdata = pdev->dev.platform_data;
-	struct gpio_led_data *leds_data;
-	int i;
-
-	leds_data = platform_get_drvdata(pdev);
-
-	for (i = 0; i < pdata->num_leds; i++)
-		led_classdev_suspend(&leds_data[i].cdev);
-
-	return 0;
-}
-
-static int gpio_led_resume(struct platform_device *pdev)
-{
-	struct gpio_led_platform_data *pdata = pdev->dev.platform_data;
-	struct gpio_led_data *leds_data;
-	int i;
-
-	leds_data = platform_get_drvdata(pdev);
-
-	for (i = 0; i < pdata->num_leds; i++)
-		led_classdev_resume(&leds_data[i].cdev);
-
-	return 0;
-}
-#else
-#define gpio_led_suspend NULL
-#define gpio_led_resume NULL
-#endif
-
 static struct platform_driver gpio_led_driver = {
 	.probe		= gpio_led_probe,
 	.remove		= __devexit_p(gpio_led_remove),
-	.suspend	= gpio_led_suspend,
-	.resume		= gpio_led_resume,
 	.driver		= {
 		.name	= "leds-gpio",
 		.owner	= THIS_MODULE,
diff --git a/drivers/leds/leds-hp-disk.c b/drivers/leds/leds-hp-disk.c
index 44fa757d8254..d786adc8c5e3 100644
--- a/drivers/leds/leds-hp-disk.c
+++ b/drivers/leds/leds-hp-disk.c
@@ -68,25 +68,9 @@ static struct led_classdev hpled_led = {
 	.name			= "hp:red:hddprotection",
 	.default_trigger	= "heartbeat",
 	.brightness_set		= hpled_set,
+	.flags			= LED_CORE_SUSPENDRESUME,
 };
 
-#ifdef CONFIG_PM
-static int hpled_suspend(struct acpi_device *dev, pm_message_t state)
-{
-	led_classdev_suspend(&hpled_led);
-	return 0;
-}
-
-static int hpled_resume(struct acpi_device *dev)
-{
-	led_classdev_resume(&hpled_led);
-	return 0;
-}
-#else
-#define hpled_suspend NULL
-#define hpled_resume NULL
-#endif
-
 static int hpled_add(struct acpi_device *device)
 {
 	int ret;
@@ -121,8 +105,6 @@ static struct acpi_driver leds_hp_driver = {
 	.ops = {
 		.add     = hpled_add,
 		.remove  = hpled_remove,
-		.suspend = hpled_suspend,
-		.resume  = hpled_resume,
 	}
 };
 
diff --git a/drivers/leds/leds-hp6xx.c b/drivers/leds/leds-hp6xx.c
index e8fb1baf8a50..e4ce1fd46338 100644
--- a/drivers/leds/leds-hp6xx.c
+++ b/drivers/leds/leds-hp6xx.c
@@ -45,30 +45,16 @@ static struct led_classdev hp6xx_red_led = {
 	.name			= "hp6xx:red",
 	.default_trigger	= "hp6xx-charge",
 	.brightness_set		= hp6xxled_red_set,
+	.flags			= LED_CORE_SUSPENDRESUME,
 };
 
 static struct led_classdev hp6xx_green_led = {
 	.name			= "hp6xx:green",
 	.default_trigger	= "ide-disk",
 	.brightness_set		= hp6xxled_green_set,
+	.flags			= LED_CORE_SUSPENDRESUME,
 };
 
-#ifdef CONFIG_PM
-static int hp6xxled_suspend(struct platform_device *dev, pm_message_t state)
-{
-	led_classdev_suspend(&hp6xx_red_led);
-	led_classdev_suspend(&hp6xx_green_led);
-	return 0;
-}
-
-static int hp6xxled_resume(struct platform_device *dev)
-{
-	led_classdev_resume(&hp6xx_red_led);
-	led_classdev_resume(&hp6xx_green_led);
-	return 0;
-}
-#endif
-
 static int hp6xxled_probe(struct platform_device *pdev)
 {
 	int ret;
@@ -98,10 +84,6 @@ MODULE_ALIAS("platform:hp6xx-led");
 static struct platform_driver hp6xxled_driver = {
 	.probe		= hp6xxled_probe,
 	.remove		= hp6xxled_remove,
-#ifdef CONFIG_PM
-	.suspend	= hp6xxled_suspend,
-	.resume		= hp6xxled_resume,
-#endif
 	.driver		= {
 		.name		= "hp6xx-led",
 		.owner		= THIS_MODULE,
diff --git a/drivers/leds/leds-net48xx.c b/drivers/leds/leds-net48xx.c
index 054360473c94..93987a12da49 100644
--- a/drivers/leds/leds-net48xx.c
+++ b/drivers/leds/leds-net48xx.c
@@ -33,26 +33,9 @@ static void net48xx_error_led_set(struct led_classdev *led_cdev,
 static struct led_classdev net48xx_error_led = {
 	.name		= "net48xx::error",
 	.brightness_set	= net48xx_error_led_set,
+	.flags		= LED_CORE_SUSPENDRESUME,
 };
 
-#ifdef CONFIG_PM
-static int net48xx_led_suspend(struct platform_device *dev,
-		pm_message_t state)
-{
-	led_classdev_suspend(&net48xx_error_led);
-	return 0;
-}
-
-static int net48xx_led_resume(struct platform_device *dev)
-{
-	led_classdev_resume(&net48xx_error_led);
-	return 0;
-}
-#else
-#define net48xx_led_suspend NULL
-#define net48xx_led_resume NULL
-#endif
-
 static int net48xx_led_probe(struct platform_device *pdev)
 {
 	return led_classdev_register(&pdev->dev, &net48xx_error_led);
@@ -67,8 +50,6 @@ static int net48xx_led_remove(struct platform_device *pdev)
 static struct platform_driver net48xx_led_driver = {
 	.probe		= net48xx_led_probe,
 	.remove		= net48xx_led_remove,
-	.suspend	= net48xx_led_suspend,
-	.resume		= net48xx_led_resume,
 	.driver		= {
 		.name		= DRVNAME,
 		.owner		= THIS_MODULE,
diff --git a/drivers/leds/leds-s3c24xx.c b/drivers/leds/leds-s3c24xx.c
index 25a07f2643ad..4d81131542ae 100644
--- a/drivers/leds/leds-s3c24xx.c
+++ b/drivers/leds/leds-s3c24xx.c
@@ -82,6 +82,7 @@ static int s3c24xx_led_probe(struct platform_device *dev)
 	led->cdev.brightness_set = s3c24xx_led_set;
 	led->cdev.default_trigger = pdata->def_trigger;
 	led->cdev.name = pdata->name;
+	led->cdev.flags |= LED_CORE_SUSPENDRESUME;
 
 	led->pdata = pdata;
 
@@ -111,33 +112,9 @@ static int s3c24xx_led_probe(struct platform_device *dev)
 	return ret;
 }
 
-
-#ifdef CONFIG_PM
-static int s3c24xx_led_suspend(struct platform_device *dev, pm_message_t state)
-{
-	struct s3c24xx_gpio_led *led = pdev_to_gpio(dev);
-
-	led_classdev_suspend(&led->cdev);
-	return 0;
-}
-
-static int s3c24xx_led_resume(struct platform_device *dev)
-{
-	struct s3c24xx_gpio_led *led = pdev_to_gpio(dev);
-
-	led_classdev_resume(&led->cdev);
-	return 0;
-}
-#else
-#define s3c24xx_led_suspend NULL
-#define s3c24xx_led_resume NULL
-#endif
-
 static struct platform_driver s3c24xx_led_driver = {
 	.probe		= s3c24xx_led_probe,
 	.remove		= s3c24xx_led_remove,
-	.suspend	= s3c24xx_led_suspend,
-	.resume		= s3c24xx_led_resume,
 	.driver		= {
 		.name		= "s3c24xx_led",
 		.owner		= THIS_MODULE,
diff --git a/drivers/leds/leds-wm8350.c b/drivers/leds/leds-wm8350.c
index 846ed978103f..38c6bcb07e6c 100644
--- a/drivers/leds/leds-wm8350.c
+++ b/drivers/leds/leds-wm8350.c
@@ -184,27 +184,6 @@ static void wm8350_led_set(struct led_classdev *led_cdev,
 	spin_unlock_irqrestore(&led->value_lock, flags);
 }
 
-#ifdef CONFIG_PM
-static int wm8350_led_suspend(struct platform_device *pdev, pm_message_t state)
-{
-	struct wm8350_led *led = platform_get_drvdata(pdev);
-
-	led_classdev_suspend(&led->cdev);
-	return 0;
-}
-
-static int wm8350_led_resume(struct platform_device *pdev)
-{
-	struct wm8350_led *led = platform_get_drvdata(pdev);
-
-	led_classdev_resume(&led->cdev);
-	return 0;
-}
-#else
-#define wm8350_led_suspend NULL
-#define wm8350_led_resume NULL
-#endif
-
 static void wm8350_led_shutdown(struct platform_device *pdev)
 {
 	struct wm8350_led *led = platform_get_drvdata(pdev);
@@ -255,6 +234,7 @@ static int wm8350_led_probe(struct platform_device *pdev)
 	led->cdev.brightness_set = wm8350_led_set;
 	led->cdev.default_trigger = pdata->default_trigger;
 	led->cdev.name = pdata->name;
+	led->cdev.flags |= LED_CORE_SUSPENDRESUME;
 	led->enabled = regulator_is_enabled(isink);
 	led->isink = isink;
 	led->dcdc = dcdc;
@@ -311,8 +291,6 @@ static struct platform_driver wm8350_led_driver = {
 	.probe = wm8350_led_probe,
 	.remove = wm8350_led_remove,
 	.shutdown = wm8350_led_shutdown,
-	.suspend = wm8350_led_suspend,
-	.resume = wm8350_led_resume,
 };
 
 static int __devinit wm8350_led_init(void)
diff --git a/drivers/leds/leds-wrap.c b/drivers/leds/leds-wrap.c
index 2f3aa87f2a1f..2982c86ac4cf 100644
--- a/drivers/leds/leds-wrap.c
+++ b/drivers/leds/leds-wrap.c
@@ -56,40 +56,21 @@ static struct led_classdev wrap_power_led = {
 	.name			= "wrap::power",
 	.brightness_set		= wrap_power_led_set,
 	.default_trigger	= "default-on",
+	.flags			= LED_CORE_SUSPENDRESUME,
 };
 
 static struct led_classdev wrap_error_led = {
 	.name		= "wrap::error",
 	.brightness_set	= wrap_error_led_set,
+	.flags			= LED_CORE_SUSPENDRESUME,
 };
 
 static struct led_classdev wrap_extra_led = {
 	.name           = "wrap::extra",
 	.brightness_set = wrap_extra_led_set,
+	.flags			= LED_CORE_SUSPENDRESUME,
 };
 
-#ifdef CONFIG_PM
-static int wrap_led_suspend(struct platform_device *dev,
-		pm_message_t state)
-{
-	led_classdev_suspend(&wrap_power_led);
-	led_classdev_suspend(&wrap_error_led);
-	led_classdev_suspend(&wrap_extra_led);
-	return 0;
-}
-
-static int wrap_led_resume(struct platform_device *dev)
-{
-	led_classdev_resume(&wrap_power_led);
-	led_classdev_resume(&wrap_error_led);
-	led_classdev_resume(&wrap_extra_led);
-	return 0;
-}
-#else
-#define wrap_led_suspend NULL
-#define wrap_led_resume NULL
-#endif
-
 static int wrap_led_probe(struct platform_device *pdev)
 {
 	int ret;
@@ -127,8 +108,6 @@ static int wrap_led_remove(struct platform_device *pdev)
 static struct platform_driver wrap_led_driver = {
 	.probe		= wrap_led_probe,
 	.remove		= wrap_led_remove,
-	.suspend	= wrap_led_suspend,
-	.resume		= wrap_led_resume,
 	.driver		= {
 		.name		= DRVNAME,
 		.owner		= THIS_MODULE,
diff --git a/include/linux/leds.h b/include/linux/leds.h
index 3c1a8ce6a5ea..24489da701e3 100644
--- a/include/linux/leds.h
+++ b/include/linux/leds.h
@@ -32,7 +32,10 @@ struct led_classdev {
 	int			 brightness;
 	int			 flags;
 
+	/* Lower 16 bits reflect status */
 #define LED_SUSPENDED		(1 << 0)
+	/* Upper 16 bits reflect control information */
+#define LED_CORE_SUSPENDRESUME	(1 << 16)
 
 	/* Set LED brightness level */
 	/* Must not sleep, use a workqueue if needed */
-- 
cgit v1.2.3


From 69279fb9a95051971ac03e558c4d46e7ba84ab3a Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Wed, 31 Dec 2008 12:52:41 +0000
Subject: regulator: Clean up kerneldoc warnings

Remove kerneldoc warnings that don't relate to missing documentation,
mostly by renaming parameters in the documentation to match their
actual names.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
---
 drivers/regulator/core.c           | 44 +++++++++++++++++++-------------------
 include/linux/regulator/consumer.h |  8 +++----
 2 files changed, 26 insertions(+), 26 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c
index 7ba9491a05cb..ea12c68c327f 100644
--- a/drivers/regulator/core.c
+++ b/drivers/regulator/core.c
@@ -655,7 +655,7 @@ static void print_constraints(struct regulator_dev *rdev)
 
 /**
  * set_machine_constraints - sets regulator constraints
- * @regulator: regulator source
+ * @rdev: regulator source
  *
  * Allows platform initialisation code to define and constrain
  * regulator circuits e.g. valid voltage/current ranges, etc.  NOTE:
@@ -730,8 +730,8 @@ out:
 
 /**
  * set_supply - set regulator supply regulator
- * @regulator: regulator name
- * @supply: supply regulator name
+ * @rdev: regulator name
+ * @supply_rdev: supply regulator name
  *
  * Called by platform initialisation code to set the supply regulator for this
  * regulator. This ensures that a regulators supply will also be enabled by the
@@ -758,9 +758,9 @@ out:
 
 /**
  * set_consumer_device_supply: Bind a regulator to a symbolic supply
- * @regulator: regulator source
- * @dev:       device the supply applies to
- * @supply:    symbolic name for supply
+ * @rdev:         regulator source
+ * @consumer_dev: device the supply applies to
+ * @supply:       symbolic name for supply
  *
  * Allows platform initialisation code to map physical regulator
  * sources to symbolic names for supplies for use by devices.  Devices
@@ -1013,9 +1013,8 @@ static int _regulator_enable(struct regulator_dev *rdev)
  *
  * Enable the regulator output at the predefined voltage or current value.
  * NOTE: the output value can be set by other drivers, boot loader or may be
- * hardwired in the regulator.
- * NOTE: calls to regulator_enable() must be balanced with calls to
- * regulator_disable().
+ * hardwired in the regulator.  Calls to regulator_enable() must be balanced
+ * with calls to regulator_disable().
  */
 int regulator_enable(struct regulator *regulator)
 {
@@ -1074,10 +1073,10 @@ static int _regulator_disable(struct regulator_dev *rdev)
  * @regulator: regulator source
  *
  * Disable the regulator output voltage or current.
+ *
  * NOTE: this will only disable the regulator output if no other consumer
- * devices have it enabled.
- * NOTE: calls to regulator_enable() must be balanced with calls to
- * regulator_disable().
+ * devices have it enabled.  Calls to regulator_enable() must be balanced with
+ * calls to regulator_disable().
  */
 int regulator_disable(struct regulator *regulator)
 {
@@ -1200,7 +1199,7 @@ EXPORT_SYMBOL_GPL(regulator_is_enabled);
  *
  * NOTE: If the regulator is shared between several devices then the lowest
  * request voltage that meets the system constraints will be used.
- * NOTE: Regulator system constraints must be set for this regulator before
+ * Regulator system constraints must be set for this regulator before
  * calling this function otherwise this call will fail.
  */
 int regulator_set_voltage(struct regulator *regulator, int min_uV, int max_uV)
@@ -1498,7 +1497,7 @@ EXPORT_SYMBOL_GPL(regulator_set_optimum_mode);
 /**
  * regulator_register_notifier - register regulator event notifier
  * @regulator: regulator source
- * @notifier_block: notifier block
+ * @nb: notifier block
  *
  * Register notifier block to receive regulator events.
  */
@@ -1513,7 +1512,7 @@ EXPORT_SYMBOL_GPL(regulator_register_notifier);
 /**
  * regulator_unregister_notifier - unregister regulator event notifier
  * @regulator: regulator source
- * @notifier_block: notifier block
+ * @nb: notifier block
  *
  * Unregister regulator event notifier block.
  */
@@ -1679,9 +1678,9 @@ EXPORT_SYMBOL_GPL(regulator_bulk_free);
 
 /**
  * regulator_notifier_call_chain - call regulator event notifier
- * @regulator: regulator source
+ * @rdev: regulator source
  * @event: notifier block
- * @data:
+ * @data: callback-specific data.
  *
  * Called by regulator drivers to notify clients a regulator event has
  * occurred. We also notify regulator clients downstream.
@@ -1808,8 +1807,9 @@ static int add_regulator_attributes(struct regulator_dev *rdev)
 
 /**
  * regulator_register - register regulator
- * @regulator: regulator source
- * @reg_data: private regulator data
+ * @regulator_desc: regulator to register
+ * @dev: struct device for the regulator
+ * @driver_data: private regulator data
  *
  * Called by regulator drivers to register a regulator.
  * Returns 0 on success.
@@ -1916,7 +1916,7 @@ EXPORT_SYMBOL_GPL(regulator_register);
 
 /**
  * regulator_unregister - unregister regulator
- * @regulator: regulator source
+ * @rdev: regulator to unregister
  *
  * Called by regulator drivers to unregister a regulator.
  */
@@ -1971,7 +1971,7 @@ EXPORT_SYMBOL_GPL(regulator_suspend_prepare);
 
 /**
  * rdev_get_drvdata - get rdev regulator driver data
- * @regulator: regulator
+ * @rdev: regulator
  *
  * Get rdev regulator driver private data. This call can be used in the
  * regulator driver context.
@@ -2008,7 +2008,7 @@ EXPORT_SYMBOL_GPL(regulator_set_drvdata);
 
 /**
  * regulator_get_id - get regulator ID
- * @regulator: regulator
+ * @rdev: regulator
  */
 int rdev_get_id(struct regulator_dev *rdev)
 {
diff --git a/include/linux/regulator/consumer.h b/include/linux/regulator/consumer.h
index afdc4558bb94..801bf77ff4e2 100644
--- a/include/linux/regulator/consumer.h
+++ b/include/linux/regulator/consumer.h
@@ -104,10 +104,10 @@ struct regulator;
 /**
  * struct regulator_bulk_data - Data used for bulk regulator operations.
  *
- * @supply   The name of the supply.  Initialised by the user before
- *           using the bulk regulator APIs.
- * @consumer The regulator consumer for the supply.  This will be managed
- *           by the bulk API.
+ * @supply:   The name of the supply.  Initialised by the user before
+ *            using the bulk regulator APIs.
+ * @consumer: The regulator consumer for the supply.  This will be managed
+ *            by the bulk API.
  *
  * The regulator APIs provide a series of regulator_bulk_() API calls as
  * a convenience to consumers which require multiple supplies.  This
-- 
cgit v1.2.3


From c8e7e4640facbe99d10a6e262523b25be129b9b9 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Wed, 31 Dec 2008 12:52:42 +0000
Subject: regulator: Add missing kerneldoc

This is only the documentation that the kerneldoc system warns about.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
---
 drivers/regulator/core.c          |  1 +
 include/linux/regulator/driver.h  | 40 +++++++++++++++++++++++++++++++++++++-
 include/linux/regulator/machine.h | 41 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 81 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c
index ea12c68c327f..fda44009024d 100644
--- a/drivers/regulator/core.c
+++ b/drivers/regulator/core.c
@@ -656,6 +656,7 @@ static void print_constraints(struct regulator_dev *rdev)
 /**
  * set_machine_constraints - sets regulator constraints
  * @rdev: regulator source
+ * @constraints: constraints to apply
  *
  * Allows platform initialisation code to define and constrain
  * regulator circuits e.g. valid voltage/current ranges, etc.  NOTE:
diff --git a/include/linux/regulator/driver.h b/include/linux/regulator/driver.h
index e37d80561985..84c3737c2d26 100644
--- a/include/linux/regulator/driver.h
+++ b/include/linux/regulator/driver.h
@@ -24,7 +24,37 @@ struct regulator_init_data;
 /**
  * struct regulator_ops - regulator operations.
  *
- * This struct describes regulator operations.
+ * This struct describes regulator operations which can be implemented by
+ * regulator chip drivers.
+ *
+ * @enable: Enable the regulator.
+ * @disable: Disable the regulator.
+ * @is_enabled: Return 1 if the reguator is enabled, 0 otherwise.
+ *
+ * @set_voltage: Set the voltage for the regulator within the range specified.
+ *               The driver should select the voltage closest to min_uV.
+ * @get_voltage: Return the currently configured voltage for the regulator.
+ *
+ * @set_current: Set the current for the regulator within the range specified.
+ *               The driver should select the current closest to min_uA.
+ * @get_current: Return the currently configured current for the regulator.
+ *
+ * @set_current_limit: Configure a limit for a current-limited regulator.
+ * @get_current_limit: Get the limit for a current-limited regulator.
+ *
+ * @set_mode: Set the operating mode for the regulator.
+ * @get_mode: Get the current operating mode for the regulator.
+ * @get_optimum_mode: Get the most efficient operating mode for the regulator
+ *                    when running with the specified parameters.
+ *
+ * @set_suspend_voltage: Set the voltage for the regulator when the system
+ *                       is suspended.
+ * @set_suspend_enable: Mark the regulator as enabled when the system is
+ *                      suspended.
+ * @set_suspend_disable: Mark the regulator as disabled when the system is
+ *                       suspended.
+ * @set_suspend_mode: Set the operating mode for the regulator when the
+ *                    system is suspended.
  */
 struct regulator_ops {
 
@@ -75,6 +105,14 @@ enum regulator_type {
 /**
  * struct regulator_desc - Regulator descriptor
  *
+ * Each regulator registered with the core is described with a structure of
+ * this type.
+ *
+ * @name: Identifying name for the regulator.
+ * @id: Numerical identifier for the regulator.
+ * @ops: Regulator operations table.
+ * @type: Indicates if the regulator is a voltage or current regulator.
+ * @owner: Module providing the regulator, used for refcounting.
  */
 struct regulator_desc {
 	const char *name;
diff --git a/include/linux/regulator/machine.h b/include/linux/regulator/machine.h
index c6d69331a81e..3794773b23d2 100644
--- a/include/linux/regulator/machine.h
+++ b/include/linux/regulator/machine.h
@@ -44,6 +44,10 @@ struct regulator;
  * struct regulator_state - regulator state during low power syatem states
  *
  * This describes a regulators state during a system wide low power state.
+ *
+ * @uV: Operating voltage during suspend.
+ * @mode: Operating mode during suspend.
+ * @enabled: Enabled during suspend.
  */
 struct regulator_state {
 	int uV;	/* suspend voltage */
@@ -55,6 +59,30 @@ struct regulator_state {
  * struct regulation_constraints - regulator operating constraints.
  *
  * This struct describes regulator and board/machine specific constraints.
+ *
+ * @name: Descriptive name for the constraints, used for display purposes.
+ *
+ * @min_uV: Smallest voltage consumers may set.
+ * @max_uV: Largest voltage consumers may set.
+ *
+ * @min_uA: Smallest consumers consumers may set.
+ * @max_uA: Largest current consumers may set.
+ *
+ * @valid_modes_mask: Mask of modes which may be configured by consumers.
+ * @valid_ops_mask: Operations which may be performed by consumers.
+ *
+ * @always_on: Set if the regulator should never be disabled.
+ * @boot_on: Set if the regulator is enabled when the system is initially
+ *           started.
+ * @apply_uV: Apply the voltage constraint when initialising.
+ *
+ * @input_uV: Input voltage for regulator when supplied by another regulator.
+ *
+ * @state_disk: State for regulator when system is suspended in disk mode.
+ * @state_mem: State for regulator when system is suspended in mem mode.
+ * @state_standby: State for regulator when system is suspended in standby
+ *                 mode.
+ * @initial_state: Suspend state to set by default.
  */
 struct regulation_constraints {
 
@@ -93,6 +121,9 @@ struct regulation_constraints {
  * struct regulator_consumer_supply - supply -> device mapping
  *
  * This maps a supply name to a device.
+ *
+ * @dev: Device structure for the consumer.
+ * @supply: Name for the supply.
  */
 struct regulator_consumer_supply {
 	struct device *dev;	/* consumer */
@@ -103,6 +134,16 @@ struct regulator_consumer_supply {
  * struct regulator_init_data - regulator platform initialisation data.
  *
  * Initialisation constraints, our supply and consumers supplies.
+ *
+ * @supply_regulator_dev: Parent regulator (if any).
+ *
+ * @constraints: Constraints.  These must be specified for the regulator to
+ *               be usable.
+ * @num_consumer_supplies: Number of consumer device supplies.
+ * @consumer_supplies: Consumer device supply configuration.
+ *
+ * @regulator_init: Callback invoked when the regulator has been registered.
+ * @driver_data: Data passed to regulator_init.
  */
 struct regulator_init_data {
 	struct device *supply_regulator_dev; /* or NULL for LINE */
-- 
cgit v1.2.3


From 0ba4887c6329043d6cee5b5b477cfe50c2b57674 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Thu, 8 Jan 2009 11:50:23 -0800
Subject: regulator: fix kernel-doc warnings

Fix kernel-doc warnings in regulator/driver.h:

Warning(linux-next-20090108//include/linux/regulator/driver.h:95): Excess struct/union/enum/typedef member 'set_current' description in 'regulator_ops'
Warning(linux-next-20090108//include/linux/regulator/driver.h:95): Excess struct/union/enum/typedef member 'get_current' description in 'regulator_ops'
Warning(linux-next-20090108//include/linux/regulator/driver.h:124): No description found for parameter 'irq'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
cc: Liam Girdwood <lrg@slimlogic.co.uk>
cc: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
---
 include/linux/regulator/driver.h | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/regulator/driver.h b/include/linux/regulator/driver.h
index 84c3737c2d26..2dae05705f13 100644
--- a/include/linux/regulator/driver.h
+++ b/include/linux/regulator/driver.h
@@ -29,16 +29,12 @@ struct regulator_init_data;
  *
  * @enable: Enable the regulator.
  * @disable: Disable the regulator.
- * @is_enabled: Return 1 if the reguator is enabled, 0 otherwise.
+ * @is_enabled: Return 1 if the regulator is enabled, 0 otherwise.
  *
  * @set_voltage: Set the voltage for the regulator within the range specified.
  *               The driver should select the voltage closest to min_uV.
  * @get_voltage: Return the currently configured voltage for the regulator.
  *
- * @set_current: Set the current for the regulator within the range specified.
- *               The driver should select the current closest to min_uA.
- * @get_current: Return the currently configured current for the regulator.
- *
  * @set_current_limit: Configure a limit for a current-limited regulator.
  * @get_current_limit: Get the limit for a current-limited regulator.
  *
@@ -111,6 +107,7 @@ enum regulator_type {
  * @name: Identifying name for the regulator.
  * @id: Numerical identifier for the regulator.
  * @ops: Regulator operations table.
+ * @irq: Interrupt number for the regulator.
  * @type: Indicates if the regulator is a voltage or current regulator.
  * @owner: Module providing the regulator, used for refcounting.
  */
-- 
cgit v1.2.3


From f4f6bda00fc6bf995a35d8246db45aacaa9b3f09 Mon Sep 17 00:00:00 2001
From: Mike Rapoport <mike@compulab.co.il>
Date: Wed, 3 Dec 2008 08:48:52 +0000
Subject: backlight: add support for Toppoly TDO35S series to tdo24m lcd driver

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Acked-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 drivers/video/backlight/Kconfig  |  4 +-
 drivers/video/backlight/tdo24m.c | 94 ++++++++++++++++++++++++++++++++++++----
 include/linux/spi/tdo24m.h       | 13 ++++++
 3 files changed, 101 insertions(+), 10 deletions(-)
 create mode 100644 include/linux/spi/tdo24m.h

(limited to 'include/linux')

diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig
index 4a4dd9adc328..7ee2313ed6b8 100644
--- a/drivers/video/backlight/Kconfig
+++ b/drivers/video/backlight/Kconfig
@@ -52,11 +52,11 @@ config LCD_ILI9320
 	  then say y to include a power driver for it.
 
 config LCD_TDO24M
-	tristate "Toppoly TDO24M LCD Panels support"
+	tristate "Toppoly TDO24M  and TDO35S LCD Panels support"
 	depends on LCD_CLASS_DEVICE && SPI_MASTER
 	default n
 	help
-	  If you have a Toppoly TDO24M series LCD panel, say y here to
+	  If you have a Toppoly TDO24M/TDO35S series LCD panel, say y here to
 	  include the support for it.
 
 config LCD_VGG2432A4
diff --git a/drivers/video/backlight/tdo24m.c b/drivers/video/backlight/tdo24m.c
index 8427669162ea..1dae7f8f3c6b 100644
--- a/drivers/video/backlight/tdo24m.c
+++ b/drivers/video/backlight/tdo24m.c
@@ -14,6 +14,7 @@
 #include <linux/init.h>
 #include <linux/device.h>
 #include <linux/spi/spi.h>
+#include <linux/spi/tdo24m.h>
 #include <linux/fb.h>
 #include <linux/lcd.h>
 
@@ -31,6 +32,9 @@ struct tdo24m {
 	struct spi_transfer	xfer;
 	uint8_t			*buf;
 
+	int (*adj_mode)(struct tdo24m *lcd, int mode);
+	int color_invert;
+
 	int			power;
 	int			mode;
 };
@@ -66,7 +70,7 @@ static uint32_t lcd_panel_off[] = {
 	CMD_NULL,
 };
 
-static uint32_t lcd_vga_pass_through[] = {
+static uint32_t lcd_vga_pass_through_tdo24m[] = {
 	CMD1(0xB0, 0x16),
 	CMD1(0xBC, 0x80),
 	CMD1(0xE1, 0x00),
@@ -75,7 +79,7 @@ static uint32_t lcd_vga_pass_through[] = {
 	CMD_NULL,
 };
 
-static uint32_t lcd_qvga_pass_through[] = {
+static uint32_t lcd_qvga_pass_through_tdo24m[] = {
 	CMD1(0xB0, 0x16),
 	CMD1(0xBC, 0x81),
 	CMD1(0xE1, 0x00),
@@ -84,7 +88,7 @@ static uint32_t lcd_qvga_pass_through[] = {
 	CMD_NULL,
 };
 
-static uint32_t lcd_vga_transfer[] = {
+static uint32_t lcd_vga_transfer_tdo24m[] = {
 	CMD1(0xcf, 0x02), 	/* Blanking period control (1) */
 	CMD2(0xd0, 0x08, 0x04),	/* Blanking period control (2) */
 	CMD1(0xd1, 0x01),	/* CKV timing control on/off */
@@ -110,6 +114,35 @@ static uint32_t lcd_qvga_transfer[] = {
 	CMD_NULL,
 };
 
+static uint32_t lcd_vga_pass_through_tdo35s[] = {
+	CMD1(0xB0, 0x16),
+	CMD1(0xBC, 0x80),
+	CMD1(0xE1, 0x00),
+	CMD1(0x3B, 0x00),
+	CMD_NULL,
+};
+
+static uint32_t lcd_qvga_pass_through_tdo35s[] = {
+	CMD1(0xB0, 0x16),
+	CMD1(0xBC, 0x81),
+	CMD1(0xE1, 0x00),
+	CMD1(0x3B, 0x22),
+	CMD_NULL,
+};
+
+static uint32_t lcd_vga_transfer_tdo35s[] = {
+	CMD1(0xcf, 0x02), 	/* Blanking period control (1) */
+	CMD2(0xd0, 0x08, 0x04),	/* Blanking period control (2) */
+	CMD1(0xd1, 0x01),	/* CKV timing control on/off */
+	CMD2(0xd2, 0x00, 0x1e),	/* CKV 1,2 timing control */
+	CMD2(0xd3, 0x14, 0x28),	/* OEV timing control */
+	CMD2(0xd4, 0x28, 0x64),	/* ASW timing control (1) */
+	CMD1(0xd5, 0x28),	/* ASW timing control (2) */
+	CMD0(0x21),		/* Invert for normally black display */
+	CMD0(0x29),		/* Display on */
+	CMD_NULL,
+};
+
 static uint32_t lcd_panel_config[] = {
 	CMD2(0xb8, 0xff, 0xf9),	/* Output control */
 	CMD0(0x11),		/* sleep out */
@@ -148,6 +181,8 @@ static int tdo24m_writes(struct tdo24m *lcd, uint32_t *array)
 	int nparams, err = 0;
 
 	for (; *p != CMD_NULL; p++) {
+		if (!lcd->color_invert && *p == CMD0(0x21))
+			continue;
 
 		nparams = (*p >> 30) & 0x3;
 
@@ -184,12 +219,33 @@ static int tdo24m_adj_mode(struct tdo24m *lcd, int mode)
 {
 	switch (mode) {
 	case MODE_VGA:
-		tdo24m_writes(lcd, lcd_vga_pass_through);
+		tdo24m_writes(lcd, lcd_vga_pass_through_tdo24m);
 		tdo24m_writes(lcd, lcd_panel_config);
-		tdo24m_writes(lcd, lcd_vga_transfer);
+		tdo24m_writes(lcd, lcd_vga_transfer_tdo24m);
 		break;
 	case MODE_QVGA:
-		tdo24m_writes(lcd, lcd_qvga_pass_through);
+		tdo24m_writes(lcd, lcd_qvga_pass_through_tdo24m);
+		tdo24m_writes(lcd, lcd_panel_config);
+		tdo24m_writes(lcd, lcd_qvga_transfer);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	lcd->mode = mode;
+	return 0;
+}
+
+static int tdo35s_adj_mode(struct tdo24m *lcd, int mode)
+{
+	switch (mode) {
+	case MODE_VGA:
+		tdo24m_writes(lcd, lcd_vga_pass_through_tdo35s);
+		tdo24m_writes(lcd, lcd_panel_config);
+		tdo24m_writes(lcd, lcd_vga_transfer_tdo35s);
+		break;
+	case MODE_QVGA:
+		tdo24m_writes(lcd, lcd_qvga_pass_through_tdo35s);
 		tdo24m_writes(lcd, lcd_panel_config);
 		tdo24m_writes(lcd, lcd_qvga_transfer);
 		break;
@@ -213,7 +269,7 @@ static int tdo24m_power_on(struct tdo24m *lcd)
 	if (err)
 		goto out;
 
-	err = tdo24m_adj_mode(lcd, lcd->mode);
+	err = lcd->adj_mode(lcd, lcd->mode);
 out:
 	return err;
 }
@@ -262,7 +318,7 @@ static int tdo24m_set_mode(struct lcd_device *ld, struct fb_videomode *m)
 	if (lcd->mode == mode)
 		return 0;
 
-	return tdo24m_adj_mode(lcd, mode);
+	return lcd->adj_mode(lcd, mode);
 }
 
 static struct lcd_ops tdo24m_ops = {
@@ -276,8 +332,16 @@ static int __devinit tdo24m_probe(struct spi_device *spi)
 	struct tdo24m *lcd;
 	struct spi_message *m;
 	struct spi_transfer *x;
+	struct tdo24m_platform_data *pdata;
+	enum tdo24m_model model;
 	int err;
 
+	pdata = spi->dev.platform_data;
+	if (pdata)
+		model = pdata->model;
+	else
+		model = TDO24M;
+
 	spi->bits_per_word = 8;
 	spi->mode = SPI_MODE_3;
 	err = spi_setup(spi);
@@ -306,6 +370,20 @@ static int __devinit tdo24m_probe(struct spi_device *spi)
 	x->tx_buf = &lcd->buf[0];
 	spi_message_add_tail(x, m);
 
+	switch (model) {
+	case TDO24M:
+		lcd->color_invert = 1;
+		lcd->adj_mode = tdo24m_adj_mode;
+		break;
+	case TDO35S:
+		lcd->adj_mode = tdo35s_adj_mode;
+		lcd->color_invert = 0;
+		break;
+	default:
+		dev_err(&spi->dev, "Unsupported model");
+		goto out_free;
+	}
+
 	lcd->lcd_dev = lcd_device_register("tdo24m", &spi->dev,
 					lcd, &tdo24m_ops);
 	if (IS_ERR(lcd->lcd_dev)) {
diff --git a/include/linux/spi/tdo24m.h b/include/linux/spi/tdo24m.h
new file mode 100644
index 000000000000..7572d4e1fe76
--- /dev/null
+++ b/include/linux/spi/tdo24m.h
@@ -0,0 +1,13 @@
+#ifndef __TDO24M_H__
+#define __TDO24M_H__
+
+enum tdo24m_model {
+	TDO24M,
+	TDO35S,
+};
+
+struct tdo24m_platform_data {
+	enum tdo24m_model model;
+};
+
+#endif /* __TDO24M_H__ */
-- 
cgit v1.2.3


From 0c3573f19d135d718264e38c46597295bd6154b7 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Fri, 9 Jan 2009 08:31:05 +1100
Subject: md: use sysfs_notify_dirent to notify changes to md/sync_action.

There is no compelling need for this, but sysfs_notify_dirent is a
nicer interface and the change is good for consistency.

Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/md.c           | 20 +++++++++++++-------
 include/linux/raid/md_k.h |  1 +
 2 files changed, 14 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 1b1d32694f6f..ee759b193e9a 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -3046,7 +3046,7 @@ action_store(mddev_t *mddev, const char *page, size_t len)
 	}
 	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 	md_wakeup_thread(mddev->thread);
-	sysfs_notify(&mddev->kobj, NULL, "sync_action");
+	sysfs_notify_dirent(mddev->sysfs_action);
 	return len;
 }
 
@@ -3684,6 +3684,7 @@ static int do_md_run(mddev_t * mddev)
 			printk(KERN_WARNING
 			       "md: cannot register extra attributes for %s\n",
 			       mdname(mddev));
+		mddev->sysfs_action = sysfs_get_dirent(mddev->kobj.sd, "sync_action");
 	} else if (mddev->ro == 2) /* auto-readonly not meaningful */
 		mddev->ro = 0;
 
@@ -3754,7 +3755,8 @@ static int do_md_run(mddev_t * mddev)
 	mddev->changed = 1;
 	md_new_event(mddev);
 	sysfs_notify_dirent(mddev->sysfs_state);
-	sysfs_notify(&mddev->kobj, NULL, "sync_action");
+	if (mddev->sysfs_action)
+		sysfs_notify_dirent(mddev->sysfs_action);
 	sysfs_notify(&mddev->kobj, NULL, "degraded");
 	kobject_uevent(&disk_to_dev(mddev->gendisk)->kobj, KOBJ_CHANGE);
 	return 0;
@@ -3854,9 +3856,12 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open)
 			mddev->queue->merge_bvec_fn = NULL;
 			mddev->queue->unplug_fn = NULL;
 			mddev->queue->backing_dev_info.congested_fn = NULL;
-			if (mddev->pers->sync_request)
+			if (mddev->pers->sync_request) {
 				sysfs_remove_group(&mddev->kobj, &md_redundancy_group);
-
+				if (mddev->sysfs_action)
+					sysfs_put(mddev->sysfs_action);
+				mddev->sysfs_action = NULL;
+			}
 			module_put(mddev->pers->owner);
 			mddev->pers = NULL;
 			/* tell userspace to handle 'inactive' */
@@ -6155,7 +6160,7 @@ void md_check_recovery(mddev_t *mddev)
 			mddev->recovery = 0;
 			/* flag recovery needed just to double check */
 			set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
-			sysfs_notify(&mddev->kobj, NULL, "sync_action");
+			sysfs_notify_dirent(mddev->sysfs_action);
 			md_new_event(mddev);
 			goto unlock;
 		}
@@ -6216,7 +6221,7 @@ void md_check_recovery(mddev_t *mddev)
 				mddev->recovery = 0;
 			} else
 				md_wakeup_thread(mddev->sync_thread);
-			sysfs_notify(&mddev->kobj, NULL, "sync_action");
+			sysfs_notify_dirent(mddev->sysfs_action);
 			md_new_event(mddev);
 		}
 	unlock:
@@ -6224,7 +6229,8 @@ void md_check_recovery(mddev_t *mddev)
 			clear_bit(MD_RECOVERY_RUNNING, &mddev->recovery);
 			if (test_and_clear_bit(MD_RECOVERY_RECOVER,
 					       &mddev->recovery))
-				sysfs_notify(&mddev->kobj, NULL, "sync_action");
+				if (mddev->sysfs_action)
+					sysfs_notify_dirent(mddev->sysfs_action);
 		}
 		mddev_unlock(mddev);
 	}
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index 8fc909ef6787..663803eaf0de 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -244,6 +244,7 @@ struct mddev_s
 	struct sysfs_dirent		*sysfs_state;	/* handle for 'array_state'
 							 * file in sysfs.
 							 */
+	struct sysfs_dirent		*sysfs_action;  /* handle for 'sync_action' */
 
 	spinlock_t			write_lock;
 	wait_queue_head_t		sb_wait;	/* for waiting on superblock updates */
-- 
cgit v1.2.3


From 019c4e2f3e02aac4b44003913b54ca4b332e4371 Mon Sep 17 00:00:00 2001
From: Andre Noll <maan@systemlinux.org>
Date: Fri, 9 Jan 2009 08:31:06 +1100
Subject: md: raid0: Represent device offset in sectors.

Rename zone->dev_offset to zone->dev_start to make sure all users
have been converted.

Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/raid0.c         | 9 ++++-----
 include/linux/raid/raid0.h | 2 +-
 2 files changed, 5 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index fd65d8806c0f..b860536dc894 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -167,7 +167,7 @@ static int create_strip_zones (mddev_t *mddev)
 		zone->dev = conf->strip_zone[i-1].dev + mddev->raid_disks;
 
 		printk("raid0: zone %d\n", i);
-		zone->dev_offset = current_offset;
+		zone->dev_start = current_offset * 2;
 		smallest = NULL;
 		c = 0;
 
@@ -452,8 +452,7 @@ static int raid0_make_request (struct request_queue *q, struct bio *bio)
 		x = sector >> chunksect_bits;
 		tmp_dev = zone->dev[sector_div(x, zone->nb_dev)];
 	}
-	rsect = (((chunk << (chunksect_bits - 1)) + zone->dev_offset)<<1)
-		+ sect_in_chunk;
+	rsect = (chunk << chunksect_bits) + zone->dev_start + sect_in_chunk;
  
 	bio->bi_bdev = tmp_dev->bdev;
 	bio->bi_sector = rsect + tmp_dev->data_offset;
@@ -490,9 +489,9 @@ static void raid0_status (struct seq_file *seq, mddev_t *mddev)
 			seq_printf(seq, "%s/", bdevname(
 				conf->strip_zone[j].dev[k]->bdev,b));
 
-		seq_printf(seq, "] zo=%d do=%d s=%d\n",
+		seq_printf(seq, "] zo=%d ds=%d s=%d\n",
 				conf->strip_zone[j].zone_offset,
-				conf->strip_zone[j].dev_offset,
+				conf->strip_zone[j].dev_start,
 				conf->strip_zone[j].size);
 	}
 #endif
diff --git a/include/linux/raid/raid0.h b/include/linux/raid/raid0.h
index 1b2dda035f8e..61c3d29dc158 100644
--- a/include/linux/raid/raid0.h
+++ b/include/linux/raid/raid0.h
@@ -6,7 +6,7 @@
 struct strip_zone
 {
 	sector_t zone_offset;	/* Zone offset in md_dev */
-	sector_t dev_offset;	/* Zone offset in real dev */
+	sector_t dev_start;	/* Zone offset in real dev (in sectors) */
 	sector_t size;		/* Zone size */
 	int nb_dev;		/* # of devices attached to the zone */
 	mdk_rdev_t **dev;	/* Devices attached to the zone */
-- 
cgit v1.2.3


From 6199d3db0fc34f8ada895879d04a353a6ae632bc Mon Sep 17 00:00:00 2001
From: Andre Noll <maan@systemlinux.org>
Date: Fri, 9 Jan 2009 08:31:07 +1100
Subject: md: raid0: Represent zone->zone_offset in sectors.

For the same reason as in the previous patch, rename it from zone_offset
to zone_start.

Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/raid0.c         | 12 ++++++------
 include/linux/raid/raid0.h |  2 +-
 2 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index b860536dc894..6e12a358f366 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -155,7 +155,7 @@ static int create_strip_zones (mddev_t *mddev)
 	}
 	zone->nb_dev = cnt;
 	zone->size = smallest->size * cnt;
-	zone->zone_offset = 0;
+	zone->zone_start = 0;
 
 	current_offset = smallest->size;
 	curr_zone_offset = zone->size;
@@ -194,7 +194,7 @@ static int create_strip_zones (mddev_t *mddev)
 		printk("raid0: zone->nb_dev: %d, size: %llu\n",
 			zone->nb_dev, (unsigned long long)zone->size);
 
-		zone->zone_offset = curr_zone_offset;
+		zone->zone_start = curr_zone_offset * 2;
 		curr_zone_offset += zone->size;
 
 		current_offset = smallest->size;
@@ -437,14 +437,14 @@ static int raid0_make_request (struct request_queue *q, struct bio *bio)
 		zone = conf->hash_table[x];
 	}
  
-	while (sector / 2 >= (zone->zone_offset + zone->size))
+	while (sector / 2 >= (zone->zone_start / 2 + zone->size))
 		zone++;
     
 	sect_in_chunk = bio->bi_sector & (chunk_sects - 1);
 
 
 	{
-		sector_t x = (sector - zone->zone_offset * 2) >> chunksect_bits;
+		sector_t x = (sector - zone->zone_start) >> chunksect_bits;
 
 		sector_div(x, zone->nb_dev);
 		chunk = x;
@@ -489,8 +489,8 @@ static void raid0_status (struct seq_file *seq, mddev_t *mddev)
 			seq_printf(seq, "%s/", bdevname(
 				conf->strip_zone[j].dev[k]->bdev,b));
 
-		seq_printf(seq, "] zo=%d ds=%d s=%d\n",
-				conf->strip_zone[j].zone_offset,
+		seq_printf(seq, "] zs=%d ds=%d s=%d\n",
+				conf->strip_zone[j].zone_start,
 				conf->strip_zone[j].dev_start,
 				conf->strip_zone[j].size);
 	}
diff --git a/include/linux/raid/raid0.h b/include/linux/raid/raid0.h
index 61c3d29dc158..eaf4f6ac55f6 100644
--- a/include/linux/raid/raid0.h
+++ b/include/linux/raid/raid0.h
@@ -5,7 +5,7 @@
 
 struct strip_zone
 {
-	sector_t zone_offset;	/* Zone offset in md_dev */
+	sector_t zone_start;	/* Zone offset in md_dev (in sectors) */
 	sector_t dev_start;	/* Zone offset in real dev (in sectors) */
 	sector_t size;		/* Zone size */
 	int nb_dev;		/* # of devices attached to the zone */
-- 
cgit v1.2.3


From 83838ed87898e0a8ff8dbf001e54e6c017f0a011 Mon Sep 17 00:00:00 2001
From: Andre Noll <maan@systemlinux.org>
Date: Fri, 9 Jan 2009 08:31:07 +1100
Subject: md: raid0: Represent the size of strip zones in sectors.

This completes the block -> sector conversion of struct strip_zone.

Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/raid0.c         | 26 +++++++++++++-------------
 include/linux/raid/raid0.h |  2 +-
 2 files changed, 14 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index c759436b5d84..6e85e88bbae9 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -156,11 +156,11 @@ static int create_strip_zones (mddev_t *mddev)
 		goto abort;
 	}
 	zone->nb_dev = cnt;
-	zone->size = smallest->size * cnt;
+	zone->sectors = smallest->size * cnt * 2;
 	zone->zone_start = 0;
 
 	current_start = smallest->size * 2;
-	curr_zone_start = zone->size * 2;
+	curr_zone_start = zone->sectors;
 
 	/* now do the other zones */
 	for (i = 1; i < conf->nr_strip_zones; i++)
@@ -193,12 +193,12 @@ static int create_strip_zones (mddev_t *mddev)
 		}
 
 		zone->nb_dev = c;
-		zone->size = (smallest->size - current_start / 2) * c;
-		printk(KERN_INFO "raid0: zone->nb_dev: %d, size: %llu\n",
-			zone->nb_dev, (unsigned long long)zone->size);
+		zone->sectors = (smallest->size * 2 - current_start) * c;
+		printk(KERN_INFO "raid0: zone->nb_dev: %d, sectors: %llu\n",
+			zone->nb_dev, (unsigned long long)zone->sectors);
 
 		zone->zone_start = curr_zone_start;
-		curr_zone_start += zone->size * 2;
+		curr_zone_start += zone->sectors;
 
 		current_start = smallest->size * 2;
 		printk(KERN_INFO "raid0: current zone start: %llu\n",
@@ -220,7 +220,7 @@ static int create_strip_zones (mddev_t *mddev)
 		sector_t sz = 0;
 		for (j=i; j<conf->nr_strip_zones-1 &&
 			     sz < min_spacing ; j++)
-			sz += conf->strip_zone[j].size;
+			sz += conf->strip_zone[j].sectors / 2;
 		if (sz >= min_spacing && sz < conf->hash_spacing)
 			conf->hash_spacing = sz;
 	}
@@ -325,13 +325,13 @@ static int raid0_run (mddev_t *mddev)
 	conf->hash_table = kmalloc (sizeof (struct strip_zone *)*nb_zone, GFP_KERNEL);
 	if (!conf->hash_table)
 		goto out_free_conf;
-	size = conf->strip_zone[cur].size;
+	size = conf->strip_zone[cur].sectors / 2;
 
 	conf->hash_table[0] = conf->strip_zone + cur;
 	for (i=1; i< nb_zone; i++) {
 		while (size <= conf->hash_spacing) {
 			cur++;
-			size += conf->strip_zone[cur].size;
+			size += conf->strip_zone[cur].sectors / 2;
 		}
 		size -= conf->hash_spacing;
 		conf->hash_table[i] = conf->strip_zone + cur;
@@ -439,10 +439,10 @@ static int raid0_make_request (struct request_queue *q, struct bio *bio)
 		sector_div(x, (u32)conf->hash_spacing);
 		zone = conf->hash_table[x];
 	}
- 
-	while (sector / 2 >= (zone->zone_start / 2 + zone->size))
+
+	while (sector >= zone->zone_start + zone->sectors)
 		zone++;
-    
+
 	sect_in_chunk = bio->bi_sector & (chunk_sects - 1);
 
 
@@ -495,7 +495,7 @@ static void raid0_status (struct seq_file *seq, mddev_t *mddev)
 		seq_printf(seq, "] zs=%d ds=%d s=%d\n",
 				conf->strip_zone[j].zone_start,
 				conf->strip_zone[j].dev_start,
-				conf->strip_zone[j].size);
+				conf->strip_zone[j].sectors);
 	}
 #endif
 	seq_printf(seq, " %dk chunks", mddev->chunk_size/1024);
diff --git a/include/linux/raid/raid0.h b/include/linux/raid/raid0.h
index eaf4f6ac55f6..c12521d027e2 100644
--- a/include/linux/raid/raid0.h
+++ b/include/linux/raid/raid0.h
@@ -7,7 +7,7 @@ struct strip_zone
 {
 	sector_t zone_start;	/* Zone offset in md_dev (in sectors) */
 	sector_t dev_start;	/* Zone offset in real dev (in sectors) */
-	sector_t size;		/* Zone size */
+	sector_t sectors;	/* Zone size in sectors */
 	int nb_dev;		/* # of devices attached to the zone */
 	mdk_rdev_t **dev;	/* Devices attached to the zone */
 };
-- 
cgit v1.2.3


From ccacc7d2cf03114a24ab903f710118e9e5d43273 Mon Sep 17 00:00:00 2001
From: Andre Noll <maan@systemlinux.org>
Date: Fri, 9 Jan 2009 08:31:08 +1100
Subject: md: raid0: make hash_spacing and preshift sector-based.

This patch renames the hash_spacing and preshift members of struct
raid0_private_data to spacing and sector_shift respectively and
changes the semantics as follows:

We always have spacing = 2 * hash_spacing. In case
sizeof(sector_t) > sizeof(u32) we also have sector_shift = preshift + 1
while sector_shift = preshift = 0 otherwise.

Note that the values of nb_zone and zone are unaffected by these changes
because in the sector_div() preceeding the assignement of these two
variables both arguments double.

Signed-off-by: Andre Noll <maan@systemlinux.org>
Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/raid0.c         | 58 +++++++++++++++++++++++-----------------------
 include/linux/raid/raid0.h |  4 ++--
 2 files changed, 31 insertions(+), 31 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index 6e85e88bbae9..90f5b24f6e6c 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -213,16 +213,16 @@ static int create_strip_zones (mddev_t *mddev)
 	 * strip though as it's size has no bearing on the efficacy of the hash
 	 * table.
 	 */
-	conf->hash_spacing = curr_zone_start / 2;
-	min_spacing = curr_zone_start / 2;
+	conf->spacing = curr_zone_start;
+	min_spacing = curr_zone_start;
 	sector_div(min_spacing, PAGE_SIZE/sizeof(struct strip_zone*));
 	for (i=0; i < conf->nr_strip_zones-1; i++) {
-		sector_t sz = 0;
-		for (j=i; j<conf->nr_strip_zones-1 &&
-			     sz < min_spacing ; j++)
-			sz += conf->strip_zone[j].sectors / 2;
-		if (sz >= min_spacing && sz < conf->hash_spacing)
-			conf->hash_spacing = sz;
+		sector_t s = 0;
+		for (j = i; j < conf->nr_strip_zones - 1 &&
+				s < min_spacing; j++)
+			s += conf->strip_zone[j].sectors;
+		if (s >= min_spacing && s < conf->spacing)
+			conf->spacing = s;
 	}
 
 	mddev->queue->unplug_fn = raid0_unplug;
@@ -265,7 +265,7 @@ static int raid0_mergeable_bvec(struct request_queue *q,
 static int raid0_run (mddev_t *mddev)
 {
 	unsigned  cur=0, i=0, nb_zone;
-	s64 size;
+	s64 sectors;
 	raid0_conf_t *conf;
 	mdk_rdev_t *rdev;
 	struct list_head *tmp;
@@ -297,51 +297,51 @@ static int raid0_run (mddev_t *mddev)
 	rdev_for_each(rdev, tmp, mddev)
 		mddev->array_sectors += rdev->size * 2;
 
-	printk("raid0 : md_size is %llu blocks.\n", 
-		(unsigned long long)mddev->array_sectors / 2);
-	printk("raid0 : conf->hash_spacing is %llu blocks.\n",
-		(unsigned long long)conf->hash_spacing);
+	printk(KERN_INFO "raid0 : md_size is %llu sectors.\n",
+		(unsigned long long)mddev->array_sectors);
+	printk(KERN_INFO "raid0 : conf->spacing is %llu sectors.\n",
+		(unsigned long long)conf->spacing);
 	{
-		sector_t s = mddev->array_sectors / 2;
-		sector_t space = conf->hash_spacing;
+		sector_t s = mddev->array_sectors;
+		sector_t space = conf->spacing;
 		int round;
-		conf->preshift = 0;
+		conf->sector_shift = 0;
 		if (sizeof(sector_t) > sizeof(u32)) {
 			/*shift down space and s so that sector_div will work */
 			while (space > (sector_t) (~(u32)0)) {
 				s >>= 1;
 				space >>= 1;
 				s += 1; /* force round-up */
-				conf->preshift++;
+				conf->sector_shift++;
 			}
 		}
 		round = sector_div(s, (u32)space) ? 1 : 0;
 		nb_zone = s + round;
 	}
-	printk("raid0 : nb_zone is %d.\n", nb_zone);
+	printk(KERN_INFO "raid0 : nb_zone is %d.\n", nb_zone);
 
-	printk("raid0 : Allocating %Zd bytes for hash.\n",
+	printk(KERN_INFO "raid0 : Allocating %zu bytes for hash.\n",
 				nb_zone*sizeof(struct strip_zone*));
 	conf->hash_table = kmalloc (sizeof (struct strip_zone *)*nb_zone, GFP_KERNEL);
 	if (!conf->hash_table)
 		goto out_free_conf;
-	size = conf->strip_zone[cur].sectors / 2;
+	sectors = conf->strip_zone[cur].sectors;
 
 	conf->hash_table[0] = conf->strip_zone + cur;
 	for (i=1; i< nb_zone; i++) {
-		while (size <= conf->hash_spacing) {
+		while (sectors <= conf->spacing) {
 			cur++;
-			size += conf->strip_zone[cur].sectors / 2;
+			sectors += conf->strip_zone[cur].sectors;
 		}
-		size -= conf->hash_spacing;
+		sectors -= conf->spacing;
 		conf->hash_table[i] = conf->strip_zone + cur;
 	}
-	if (conf->preshift) {
-		conf->hash_spacing >>= conf->preshift;
-		/* round hash_spacing up so when we divide by it, we
+	if (conf->sector_shift) {
+		conf->spacing >>= conf->sector_shift;
+		/* round spacing up so when we divide by it, we
 		 * err on the side of too-low, which is safest
 		 */
-		conf->hash_spacing++;
+		conf->spacing++;
 	}
 
 	/* calculate the max read-ahead size.
@@ -435,8 +435,8 @@ static int raid0_make_request (struct request_queue *q, struct bio *bio)
  
 
 	{
-		sector_t x = sector >> (conf->preshift + 1);
-		sector_div(x, (u32)conf->hash_spacing);
+		sector_t x = sector >> conf->sector_shift;
+		sector_div(x, (u32)conf->spacing);
 		zone = conf->hash_table[x];
 	}
 
diff --git a/include/linux/raid/raid0.h b/include/linux/raid/raid0.h
index c12521d027e2..fd42aa87c391 100644
--- a/include/linux/raid/raid0.h
+++ b/include/linux/raid/raid0.h
@@ -19,8 +19,8 @@ struct raid0_private_data
 	mdk_rdev_t **devlist; /* lists of rdevs, pointed to by strip_zone->dev */
 	int nr_strip_zones;
 
-	sector_t hash_spacing;
-	int preshift;			/* shift this before divide by hash_spacing */
+	sector_t spacing;
+	int sector_shift; /* shift this before divide by spacing */
 };
 
 typedef struct raid0_private_data raid0_conf_t;
-- 
cgit v1.2.3


From 159ec1fc060ab22b157a62364045f5e98749c4d3 Mon Sep 17 00:00:00 2001
From: Cheng Renquan <crquan@gmail.com>
Date: Fri, 9 Jan 2009 08:31:08 +1100
Subject: md: use list_for_each_entry macro directly

The rdev_for_each macro defined in <linux/raid/md_k.h> is identical to
list_for_each_entry_safe, from <linux/list.h>, it should be defined to
use list_for_each_entry_safe, instead of reinventing the wheel.

But some calls to each_entry_safe don't really need a safe version,
just a direct list_for_each_entry is enough, this could save a temp
variable (tmp) in every function that used rdev_for_each.

In this patch, most rdev_for_each loops are replaced by list_for_each_entry,
totally save many tmp vars; and only in the other situations that will call
list_del to delete an entry, the safe version is used.

Signed-off-by: Cheng Renquan <crquan@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/bitmap.c       |   3 +-
 drivers/md/faulty.c       |   3 +-
 drivers/md/linear.c       |   3 +-
 drivers/md/md.c           | 107 +++++++++++++++++++---------------------------
 drivers/md/multipath.c    |   3 +-
 drivers/md/raid0.c        |  10 ++---
 drivers/md/raid1.c        |   3 +-
 drivers/md/raid10.c       |   3 +-
 drivers/md/raid5.c        |   8 ++--
 include/linux/raid/md_k.h |  11 ++---
 10 files changed, 60 insertions(+), 94 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index 666b7ba47ec5..719943763391 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -215,7 +215,6 @@ static struct page *read_sb_page(mddev_t *mddev, long offset,
 	/* choose a good rdev and read the page from there */
 
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 	sector_t target;
 
 	if (!page)
@@ -223,7 +222,7 @@ static struct page *read_sb_page(mddev_t *mddev, long offset,
 	if (!page)
 		return ERR_PTR(-ENOMEM);
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		if (! test_bit(In_sync, &rdev->flags)
 		    || test_bit(Faulty, &rdev->flags))
 			continue;
diff --git a/drivers/md/faulty.c b/drivers/md/faulty.c
index f26c1f9a475b..86d9adf90e79 100644
--- a/drivers/md/faulty.c
+++ b/drivers/md/faulty.c
@@ -283,7 +283,6 @@ static int reconfig(mddev_t *mddev, int layout, int chunk_size)
 static int run(mddev_t *mddev)
 {
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 	int i;
 
 	conf_t *conf = kmalloc(sizeof(*conf), GFP_KERNEL);
@@ -296,7 +295,7 @@ static int run(mddev_t *mddev)
 	}
 	conf->nfaults = 0;
 
-	rdev_for_each(rdev, tmp, mddev)
+	list_for_each_entry(rdev, &mddev->disks, same_set)
 		conf->rdev = rdev;
 
 	mddev->array_sectors = mddev->size * 2;
diff --git a/drivers/md/linear.c b/drivers/md/linear.c
index 3b90c5c924ec..1e3aea9eecf1 100644
--- a/drivers/md/linear.c
+++ b/drivers/md/linear.c
@@ -105,7 +105,6 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
 	int i, nb_zone, cnt;
 	sector_t min_sectors;
 	sector_t curr_sector;
-	struct list_head *tmp;
 
 	conf = kzalloc (sizeof (*conf) + raid_disks*sizeof(dev_info_t),
 			GFP_KERNEL);
@@ -115,7 +114,7 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
 	cnt = 0;
 	conf->array_sectors = 0;
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		int j = rdev->raid_disk;
 		dev_info_t *disk = conf->disks + j;
 
diff --git a/drivers/md/md.c b/drivers/md/md.c
index ee759b193e9a..58be665bb46b 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -307,25 +307,23 @@ static inline void mddev_unlock(mddev_t * mddev)
 
 static mdk_rdev_t * find_rdev_nr(mddev_t *mddev, int nr)
 {
-	mdk_rdev_t * rdev;
-	struct list_head *tmp;
+	mdk_rdev_t *rdev;
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set)
 		if (rdev->desc_nr == nr)
 			return rdev;
-	}
+
 	return NULL;
 }
 
 static mdk_rdev_t * find_rdev(mddev_t * mddev, dev_t dev)
 {
-	struct list_head *tmp;
 	mdk_rdev_t *rdev;
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set)
 		if (rdev->bdev->bd_dev == dev)
 			return rdev;
-	}
+
 	return NULL;
 }
 
@@ -861,7 +859,6 @@ static int super_90_validate(mddev_t *mddev, mdk_rdev_t *rdev)
 static void super_90_sync(mddev_t *mddev, mdk_rdev_t *rdev)
 {
 	mdp_super_t *sb;
-	struct list_head *tmp;
 	mdk_rdev_t *rdev2;
 	int next_spare = mddev->raid_disks;
 
@@ -933,7 +930,7 @@ static void super_90_sync(mddev_t *mddev, mdk_rdev_t *rdev)
 		sb->state |= (1<<MD_SB_BITMAP_PRESENT);
 
 	sb->disks[0].state = (1<<MD_DISK_REMOVED);
-	rdev_for_each(rdev2, tmp, mddev) {
+	list_for_each_entry(rdev2, &mddev->disks, same_set) {
 		mdp_disk_t *d;
 		int desc_nr;
 		if (rdev2->raid_disk >= 0 && test_bit(In_sync, &rdev2->flags)
@@ -1259,7 +1256,6 @@ static int super_1_validate(mddev_t *mddev, mdk_rdev_t *rdev)
 static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev)
 {
 	struct mdp_superblock_1 *sb;
-	struct list_head *tmp;
 	mdk_rdev_t *rdev2;
 	int max_dev, i;
 	/* make rdev->sb match mddev and rdev data. */
@@ -1307,7 +1303,7 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev)
 	}
 
 	max_dev = 0;
-	rdev_for_each(rdev2, tmp, mddev)
+	list_for_each_entry(rdev2, &mddev->disks, same_set)
 		if (rdev2->desc_nr+1 > max_dev)
 			max_dev = rdev2->desc_nr+1;
 
@@ -1316,7 +1312,7 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev)
 	for (i=0; i<max_dev;i++)
 		sb->dev_roles[i] = cpu_to_le16(0xfffe);
 	
-	rdev_for_each(rdev2, tmp, mddev) {
+	list_for_each_entry(rdev2, &mddev->disks, same_set) {
 		i = rdev2->desc_nr;
 		if (test_bit(Faulty, &rdev2->flags))
 			sb->dev_roles[i] = cpu_to_le16(0xfffe);
@@ -1571,8 +1567,7 @@ static void kick_rdev_from_array(mdk_rdev_t * rdev)
 
 static void export_array(mddev_t *mddev)
 {
-	struct list_head *tmp;
-	mdk_rdev_t *rdev;
+	mdk_rdev_t *rdev, *tmp;
 
 	rdev_for_each(rdev, tmp, mddev) {
 		if (!rdev->mddev) {
@@ -1643,7 +1638,7 @@ static void print_rdev(mdk_rdev_t *rdev)
 
 static void md_print_devices(void)
 {
-	struct list_head *tmp, *tmp2;
+	struct list_head *tmp;
 	mdk_rdev_t *rdev;
 	mddev_t *mddev;
 	char b[BDEVNAME_SIZE];
@@ -1658,11 +1653,11 @@ static void md_print_devices(void)
 			bitmap_print_sb(mddev->bitmap);
 		else
 			printk("%s: ", mdname(mddev));
-		rdev_for_each(rdev, tmp2, mddev)
+		list_for_each_entry(rdev, &mddev->disks, same_set)
 			printk("<%s>", bdevname(rdev->bdev,b));
 		printk("\n");
 
-		rdev_for_each(rdev, tmp2, mddev)
+		list_for_each_entry(rdev, &mddev->disks, same_set)
 			print_rdev(rdev);
 	}
 	printk("md:	**********************************\n");
@@ -1679,9 +1674,8 @@ static void sync_sbs(mddev_t * mddev, int nospares)
 	 * with the rest of the array)
 	 */
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		if (rdev->sb_events == mddev->events ||
 		    (nospares &&
 		     rdev->raid_disk < 0 &&
@@ -1699,7 +1693,6 @@ static void sync_sbs(mddev_t * mddev, int nospares)
 
 static void md_update_sb(mddev_t * mddev, int force_change)
 {
-	struct list_head *tmp;
 	mdk_rdev_t *rdev;
 	int sync_req;
 	int nospares = 0;
@@ -1790,7 +1783,7 @@ repeat:
 		mdname(mddev),mddev->in_sync);
 
 	bitmap_update_sb(mddev->bitmap);
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		char b[BDEVNAME_SIZE];
 		dprintk(KERN_INFO "md: ");
 		if (rdev->sb_loaded != 1)
@@ -1999,7 +1992,6 @@ slot_store(mdk_rdev_t *rdev, const char *buf, size_t len)
 		md_wakeup_thread(rdev->mddev->thread);
 	} else if (rdev->mddev->pers) {
 		mdk_rdev_t *rdev2;
-		struct list_head *tmp;
 		/* Activating a spare .. or possibly reactivating
 		 * if we every get bitmaps working here.
 		 */
@@ -2010,7 +2002,7 @@ slot_store(mdk_rdev_t *rdev, const char *buf, size_t len)
 		if (rdev->mddev->pers->hot_add_disk == NULL)
 			return -EINVAL;
 
-		rdev_for_each(rdev2, tmp, rdev->mddev)
+		list_for_each_entry(rdev2, &rdev->mddev->disks, same_set)
 			if (rdev2->raid_disk == slot)
 				return -EEXIST;
 
@@ -2125,14 +2117,14 @@ rdev_size_store(mdk_rdev_t *rdev, const char *buf, size_t len)
 		 */
 		mddev_t *mddev;
 		int overlap = 0;
-		struct list_head *tmp, *tmp2;
+		struct list_head *tmp;
 
 		mddev_unlock(my_mddev);
 		for_each_mddev(mddev, tmp) {
 			mdk_rdev_t *rdev2;
 
 			mddev_lock(mddev);
-			rdev_for_each(rdev2, tmp2, mddev)
+			list_for_each_entry(rdev2, &mddev->disks, same_set)
 				if (test_bit(AllReserved, &rdev2->flags) ||
 				    (rdev->bdev == rdev2->bdev &&
 				     rdev != rdev2 &&
@@ -2328,8 +2320,7 @@ abort_free:
 static void analyze_sbs(mddev_t * mddev)
 {
 	int i;
-	struct list_head *tmp;
-	mdk_rdev_t *rdev, *freshest;
+	mdk_rdev_t *rdev, *freshest, *tmp;
 	char b[BDEVNAME_SIZE];
 
 	freshest = NULL;
@@ -3501,7 +3492,6 @@ static int do_md_run(mddev_t * mddev)
 {
 	int err;
 	int chunk_size;
-	struct list_head *tmp;
 	mdk_rdev_t *rdev;
 	struct gendisk *disk;
 	struct mdk_personality *pers;
@@ -3540,7 +3530,7 @@ static int do_md_run(mddev_t * mddev)
 		}
 
 		/* devices must have minimum size of one chunk */
-		rdev_for_each(rdev, tmp, mddev) {
+		list_for_each_entry(rdev, &mddev->disks, same_set) {
 			if (test_bit(Faulty, &rdev->flags))
 				continue;
 			if (rdev->size < chunk_size / 1024) {
@@ -3565,7 +3555,7 @@ static int do_md_run(mddev_t * mddev)
 	 * the only valid external interface is through the md
 	 * device.
 	 */
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		if (test_bit(Faulty, &rdev->flags))
 			continue;
 		sync_blockdev(rdev->bdev);
@@ -3630,10 +3620,10 @@ static int do_md_run(mddev_t * mddev)
 		 */
 		char b[BDEVNAME_SIZE], b2[BDEVNAME_SIZE];
 		mdk_rdev_t *rdev2;
-		struct list_head *tmp2;
 		int warned = 0;
-		rdev_for_each(rdev, tmp, mddev) {
-			rdev_for_each(rdev2, tmp2, mddev) {
+
+		list_for_each_entry(rdev, &mddev->disks, same_set)
+			list_for_each_entry(rdev2, &mddev->disks, same_set) {
 				if (rdev < rdev2 &&
 				    rdev->bdev->bd_contains ==
 				    rdev2->bdev->bd_contains) {
@@ -3647,7 +3637,7 @@ static int do_md_run(mddev_t * mddev)
 					warned = 1;
 				}
 			}
-		}
+
 		if (warned)
 			printk(KERN_WARNING
 			       "True protection against single-disk"
@@ -3695,7 +3685,7 @@ static int do_md_run(mddev_t * mddev)
 	mddev->safemode_delay = (200 * HZ)/1000 +1; /* 200 msec delay */
 	mddev->in_sync = 1;
 
-	rdev_for_each(rdev, tmp, mddev)
+	list_for_each_entry(rdev, &mddev->disks, same_set)
 		if (rdev->raid_disk >= 0) {
 			char nm[20];
 			sprintf(nm, "rd%d", rdev->raid_disk);
@@ -3726,9 +3716,8 @@ static int do_md_run(mddev_t * mddev)
 	 * it will remove the drives and not do the right thing
 	 */
 	if (mddev->degraded && !mddev->sync_thread) {
-		struct list_head *rtmp;
 		int spares = 0;
-		rdev_for_each(rdev, rtmp, mddev)
+		list_for_each_entry(rdev, &mddev->disks, same_set)
 			if (rdev->raid_disk >= 0 &&
 			    !test_bit(In_sync, &rdev->flags) &&
 			    !test_bit(Faulty, &rdev->flags))
@@ -3888,7 +3877,6 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open)
 	 */
 	if (mode == 0) {
 		mdk_rdev_t *rdev;
-		struct list_head *tmp;
 
 		printk(KERN_INFO "md: %s stopped.\n", mdname(mddev));
 
@@ -3900,7 +3888,7 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open)
 		}
 		mddev->bitmap_offset = 0;
 
-		rdev_for_each(rdev, tmp, mddev)
+		list_for_each_entry(rdev, &mddev->disks, same_set)
 			if (rdev->raid_disk >= 0) {
 				char nm[20];
 				sprintf(nm, "rd%d", rdev->raid_disk);
@@ -3961,7 +3949,6 @@ out:
 static void autorun_array(mddev_t *mddev)
 {
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 	int err;
 
 	if (list_empty(&mddev->disks))
@@ -3969,7 +3956,7 @@ static void autorun_array(mddev_t *mddev)
 
 	printk(KERN_INFO "md: running: ");
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		char b[BDEVNAME_SIZE];
 		printk("<%s>", bdevname(rdev->bdev,b));
 	}
@@ -3996,8 +3983,7 @@ static void autorun_array(mddev_t *mddev)
  */
 static void autorun_devices(int part)
 {
-	struct list_head *tmp;
-	mdk_rdev_t *rdev0, *rdev;
+	mdk_rdev_t *rdev0, *rdev, *tmp;
 	mddev_t *mddev;
 	char b[BDEVNAME_SIZE];
 
@@ -4012,7 +3998,7 @@ static void autorun_devices(int part)
 		printk(KERN_INFO "md: considering %s ...\n",
 			bdevname(rdev0->bdev,b));
 		INIT_LIST_HEAD(&candidates);
-		rdev_for_each_list(rdev, tmp, pending_raid_disks)
+		rdev_for_each_list(rdev, tmp, &pending_raid_disks)
 			if (super_90_load(rdev, rdev0, 0) >= 0) {
 				printk(KERN_INFO "md:  adding %s ...\n",
 					bdevname(rdev->bdev,b));
@@ -4058,7 +4044,7 @@ static void autorun_devices(int part)
 		} else {
 			printk(KERN_INFO "md: created %s\n", mdname(mddev));
 			mddev->persistent = 1;
-			rdev_for_each_list(rdev, tmp, candidates) {
+			rdev_for_each_list(rdev, tmp, &candidates) {
 				list_del_init(&rdev->same_set);
 				if (bind_rdev_to_array(rdev, mddev))
 					export_rdev(rdev);
@@ -4069,7 +4055,7 @@ static void autorun_devices(int part)
 		/* on success, candidates will be empty, on error
 		 * it won't...
 		 */
-		rdev_for_each_list(rdev, tmp, candidates) {
+		rdev_for_each_list(rdev, tmp, &candidates) {
 			list_del_init(&rdev->same_set);
 			export_rdev(rdev);
 		}
@@ -4098,10 +4084,9 @@ static int get_array_info(mddev_t * mddev, void __user * arg)
 	mdu_array_info_t info;
 	int nr,working,active,failed,spare;
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 
 	nr=working=active=failed=spare=0;
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		nr++;
 		if (test_bit(Faulty, &rdev->flags))
 			failed++;
@@ -4619,9 +4604,8 @@ static int set_array_info(mddev_t * mddev, mdu_array_info_t *info)
 
 static int update_size(mddev_t *mddev, sector_t num_sectors)
 {
-	mdk_rdev_t * rdev;
+	mdk_rdev_t *rdev;
 	int rv;
-	struct list_head *tmp;
 	int fit = (num_sectors == 0);
 
 	if (mddev->pers->resize == NULL)
@@ -4643,7 +4627,7 @@ static int update_size(mddev_t *mddev, sector_t num_sectors)
 		 * grow, and re-add.
 		 */
 		return -EBUSY;
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		sector_t avail;
 		avail = rdev->size * 2;
 
@@ -5192,11 +5176,10 @@ static void status_unused(struct seq_file *seq)
 {
 	int i = 0;
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 
 	seq_printf(seq, "unused devices: ");
 
-	rdev_for_each_list(rdev, tmp, pending_raid_disks) {
+	list_for_each_entry(rdev, &pending_raid_disks, same_set) {
 		char b[BDEVNAME_SIZE];
 		i++;
 		seq_printf(seq, "%s ",
@@ -5355,7 +5338,6 @@ static int md_seq_show(struct seq_file *seq, void *v)
 {
 	mddev_t *mddev = v;
 	sector_t size;
-	struct list_head *tmp2;
 	mdk_rdev_t *rdev;
 	struct mdstat_info *mi = seq->private;
 	struct bitmap *bitmap;
@@ -5392,7 +5374,7 @@ static int md_seq_show(struct seq_file *seq, void *v)
 		}
 
 		size = 0;
-		rdev_for_each(rdev, tmp2, mddev) {
+		list_for_each_entry(rdev, &mddev->disks, same_set) {
 			char b[BDEVNAME_SIZE];
 			seq_printf(seq, " %s[%d]",
 				bdevname(rdev->bdev,b), rdev->desc_nr);
@@ -5699,7 +5681,6 @@ void md_do_sync(mddev_t *mddev)
 	struct list_head *tmp;
 	sector_t last_check;
 	int skipped = 0;
-	struct list_head *rtmp;
 	mdk_rdev_t *rdev;
 	char *desc;
 
@@ -5804,7 +5785,7 @@ void md_do_sync(mddev_t *mddev)
 		/* recovery follows the physical size of devices */
 		max_sectors = mddev->size << 1;
 		j = MaxSector;
-		rdev_for_each(rdev, rtmp, mddev)
+		list_for_each_entry(rdev, &mddev->disks, same_set)
 			if (rdev->raid_disk >= 0 &&
 			    !test_bit(Faulty, &rdev->flags) &&
 			    !test_bit(In_sync, &rdev->flags) &&
@@ -5954,7 +5935,7 @@ void md_do_sync(mddev_t *mddev)
 		} else {
 			if (!test_bit(MD_RECOVERY_INTR, &mddev->recovery))
 				mddev->curr_resync = MaxSector;
-			rdev_for_each(rdev, rtmp, mddev)
+			list_for_each_entry(rdev, &mddev->disks, same_set)
 				if (rdev->raid_disk >= 0 &&
 				    !test_bit(Faulty, &rdev->flags) &&
 				    !test_bit(In_sync, &rdev->flags) &&
@@ -5990,10 +5971,9 @@ EXPORT_SYMBOL_GPL(md_do_sync);
 static int remove_and_add_spares(mddev_t *mddev)
 {
 	mdk_rdev_t *rdev;
-	struct list_head *rtmp;
 	int spares = 0;
 
-	rdev_for_each(rdev, rtmp, mddev)
+	list_for_each_entry(rdev, &mddev->disks, same_set)
 		if (rdev->raid_disk >= 0 &&
 		    !test_bit(Blocked, &rdev->flags) &&
 		    (test_bit(Faulty, &rdev->flags) ||
@@ -6009,7 +5989,7 @@ static int remove_and_add_spares(mddev_t *mddev)
 		}
 
 	if (mddev->degraded && ! mddev->ro) {
-		rdev_for_each(rdev, rtmp, mddev) {
+		list_for_each_entry(rdev, &mddev->disks, same_set) {
 			if (rdev->raid_disk >= 0 &&
 			    !test_bit(In_sync, &rdev->flags) &&
 			    !test_bit(Blocked, &rdev->flags))
@@ -6061,7 +6041,6 @@ static int remove_and_add_spares(mddev_t *mddev)
 void md_check_recovery(mddev_t *mddev)
 {
 	mdk_rdev_t *rdev;
-	struct list_head *rtmp;
 
 
 	if (mddev->bitmap)
@@ -6125,7 +6104,7 @@ void md_check_recovery(mddev_t *mddev)
 		if (mddev->flags)
 			md_update_sb(mddev, 0);
 
-		rdev_for_each(rdev, rtmp, mddev)
+		list_for_each_entry(rdev, &mddev->disks, same_set)
 			if (test_and_clear_bit(StateChanged, &rdev->flags))
 				sysfs_notify_dirent(rdev->sysfs_state);
 
@@ -6154,7 +6133,7 @@ void md_check_recovery(mddev_t *mddev)
 			 * information must be scrapped
 			 */
 			if (!mddev->degraded)
-				rdev_for_each(rdev, rtmp, mddev)
+				list_for_each_entry(rdev, &mddev->disks, same_set)
 					rdev->saved_raid_disk = -1;
 
 			mddev->recovery = 0;
diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c
index d4ac47d11279..f6d08f241671 100644
--- a/drivers/md/multipath.c
+++ b/drivers/md/multipath.c
@@ -408,7 +408,6 @@ static int multipath_run (mddev_t *mddev)
 	int disk_idx;
 	struct multipath_info *disk;
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 
 	if (mddev->level != LEVEL_MULTIPATH) {
 		printk("multipath: %s: raid level not set to multipath IO (%d)\n",
@@ -441,7 +440,7 @@ static int multipath_run (mddev_t *mddev)
 	}
 
 	conf->working_disks = 0;
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		disk_idx = rdev->raid_disk;
 		if (disk_idx < 0 ||
 		    disk_idx >= mddev->raid_disks)
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index 90f5b24f6e6c..c605ba805586 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -57,7 +57,6 @@ static int create_strip_zones (mddev_t *mddev)
 	sector_t min_spacing;
 	raid0_conf_t *conf = mddev_to_conf(mddev);
 	mdk_rdev_t *smallest, *rdev1, *rdev2, *rdev;
-	struct list_head *tmp1, *tmp2;
 	struct strip_zone *zone;
 	int cnt;
 	char b[BDEVNAME_SIZE];
@@ -67,11 +66,11 @@ static int create_strip_zones (mddev_t *mddev)
 	 */
 	conf->nr_strip_zones = 0;
  
-	rdev_for_each(rdev1, tmp1, mddev) {
+	list_for_each_entry(rdev1, &mddev->disks, same_set) {
 		printk(KERN_INFO "raid0: looking at %s\n",
 			bdevname(rdev1->bdev,b));
 		c = 0;
-		rdev_for_each(rdev2, tmp2, mddev) {
+		list_for_each_entry(rdev2, &mddev->disks, same_set) {
 			printk(KERN_INFO "raid0:   comparing %s(%llu)",
 			       bdevname(rdev1->bdev,b),
 			       (unsigned long long)rdev1->size);
@@ -120,7 +119,7 @@ static int create_strip_zones (mddev_t *mddev)
 	cnt = 0;
 	smallest = NULL;
 	zone->dev = conf->devlist;
-	rdev_for_each(rdev1, tmp1, mddev) {
+	list_for_each_entry(rdev1, &mddev->disks, same_set) {
 		int j = rdev1->raid_disk;
 
 		if (j < 0 || j >= mddev->raid_disks) {
@@ -268,7 +267,6 @@ static int raid0_run (mddev_t *mddev)
 	s64 sectors;
 	raid0_conf_t *conf;
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 
 	if (mddev->chunk_size == 0) {
 		printk(KERN_ERR "md/raid0: non-zero chunk size required.\n");
@@ -294,7 +292,7 @@ static int raid0_run (mddev_t *mddev)
 
 	/* calculate array device size */
 	mddev->array_sectors = 0;
-	rdev_for_each(rdev, tmp, mddev)
+	list_for_each_entry(rdev, &mddev->disks, same_set)
 		mddev->array_sectors += rdev->size * 2;
 
 	printk(KERN_INFO "raid0 : md_size is %llu sectors.\n",
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 9c788e2489b1..c165b1eed8bb 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1919,7 +1919,6 @@ static int run(mddev_t *mddev)
 	int i, j, disk_idx;
 	mirror_info_t *disk;
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 
 	if (mddev->level != 1) {
 		printk("raid1: %s: raid level not set to mirroring (%d)\n",
@@ -1964,7 +1963,7 @@ static int run(mddev_t *mddev)
 	spin_lock_init(&conf->device_lock);
 	mddev->queue->queue_lock = &conf->device_lock;
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		disk_idx = rdev->raid_disk;
 		if (disk_idx >= mddev->raid_disks
 		    || disk_idx < 0)
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 970a96ef9b18..6736d6dff981 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -2025,7 +2025,6 @@ static int run(mddev_t *mddev)
 	int i, disk_idx;
 	mirror_info_t *disk;
 	mdk_rdev_t *rdev;
-	struct list_head *tmp;
 	int nc, fc, fo;
 	sector_t stride, size;
 
@@ -2108,7 +2107,7 @@ static int run(mddev_t *mddev)
 	spin_lock_init(&conf->device_lock);
 	mddev->queue->queue_lock = &conf->device_lock;
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		disk_idx = rdev->raid_disk;
 		if (disk_idx >= mddev->raid_disks
 		    || disk_idx < 0)
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index a36a7435edf5..a5ba080d303b 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -3998,7 +3998,6 @@ static int run(mddev_t *mddev)
 	int raid_disk, memory;
 	mdk_rdev_t *rdev;
 	struct disk_info *disk;
-	struct list_head *tmp;
 	int working_disks = 0;
 
 	if (mddev->level != 5 && mddev->level != 4 && mddev->level != 6) {
@@ -4108,7 +4107,7 @@ static int run(mddev_t *mddev)
 
 	pr_debug("raid5: run(%s) called.\n", mdname(mddev));
 
-	rdev_for_each(rdev, tmp, mddev) {
+	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		raid_disk = rdev->raid_disk;
 		if (raid_disk >= conf->raid_disks
 		    || raid_disk < 0)
@@ -4533,7 +4532,6 @@ static int raid5_start_reshape(mddev_t *mddev)
 {
 	raid5_conf_t *conf = mddev_to_conf(mddev);
 	mdk_rdev_t *rdev;
-	struct list_head *rtmp;
 	int spares = 0;
 	int added_devices = 0;
 	unsigned long flags;
@@ -4541,7 +4539,7 @@ static int raid5_start_reshape(mddev_t *mddev)
 	if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
 		return -EBUSY;
 
-	rdev_for_each(rdev, rtmp, mddev)
+	list_for_each_entry(rdev, &mddev->disks, same_set)
 		if (rdev->raid_disk < 0 &&
 		    !test_bit(Faulty, &rdev->flags))
 			spares++;
@@ -4563,7 +4561,7 @@ static int raid5_start_reshape(mddev_t *mddev)
 	/* Add some new drives, as many as will fit.
 	 * We know there are enough to make the newly sized array work.
 	 */
-	rdev_for_each(rdev, rtmp, mddev)
+	list_for_each_entry(rdev, &mddev->disks, same_set)
 		if (rdev->raid_disk < 0 &&
 		    !test_bit(Faulty, &rdev->flags)) {
 			if (raid5_add_disk(mddev, rdev) == 0) {
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index 663803eaf0de..8f9a54c1fb0e 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -335,17 +335,14 @@ static inline char * mdname (mddev_t * mddev)
  * iterates through some rdev ringlist. It's safe to remove the
  * current 'rdev'. Dont touch 'tmp' though.
  */
-#define rdev_for_each_list(rdev, tmp, list)				\
-									\
-	for ((tmp) = (list).next;					\
-		(rdev) = (list_entry((tmp), mdk_rdev_t, same_set)),	\
-			(tmp) = (tmp)->next, (tmp)->prev != &(list)	\
-		; )
+#define rdev_for_each_list(rdev, tmp, head)				\
+	list_for_each_entry_safe(rdev, tmp, head, same_set)
+
 /*
  * iterates through the 'same array disks' ringlist
  */
 #define rdev_for_each(rdev, tmp, mddev)				\
-	rdev_for_each_list(rdev, tmp, (mddev)->disks)
+	list_for_each_entry_safe(rdev, tmp, &((mddev)->disks), same_set)
 
 #define rdev_for_each_rcu(rdev, mddev)				\
 	list_for_each_entry_rcu(rdev, &((mddev)->disks), same_set)
-- 
cgit v1.2.3


From cd2ac9321c26dc7a76455cd2a4df89123fa2b73e Mon Sep 17 00:00:00 2001
From: Cheng Renquan <crquan@gmail.com>
Date: Fri, 9 Jan 2009 08:31:08 +1100
Subject: md: need another print_sb for mdp_superblock_1

md_print_devices is called in two code path: MD_BUG(...), and md_ioctl
with PRINT_RAID_DEBUG.  it will dump out all in use md devices
information;

However, it wrongly processed two types of superblock in one:

The header file <linux/raid/md_p.h> has defined two types of superblock,
struct mdp_superblock_s (typedefed with mdp_super_t) according to md with
metadata 0.90, and struct mdp_superblock_1 according to md with metadata
1.0 and later,

These two types of superblock are very different,

The md_print_devices code processed them both in mdp_super_t, that would
lead to wrong informaton dump like:

	[ 6742.345877]
	[ 6742.345887] md:	**********************************
	[ 6742.345890] md:	* <COMPLETE RAID STATE PRINTOUT> *
	[ 6742.345892] md:	**********************************
	[ 6742.345896] md1: <ram7><ram6><ram5><ram4>
	[ 6742.345907] md: rdev ram7, SZ:00065472 F:0 S:1 DN:3
	[ 6742.345909] md: rdev superblock:
	[ 6742.345914] md:  SB: (V:0.90.0) ID:<42ef13c7.598c059a.5f9f1645.801e9ee6> CT:4919856d
	[ 6742.345918] md:     L5 S00065472 ND:4 RD:4 md1 LO:2 CS:65536
	[ 6742.345922] md:     UT:4919856d ST:1 AD:4 WD:4 FD:0 SD:0 CSUM:b7992907 E:00000001
	[ 6742.345924]      D  0:  DISK<N:0,(1,8),R:0,S:6>
	[ 6742.345930]      D  1:  DISK<N:1,(1,10),R:1,S:6>
	[ 6742.345933]      D  2:  DISK<N:2,(1,12),R:2,S:6>
	[ 6742.345937]      D  3:  DISK<N:3,(1,14),R:3,S:6>
	[ 6742.345942] md:     THIS:  DISK<N:3,(1,14),R:3,S:6>
	...
	[ 6742.346058] md0: <ram3><ram2><ram1><ram0>
	[ 6742.346067] md: rdev ram3, SZ:00065472 F:0 S:1 DN:3
	[ 6742.346070] md: rdev superblock:
	[ 6742.346073] md:  SB: (V:1.0.0) ID:<369aad81.00000000.00000000.00000000> CT:9a322a9c
	[ 6742.346077] md:     L-1507699579 S976570180 ND:48 RD:0 md0 LO:65536 CS:196610
	[ 6742.346081] md:     UT:00000018 ST:0 AD:131048 WD:0 FD:8 SD:0 CSUM:00000000 E:00000000
	[ 6742.346084]      D  0:  DISK<N:-1,(-1,-1),R:-1,S:-1>
	[ 6742.346089]      D  1:  DISK<N:-1,(-1,-1),R:-1,S:-1>
	[ 6742.346092]      D  2:  DISK<N:-1,(-1,-1),R:-1,S:-1>
	[ 6742.346096]      D  3:  DISK<N:-1,(-1,-1),R:-1,S:-1>
	[ 6742.346102] md:     THIS:  DISK<N:0,(0,0),R:0,S:0>
	...
	[ 6742.346219] md:	**********************************
	[ 6742.346221]

Here md1 is metadata 0.90.0, and md0 is metadata 1.2

After some more code to distinguish these two types of superblock, in this patch,

it will generate dump information like:

	[ 7906.755790]
	[ 7906.755799] md:	**********************************
	[ 7906.755802] md:	* <COMPLETE RAID STATE PRINTOUT> *
	[ 7906.755804] md:	**********************************
	[ 7906.755808] md1: <ram7><ram6><ram5><ram4>
	[ 7906.755819] md: rdev ram7, SZ:00065472 F:0 S:1 DN:3
	[ 7906.755821] md: rdev superblock (MJ:0):
	[ 7906.755826] md:  SB: (V:0.90.0) ID:<3fca7a0d.a612bfed.5f9f1645.801e9ee6> CT:491989f3
	[ 7906.755830] md:     L5 S00065472 ND:4 RD:4 md1 LO:2 CS:65536
	[ 7906.755834] md:     UT:491989f3 ST:1 AD:4 WD:4 FD:0 SD:0 CSUM:00fb52ad E:00000001
	[ 7906.755836]      D  0:  DISK<N:0,(1,8),R:0,S:6>
	[ 7906.755842]      D  1:  DISK<N:1,(1,10),R:1,S:6>
	[ 7906.755845]      D  2:  DISK<N:2,(1,12),R:2,S:6>
	[ 7906.755849]      D  3:  DISK<N:3,(1,14),R:3,S:6>
	[ 7906.755855] md:     THIS:  DISK<N:3,(1,14),R:3,S:6>
	...
	[ 7906.755972] md0: <ram3><ram2><ram1><ram0>
	[ 7906.755981] md: rdev ram3, SZ:00065472 F:0 S:1 DN:3
	[ 7906.755984] md: rdev superblock (MJ:1):
	[ 7906.755989] md:  SB: (V:1) (F:0) Array-ID:<5fbcf158:55aa:5fbe:9a79:1e939880dcbd>
	[ 7906.755990] md:    Name: "DG5:0" CT:1226410480
	[ 7906.755998] md:       L5 SZ130944 RD:4 LO:2 CS:128 DO:24 DS:131048 SO:8 RO:0
	[ 7906.755999] md:     Dev:00000003 UUID: 9194d744:87f7:a448:85f2:7497b84ce30a
	[ 7906.756001] md:       (F:0) UT:1226410480 Events:0 ResyncOffset:-1 CSUM:0dbcd829
	[ 7906.756003] md:         (MaxDev:384)
	...
	[ 7906.756113] md:	**********************************
	[ 7906.756116]

this md0 (metadata 1.2) information dumping is exactly according to struct
mdp_superblock_1.

Signed-off-by: Cheng Renquan <crquan@gmail.com>
Cc: Neil Brown <neilb@suse.de>
Cc: Dan Williams <dan.j.williams@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/md.c           | 66 ++++++++++++++++++++++++++++++++++++++++++-----
 include/linux/raid/md_p.h |  2 ++
 2 files changed, 62 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 58be665bb46b..1f770c16d435 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -1588,7 +1588,7 @@ static void print_desc(mdp_disk_t *desc)
 		desc->major,desc->minor,desc->raid_disk,desc->state);
 }
 
-static void print_sb(mdp_super_t *sb)
+static void print_sb_90(mdp_super_t *sb)
 {
 	int i;
 
@@ -1619,10 +1619,57 @@ static void print_sb(mdp_super_t *sb)
 	}
 	printk(KERN_INFO "md:     THIS: ");
 	print_desc(&sb->this_disk);
-
 }
 
-static void print_rdev(mdk_rdev_t *rdev)
+static void print_sb_1(struct mdp_superblock_1 *sb)
+{
+	__u8 *uuid;
+
+	uuid = sb->set_uuid;
+	printk(KERN_INFO "md:  SB: (V:%u) (F:0x%08x) Array-ID:<%02x%02x%02x%02x"
+			":%02x%02x:%02x%02x:%02x%02x:%02x%02x%02x%02x%02x%02x>\n"
+	       KERN_INFO "md:    Name: \"%s\" CT:%llu\n",
+		le32_to_cpu(sb->major_version),
+		le32_to_cpu(sb->feature_map),
+		uuid[0], uuid[1], uuid[2], uuid[3],
+		uuid[4], uuid[5], uuid[6], uuid[7],
+		uuid[8], uuid[9], uuid[10], uuid[11],
+		uuid[12], uuid[13], uuid[14], uuid[15],
+		sb->set_name,
+		(unsigned long long)le64_to_cpu(sb->ctime)
+		       & MD_SUPERBLOCK_1_TIME_SEC_MASK);
+
+	uuid = sb->device_uuid;
+	printk(KERN_INFO "md:       L%u SZ%llu RD:%u LO:%u CS:%u DO:%llu DS:%llu SO:%llu"
+			" RO:%llu\n"
+	       KERN_INFO "md:     Dev:%08x UUID: %02x%02x%02x%02x:%02x%02x:%02x%02x:%02x%02x"
+			":%02x%02x%02x%02x%02x%02x\n"
+	       KERN_INFO "md:       (F:0x%08x) UT:%llu Events:%llu ResyncOffset:%llu CSUM:0x%08x\n"
+	       KERN_INFO "md:         (MaxDev:%u) \n",
+		le32_to_cpu(sb->level),
+		(unsigned long long)le64_to_cpu(sb->size),
+		le32_to_cpu(sb->raid_disks),
+		le32_to_cpu(sb->layout),
+		le32_to_cpu(sb->chunksize),
+		(unsigned long long)le64_to_cpu(sb->data_offset),
+		(unsigned long long)le64_to_cpu(sb->data_size),
+		(unsigned long long)le64_to_cpu(sb->super_offset),
+		(unsigned long long)le64_to_cpu(sb->recovery_offset),
+		le32_to_cpu(sb->dev_number),
+		uuid[0], uuid[1], uuid[2], uuid[3],
+		uuid[4], uuid[5], uuid[6], uuid[7],
+		uuid[8], uuid[9], uuid[10], uuid[11],
+		uuid[12], uuid[13], uuid[14], uuid[15],
+		sb->devflags,
+		(unsigned long long)le64_to_cpu(sb->utime) & MD_SUPERBLOCK_1_TIME_SEC_MASK,
+		(unsigned long long)le64_to_cpu(sb->events),
+		(unsigned long long)le64_to_cpu(sb->resync_offset),
+		le32_to_cpu(sb->sb_csum),
+		le32_to_cpu(sb->max_dev)
+		);
+}
+
+static void print_rdev(mdk_rdev_t *rdev, int major_version)
 {
 	char b[BDEVNAME_SIZE];
 	printk(KERN_INFO "md: rdev %s, SZ:%08llu F:%d S:%d DN:%u\n",
@@ -1630,8 +1677,15 @@ static void print_rdev(mdk_rdev_t *rdev)
 	        test_bit(Faulty, &rdev->flags), test_bit(In_sync, &rdev->flags),
 	        rdev->desc_nr);
 	if (rdev->sb_loaded) {
-		printk(KERN_INFO "md: rdev superblock:\n");
-		print_sb((mdp_super_t*)page_address(rdev->sb_page));
+		printk(KERN_INFO "md: rdev superblock (MJ:%d):\n", major_version);
+		switch (major_version) {
+		case 0:
+			print_sb_90((mdp_super_t*)page_address(rdev->sb_page));
+			break;
+		case 1:
+			print_sb_1((struct mdp_superblock_1 *)page_address(rdev->sb_page));
+			break;
+		}
 	} else
 		printk(KERN_INFO "md: no rdev superblock!\n");
 }
@@ -1658,7 +1712,7 @@ static void md_print_devices(void)
 		printk("\n");
 
 		list_for_each_entry(rdev, &mddev->disks, same_set)
-			print_rdev(rdev);
+			print_rdev(rdev, mddev->major_version);
 	}
 	printk("md:	**********************************\n");
 	printk("\n");
diff --git a/include/linux/raid/md_p.h b/include/linux/raid/md_p.h
index 8b4de4a41ff1..9491026afe66 100644
--- a/include/linux/raid/md_p.h
+++ b/include/linux/raid/md_p.h
@@ -194,6 +194,8 @@ static inline __u64 md_event(mdp_super_t *sb) {
 	return (ev<<32)| sb->events_lo;
 }
 
+#define MD_SUPERBLOCK_1_TIME_SEC_MASK ((1ULL<<40) - 1)
+
 /*
  * The version-1 superblock :
  * All numeric fields are little-endian.
-- 
cgit v1.2.3


From d3374825ce57ba2214d375023979f6197ccc1385 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Fri, 9 Jan 2009 08:31:10 +1100
Subject: md: make devices disappear when they are no longer needed.

Currently md devices, once created, never disappear until the module
is unloaded.  This is essentially because the gendisk holds a
reference to the mddev, and the mddev holds a reference to the
gendisk, this a circular reference.

If we drop the reference from mddev to gendisk, then we need to ensure
that the mddev is destroyed when the gendisk is destroyed.  However it
is not possible to hook into the gendisk destruction process to enable
this.

So we drop the reference from the gendisk to the mddev and destroy the
gendisk when the mddev gets destroyed.  However this has a
complication.
Between the call
   __blkdev_get->get_gendisk->kobj_lookup->md_probe
and the call
   __blkdev_get->md_open

there is no obvious way to hold a reference on the mddev any more, so
unless something is done, it will disappear and gendisk will be
destroyed prematurely.

Also, once we decide to destroy the mddev, there will be an unlockable
moment before the gendisk is unlinked (blk_unregister_region) during
which a new reference to the gendisk can be created.  We need to
ensure that this reference can not be used.  i.e. the ->open must
fail.

So:
 1/  in md_probe we set a flag in the mddev (hold_active) which
     indicates that the array should be treated as active, even
     though there are no references, and no appearance of activity.
     This is cleared by md_release when the device is closed if it
     is no longer needed.
     This ensures that the gendisk will survive between md_probe and
     md_open.

 2/  In md_open we check if the mddev we expect to open matches
     the gendisk that we did open.
     If there is a mismatch we return -ERESTARTSYS and modify
     __blkdev_get to retry from the top in that case.
     In the -ERESTARTSYS sys case we make sure to wait until
     the old gendisk (that we succeeded in opening) is really gone so
     we loop at most once.

Some udev configurations will always open an md device when it first
appears.   If we allow an md device that was just created by an open
to disappear on an immediate close, then this can race with such udev
configurations and result in an infinite loop the device being opened
and closed, then re-open due to the 'ADD' even from the first open,
and then close and so on.
So we make sure an md device, once created by an open, remains active
at least until some md 'ioctl' has been made on it.  This means that
all normal usage of md devices will allow them to disappear promptly
when not needed, but the worst that an incorrect usage will do it
cause an inactive md device to be left in existence (it can easily be
removed).

As an array can be stopped by writing to a sysfs attribute
  echo clear > /sys/block/mdXXX/md/array_state
we need to use scheduled work for deleting the gendisk and other
kobjects.  This allows us to wait for any pending gendisk deletion to
complete by simply calling flush_scheduled_work().



Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/md.c           | 61 +++++++++++++++++++++++++++++++++++++----------
 fs/block_dev.c            | 14 +++++++++++
 include/linux/raid/md_k.h |  4 ++++
 3 files changed, 67 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 970a8c42ba92..38697283aaf4 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -214,16 +214,33 @@ static inline mddev_t *mddev_get(mddev_t *mddev)
 	return mddev;
 }
 
+static void mddev_delayed_delete(struct work_struct *ws)
+{
+	mddev_t *mddev = container_of(ws, mddev_t, del_work);
+	kobject_del(&mddev->kobj);
+	kobject_put(&mddev->kobj);
+}
+
 static void mddev_put(mddev_t *mddev)
 {
 	if (!atomic_dec_and_lock(&mddev->active, &all_mddevs_lock))
 		return;
-	if (!mddev->raid_disks && list_empty(&mddev->disks)) {
+	if (!mddev->raid_disks && list_empty(&mddev->disks) &&
+	    !mddev->hold_active) {
 		list_del(&mddev->all_mddevs);
-		spin_unlock(&all_mddevs_lock);
-		kobject_put(&mddev->kobj);
-	} else
-		spin_unlock(&all_mddevs_lock);
+		if (mddev->gendisk) {
+			/* we did a probe so need to clean up.
+			 * Call schedule_work inside the spinlock
+			 * so that flush_scheduled_work() after
+			 * mddev_find will succeed in waiting for the
+			 * work to be done.
+			 */
+			INIT_WORK(&mddev->del_work, mddev_delayed_delete);
+			schedule_work(&mddev->del_work);
+		} else
+			kfree(mddev);
+	}
+	spin_unlock(&all_mddevs_lock);
 }
 
 static mddev_t * mddev_find(dev_t unit)
@@ -242,6 +259,7 @@ static mddev_t * mddev_find(dev_t unit)
 
 	if (new) {
 		list_add(&new->all_mddevs, &all_mddevs);
+		mddev->hold_active = UNTIL_IOCTL;
 		spin_unlock(&all_mddevs_lock);
 		return new;
 	}
@@ -3435,6 +3453,8 @@ md_attr_store(struct kobject *kobj, struct attribute *attr,
 	if (!capable(CAP_SYS_ADMIN))
 		return -EACCES;
 	rv = mddev_lock(mddev);
+	if (mddev->hold_active == UNTIL_IOCTL)
+		mddev->hold_active = 0;
 	if (!rv) {
 		rv = entry->store(mddev, page, length);
 		mddev_unlock(mddev);
@@ -3484,6 +3504,11 @@ static struct kobject *md_probe(dev_t dev, int *part, void *data)
 	if (!mddev)
 		return NULL;
 
+	/* wait for any previous instance if this device
+	 * to be completed removed (mddev_delayed_delete).
+	 */
+	flush_scheduled_work();
+
 	mutex_lock(&disks_mutex);
 	if (mddev->gendisk) {
 		mutex_unlock(&disks_mutex);
@@ -3520,7 +3545,7 @@ static struct kobject *md_probe(dev_t dev, int *part, void *data)
 	disk->private_data = mddev;
 	disk->queue = mddev->queue;
 	/* Allow extended partitions.  This makes the
-	 * 'mdp' device redundant, but we can really
+	 * 'mdp' device redundant, but we can't really
 	 * remove it now.
 	 */
 	disk->flags |= GENHD_FL_EXT_DEVT;
@@ -3536,6 +3561,7 @@ static struct kobject *md_probe(dev_t dev, int *part, void *data)
 		kobject_uevent(&mddev->kobj, KOBJ_ADD);
 		mddev->sysfs_state = sysfs_get_dirent(mddev->kobj.sd, "array_state");
 	}
+	mddev_put(mddev);
 	return NULL;
 }
 
@@ -5054,6 +5080,9 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode,
 
 done_unlock:
 abort_unlock:
+	if (mddev->hold_active == UNTIL_IOCTL &&
+	    err != -EINVAL)
+		mddev->hold_active = 0;
 	mddev_unlock(mddev);
 
 	return err;
@@ -5070,14 +5099,25 @@ static int md_open(struct block_device *bdev, fmode_t mode)
 	 * Succeed if we can lock the mddev, which confirms that
 	 * it isn't being stopped right now.
 	 */
-	mddev_t *mddev = bdev->bd_disk->private_data;
+	mddev_t *mddev = mddev_find(bdev->bd_dev);
 	int err;
 
+	if (mddev->gendisk != bdev->bd_disk) {
+		/* we are racing with mddev_put which is discarding this
+		 * bd_disk.
+		 */
+		mddev_put(mddev);
+		/* Wait until bdev->bd_disk is definitely gone */
+		flush_scheduled_work();
+		/* Then retry the open from the top */
+		return -ERESTARTSYS;
+	}
+	BUG_ON(mddev != bdev->bd_disk->private_data);
+
 	if ((err = mutex_lock_interruptible_nested(&mddev->reconfig_mutex, 1)))
 		goto out;
 
 	err = 0;
-	mddev_get(mddev);
 	atomic_inc(&mddev->openers);
 	mddev_unlock(mddev);
 
@@ -6436,11 +6476,8 @@ static __exit void md_exit(void)
 	unregister_sysctl_table(raid_table_header);
 	remove_proc_entry("mdstat", NULL);
 	for_each_mddev(mddev, tmp) {
-		struct gendisk *disk = mddev->gendisk;
-		if (!disk)
-			continue;
 		export_array(mddev);
-		mddev_put(mddev);
+		mddev->hold_active = 0;
 	}
 }
 
diff --git a/fs/block_dev.c b/fs/block_dev.c
index b957717e25ab..8ebbfdf708c2 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1005,6 +1005,7 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 	}
 
 	lock_kernel();
+ restart:
 
 	ret = -ENXIO;
 	disk = get_gendisk(bdev->bd_dev, &partno);
@@ -1025,6 +1026,19 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 
 			if (disk->fops->open) {
 				ret = disk->fops->open(bdev, mode);
+				if (ret == -ERESTARTSYS) {
+					/* Lost a race with 'disk' being
+					 * deleted, try again.
+					 * See md.c
+					 */
+					disk_put_part(bdev->bd_part);
+					bdev->bd_part = NULL;
+					module_put(disk->fops->owner);
+					put_disk(disk);
+					bdev->bd_disk = NULL;
+					mutex_unlock(&bdev->bd_mutex);
+					goto restart;
+				}
 				if (ret)
 					goto out_clear;
 			}
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index 8f9a54c1fb0e..e3d17c7f954e 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -137,6 +137,8 @@ struct mddev_s
 	struct gendisk			*gendisk;
 
 	struct kobject			kobj;
+	int				hold_active;
+#define	UNTIL_IOCTL	1
 
 	/* Superblock information */
 	int				major_version,
@@ -246,6 +248,8 @@ struct mddev_s
 							 */
 	struct sysfs_dirent		*sysfs_action;  /* handle for 'sync_action' */
 
+	struct work_struct del_work;	/* used for delayed sysfs removal */
+
 	spinlock_t			write_lock;
 	wait_queue_head_t		sb_wait;	/* for waiting on superblock updates */
 	atomic_t			pending_writes;	/* number of active superblock writes */
-- 
cgit v1.2.3


From efeb53c0e57213e843b7ef3cc6ebcdea7d6186ac Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Fri, 9 Jan 2009 08:31:10 +1100
Subject: md: Allow md devices to be created by name.

Using sequential numbers to identify md devices is somewhat artificial.
Using names can be a lot more user-friendly.

Also, creating md devices by opening the device special file is a bit
awkward.

So this patch provides a new option for creating and naming devices.

Writing a name such as "md_home" to
    /sys/modules/md_mod/parameters/new_array
will cause an array with that name to be created.  It will appear in
/sys/block/ /proc/partitions and /proc/mdstat as 'md_home'.
It will have an arbitrary minor number allocated.

md devices that a created by an open are destroyed on the last
close when the device is inactive.
For named md devices, they will not be destroyed until the array
is explicitly stopped, either with the STOP_ARRAY ioctl or by
writing 'clear' to /sys/block/md_XXXX/md/array_state.

The name of the array must start 'md_' to avoid conflict with
other devices.

Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/md.c           | 119 +++++++++++++++++++++++++++++++++++++++-------
 include/linux/raid/md_k.h |   1 +
 2 files changed, 102 insertions(+), 18 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 38697283aaf4..f5cbb9d2371a 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -249,17 +249,51 @@ static mddev_t * mddev_find(dev_t unit)
 
  retry:
 	spin_lock(&all_mddevs_lock);
-	list_for_each_entry(mddev, &all_mddevs, all_mddevs)
-		if (mddev->unit == unit) {
-			mddev_get(mddev);
+
+	if (unit) {
+		list_for_each_entry(mddev, &all_mddevs, all_mddevs)
+			if (mddev->unit == unit) {
+				mddev_get(mddev);
+				spin_unlock(&all_mddevs_lock);
+				kfree(new);
+				return mddev;
+			}
+
+		if (new) {
+			list_add(&new->all_mddevs, &all_mddevs);
 			spin_unlock(&all_mddevs_lock);
-			kfree(new);
-			return mddev;
+			new->hold_active = UNTIL_IOCTL;
+			return new;
 		}
-
-	if (new) {
+	} else if (new) {
+		/* find an unused unit number */
+		static int next_minor = 512;
+		int start = next_minor;
+		int is_free = 0;
+		int dev = 0;
+		while (!is_free) {
+			dev = MKDEV(MD_MAJOR, next_minor);
+			next_minor++;
+			if (next_minor > MINORMASK)
+				next_minor = 0;
+			if (next_minor == start) {
+				/* Oh dear, all in use. */
+				spin_unlock(&all_mddevs_lock);
+				kfree(new);
+				return NULL;
+			}
+				
+			is_free = 1;
+			list_for_each_entry(mddev, &all_mddevs, all_mddevs)
+				if (mddev->unit == dev) {
+					is_free = 0;
+					break;
+				}
+		}
+		new->unit = dev;
+		new->md_minor = MINOR(dev);
+		new->hold_active = UNTIL_STOP;
 		list_add(&new->all_mddevs, &all_mddevs);
-		mddev->hold_active = UNTIL_IOCTL;
 		spin_unlock(&all_mddevs_lock);
 		return new;
 	}
@@ -3491,18 +3525,22 @@ static struct kobj_type md_ktype = {
 
 int mdp_major = 0;
 
-static struct kobject *md_probe(dev_t dev, int *part, void *data)
+static int md_alloc(dev_t dev, char *name)
 {
 	static DEFINE_MUTEX(disks_mutex);
 	mddev_t *mddev = mddev_find(dev);
 	struct gendisk *disk;
-	int partitioned = (MAJOR(dev) != MD_MAJOR);
-	int shift = partitioned ? MdpMinorShift : 0;
-	int unit = MINOR(dev) >> shift;
+	int partitioned;
+	int shift;
+	int unit;
 	int error;
 
 	if (!mddev)
-		return NULL;
+		return -ENODEV;
+
+	partitioned = (MAJOR(mddev->unit) != MD_MAJOR);
+	shift = partitioned ? MdpMinorShift : 0;
+	unit = MINOR(mddev->unit) >> shift;
 
 	/* wait for any previous instance if this device
 	 * to be completed removed (mddev_delayed_delete).
@@ -3513,14 +3551,29 @@ static struct kobject *md_probe(dev_t dev, int *part, void *data)
 	if (mddev->gendisk) {
 		mutex_unlock(&disks_mutex);
 		mddev_put(mddev);
-		return NULL;
+		return -EEXIST;
+	}
+
+	if (name) {
+		/* Need to ensure that 'name' is not a duplicate.
+		 */
+		mddev_t *mddev2;
+		spin_lock(&all_mddevs_lock);
+
+		list_for_each_entry(mddev2, &all_mddevs, all_mddevs)
+			if (mddev2->gendisk &&
+			    strcmp(mddev2->gendisk->disk_name, name) == 0) {
+				spin_unlock(&all_mddevs_lock);
+				return -EEXIST;
+			}
+		spin_unlock(&all_mddevs_lock);
 	}
 
 	mddev->queue = blk_alloc_queue(GFP_KERNEL);
 	if (!mddev->queue) {
 		mutex_unlock(&disks_mutex);
 		mddev_put(mddev);
-		return NULL;
+		return -ENOMEM;
 	}
 	/* Can be unlocked because the queue is new: no concurrency */
 	queue_flag_set_unlocked(QUEUE_FLAG_CLUSTER, mddev->queue);
@@ -3533,11 +3586,13 @@ static struct kobject *md_probe(dev_t dev, int *part, void *data)
 		blk_cleanup_queue(mddev->queue);
 		mddev->queue = NULL;
 		mddev_put(mddev);
-		return NULL;
+		return -ENOMEM;
 	}
-	disk->major = MAJOR(dev);
+	disk->major = MAJOR(mddev->unit);
 	disk->first_minor = unit << shift;
-	if (partitioned)
+	if (name)
+		strcpy(disk->disk_name, name);
+	else if (partitioned)
 		sprintf(disk->disk_name, "md_d%d", unit);
 	else
 		sprintf(disk->disk_name, "md%d", unit);
@@ -3562,9 +3617,34 @@ static struct kobject *md_probe(dev_t dev, int *part, void *data)
 		mddev->sysfs_state = sysfs_get_dirent(mddev->kobj.sd, "array_state");
 	}
 	mddev_put(mddev);
+	return 0;
+}
+
+static struct kobject *md_probe(dev_t dev, int *part, void *data)
+{
+	md_alloc(dev, NULL);
 	return NULL;
 }
 
+static int add_named_array(const char *val, struct kernel_param *kp)
+{
+	/* val must be "md_*" where * is not all digits.
+	 * We allocate an array with a large free minor number, and
+	 * set the name to val.  val must not already be an active name.
+	 */
+	int len = strlen(val);
+	char buf[DISK_NAME_LEN];
+
+	while (len && val[len-1] == '\n')
+		len--;
+	if (len >= DISK_NAME_LEN)
+		return -E2BIG;
+	strlcpy(buf, val, len+1);
+	if (strncmp(buf, "md_", 3) != 0)
+		return -EINVAL;
+	return md_alloc(0, buf);
+}
+
 static void md_safemode_timeout(unsigned long data)
 {
 	mddev_t *mddev = (mddev_t *) data;
@@ -4025,6 +4105,8 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open)
 		mddev->barriers_work = 0;
 		mddev->safemode = 0;
 		kobject_uevent(&disk_to_dev(mddev->gendisk)->kobj, KOBJ_CHANGE);
+		if (mddev->hold_active == UNTIL_STOP)
+			mddev->hold_active = 0;
 
 	} else if (mddev->pers)
 		printk(KERN_INFO "md: %s switched to read-only mode.\n",
@@ -6502,6 +6584,7 @@ static int set_ro(const char *val, struct kernel_param *kp)
 module_param_call(start_ro, set_ro, get_ro, NULL, S_IRUSR|S_IWUSR);
 module_param(start_dirty_degraded, int, S_IRUGO|S_IWUSR);
 
+module_param_call(new_array, add_named_array, NULL, NULL, S_IWUSR);
 
 EXPORT_SYMBOL(register_md_personality);
 EXPORT_SYMBOL(unregister_md_personality);
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index e3d17c7f954e..dac4217194b8 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -139,6 +139,7 @@ struct mddev_s
 	struct kobject			kobj;
 	int				hold_active;
 #define	UNTIL_IOCTL	1
+#define	UNTIL_STOP	2
 
 	/* Superblock information */
 	int				major_version,
-- 
cgit v1.2.3


From 4044ba58dd15cb01797c4fd034f39ef4a75f7cc3 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Fri, 9 Jan 2009 08:31:11 +1100
Subject: md: don't retry recovery of raid1 that fails due to error on source
 drive.

If a raid1 has only one working drive and it has a sector which
gives an error on read, then an attempt to recover onto a spare will
fail, but as the single remaining drive is not removed from the
array, the recovery will be immediately re-attempted, resulting
in an infinite recovery loop.

So detect this situation and don't retry recovery once an error
on the lone remaining drive is detected.

Allow recovery to be retried once every time a spare is added
in case the problem wasn't actually a media error.

Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/md.c           | 5 ++++-
 drivers/md/raid1.c        | 8 ++++++--
 include/linux/raid/md_k.h | 3 +++
 3 files changed, 13 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index f5cbb9d2371a..41e2509bf896 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -1500,6 +1500,9 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev)
 
 	list_add_rcu(&rdev->same_set, &mddev->disks);
 	bd_claim_by_disk(rdev->bdev, rdev->bdev->bd_holder, mddev->gendisk);
+
+	/* May as well allow recovery to be retried once */
+	mddev->recovery_disabled = 0;
 	return 0;
 
  fail:
@@ -6175,7 +6178,7 @@ static int remove_and_add_spares(mddev_t *mddev)
 			}
 		}
 
-	if (mddev->degraded && ! mddev->ro) {
+	if (mddev->degraded && ! mddev->ro && !mddev->recovery_disabled) {
 		list_for_each_entry(rdev, &mddev->disks, same_set) {
 			if (rdev->raid_disk >= 0 &&
 			    !test_bit(In_sync, &rdev->flags) &&
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index c165b1eed8bb..7b4f5f7155d8 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1016,12 +1016,16 @@ static void error(mddev_t *mddev, mdk_rdev_t *rdev)
 	 * else mark the drive as failed
 	 */
 	if (test_bit(In_sync, &rdev->flags)
-	    && (conf->raid_disks - mddev->degraded) == 1)
+	    && (conf->raid_disks - mddev->degraded) == 1) {
 		/*
 		 * Don't fail the drive, act as though we were just a
-		 * normal single drive
+		 * normal single drive.
+		 * However don't try a recovery from this drive as
+		 * it is very likely to fail.
 		 */
+		mddev->recovery_disabled = 1;
 		return;
+	}
 	if (test_and_clear_bit(In_sync, &rdev->flags)) {
 		unsigned long flags;
 		spin_lock_irqsave(&conf->device_lock, flags);
diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h
index dac4217194b8..9743e4dbc918 100644
--- a/include/linux/raid/md_k.h
+++ b/include/linux/raid/md_k.h
@@ -218,6 +218,9 @@ struct mddev_s
 #define	MD_RECOVERY_FROZEN	9
 
 	unsigned long			recovery;
+	int				recovery_disabled; /* if we detect that recovery
+							    * will always fail, set this
+							    * so we don't loop trying */
 
 	int				in_sync;	/* know to not need resync */
 	struct mutex			reconfig_mutex;
-- 
cgit v1.2.3


From 871af1210f13966ab911ed2166e4ab2ce775b99d Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@redhat.com>
Date: Mon, 5 Jan 2009 14:16:39 +0000
Subject: libata: Add 32bit PIO support

This matters for some controllers and in one or two cases almost doubles
PIO performance. Add a bmdma32 operations set we can inherit and activate
it for some controllers

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/ata_piix.c    |  2 +-
 drivers/ata/libata-sff.c  | 53 +++++++++++++++++++++++++++++++++++++++++++++++
 drivers/ata/pata_ali.c    |  6 +++---
 drivers/ata/pata_amd.c    |  4 ++--
 drivers/ata/pata_mpiix.c  |  3 ++-
 drivers/ata/pata_sil680.c |  4 ++--
 include/linux/libata.h    |  3 +++
 7 files changed, 66 insertions(+), 9 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index 78659546130c..887d8f46a287 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -310,7 +310,7 @@ static struct scsi_host_template piix_sht = {
 };
 
 static struct ata_port_operations piix_pata_ops = {
-	.inherits		= &ata_bmdma_port_ops,
+	.inherits		= &ata_bmdma32_port_ops,
 	.cable_detect		= ata_cable_40wire,
 	.set_piomode		= piix_set_piomode,
 	.set_dmamode		= piix_set_dmamode,
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 9033d164c4ec..b58549fac460 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -78,6 +78,13 @@ const struct ata_port_operations ata_bmdma_port_ops = {
 	.bmdma_status		= ata_bmdma_status,
 };
 
+const struct ata_port_operations ata_bmdma32_port_ops = {
+	.inherits		= &ata_bmdma_port_ops,
+
+	.sff_data_xfer		= ata_sff_data_xfer32,
+};
+EXPORT_SYMBOL_GPL(ata_bmdma32_port_ops);
+
 /**
  *	ata_fill_sg - Fill PCI IDE PRD table
  *	@qc: Metadata associated with taskfile to be transferred
@@ -718,6 +725,52 @@ unsigned int ata_sff_data_xfer(struct ata_device *dev, unsigned char *buf,
 	return words << 1;
 }
 
+/**
+ *	ata_sff_data_xfer32 - Transfer data by PIO
+ *	@dev: device to target
+ *	@buf: data buffer
+ *	@buflen: buffer length
+ *	@rw: read/write
+ *
+ *	Transfer data from/to the device data register by PIO using 32bit
+ *	I/O operations.
+ *
+ *	LOCKING:
+ *	Inherited from caller.
+ *
+ *	RETURNS:
+ *	Bytes consumed.
+ */
+
+unsigned int ata_sff_data_xfer32(struct ata_device *dev, unsigned char *buf,
+			       unsigned int buflen, int rw)
+{
+	struct ata_port *ap = dev->link->ap;
+	void __iomem *data_addr = ap->ioaddr.data_addr;
+	unsigned int words = buflen >> 2;
+	int slop = buflen & 3;
+
+	/* Transfer multiple of 4 bytes */
+	if (rw == READ)
+		ioread32_rep(data_addr, buf, words);
+	else
+		iowrite32_rep(data_addr, buf, words);
+
+	if (unlikely(slop)) {
+		__le32 pad;
+		if (rw == READ) {
+			pad = cpu_to_le32(ioread32(ap->ioaddr.data_addr));
+			memcpy(buf + buflen - slop, &pad, slop);
+		} else {
+			memcpy(&pad, buf + buflen - slop, slop);
+			iowrite32(le32_to_cpu(pad), ap->ioaddr.data_addr);
+		}
+		words++;
+	}
+	return words << 2;
+}
+EXPORT_SYMBOL_GPL(ata_sff_data_xfer32);
+
 /**
  *	ata_sff_data_xfer_noirq - Transfer data by PIO
  *	@dev: device to target
diff --git a/drivers/ata/pata_ali.c b/drivers/ata/pata_ali.c
index a4f9e39442c6..a7999c19f0c9 100644
--- a/drivers/ata/pata_ali.c
+++ b/drivers/ata/pata_ali.c
@@ -151,8 +151,7 @@ static void ali_fifo_control(struct ata_port *ap, struct ata_device *adev, int o
 
 	pci_read_config_byte(pdev, pio_fifo, &fifo);
 	fifo &= ~(0x0F << shift);
-	if (on)
-		fifo |= (on << shift);
+	fifo |= (on << shift);
 	pci_write_config_byte(pdev, pio_fifo, fifo);
 }
 
@@ -370,10 +369,11 @@ static struct ata_port_operations ali_early_port_ops = {
 	.inherits	= &ata_sff_port_ops,
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= ali_set_piomode,
+	.sff_data_xfer  = ata_sff_data_xfer32,
 };
 
 static const struct ata_port_operations ali_dma_base_ops = {
-	.inherits	= &ata_bmdma_port_ops,
+	.inherits	= &ata_bmdma32_port_ops,
 	.set_piomode	= ali_set_piomode,
 	.set_dmamode	= ali_set_dmamode,
 };
diff --git a/drivers/ata/pata_amd.c b/drivers/ata/pata_amd.c
index 0ec9c7d9fe9d..63719ab9ea44 100644
--- a/drivers/ata/pata_amd.c
+++ b/drivers/ata/pata_amd.c
@@ -24,7 +24,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME "pata_amd"
-#define DRV_VERSION "0.3.10"
+#define DRV_VERSION "0.3.11"
 
 /**
  *	timing_setup		-	shared timing computation and load
@@ -345,7 +345,7 @@ static struct scsi_host_template amd_sht = {
 };
 
 static const struct ata_port_operations amd_base_port_ops = {
-	.inherits	= &ata_bmdma_port_ops,
+	.inherits	= &ata_bmdma32_port_ops,
 	.prereset	= amd_pre_reset,
 };
 
diff --git a/drivers/ata/pata_mpiix.c b/drivers/ata/pata_mpiix.c
index 7c8faa48b5f3..aa576cac4d17 100644
--- a/drivers/ata/pata_mpiix.c
+++ b/drivers/ata/pata_mpiix.c
@@ -35,7 +35,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME "pata_mpiix"
-#define DRV_VERSION "0.7.6"
+#define DRV_VERSION "0.7.7"
 
 enum {
 	IDETIM = 0x6C,		/* IDE control register */
@@ -146,6 +146,7 @@ static struct ata_port_operations mpiix_port_ops = {
 	.cable_detect	= ata_cable_40wire,
 	.set_piomode	= mpiix_set_piomode,
 	.prereset	= mpiix_pre_reset,
+	.sff_data_xfer	= ata_sff_data_xfer32,
 };
 
 static int mpiix_init_one(struct pci_dev *dev, const struct pci_device_id *id)
diff --git a/drivers/ata/pata_sil680.c b/drivers/ata/pata_sil680.c
index 83580a59db58..9e764e5747e6 100644
--- a/drivers/ata/pata_sil680.c
+++ b/drivers/ata/pata_sil680.c
@@ -32,7 +32,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME "pata_sil680"
-#define DRV_VERSION "0.4.8"
+#define DRV_VERSION "0.4.9"
 
 #define SIL680_MMIO_BAR		5
 
@@ -195,7 +195,7 @@ static struct scsi_host_template sil680_sht = {
 };
 
 static struct ata_port_operations sil680_port_ops = {
-	.inherits	= &ata_bmdma_port_ops,
+	.inherits	= &ata_bmdma32_port_ops,
 	.cable_detect	= sil680_cable_detect,
 	.set_piomode	= sil680_set_piomode,
 	.set_dmamode	= sil680_set_dmamode,
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 3449de597eff..4f7c8fb4d3fe 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1518,6 +1518,7 @@ extern void sata_pmp_error_handler(struct ata_port *ap);
 
 extern const struct ata_port_operations ata_sff_port_ops;
 extern const struct ata_port_operations ata_bmdma_port_ops;
+extern const struct ata_port_operations ata_bmdma32_port_ops;
 
 /* PIO only, sg_tablesize and dma_boundary limits can be removed */
 #define ATA_PIO_SHT(drv_name)					\
@@ -1545,6 +1546,8 @@ extern void ata_sff_exec_command(struct ata_port *ap,
 				 const struct ata_taskfile *tf);
 extern unsigned int ata_sff_data_xfer(struct ata_device *dev,
 			unsigned char *buf, unsigned int buflen, int rw);
+extern unsigned int ata_sff_data_xfer32(struct ata_device *dev,
+			unsigned char *buf, unsigned int buflen, int rw);
 extern unsigned int ata_sff_data_xfer_noirq(struct ata_device *dev,
 			unsigned char *buf, unsigned int buflen, int rw);
 extern u8 ata_sff_irq_on(struct ata_port *ap);
-- 
cgit v1.2.3


From fefae48bf8caab7d56ee4f8181f06602cf73d29e Mon Sep 17 00:00:00 2001
From: Wolfgang Grandegger <wg@grandegger.com>
Date: Thu, 8 Jan 2009 19:21:27 +0100
Subject: [MTD] CFI: remove major/minor version check for command set 0x0002

The NOR Flash memory K8P2815UQB from Samsung uses the major version
number '0'. Add a quirk to cope with it.

Signed-off-by: Wolfgang Grandegger <wg@grandegger.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/chips/cfi_cmdset_0002.c | 10 ++++++++++
 include/linux/mtd/cfi.h             |  1 +
 2 files changed, 11 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/mtd/chips/cfi_cmdset_0002.c b/drivers/mtd/chips/cfi_cmdset_0002.c
index f9c435a42670..94bb61e19047 100644
--- a/drivers/mtd/chips/cfi_cmdset_0002.c
+++ b/drivers/mtd/chips/cfi_cmdset_0002.c
@@ -322,6 +322,14 @@ static struct cfi_fixup fixup_table[] = {
 };
 
 
+static void cfi_fixup_major_minor(struct cfi_private *cfi,
+				  struct cfi_pri_amdstd *extp)
+{
+	if (cfi->mfr == CFI_MFR_SAMSUNG && cfi->id == 0x257e &&
+	    extp->MajorVersion == '0')
+		extp->MajorVersion = '1';
+}
+
 struct mtd_info *cfi_cmdset_0002(struct map_info *map, int primary)
 {
 	struct cfi_private *cfi = map->fldrv_priv;
@@ -363,6 +371,8 @@ struct mtd_info *cfi_cmdset_0002(struct map_info *map, int primary)
 			return NULL;
 		}
 
+		cfi_fixup_major_minor(cfi, extp);
+
 		if (extp->MajorVersion != '1' ||
 		    (extp->MinorVersion < '0' || extp->MinorVersion > '4')) {
 			printk(KERN_ERR "  Unknown Amd/Fujitsu Extended Query "
diff --git a/include/linux/mtd/cfi.h b/include/linux/mtd/cfi.h
index 00e2b575021f..88d3d8fbf9f2 100644
--- a/include/linux/mtd/cfi.h
+++ b/include/linux/mtd/cfi.h
@@ -520,6 +520,7 @@ struct cfi_fixup {
 
 #define CFI_MFR_AMD 0x0001
 #define CFI_MFR_ATMEL 0x001F
+#define CFI_MFR_SAMSUNG 0x00EC
 #define CFI_MFR_ST  0x0020 	/* STMicroelectronics */
 
 void cfi_fixup(struct mtd_info *mtd, struct cfi_fixup* fixups);
-- 
cgit v1.2.3


From 8dd2f36f317569665e454268a2677cfba3e848f1 Mon Sep 17 00:00:00 2001
From: Andreas Eversberg <andreas@eversberg.eu>
Date: Sat, 2 Aug 2008 22:51:52 +0200
Subject: mISDN: Add feature via MISDN_CTRL_FILL_EMPTY to fill fifo if empty

This prevents underrun of fifo when filled and in case of an underrun it
prevents subsequent underruns due to jitter.
Improve dsp, so buffers are kept filled with a certain delay, so moderate
jitter will not cause underrun all the time -> the audio quality is highly
improved. tones are not interrupted by gaps anymore, except when CPU is
stalling or in high load.

Signed-off-by: Andreas Eversberg <andreas@eversberg.eu>
Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 drivers/isdn/hardware/mISDN/hfc_multi.h |  3 ++
 drivers/isdn/hardware/mISDN/hfc_pci.h   |  1 +
 drivers/isdn/hardware/mISDN/hfcmulti.c  | 45 ++++++++++++++++++++------
 drivers/isdn/hardware/mISDN/hfcpci.c    | 41 ++++++++++++++++++++++--
 drivers/isdn/mISDN/dsp.h                |  1 +
 drivers/isdn/mISDN/dsp_cmx.c            | 56 ++++++++++++++++++++++-----------
 drivers/isdn/mISDN/dsp_core.c           | 37 ++++++++++++++++++++--
 include/linux/mISDNhw.h                 | 25 ++++++++-------
 include/linux/mISDNif.h                 |  1 +
 9 files changed, 165 insertions(+), 45 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/isdn/hardware/mISDN/hfc_multi.h b/drivers/isdn/hardware/mISDN/hfc_multi.h
index 6d74951a1797..4aa6a8b41f50 100644
--- a/drivers/isdn/hardware/mISDN/hfc_multi.h
+++ b/drivers/isdn/hardware/mISDN/hfc_multi.h
@@ -9,6 +9,7 @@
 #define	DEBUG_HFCMULTI_MODE	0x00100000
 #define	DEBUG_HFCMULTI_MSG	0x00200000
 #define	DEBUG_HFCMULTI_STATE	0x00400000
+#define	DEBUG_HFCMULTI_FILL	0x00800000
 #define	DEBUG_HFCMULTI_SYNC	0x01000000
 #define	DEBUG_HFCMULTI_DTMF	0x02000000
 #define	DEBUG_HFCMULTI_LOCK	0x80000000
@@ -166,6 +167,8 @@ struct hfc_multi {
 
 	u_long		chip;	/* chip configuration */
 	int		masterclk; /* port that provides master clock -1=off */
+	unsigned char	silence;/* silence byte */
+	unsigned char	silence_data[128];/* silence block */
 	int		dtmf;	/* flag that dtmf is currently in process */
 	int		Flen;	/* F-buffer size */
 	int		Zlen;	/* Z-buffer size (must be int for calculation)*/
diff --git a/drivers/isdn/hardware/mISDN/hfc_pci.h b/drivers/isdn/hardware/mISDN/hfc_pci.h
index 5783d22a18fe..fd9241ab1802 100644
--- a/drivers/isdn/hardware/mISDN/hfc_pci.h
+++ b/drivers/isdn/hardware/mISDN/hfc_pci.h
@@ -27,6 +27,7 @@
  */
 #define HFCPCI_BTRANS_THRESHOLD 128
 #define HFCPCI_BTRANS_MAX	256
+#define HFCPCI_FILLEMPTY	64
 #define HFCPCI_BTRANS_THRESMASK 0x00
 
 /* defines for PCI config */
diff --git a/drivers/isdn/hardware/mISDN/hfcmulti.c b/drivers/isdn/hardware/mISDN/hfcmulti.c
index ff5ec3cbeb77..3fc2e9d95341 100644
--- a/drivers/isdn/hardware/mISDN/hfcmulti.c
+++ b/drivers/isdn/hardware/mISDN/hfcmulti.c
@@ -180,7 +180,6 @@ static int nt_t1_count[] = { 3840, 1920, 960, 480, 240, 120, 60, 30  };
 #define	CLKDEL_TE	0x0f	/* CLKDEL in TE mode */
 #define	CLKDEL_NT	0x6c	/* CLKDEL in NT mode
 				   (0x60 MUST be included!) */
-static u_char silence =	0xff;	/* silence by LAW */
 
 #define	DIP_4S	0x1		/* DIP Switches for Beronet 1S/2S/4S cards */
 #define	DIP_8S	0x2		/* DIP Switches for Beronet 8S+ cards */
@@ -1975,6 +1974,17 @@ next_frame:
 		return; /* no data */
 	}
 
+	/* "fill fifo if empty" feature */
+	if (bch && test_bit(FLG_FILLEMPTY, &bch->Flags)
+		&& !test_bit(FLG_HDLC, &bch->Flags) && z2 == z1) {
+		if (debug & DEBUG_HFCMULTI_FILL)
+			printk(KERN_DEBUG "%s: buffer empty, so we have "
+				"underrun\n", __func__);
+		/* fill buffer, to prevent future underrun */
+		hc->write_fifo(hc, hc->silence_data, poll >> 1);
+		Zspace -= (poll >> 1);
+	}
+
 	/* if audio data and connected slot */
 	if (bch && (!test_bit(FLG_HDLC, &bch->Flags)) && (!*txpending)
 		&& slot_tx >= 0) {
@@ -2011,7 +2021,6 @@ next_frame:
 			__func__, hc->id + 1, ch, Zspace, z1, z2, ii-i, len-i,
 			temp ? "HDLC":"TRANS");
 
-
 	/* Have to prep the audio data */
 	hc->write_fifo(hc, d, ii - i);
 	*idxp = ii;
@@ -2050,7 +2059,7 @@ next_frame:
 	 * no more data at all. this prevents sending an undefined value.
 	 */
 	if (bch && test_bit(FLG_TRANSPARENT, &bch->Flags))
-		HFC_outb_nodebug(hc, A_FIFO_DATA0_NOINC, silence);
+		HFC_outb_nodebug(hc, A_FIFO_DATA0_NOINC, hc->silence);
 }
 
 
@@ -2932,7 +2941,7 @@ mode_hfcmulti(struct hfc_multi *hc, int ch, int protocol, int slot_tx,
 			HFC_outb(hc, R_INC_RES_FIFO, V_RES_F);
 			HFC_wait(hc);
 			/* tx silence */
-			HFC_outb_nodebug(hc, A_FIFO_DATA0_NOINC, silence);
+			HFC_outb_nodebug(hc, A_FIFO_DATA0_NOINC, hc->silence);
 			HFC_outb(hc, R_SLOT, (((ch / 4) * 8) +
 			    ((ch % 4) * 4)) << 1);
 			HFC_outb(hc, A_SL_CFG, 0x80 | 0x20 | (ch << 1));
@@ -2947,7 +2956,7 @@ mode_hfcmulti(struct hfc_multi *hc, int ch, int protocol, int slot_tx,
 			HFC_outb(hc, R_INC_RES_FIFO, V_RES_F);
 			HFC_wait(hc);
 			/* tx silence */
-			HFC_outb_nodebug(hc, A_FIFO_DATA0_NOINC, silence);
+			HFC_outb_nodebug(hc, A_FIFO_DATA0_NOINC, hc->silence);
 			/* enable RX fifo */
 			HFC_outb(hc, R_FIFO, (ch<<1)|1);
 			HFC_wait(hc);
@@ -3439,7 +3448,7 @@ channel_bctrl(struct bchannel *bch, struct mISDN_ctrl_req *cq)
 	switch (cq->op) {
 	case MISDN_CTRL_GETOP:
 		cq->op = MISDN_CTRL_HFC_OP | MISDN_CTRL_HW_FEATURES_OP
-			| MISDN_CTRL_RX_OFF;
+			| MISDN_CTRL_RX_OFF | MISDN_CTRL_FILL_EMPTY;
 		break;
 	case MISDN_CTRL_RX_OFF: /* turn off / on rx stream */
 		hc->chan[bch->slot].rx_off = !!cq->p1;
@@ -3454,6 +3463,12 @@ channel_bctrl(struct bchannel *bch, struct mISDN_ctrl_req *cq)
 			printk(KERN_DEBUG "%s: RX_OFF request (nr=%d off=%d)\n",
 			    __func__, bch->nr, hc->chan[bch->slot].rx_off);
 		break;
+	case MISDN_CTRL_FILL_EMPTY: /* fill fifo, if empty */
+		test_and_set_bit(FLG_FILLEMPTY, &bch->Flags);
+		if (debug & DEBUG_HFCMULTI_MSG)
+			printk(KERN_DEBUG "%s: FILL_EMPTY request (nr=%d "
+				"off=%d)\n", __func__, bch->nr, !!cq->p1);
+		break;
 	case MISDN_CTRL_HW_FEATURES: /* fill features structure */
 		if (debug & DEBUG_HFCMULTI_MSG)
 			printk(KERN_DEBUG "%s: HW_FEATURE request\n",
@@ -3970,6 +3985,7 @@ open_bchannel(struct hfc_multi *hc, struct dchannel *dch,
 	}
 	if (test_and_set_bit(FLG_OPEN, &bch->Flags))
 		return -EBUSY; /* b-channel can be only open once */
+	test_and_clear_bit(FLG_FILLEMPTY, &bch->Flags);
 	bch->ch.protocol = rq->protocol;
 	hc->chan[ch].rx_off = 0;
 	rq->ch = &bch->ch;
@@ -4806,6 +4822,7 @@ hfcmulti_init(struct pci_dev *pdev, const struct pci_device_id *ent)
 	struct hfc_multi	*hc;
 	u_long		flags;
 	u_char		dips = 0, pmj = 0; /* dip settings, port mode Jumpers */
+	int		i;
 
 	if (HFC_cnt >= MAX_CARDS) {
 		printk(KERN_ERR "too many cards (max=%d).\n",
@@ -4839,11 +4856,11 @@ hfcmulti_init(struct pci_dev *pdev, const struct pci_device_id *ent)
 	hc->id = HFC_cnt;
 	hc->pcm = pcm[HFC_cnt];
 	hc->io_mode = iomode[HFC_cnt];
-	if (dslot[HFC_cnt] < 0) {
+	if (dslot[HFC_cnt] < 0 && hc->type == 1) {
 		hc->dslot = 0;
 		printk(KERN_INFO "HFC-E1 card has disabled D-channel, but "
 			"31 B-channels\n");
-	} if (dslot[HFC_cnt] > 0 && dslot[HFC_cnt] < 32) {
+	} if (dslot[HFC_cnt] > 0 && dslot[HFC_cnt] < 32 && hc->type == 1) {
 		hc->dslot = dslot[HFC_cnt];
 		printk(KERN_INFO "HFC-E1 card has alternating D-channel on "
 			"time slot %d\n", dslot[HFC_cnt]);
@@ -4854,9 +4871,17 @@ hfcmulti_init(struct pci_dev *pdev, const struct pci_device_id *ent)
 	hc->masterclk = -1;
 	if (type[HFC_cnt] & 0x100) {
 		test_and_set_bit(HFC_CHIP_ULAW, &hc->chip);
-		silence = 0xff; /* ulaw silence */
+		hc->silence = 0xff; /* ulaw silence */
 	} else
-		silence = 0x2a; /* alaw silence */
+		hc->silence = 0x2a; /* alaw silence */
+	if ((poll >> 1) > sizeof(hc->silence_data)) {
+		printk(KERN_ERR "HFCMULTI error: silence_data too small, "
+			"please fix\n");
+		return -EINVAL;
+	}
+	for (i = 0; i < (poll >> 1); i++)
+		hc->silence_data[i] = hc->silence;
+
 	if (!(type[HFC_cnt] & 0x200))
 		test_and_set_bit(HFC_CHIP_DTMF, &hc->chip);
 
diff --git a/drivers/isdn/hardware/mISDN/hfcpci.c b/drivers/isdn/hardware/mISDN/hfcpci.c
index cd8302af40eb..80c356e5dbe1 100644
--- a/drivers/isdn/hardware/mISDN/hfcpci.c
+++ b/drivers/isdn/hardware/mISDN/hfcpci.c
@@ -751,6 +751,36 @@ hfcpci_fill_fifo(struct bchannel *bch)
 			    /* fcnt contains available bytes in fifo */
 		fcnt = B_FIFO_SIZE - fcnt;
 		    /* remaining bytes to send (bytes in fifo) */
+
+		/* "fill fifo if empty" feature */
+		if (test_bit(FLG_FILLEMPTY, &bch->Flags) && !fcnt) {
+			/* printk(KERN_DEBUG "%s: buffer empty, so we have "
+				"underrun\n", __func__); */
+			/* fill buffer, to prevent future underrun */
+			count = HFCPCI_FILLEMPTY;
+			new_z1 = le16_to_cpu(*z1t) + count;
+			   /* new buffer Position */
+			if (new_z1 >= (B_FIFO_SIZE + B_SUB_VAL))
+				new_z1 -= B_FIFO_SIZE;	/* buffer wrap */
+			dst = bdata + (le16_to_cpu(*z1t) - B_SUB_VAL);
+			maxlen = (B_FIFO_SIZE + B_SUB_VAL) - le16_to_cpu(*z1t);
+			    /* end of fifo */
+			if (bch->debug & DEBUG_HW_BFIFO)
+				printk(KERN_DEBUG "hfcpci_FFt fillempty "
+				    "fcnt(%d) maxl(%d) nz1(%x) dst(%p)\n",
+				    fcnt, maxlen, new_z1, dst);
+			fcnt += count;
+			if (maxlen > count)
+				maxlen = count; 	/* limit size */
+			memset(dst, 0x2a, maxlen);	/* first copy */
+			count -= maxlen;		/* remaining bytes */
+			if (count) {
+				dst = bdata;		/* start of buffer */
+				memset(dst, 0x2a, count);
+			}
+			*z1t = cpu_to_le16(new_z1);	/* now send data */
+		}
+
 next_t_frame:
 		count = bch->tx_skb->len - bch->tx_idx;
 		/* maximum fill shall be HFCPCI_BTRANS_MAX */
@@ -1481,11 +1511,17 @@ deactivate_bchannel(struct bchannel *bch)
 static int
 channel_bctrl(struct bchannel *bch, struct mISDN_ctrl_req *cq)
 {
-	int			ret = 0;
+	int	ret = 0;
 
 	switch (cq->op) {
 	case MISDN_CTRL_GETOP:
-		cq->op = 0;
+		cq->op = MISDN_CTRL_FILL_EMPTY;
+		break;
+	case MISDN_CTRL_FILL_EMPTY: /* fill fifo, if empty */
+		test_and_set_bit(FLG_FILLEMPTY, &bch->Flags);
+		if (debug & DEBUG_HW_OPEN)
+			printk(KERN_DEBUG "%s: FILL_EMPTY request (nr=%d "
+				"off=%d)\n", __func__, bch->nr, !!cq->p1);
 		break;
 	default:
 		printk(KERN_WARNING "%s: unknown Op %x\n", __func__, cq->op);
@@ -1903,6 +1939,7 @@ open_bchannel(struct hfc_pci *hc, struct channel_req *rq)
 	bch = &hc->bch[rq->adr.channel - 1];
 	if (test_and_set_bit(FLG_OPEN, &bch->Flags))
 		return -EBUSY; /* b-channel can be only open once */
+	test_and_clear_bit(FLG_FILLEMPTY, &bch->Flags);
 	bch->ch.protocol = rq->protocol;
 	rq->ch = &bch->ch; /* TODO: E-channel */
 	if (!try_module_get(THIS_MODULE))
diff --git a/drivers/isdn/mISDN/dsp.h b/drivers/isdn/mISDN/dsp.h
index 6c3fed6b8d4f..ded948069335 100644
--- a/drivers/isdn/mISDN/dsp.h
+++ b/drivers/isdn/mISDN/dsp.h
@@ -198,6 +198,7 @@ struct dsp {
 	/* hardware stuff */
 	struct dsp_features features;
 	int		features_rx_off; /* set if rx_off is featured */
+	int		features_fill_empty; /* set if fill_empty is featured */
 	int		pcm_slot_rx; /* current PCM slot (or -1) */
 	int		pcm_bank_rx;
 	int		pcm_slot_tx;
diff --git a/drivers/isdn/mISDN/dsp_cmx.c b/drivers/isdn/mISDN/dsp_cmx.c
index c884511e2d49..fc8ea41ae6a2 100644
--- a/drivers/isdn/mISDN/dsp_cmx.c
+++ b/drivers/isdn/mISDN/dsp_cmx.c
@@ -1168,11 +1168,18 @@ dsp_cmx_receive(struct dsp *dsp, struct sk_buff *skb)
 		dsp->rx_init = 0;
 		if (dsp->features.unordered) {
 			dsp->rx_R = (hh->id & CMX_BUFF_MASK);
-			dsp->rx_W = (dsp->rx_R + dsp->cmx_delay)
-				& CMX_BUFF_MASK;
+			if (dsp->cmx_delay)
+				dsp->rx_W = (dsp->rx_R + dsp->cmx_delay)
+					& CMX_BUFF_MASK;
+			else
+				dsp->rx_W = (dsp->rx_R + (dsp_poll >> 1))
+					& CMX_BUFF_MASK;
 		} else {
 			dsp->rx_R = 0;
-			dsp->rx_W = dsp->cmx_delay;
+			if (dsp->cmx_delay)
+				dsp->rx_W = dsp->cmx_delay;
+			else
+				dsp->rx_W = dsp_poll >> 1;
 		}
 	}
 	/* if frame contains time code, write directly */
@@ -1190,14 +1197,20 @@ dsp_cmx_receive(struct dsp *dsp, struct sk_buff *skb)
 			    "cmx_receive(dsp=%lx): UNDERRUN (or overrun the "
 			    "maximum delay), adjusting read pointer! "
 			    "(inst %s)\n", (u_long)dsp, dsp->name);
-		/* flush buffer */
+		/* flush rx buffer and set delay to dsp_poll / 2 */
 		if (dsp->features.unordered) {
 			dsp->rx_R = (hh->id & CMX_BUFF_MASK);
-			dsp->rx_W = (dsp->rx_R + dsp->cmx_delay)
-				& CMX_BUFF_MASK;
+			if (dsp->cmx_delay)
+				dsp->rx_W = (dsp->rx_R + dsp->cmx_delay)
+					& CMX_BUFF_MASK;
+				dsp->rx_W = (dsp->rx_R + (dsp_poll >> 1))
+					& CMX_BUFF_MASK;
 		} else {
 			dsp->rx_R = 0;
-			dsp->rx_W = dsp->cmx_delay;
+			if (dsp->cmx_delay)
+				dsp->rx_W = dsp->cmx_delay;
+			else
+				dsp->rx_W = dsp_poll >> 1;
 		}
 		memset(dsp->rx_buff, dsp_silence, sizeof(dsp->rx_buff));
 	}
@@ -1360,8 +1373,11 @@ dsp_cmx_send_member(struct dsp *dsp, int len, s32 *c, int members)
 				t = (t+1) & CMX_BUFF_MASK;
 				r = (r+1) & CMX_BUFF_MASK;
 			}
-			if (r != rr)
+			if (r != rr) {
+				printk(KERN_DEBUG "%s: buffer empty\n",
+					__func__);
 				memset(d, dsp_silence, (rr-r)&CMX_BUFF_MASK);
+			}
 		/* -> if echo is enabled */
 		} else {
 			/*
@@ -1704,9 +1720,10 @@ dsp_cmx_send(void *arg)
 			}
 			/*
 			 * remove rx_delay only if we have delay AND we
-			 * have not preset cmx_delay
+			 * have not preset cmx_delay AND
+			 * the delay is greater dsp_poll
 			 */
-			if (delay && !dsp->cmx_delay) {
+			if (delay > dsp_poll && !dsp->cmx_delay) {
 				if (dsp_debug & DEBUG_DSP_CMX)
 					printk(KERN_DEBUG
 					    "%s lowest rx_delay of %d bytes for"
@@ -1714,7 +1731,8 @@ dsp_cmx_send(void *arg)
 					    __func__, delay,
 					    dsp->name);
 				r = dsp->rx_R;
-				rr = (r + delay) & CMX_BUFF_MASK;
+				rr = (r + delay - (dsp_poll >> 1))
+					& CMX_BUFF_MASK;
 				/* delete rx-data */
 				while (r != rr) {
 					p[r] = dsp_silence;
@@ -1736,7 +1754,7 @@ dsp_cmx_send(void *arg)
 			 * remove delay only if we have delay AND we
 			 * have enabled tx_dejitter
 			 */
-			if (delay && dsp->tx_dejitter) {
+			if (delay > dsp_poll && dsp->tx_dejitter) {
 				if (dsp_debug & DEBUG_DSP_CMX)
 					printk(KERN_DEBUG
 					    "%s lowest tx_delay of %d bytes for"
@@ -1744,7 +1762,8 @@ dsp_cmx_send(void *arg)
 					    __func__, delay,
 					    dsp->name);
 				r = dsp->tx_R;
-				rr = (r + delay) & CMX_BUFF_MASK;
+				rr = (r + delay - (dsp_poll >> 1))
+					& CMX_BUFF_MASK;
 				/* delete tx-data */
 				while (r != rr) {
 					q[r] = dsp_silence;
@@ -1797,14 +1816,13 @@ dsp_cmx_transmit(struct dsp *dsp, struct sk_buff *skb)
 	ww = dsp->tx_R;
 	p = dsp->tx_buff;
 	d = skb->data;
-	space = ww-w;
-	if (space <= 0)
-		space += CMX_BUFF_SIZE;
+	space = (ww - w - 1) & CMX_BUFF_MASK;
 	/* write-pointer should not overrun nor reach read pointer */
-	if (space-1 < skb->len)
+	if (space < skb->len) {
 		/* write to the space we have left */
-		ww = (ww - 1) & CMX_BUFF_MASK;
-	else
+		ww = (ww - 1) & CMX_BUFF_MASK; /* end one byte prior tx_R */
+		printk(KERN_DEBUG "%s: buffer overflow\n", __func__);
+	} else
 		/* write until all byte are copied */
 		ww = (w + skb->len) & CMX_BUFF_MASK;
 	dsp->tx_W = ww;
diff --git a/drivers/isdn/mISDN/dsp_core.c b/drivers/isdn/mISDN/dsp_core.c
index 1dc21d803410..1d504ba954f5 100644
--- a/drivers/isdn/mISDN/dsp_core.c
+++ b/drivers/isdn/mISDN/dsp_core.c
@@ -191,6 +191,8 @@ dsp_rx_off_member(struct dsp *dsp)
 	struct mISDN_ctrl_req	cq;
 	int rx_off = 1;
 
+	memset(&cq, 0, sizeof(cq));
+
 	if (!dsp->features_rx_off)
 		return;
 
@@ -249,6 +251,32 @@ dsp_rx_off(struct dsp *dsp)
 	}
 }
 
+/* enable "fill empty" feature */
+static void
+dsp_fill_empty(struct dsp *dsp)
+{
+	struct mISDN_ctrl_req	cq;
+
+	memset(&cq, 0, sizeof(cq));
+
+	if (!dsp->ch.peer) {
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: no peer, no fill_empty\n",
+				__func__);
+		return;
+	}
+	cq.op = MISDN_CTRL_FILL_EMPTY;
+	cq.p1 = 1;
+	if (dsp->ch.peer->ctrl(dsp->ch.peer, CONTROL_CHANNEL, &cq)) {
+		printk(KERN_DEBUG "%s: CONTROL_CHANNEL failed\n",
+			__func__);
+		return;
+	}
+	if (dsp_debug & DEBUG_DSP_CORE)
+		printk(KERN_DEBUG "%s: %s set fill_empty = 1\n",
+			__func__, dsp->name);
+}
+
 static int
 dsp_control_req(struct dsp *dsp, struct mISDNhead *hh, struct sk_buff *skb)
 {
@@ -593,8 +621,6 @@ get_features(struct mISDNchannel *ch)
 	struct dsp		*dsp = container_of(ch, struct dsp, ch);
 	struct mISDN_ctrl_req	cq;
 
-	if (dsp_options & DSP_OPT_NOHARDWARE)
-		return;
 	if (!ch->peer) {
 		if (dsp_debug & DEBUG_DSP_CORE)
 			printk(KERN_DEBUG "%s: no peer, no features\n",
@@ -610,6 +636,10 @@ get_features(struct mISDNchannel *ch)
 	}
 	if (cq.op & MISDN_CTRL_RX_OFF)
 		dsp->features_rx_off = 1;
+	if (cq.op & MISDN_CTRL_FILL_EMPTY)
+		dsp->features_fill_empty = 1;
+	if (dsp_options & DSP_OPT_NOHARDWARE)
+		return;
 	if ((cq.op & MISDN_CTRL_HW_FEATURES_OP)) {
 		cq.op = MISDN_CTRL_HW_FEATURES;
 		*((u_long *)&cq.p1) = (u_long)&dsp->features;
@@ -865,6 +895,9 @@ dsp_function(struct mISDNchannel *ch,  struct sk_buff *skb)
 		if (dsp->dtmf.hardware || dsp->dtmf.software)
 			dsp_dtmf_goertzel_init(dsp);
 		get_features(ch);
+		/* enable fill_empty feature */
+		if (dsp->features_fill_empty)
+			dsp_fill_empty(dsp);
 		/* send ph_activate */
 		hh->prim = PH_ACTIVATE_REQ;
 		if (ch->peer)
diff --git a/include/linux/mISDNhw.h b/include/linux/mISDNhw.h
index e794dfb87504..9384b92dfc65 100644
--- a/include/linux/mISDNhw.h
+++ b/include/linux/mISDNhw.h
@@ -57,20 +57,21 @@
 #define FLG_L2DATA		14	/* channel use L2 DATA primitivs */
 #define FLG_ORIGIN		15	/* channel is on origin site */
 /* channel specific stuff */
+#define FLG_FILLEMPTY		16	/* fill fifo on first frame (empty) */
 /* arcofi specific */
-#define FLG_ARCOFI_TIMER	16
-#define FLG_ARCOFI_ERROR	17
+#define FLG_ARCOFI_TIMER	17
+#define FLG_ARCOFI_ERROR	18
 /* isar specific */
-#define FLG_INITIALIZED		16
-#define FLG_DLEETX		17
-#define FLG_LASTDLE		18
-#define FLG_FIRST		19
-#define FLG_LASTDATA		20
-#define FLG_NMD_DATA		21
-#define FLG_FTI_RUN		22
-#define FLG_LL_OK		23
-#define FLG_LL_CONN		24
-#define FLG_DTMFSEND		25
+#define FLG_INITIALIZED		17
+#define FLG_DLEETX		18
+#define FLG_LASTDLE		19
+#define FLG_FIRST		20
+#define FLG_LASTDATA		21
+#define FLG_NMD_DATA		22
+#define FLG_FTI_RUN		23
+#define FLG_LL_OK		24
+#define FLG_LL_CONN		25
+#define FLG_DTMFSEND		26
 
 /* workq events */
 #define FLG_RECVQUEUE		30
diff --git a/include/linux/mISDNif.h b/include/linux/mISDNif.h
index 8f2d60da04e7..74c903cd7a0a 100644
--- a/include/linux/mISDNif.h
+++ b/include/linux/mISDNif.h
@@ -312,6 +312,7 @@ clear_channelmap(u_int nr, u_char *map)
 #define MISDN_CTRL_SETPEER		0x0040
 #define MISDN_CTRL_UNSETPEER		0x0080
 #define MISDN_CTRL_RX_OFF		0x0100
+#define MISDN_CTRL_FILL_EMPTY		0x0200
 #define MISDN_CTRL_HW_FEATURES_OP	0x2000
 #define MISDN_CTRL_HW_FEATURES		0x2001
 #define MISDN_CTRL_HFC_OP		0x4000
-- 
cgit v1.2.3


From 8b6015f736125050722dbe59c4f943e78cd626f0 Mon Sep 17 00:00:00 2001
From: Matthias Urlichs <smurf@smurf.noris.de>
Date: Tue, 12 Aug 2008 10:12:09 +0200
Subject: mISDN: Added an ioctl to change the device name

To get persistent device names with hotplug we need to rename devices
sometime.

Signed-off-by: Matthias Urlichs <matthias@urlichs.de>
Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 drivers/isdn/mISDN/socket.c | 15 +++++++++++++++
 include/linux/mISDNif.h     | 31 +++++++++++++++++++------------
 2 files changed, 34 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/isdn/mISDN/socket.c b/drivers/isdn/mISDN/socket.c
index 37a2de18cfd0..98fa6c4a4bd6 100644
--- a/drivers/isdn/mISDN/socket.c
+++ b/drivers/isdn/mISDN/socket.c
@@ -645,6 +645,21 @@ base_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 		} else
 			err = -ENODEV;
 		break;
+	case IMSETDEVNAME:
+		{
+			struct mISDN_devrename dn;
+			if (copy_from_user(&dn, (void __user *)arg,
+			    sizeof(dn))) {
+				err = -EFAULT;
+				break;
+			}
+			dev = get_mdevice(dn.id);
+			if (dev)
+				strlcpy(dev->name, dn.name, MISDN_MAX_IDLEN);
+			else
+				err = -ENODEV;
+		}
+		break;
 	default:
 		err = -EINVAL;
 	}
diff --git a/include/linux/mISDNif.h b/include/linux/mISDNif.h
index 74c903cd7a0a..be09476ed854 100644
--- a/include/linux/mISDNif.h
+++ b/include/linux/mISDNif.h
@@ -36,8 +36,8 @@
  *              - should be incremented on every checkin
  */
 #define	MISDN_MAJOR_VERSION	1
-#define	MISDN_MINOR_VERSION	0
-#define MISDN_RELEASE		19
+#define	MISDN_MINOR_VERSION	1
+#define MISDN_RELEASE		20
 
 /* primitives for information exchange
  * generell format
@@ -255,16 +255,6 @@ struct sockaddr_mISDN {
 	unsigned char	tei;
 };
 
-/* timer device ioctl */
-#define IMADDTIMER	_IOR('I', 64, int)
-#define IMDELTIMER	_IOR('I', 65, int)
-/* socket ioctls */
-#define	IMGETVERSION	_IOR('I', 66, int)
-#define	IMGETCOUNT	_IOR('I', 67, int)
-#define IMGETDEVINFO	_IOR('I', 68, int)
-#define IMCTRLREQ	_IOR('I', 69, int)
-#define IMCLEAR_L2	_IOR('I', 70, int)
-
 struct mISDNversion {
 	unsigned char	major;
 	unsigned char	minor;
@@ -281,6 +271,23 @@ struct mISDN_devinfo {
 	char			name[MISDN_MAX_IDLEN];
 };
 
+struct mISDN_devrename {
+	u_int			id;
+	char			name[MISDN_MAX_IDLEN]; /* new name */
+};
+
+/* timer device ioctl */
+#define IMADDTIMER	_IOR('I', 64, int)
+#define IMDELTIMER	_IOR('I', 65, int)
+
+/* socket ioctls */
+#define	IMGETVERSION	_IOR('I', 66, int)
+#define	IMGETCOUNT	_IOR('I', 67, int)
+#define IMGETDEVINFO	_IOR('I', 68, int)
+#define IMCTRLREQ	_IOR('I', 69, int)
+#define IMCLEAR_L2	_IOR('I', 70, int)
+#define IMSETDEVNAME	_IOR('I', 71, struct mISDN_devrename)
+
 static inline int
 test_channelmap(u_int nr, u_char *map)
 {
-- 
cgit v1.2.3


From 837468d135dcc49cdabc9fa92fc9550479f60704 Mon Sep 17 00:00:00 2001
From: Matthias Urlichs <smurf@smurf.noris.de>
Date: Sat, 16 Aug 2008 00:04:33 +0200
Subject: mISDN: Use struct device name field

struct device already has a 'name' member, use it.

Signed-off-by: Matthias Urlichs <matthias@urlichs.de>
Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 drivers/isdn/mISDN/core.c   |  8 ++++----
 drivers/isdn/mISDN/layer1.c |  2 +-
 drivers/isdn/mISDN/socket.c |  6 +++---
 drivers/isdn/mISDN/stack.c  | 43 ++++++++++++++++++++++++-------------------
 drivers/isdn/mISDN/tei.c    |  6 +++---
 include/linux/mISDNif.h     |  1 -
 6 files changed, 35 insertions(+), 31 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/isdn/mISDN/core.c b/drivers/isdn/mISDN/core.c
index 751665c448d0..43fd97b0fe42 100644
--- a/drivers/isdn/mISDN/core.c
+++ b/drivers/isdn/mISDN/core.c
@@ -82,12 +82,12 @@ mISDN_register_device(struct mISDNdevice *dev, char *name)
 	if (dev->id < 0)
 		return -EBUSY;
 	if (name && name[0])
-		strcpy(dev->name, name);
+		dev_set_name(&dev->dev, "%s", name);
 	else
-		sprintf(dev->name, "mISDN%d", dev->id);
+		dev_set_name(&dev->dev, "mISDN%d", dev->id);
 	if (debug & DEBUG_CORE)
 		printk(KERN_DEBUG "mISDN_register %s %d\n",
-			dev->name, dev->id);
+			dev_name(&dev->dev), dev->id);
 	err = create_stack(dev);
 	if (err)
 		return err;
@@ -104,7 +104,7 @@ mISDN_unregister_device(struct mISDNdevice *dev) {
 
 	if (debug & DEBUG_CORE)
 		printk(KERN_DEBUG "mISDN_unregister %s %d\n",
-			dev->name, dev->id);
+			dev_name(&dev->dev), dev->id);
 	write_lock_irqsave(&device_lock, flags);
 	list_del(&dev->D.list);
 	write_unlock_irqrestore(&device_lock, flags);
diff --git a/drivers/isdn/mISDN/layer1.c b/drivers/isdn/mISDN/layer1.c
index b73e952d12cf..e826eeb1ecec 100644
--- a/drivers/isdn/mISDN/layer1.c
+++ b/drivers/isdn/mISDN/layer1.c
@@ -101,7 +101,7 @@ l1m_debug(struct FsmInst *fi, char *fmt, ...)
 	va_list va;
 
 	va_start(va, fmt);
-	printk(KERN_DEBUG "%s: ", l1->dch->dev.name);
+	printk(KERN_DEBUG "%s: ", dev_name(&l1->dch->dev.dev));
 	vprintk(fmt, va);
 	printk("\n");
 	va_end(va);
diff --git a/drivers/isdn/mISDN/socket.c b/drivers/isdn/mISDN/socket.c
index 98fa6c4a4bd6..2f6d6e88ff2c 100644
--- a/drivers/isdn/mISDN/socket.c
+++ b/drivers/isdn/mISDN/socket.c
@@ -381,7 +381,7 @@ data_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 			memcpy(di.channelmap, dev->channelmap,
 				sizeof(di.channelmap));
 			di.nrbchan = dev->nrbchan;
-			strcpy(di.name, dev->name);
+			strcpy(di.name, dev_name(&dev->dev));
 			if (copy_to_user((void __user *)arg, &di, sizeof(di)))
 				err = -EFAULT;
 		} else
@@ -639,7 +639,7 @@ base_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 			memcpy(di.channelmap, dev->channelmap,
 				sizeof(di.channelmap));
 			di.nrbchan = dev->nrbchan;
-			strcpy(di.name, dev->name);
+			strcpy(di.name, dev_name(&dev->dev));
 			if (copy_to_user((void __user *)arg, &di, sizeof(di)))
 				err = -EFAULT;
 		} else
@@ -655,7 +655,7 @@ base_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 			}
 			dev = get_mdevice(dn.id);
 			if (dev)
-				strlcpy(dev->name, dn.name, MISDN_MAX_IDLEN);
+				err = device_rename(&dev->dev, dn.name);
 			else
 				err = -ENODEV;
 		}
diff --git a/drivers/isdn/mISDN/stack.c b/drivers/isdn/mISDN/stack.c
index d55b14ae4e99..9b9fab47cb2b 100644
--- a/drivers/isdn/mISDN/stack.c
+++ b/drivers/isdn/mISDN/stack.c
@@ -172,7 +172,8 @@ send_msg_to_layer(struct mISDNstack *st, struct sk_buff *skb)
 		else
 			printk(KERN_WARNING
 			    "%s: dev(%s) prim(%x) id(%x) no channel\n",
-			    __func__, st->dev->name, hh->prim, hh->id);
+			    __func__, dev_name(&st->dev->dev), hh->prim,
+			    hh->id);
 	} else if (lm == 0x8) {
 		WARN_ON(lm == 0x8);
 		ch = get_channel4id(st, hh->id);
@@ -181,11 +182,12 @@ send_msg_to_layer(struct mISDNstack *st, struct sk_buff *skb)
 		else
 			printk(KERN_WARNING
 			    "%s: dev(%s) prim(%x) id(%x) no channel\n",
-			    __func__, st->dev->name, hh->prim, hh->id);
+			    __func__, dev_name(&st->dev->dev), hh->prim,
+			    hh->id);
 	} else {
 		/* broadcast not handled yet */
 		printk(KERN_WARNING "%s: dev(%s) prim %x not delivered\n",
-		    __func__, st->dev->name, hh->prim);
+		    __func__, dev_name(&st->dev->dev), hh->prim);
 	}
 	return -ESRCH;
 }
@@ -209,7 +211,8 @@ mISDNStackd(void *data)
 	unlock_kernel();
 #endif
 	if (*debug & DEBUG_MSG_THREAD)
-		printk(KERN_DEBUG "mISDNStackd %s started\n", st->dev->name);
+		printk(KERN_DEBUG "mISDNStackd %s started\n",
+		    dev_name(&st->dev->dev));
 
 	if (st->notify != NULL) {
 		complete(st->notify);
@@ -245,7 +248,7 @@ mISDNStackd(void *data)
 					printk(KERN_DEBUG
 					    "%s: %s prim(%x) id(%x) "
 					    "send call(%d)\n",
-					    __func__, st->dev->name,
+					    __func__, dev_name(&st->dev->dev),
 					    mISDN_HEAD_PRIM(skb),
 					    mISDN_HEAD_ID(skb), err);
 				dev_kfree_skb(skb);
@@ -288,7 +291,7 @@ mISDNStackd(void *data)
 		    mISDN_STACK_ACTION_MASK));
 		if (*debug & DEBUG_MSG_THREAD)
 			printk(KERN_DEBUG "%s: %s wake status %08lx\n",
-			    __func__, st->dev->name, st->status);
+			    __func__, dev_name(&st->dev->dev), st->status);
 		test_and_set_bit(mISDN_STACK_ACTIVE, &st->status);
 
 		test_and_clear_bit(mISDN_STACK_WAKEUP, &st->status);
@@ -303,15 +306,16 @@ mISDNStackd(void *data)
 #ifdef MISDN_MSG_STATS
 	printk(KERN_DEBUG "mISDNStackd daemon for %s proceed %d "
 	    "msg %d sleep %d stopped\n",
-	    st->dev->name, st->msg_cnt, st->sleep_cnt, st->stopped_cnt);
+	    dev_name(&st->dev->dev), st->msg_cnt, st->sleep_cnt,
+	    st->stopped_cnt);
 	printk(KERN_DEBUG
 	    "mISDNStackd daemon for %s utime(%ld) stime(%ld)\n",
-	    st->dev->name, st->thread->utime, st->thread->stime);
+	    dev_name(&st->dev->dev), st->thread->utime, st->thread->stime);
 	printk(KERN_DEBUG
 	    "mISDNStackd daemon for %s nvcsw(%ld) nivcsw(%ld)\n",
-	    st->dev->name, st->thread->nvcsw, st->thread->nivcsw);
+	    dev_name(&st->dev->dev), st->thread->nvcsw, st->thread->nivcsw);
 	printk(KERN_DEBUG "mISDNStackd daemon for %s killed now\n",
-	    st->dev->name);
+	    dev_name(&st->dev->dev));
 #endif
 	test_and_set_bit(mISDN_STACK_KILLED, &st->status);
 	test_and_clear_bit(mISDN_STACK_RUNNING, &st->status);
@@ -401,15 +405,16 @@ create_stack(struct mISDNdevice *dev)
 	newst->own.send = mISDN_queue_message;
 	newst->own.recv = mISDN_queue_message;
 	if (*debug & DEBUG_CORE_FUNC)
-		printk(KERN_DEBUG "%s: st(%s)\n", __func__, newst->dev->name);
+		printk(KERN_DEBUG "%s: st(%s)\n", __func__,
+		    dev_name(&newst->dev->dev));
 	newst->notify = &done;
 	newst->thread = kthread_run(mISDNStackd, (void *)newst, "mISDN_%s",
-		newst->dev->name);
+		dev_name(&newst->dev->dev));
 	if (IS_ERR(newst->thread)) {
 		err = PTR_ERR(newst->thread);
 		printk(KERN_ERR
 			"mISDN:cannot create kernel thread for %s (%d)\n",
-			newst->dev->name, err);
+			dev_name(&newst->dev->dev), err);
 		delete_teimanager(dev->teimgr);
 		kfree(newst);
 	} else
@@ -428,8 +433,8 @@ connect_layer1(struct mISDNdevice *dev, struct mISDNchannel *ch,
 
 	if (*debug &  DEBUG_CORE_FUNC)
 		printk(KERN_DEBUG "%s: %s proto(%x) adr(%d %d %d %d)\n",
-			__func__, dev->name, protocol, adr->dev, adr->channel,
-			 adr->sapi, adr->tei);
+			__func__, dev_name(&dev->dev), protocol, adr->dev,
+			adr->channel, adr->sapi, adr->tei);
 	switch (protocol) {
 	case ISDN_P_NT_S0:
 	case ISDN_P_NT_E1:
@@ -473,7 +478,7 @@ connect_Bstack(struct mISDNdevice *dev, struct mISDNchannel *ch,
 
 	if (*debug &  DEBUG_CORE_FUNC)
 		printk(KERN_DEBUG "%s: %s proto(%x) adr(%d %d %d %d)\n",
-			__func__, dev->name, protocol,
+			__func__, dev_name(&dev->dev), protocol,
 			adr->dev, adr->channel, adr->sapi,
 			adr->tei);
 	ch->st = dev->D.st;
@@ -529,7 +534,7 @@ create_l2entity(struct mISDNdevice *dev, struct mISDNchannel *ch,
 
 	if (*debug &  DEBUG_CORE_FUNC)
 		printk(KERN_DEBUG "%s: %s proto(%x) adr(%d %d %d %d)\n",
-			__func__, dev->name, protocol,
+			__func__, dev_name(&dev->dev), protocol,
 			adr->dev, adr->channel, adr->sapi,
 			adr->tei);
 	rq.protocol = ISDN_P_TE_S0;
@@ -590,7 +595,7 @@ delete_channel(struct mISDNchannel *ch)
 	}
 	if (*debug & DEBUG_CORE_FUNC)
 		printk(KERN_DEBUG "%s: st(%s) protocol(%x)\n", __func__,
-		    ch->st->dev->name, ch->protocol);
+		    dev_name(&ch->st->dev->dev), ch->protocol);
 	if (ch->protocol >= ISDN_P_B_START) {
 		if (ch->peer) {
 			ch->peer->ctrl(ch->peer, CLOSE_CHANNEL, NULL);
@@ -643,7 +648,7 @@ delete_stack(struct mISDNdevice *dev)
 
 	if (*debug & DEBUG_CORE_FUNC)
 		printk(KERN_DEBUG "%s: st(%s)\n", __func__,
-		    st->dev->name);
+		    dev_name(&st->dev->dev));
 	if (dev->teimgr)
 		delete_teimanager(dev->teimgr);
 	if (st->thread) {
diff --git a/drivers/isdn/mISDN/tei.c b/drivers/isdn/mISDN/tei.c
index 5c43d19e7c11..b452dead8fd0 100644
--- a/drivers/isdn/mISDN/tei.c
+++ b/drivers/isdn/mISDN/tei.c
@@ -968,9 +968,9 @@ create_teimgr(struct manager *mgr, struct channel_req *crq)
 
 	if (*debug & DEBUG_L2_TEI)
 		printk(KERN_DEBUG "%s: %s proto(%x) adr(%d %d %d %d)\n",
-			__func__, mgr->ch.st->dev->name, crq->protocol,
-			crq->adr.dev, crq->adr.channel, crq->adr.sapi,
-			crq->adr.tei);
+			__func__, dev_name(&mgr->ch.st->dev->dev),
+			crq->protocol, crq->adr.dev, crq->adr.channel,
+			crq->adr.sapi, crq->adr.tei);
 	if (crq->adr.sapi != 0) /* not supported yet */
 		return -EINVAL;
 	if (crq->adr.tei > GROUP_TEI)
diff --git a/include/linux/mISDNif.h b/include/linux/mISDNif.h
index be09476ed854..a59febb6143a 100644
--- a/include/linux/mISDNif.h
+++ b/include/linux/mISDNif.h
@@ -431,7 +431,6 @@ struct mISDN_sock {
 struct mISDNdevice {
 	struct mISDNchannel	D;
 	u_int			id;
-	char			name[MISDN_MAX_IDLEN];
 	u_int			Dprotocols;
 	u_int			Bprotocols;
 	u_int			nrbchan;
-- 
cgit v1.2.3


From 1f28fa19d34c0d9186f274e61e4b3dcfc6428c5c Mon Sep 17 00:00:00 2001
From: Martin Bachem <m.bachem@gmx.de>
Date: Wed, 3 Sep 2008 15:17:45 +0200
Subject: mISDN: Add E-Channel logging features

New prim PH_DATA_E_IND.

 - all E-ch frames are indicated by recv_Echannel(), which pushes E-Channel
   frames into dch's rqueue
 - if dchannel is opened with channel nr 0, no E-Channel logging
   is requested
 - if dchannel is opened with channel nr 1, E-Channel logging
   is requested. if layer1 does not support that, -EINVAL
   in return is appropriate

Signed-off-by: Martin Bachem <m.bachem@gmx.de>
Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 drivers/isdn/mISDN/hwchannel.c | 19 +++++++++++++++++++
 drivers/isdn/mISDN/stack.c     |  2 +-
 include/linux/mISDNhw.h        |  1 +
 include/linux/mISDNif.h        |  1 +
 4 files changed, 22 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/isdn/mISDN/hwchannel.c b/drivers/isdn/mISDN/hwchannel.c
index 535ceacc05b9..ab1168a110ae 100644
--- a/drivers/isdn/mISDN/hwchannel.c
+++ b/drivers/isdn/mISDN/hwchannel.c
@@ -165,6 +165,25 @@ recv_Dchannel(struct dchannel *dch)
 }
 EXPORT_SYMBOL(recv_Dchannel);
 
+void
+recv_Echannel(struct dchannel *ech, struct dchannel *dch)
+{
+	struct mISDNhead *hh;
+
+	if (ech->rx_skb->len < 2) { /* at least 2 for sapi / tei */
+		dev_kfree_skb(ech->rx_skb);
+		ech->rx_skb = NULL;
+		return;
+	}
+	hh = mISDN_HEAD_P(ech->rx_skb);
+	hh->prim = PH_DATA_E_IND;
+	hh->id = get_sapi_tei(ech->rx_skb->data);
+	skb_queue_tail(&dch->rqueue, ech->rx_skb);
+	ech->rx_skb = NULL;
+	schedule_event(dch, FLG_RECVQUEUE);
+}
+EXPORT_SYMBOL(recv_Echannel);
+
 void
 recv_Bchannel(struct bchannel *bch)
 {
diff --git a/drivers/isdn/mISDN/stack.c b/drivers/isdn/mISDN/stack.c
index 9b9fab47cb2b..8cff570bb8df 100644
--- a/drivers/isdn/mISDN/stack.c
+++ b/drivers/isdn/mISDN/stack.c
@@ -453,7 +453,7 @@ connect_layer1(struct mISDNdevice *dev, struct mISDNchannel *ch,
 		ch->peer = &dev->D.st->own;
 		ch->st = dev->D.st;
 		rq.protocol = protocol;
-		rq.adr.channel = 0;
+		rq.adr.channel = adr->channel;
 		err = dev->D.ctrl(&dev->D, OPEN_CHANNEL, &rq);
 		printk(KERN_DEBUG "%s: ret 1 %d\n", __func__, err);
 		if (err)
diff --git a/include/linux/mISDNhw.h b/include/linux/mISDNhw.h
index 9384b92dfc65..97ffdc1d3442 100644
--- a/include/linux/mISDNhw.h
+++ b/include/linux/mISDNhw.h
@@ -184,6 +184,7 @@ extern void	queue_ch_frame(struct mISDNchannel *, u_int,
 extern int	dchannel_senddata(struct dchannel *, struct sk_buff *);
 extern int	bchannel_senddata(struct bchannel *, struct sk_buff *);
 extern void	recv_Dchannel(struct dchannel *);
+extern void	recv_Echannel(struct dchannel *, struct dchannel *);
 extern void	recv_Bchannel(struct bchannel *);
 extern void	recv_Dchannel_skb(struct dchannel *, struct sk_buff *);
 extern void	recv_Bchannel_skb(struct bchannel *, struct sk_buff *);
diff --git a/include/linux/mISDNif.h b/include/linux/mISDNif.h
index a59febb6143a..f75d596c5316 100644
--- a/include/linux/mISDNif.h
+++ b/include/linux/mISDNif.h
@@ -80,6 +80,7 @@
 #define PH_DEACTIVATE_IND	0x0202
 #define PH_DEACTIVATE_CNF	0x4202
 #define PH_DATA_IND		0x2002
+#define PH_DATA_E_IND		0x3002
 #define MPH_ACTIVATE_IND	0x0502
 #define MPH_DEACTIVATE_IND	0x0602
 #define MPH_INFORMATION_IND	0x0702
-- 
cgit v1.2.3


From 3bd69ad197a4a3d0085a5dc3b5796111bf176b12 Mon Sep 17 00:00:00 2001
From: Andreas Eversberg <andreas@eversberg.eu>
Date: Sat, 6 Sep 2008 09:03:46 +0200
Subject: mISDN: Add ISDN sample clock API to mISDN core

Add ISDN sample clock API to mISDN core (new file clock.c)
hfcmulti and mISDNdsp use clock API.

Signed-off-by: Andreas Eversberg <andreas@eversberg.eu>
Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 drivers/isdn/hardware/mISDN/hfc_multi.h |   3 +
 drivers/isdn/hardware/mISDN/hfcmulti.c  |  45 +++++--
 drivers/isdn/mISDN/Makefile             |   2 +-
 drivers/isdn/mISDN/clock.c              | 216 ++++++++++++++++++++++++++++++++
 drivers/isdn/mISDN/core.c               |   1 +
 drivers/isdn/mISDN/core.h               |   2 +
 drivers/isdn/mISDN/dsp_cmx.c            |  40 +++---
 include/linux/mISDNif.h                 |  17 +++
 8 files changed, 294 insertions(+), 32 deletions(-)
 create mode 100644 drivers/isdn/mISDN/clock.c

(limited to 'include/linux')

diff --git a/drivers/isdn/hardware/mISDN/hfc_multi.h b/drivers/isdn/hardware/mISDN/hfc_multi.h
index 4aa6a8b41f50..663b77f578be 100644
--- a/drivers/isdn/hardware/mISDN/hfc_multi.h
+++ b/drivers/isdn/hardware/mISDN/hfc_multi.h
@@ -197,6 +197,9 @@ struct hfc_multi {
 
 	spinlock_t	lock;	/* the lock */
 
+	struct mISDNclock *iclock; /* isdn clock support */
+	int		iclock_on;
+
 	/*
 	 * the channel index is counted from 0, regardless where the channel
 	 * is located on the hfc-channel.
diff --git a/drivers/isdn/hardware/mISDN/hfcmulti.c b/drivers/isdn/hardware/mISDN/hfcmulti.c
index 3fc2e9d95341..592db93105f9 100644
--- a/drivers/isdn/hardware/mISDN/hfcmulti.c
+++ b/drivers/isdn/hardware/mISDN/hfcmulti.c
@@ -133,6 +133,12 @@
  *	Give the value of the clock control register (A_ST_CLK_DLY)
  *	of the S/T interfaces in TE mode.
  *	This register is needed for the TBR3 certification, so don't change it.
+ *
+ * clock:
+ *	NOTE: only one clockdelay_te value must be given once
+ *	Selects interface with clock source for mISDN and applications.
+ *	Set to card number starting with 1. Set to -1 to disable.
+ *	By default, the first card is used as clock source.
  */
 
 /*
@@ -190,12 +196,13 @@ static int nt_t1_count[] = { 3840, 1920, 960, 480, 240, 120, 60, 30  };
  */
 
 static uint	type[MAX_CARDS];
-static uint	pcm[MAX_CARDS];
-static uint	dslot[MAX_CARDS];
+static int	pcm[MAX_CARDS];
+static int	dslot[MAX_CARDS];
 static uint	iomode[MAX_CARDS];
 static uint	port[MAX_PORTS];
 static uint	debug;
 static uint	poll;
+static int	clock;
 static uint	timer;
 static uint	clockdelay_te = CLKDEL_TE;
 static uint	clockdelay_nt = CLKDEL_NT;
@@ -207,12 +214,13 @@ MODULE_LICENSE("GPL");
 MODULE_VERSION(HFC_MULTI_VERSION);
 module_param(debug, uint, S_IRUGO | S_IWUSR);
 module_param(poll, uint, S_IRUGO | S_IWUSR);
+module_param(clock, int, S_IRUGO | S_IWUSR);
 module_param(timer, uint, S_IRUGO | S_IWUSR);
 module_param(clockdelay_te, uint, S_IRUGO | S_IWUSR);
 module_param(clockdelay_nt, uint, S_IRUGO | S_IWUSR);
 module_param_array(type, uint, NULL, S_IRUGO | S_IWUSR);
-module_param_array(pcm, uint, NULL, S_IRUGO | S_IWUSR);
-module_param_array(dslot, uint, NULL, S_IRUGO | S_IWUSR);
+module_param_array(pcm, int, NULL, S_IRUGO | S_IWUSR);
+module_param_array(dslot, int, NULL, S_IRUGO | S_IWUSR);
 module_param_array(iomode, uint, NULL, S_IRUGO | S_IWUSR);
 module_param_array(port, uint, NULL, S_IRUGO | S_IWUSR);
 
@@ -2629,6 +2637,7 @@ hfcmulti_interrupt(int intno, void *dev_id)
 		iqcnt = 0;
 	}
 #endif
+
 	if (!r_irq_statech &&
 	    !(status & (V_DTMF_STA | V_LOST_STA | V_EXT_IRQSTA |
 	    V_MISC_IRQSTA | V_FR_IRQSTA))) {
@@ -2683,11 +2692,13 @@ hfcmulti_interrupt(int intno, void *dev_id)
 					plxsd_checksync(hc, 0);
 			}
 		}
-		if (r_irq_misc & V_TI_IRQ)
+		if (r_irq_misc & V_TI_IRQ) {
+			if (hc->iclock_on)
+				mISDN_clock_update(hc->iclock, poll, NULL);
 			handle_timer_irq(hc);
+		}
 
 		if (r_irq_misc & V_DTMF_IRQ) {
-			/* -> DTMF IRQ */
 			hfcmulti_dtmf(hc);
 		}
 		if (r_irq_misc & V_IRQ_PROC) {
@@ -4075,6 +4086,15 @@ hfcm_dctrl(struct mISDNchannel *ch, u_int cmd, void *arg)
 	return err;
 }
 
+static int
+clockctl(void *priv, int enable)
+{
+	struct hfc_multi *hc = priv;
+
+	hc->iclock_on = enable;
+	return 0;
+}
+
 /*
  * initialize the card
  */
@@ -4489,10 +4509,14 @@ release_card(struct hfc_multi *hc)
 		printk(KERN_WARNING "%s: release card (%d) entered\n",
 		    __func__, hc->id);
 
+	/* unregister clock source */
+	if (hc->iclock)
+		mISDN_unregister_clock(hc->iclock);
+
+	/* disable irq */
 	spin_lock_irqsave(&hc->lock, flags);
 	disable_hwirq(hc);
 	spin_unlock_irqrestore(&hc->lock, flags);
-
 	udelay(1000);
 
 	/* dimm leds */
@@ -5003,6 +5027,10 @@ hfcmulti_init(struct pci_dev *pdev, const struct pci_device_id *ent)
 	list_add_tail(&hc->list, &HFClist);
 	spin_unlock_irqrestore(&HFClock, flags);
 
+	/* use as clock source */
+	if (clock == HFC_cnt + 1)
+		hc->iclock = mISDN_register_clock("HFCMulti", 0, clockctl, hc);
+
 	/* initialize hardware */
 	ret_err = init_card(hc);
 	if (ret_err) {
@@ -5273,6 +5301,9 @@ HFCmulti_init(void)
 
 	}
 
+	if (!clock)
+		clock = 1;
+
 	err = pci_register_driver(&hfcmultipci_driver);
 	if (err < 0) {
 		printk(KERN_ERR "error registering pci driver: %x\n", err);
diff --git a/drivers/isdn/mISDN/Makefile b/drivers/isdn/mISDN/Makefile
index 1cb5e633cf75..0a6bd2a9e730 100644
--- a/drivers/isdn/mISDN/Makefile
+++ b/drivers/isdn/mISDN/Makefile
@@ -8,6 +8,6 @@ obj-$(CONFIG_MISDN_L1OIP) += l1oip.o
 
 # multi objects
 
-mISDN_core-objs := core.o fsm.o socket.o hwchannel.o stack.o layer1.o layer2.o tei.o timerdev.o
+mISDN_core-objs := core.o fsm.o socket.o clock.o hwchannel.o stack.o layer1.o layer2.o tei.o timerdev.o
 mISDN_dsp-objs := dsp_core.o dsp_cmx.o dsp_tones.o dsp_dtmf.o dsp_audio.o dsp_blowfish.o dsp_pipeline.o dsp_hwec.o
 l1oip-objs := l1oip_core.o l1oip_codec.o
diff --git a/drivers/isdn/mISDN/clock.c b/drivers/isdn/mISDN/clock.c
new file mode 100644
index 000000000000..44d9c3d5d33d
--- /dev/null
+++ b/drivers/isdn/mISDN/clock.c
@@ -0,0 +1,216 @@
+/*
+ * Copyright 2008  by Andreas Eversberg <andreas@eversberg.eu>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * Quick API description:
+ *
+ * A clock source registers using mISDN_register_clock:
+ * 	name = text string to name clock source
+ *	priority = value to priorize clock sources (0 = default)
+ *	ctl = callback function to enable/disable clock source
+ *	priv = private pointer of clock source
+ * 	return = pointer to clock source structure;
+ *
+ * Note: Callback 'ctl' can be called before mISDN_register_clock returns!
+ *       Also it can be called during mISDN_unregister_clock.
+ *
+ * A clock source calls mISDN_clock_update with given samples elapsed, if
+ * enabled. If function call is delayed, tv must be set with the timestamp
+ * of the actual event.
+ *
+ * A clock source unregisters using mISDN_unregister_clock.
+ *
+ * To get current clock, call mISDN_clock_get. The signed short value
+ * counts the number of samples since. Time since last clock event is added.
+ *
+ */
+
+#include <linux/types.h>
+#include <linux/stddef.h>
+#include <linux/spinlock.h>
+#include <linux/mISDNif.h>
+#include "core.h"
+
+static u_int *debug;
+static LIST_HEAD(iclock_list);
+DEFINE_RWLOCK(iclock_lock);
+u16	iclock_count;		/* counter of last clock */
+struct	timeval iclock_tv;	/* time stamp of last clock */
+int	iclock_tv_valid;	/* already received one timestamp */
+struct	mISDNclock *iclock_current;
+
+void
+mISDN_init_clock(u_int *dp)
+{
+	debug = dp;
+	do_gettimeofday(&iclock_tv);
+}
+
+static void
+select_iclock(void)
+{
+	struct mISDNclock *iclock, *bestclock = NULL, *lastclock = NULL;
+	int pri = -128;
+
+	list_for_each_entry(iclock, &iclock_list, list) {
+		if (iclock->pri > pri) {
+			pri = iclock->pri;
+			bestclock = iclock;
+		}
+		if (iclock_current == iclock)
+			lastclock = iclock;
+	}
+	if (lastclock && bestclock != lastclock) {
+		/* last used clock source still exists but changes, disable */
+		if (*debug & DEBUG_CLOCK)
+			printk(KERN_DEBUG "Old clock source '%s' disable.\n",
+				lastclock->name);
+		lastclock->ctl(lastclock->priv, 0);
+	}
+	if (bestclock && bestclock != iclock_current) {
+		/* new clock source selected, enable */
+		if (*debug & DEBUG_CLOCK)
+			printk(KERN_DEBUG "New clock source '%s' enable.\n",
+				bestclock->name);
+		bestclock->ctl(bestclock->priv, 1);
+	}
+	if (bestclock != iclock_current) {
+		/* no clock received yet */
+		iclock_tv_valid = 0;
+	}
+	iclock_current = bestclock;
+}
+
+struct mISDNclock
+*mISDN_register_clock(char *name, int pri, clockctl_func_t *ctl, void *priv)
+{
+	u_long			flags;
+	struct mISDNclock	*iclock;
+
+	if (*debug & (DEBUG_CORE | DEBUG_CLOCK))
+		printk(KERN_DEBUG "%s: %s %d\n", __func__, name, pri);
+	iclock = kzalloc(sizeof(struct mISDNclock), GFP_ATOMIC);
+	if (!iclock) {
+		printk(KERN_ERR "%s: No memory for clock entry.\n", __func__);
+		return NULL;
+	}
+	strncpy(iclock->name, name, sizeof(iclock->name)-1);
+	iclock->pri = pri;
+	iclock->priv = priv;
+	iclock->ctl = ctl;
+	write_lock_irqsave(&iclock_lock, flags);
+	list_add_tail(&iclock->list, &iclock_list);
+	select_iclock();
+	write_unlock_irqrestore(&iclock_lock, flags);
+	return iclock;
+}
+EXPORT_SYMBOL(mISDN_register_clock);
+
+void
+mISDN_unregister_clock(struct mISDNclock *iclock)
+{
+	u_long	flags;
+
+	if (*debug & (DEBUG_CORE | DEBUG_CLOCK))
+		printk(KERN_DEBUG "%s: %s %d\n", __func__, iclock->name,
+			iclock->pri);
+	write_lock_irqsave(&iclock_lock, flags);
+	if (iclock_current == iclock) {
+		if (*debug & DEBUG_CLOCK)
+			printk(KERN_DEBUG
+				"Current clock source '%s' unregisters.\n",
+				iclock->name);
+		iclock->ctl(iclock->priv, 0);
+	}
+	list_del(&iclock->list);
+	select_iclock();
+	write_unlock_irqrestore(&iclock_lock, flags);
+}
+EXPORT_SYMBOL(mISDN_unregister_clock);
+
+void
+mISDN_clock_update(struct mISDNclock *iclock, int samples, struct timeval *tv)
+{
+	u_long		flags;
+	struct timeval	tv_now;
+	time_t		elapsed_sec;
+	int		elapsed_8000th;
+
+	write_lock_irqsave(&iclock_lock, flags);
+	if (iclock_current != iclock) {
+		printk(KERN_ERR "%s: '%s' sends us clock updates, but we do "
+			"listen to '%s'. This is a bug!\n", __func__,
+			iclock->name,
+			iclock_current ? iclock_current->name : "nothing");
+		iclock->ctl(iclock->priv, 0);
+		write_unlock_irqrestore(&iclock_lock, flags);
+		return;
+	}
+	if (iclock_tv_valid) {
+		/* increment sample counter by given samples */
+		iclock_count += samples;
+		if (tv) { /* tv must be set, if function call is delayed */
+			iclock_tv.tv_sec = tv->tv_sec;
+			iclock_tv.tv_usec = tv->tv_usec;
+		} else
+			do_gettimeofday(&iclock_tv);
+	} else {
+		/* calc elapsed time by system clock */
+		if (tv) { /* tv must be set, if function call is delayed */
+			tv_now.tv_sec = tv->tv_sec;
+			tv_now.tv_usec = tv->tv_usec;
+		} else
+			do_gettimeofday(&tv_now);
+		elapsed_sec = tv_now.tv_sec - iclock_tv.tv_sec;
+		elapsed_8000th = (tv_now.tv_usec / 125)
+			- (iclock_tv.tv_usec / 125);
+		if (elapsed_8000th < 0) {
+			elapsed_sec -= 1;
+			elapsed_8000th += 8000;
+		}
+		/* add elapsed time to counter and set new timestamp */
+		iclock_count += elapsed_sec * 8000 + elapsed_8000th;
+		iclock_tv.tv_sec = tv_now.tv_sec;
+		iclock_tv.tv_usec = tv_now.tv_usec;
+		iclock_tv_valid = 1;
+		if (*debug & DEBUG_CLOCK)
+			printk("Received first clock from source '%s'.\n",
+			    iclock_current ? iclock_current->name : "nothing");
+	}
+	write_unlock_irqrestore(&iclock_lock, flags);
+}
+EXPORT_SYMBOL(mISDN_clock_update);
+
+unsigned short
+mISDN_clock_get(void)
+{
+	u_long		flags;
+	struct timeval	tv_now;
+	time_t		elapsed_sec;
+	int		elapsed_8000th;
+	u16		count;
+
+	read_lock_irqsave(&iclock_lock, flags);
+	/* calc elapsed time by system clock */
+	do_gettimeofday(&tv_now);
+	elapsed_sec = tv_now.tv_sec - iclock_tv.tv_sec;
+	elapsed_8000th = (tv_now.tv_usec / 125) - (iclock_tv.tv_usec / 125);
+	if (elapsed_8000th < 0) {
+		elapsed_sec -= 1;
+		elapsed_8000th += 8000;
+	}
+	/* add elapsed time to counter */
+	count =	iclock_count + elapsed_sec * 8000 + elapsed_8000th;
+	read_unlock_irqrestore(&iclock_lock, flags);
+	return count;
+}
+EXPORT_SYMBOL(mISDN_clock_get);
+
diff --git a/drivers/isdn/mISDN/core.c b/drivers/isdn/mISDN/core.c
index 43fd97b0fe42..9116f54def2c 100644
--- a/drivers/isdn/mISDN/core.c
+++ b/drivers/isdn/mISDN/core.c
@@ -199,6 +199,7 @@ mISDNInit(void)
 
 	printk(KERN_INFO "Modular ISDN core version %d.%d.%d\n",
 		MISDN_MAJOR_VERSION, MISDN_MINOR_VERSION, MISDN_RELEASE);
+	mISDN_init_clock(&debug);
 	mISDN_initstack(&debug);
 	err = mISDN_inittimer(&debug);
 	if (err)
diff --git a/drivers/isdn/mISDN/core.h b/drivers/isdn/mISDN/core.h
index 7da7233b4c1a..7ac2f81a812b 100644
--- a/drivers/isdn/mISDN/core.h
+++ b/drivers/isdn/mISDN/core.h
@@ -74,4 +74,6 @@ extern void	l1_cleanup(void);
 extern int 	Isdnl2_Init(u_int *);
 extern void	Isdnl2_cleanup(void);
 
+extern void	mISDN_init_clock(u_int *);
+
 #endif
diff --git a/drivers/isdn/mISDN/dsp_cmx.c b/drivers/isdn/mISDN/dsp_cmx.c
index 04dbb407f7a0..efe4c7430e6d 100644
--- a/drivers/isdn/mISDN/dsp_cmx.c
+++ b/drivers/isdn/mISDN/dsp_cmx.c
@@ -1557,13 +1557,11 @@ send_packet:
 	schedule_work(&dsp->workq);
 }
 
-static u32	samplecount;
+static u32	jittercount; /* counter for jitter check */;
 struct timer_list dsp_spl_tl;
 u32	dsp_spl_jiffies; /* calculate the next time to fire */
-#ifdef UNUSED
-static u32	dsp_start_jiffies; /* jiffies at the time, the calculation begins */
-#endif /* UNUSED */
-static struct timeval dsp_start_tv; /* time at start of calculation */
+static u16	dsp_count; /* last sample count */
+static int	dsp_count_valid ; /* if we have last sample count */
 
 void
 dsp_cmx_send(void *arg)
@@ -1577,38 +1575,32 @@ dsp_cmx_send(void *arg)
 	int r, rr;
 	int jittercheck = 0, delay, i;
 	u_long flags;
-	struct timeval tv;
-	u32 elapsed;
-	s16 length;
+	u16 length, count;
 
 	/* lock */
 	spin_lock_irqsave(&dsp_lock, flags);
 
-	if (!dsp_start_tv.tv_sec) {
-		do_gettimeofday(&dsp_start_tv);
+	if (!dsp_count_valid) {
+		dsp_count = mISDN_clock_get();
 		length = dsp_poll;
+		dsp_count_valid = 1;
 	} else {
-		do_gettimeofday(&tv);
-		elapsed = ((tv.tv_sec - dsp_start_tv.tv_sec) * 8000)
-		    + ((s32)(tv.tv_usec / 125) - (dsp_start_tv.tv_usec / 125));
-		dsp_start_tv.tv_sec = tv.tv_sec;
-		dsp_start_tv.tv_usec = tv.tv_usec;
-		length = elapsed;
+		count = mISDN_clock_get();
+		length = count - dsp_count;
+		dsp_count = count;
 	}
 	if (length > MAX_POLL + 100)
 		length = MAX_POLL + 100;
-/* printk(KERN_DEBUG "len=%d dsp_count=0x%x.%04x dsp_poll_diff=0x%x.%04x\n",
- length, dsp_count >> 16, dsp_count & 0xffff, dsp_poll_diff >> 16,
- dsp_poll_diff & 0xffff);
- */
+	/* printk(KERN_DEBUG "len=%d dsp_count=0x%x\n", length, dsp_count); */
 
 	/*
-	 * check if jitter needs to be checked
-	 * (this is about every second = 8192 samples)
+	 * check if jitter needs to be checked (this is every second)
 	 */
-	samplecount += length;
-	if ((samplecount & 8191) < length)
+	jittercount += length;
+	if (jittercount >= 8000) {
+		jittercount -= 8000;
 		jittercheck = 1;
+	}
 
 	/* loop all members that do not require conference mixing */
 	list_for_each_entry(dsp, &dsp_ilist, list) {
diff --git a/include/linux/mISDNif.h b/include/linux/mISDNif.h
index f75d596c5316..364f1018f0d1 100644
--- a/include/linux/mISDNif.h
+++ b/include/linux/mISDNif.h
@@ -371,6 +371,7 @@ struct mISDN_ctrl_req {
 #define DEBUG_L2_TEI		0x00100000
 #define DEBUG_L2_TEIFSM		0x00200000
 #define DEBUG_TIMER		0x01000000
+#define DEBUG_CLOCK		0x02000000
 
 #define mISDN_HEAD_P(s)		((struct mISDNhead *)&s->cb[0])
 #define mISDN_HEAD_PRIM(s)	(((struct mISDNhead *)&s->cb[0])->prim)
@@ -384,6 +385,7 @@ struct mISDN_ctrl_req {
 struct mISDNchannel;
 struct mISDNdevice;
 struct mISDNstack;
+struct mISDNclock;
 
 struct channel_req {
 	u_int			protocol;
@@ -460,6 +462,16 @@ struct mISDNstack {
 #endif
 };
 
+typedef	int	(clockctl_func_t)(void *, int);
+
+struct	mISDNclock {
+	struct list_head	list;
+	char			name[64];
+	int			pri;
+	clockctl_func_t		*ctl;
+	void			*priv;
+};
+
 /* global alloc/queue functions */
 
 static inline struct sk_buff *
@@ -510,8 +522,13 @@ extern int	mISDN_register_device(struct mISDNdevice *, char *name);
 extern void	mISDN_unregister_device(struct mISDNdevice *);
 extern int	mISDN_register_Bprotocol(struct Bprotocol *);
 extern void	mISDN_unregister_Bprotocol(struct Bprotocol *);
+extern struct mISDNclock *mISDN_register_clock(char *, int, clockctl_func_t *,
+						void *);
+extern void	mISDN_unregister_clock(struct mISDNclock *);
 
 extern void	set_channel_address(struct mISDNchannel *, u_int, u_int);
+extern void	mISDN_clock_update(struct mISDNclock *, int, struct timeval *);
+extern unsigned short mISDN_clock_get(void);
 
 #endif /* __KERNEL__ */
 #endif /* mISDNIF_H */
-- 
cgit v1.2.3


From 02282eee56b75a35e6bbc42cc34c9005eb1653f4 Mon Sep 17 00:00:00 2001
From: Martin Bachem <m.bachem@gmx.de>
Date: Mon, 8 Sep 2008 15:57:48 +0200
Subject: mISDN: Add ISDN_P_TE_UP0 / ISDN_P_NT_UP0

- new layer1 protocols for UP0 bus
- helper #defines to test for TE/NT/S0/E1/UP0

Signed-off-by: Martin Bachem <m.bachem@gmx.de>
Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 include/linux/mISDNif.h | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/mISDNif.h b/include/linux/mISDNif.h
index 364f1018f0d1..7f65aa0c1cc5 100644
--- a/include/linux/mISDNif.h
+++ b/include/linux/mISDNif.h
@@ -200,6 +200,18 @@
 #define ISDN_P_NT_S0  		0x02
 #define ISDN_P_TE_E1		0x03
 #define ISDN_P_NT_E1  		0x04
+#define ISDN_P_TE_UP0		0x05
+#define ISDN_P_NT_UP0		0x06
+
+#define IS_ISDN_P_TE(p) ((p == ISDN_P_TE_S0) || (p == ISDN_P_TE_E1) || \
+				(p == ISDN_P_TE_UP0))
+#define IS_ISDN_P_NT(p) ((p == ISDN_P_NT_S0) || (p == ISDN_P_NT_E1) || \
+				(p == ISDN_P_NT_UP0))
+#define IS_ISDN_P_S0(p) ((p == ISDN_P_TE_S0) || (p == ISDN_P_NT_S0))
+#define IS_ISDN_P_E1(p) ((p == ISDN_P_TE_E1) || (p == ISDN_P_NT_E1))
+#define IS_ISDN_P_UP0(p) ((p == ISDN_P_TE_UP0) || (p == ISDN_P_NT_UP0))
+
+
 #define ISDN_P_LAPD_TE		0x10
 #define	ISDN_P_LAPD_NT		0x11
 
-- 
cgit v1.2.3


From 1b4d33121f1d991f6ae226cc3333428ff87627bb Mon Sep 17 00:00:00 2001
From: Andreas Eversberg <andreas@eversberg.eu>
Date: Sun, 14 Sep 2008 12:30:18 +0200
Subject: mISDN: Fix deactivation, if peer IP is removed from l1oip instance.

 Added GETPEER operation.
 Socket now checks if device is already busy at a differen mode.

Signed-off-by: Andreas Eversberg <andreas@eversberg.eu>
Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 drivers/isdn/mISDN/l1oip_core.c | 22 +++++++++++++++++++++-
 drivers/isdn/mISDN/socket.c     | 20 ++++++++++++++++++++
 drivers/isdn/mISDN/stack.c      | 18 ------------------
 include/linux/mISDNif.h         |  5 +++--
 4 files changed, 44 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/isdn/mISDN/l1oip_core.c b/drivers/isdn/mISDN/l1oip_core.c
index 0884dd6892f8..3ddcd2e09dc9 100644
--- a/drivers/isdn/mISDN/l1oip_core.c
+++ b/drivers/isdn/mISDN/l1oip_core.c
@@ -777,6 +777,8 @@ fail:
 static void
 l1oip_socket_close(struct l1oip *hc)
 {
+	struct dchannel *dch = hc->chan[hc->d_idx].dch;
+
 	/* kill thread */
 	if (hc->socket_thread) {
 		if (debug & DEBUG_L1OIP_SOCKET)
@@ -785,6 +787,16 @@ l1oip_socket_close(struct l1oip *hc)
 		send_sig(SIGTERM, hc->socket_thread, 0);
 		wait_for_completion(&hc->socket_complete);
 	}
+
+	/* if active, we send up a PH_DEACTIVATE and deactivate */
+	if (test_bit(FLG_ACTIVE, &dch->Flags)) {
+		if (debug & (DEBUG_L1OIP_MSG|DEBUG_L1OIP_SOCKET))
+			printk(KERN_DEBUG "%s: interface become deactivated "
+				"due to timeout\n", __func__);
+		test_and_clear_bit(FLG_ACTIVE, &dch->Flags);
+		_queue_data(&dch->dev.D, PH_DEACTIVATE_IND, MISDN_ID_ANY, 0,
+			NULL, GFP_ATOMIC);
+	}
 }
 
 static int
@@ -944,7 +956,8 @@ channel_dctrl(struct dchannel *dch, struct mISDN_ctrl_req *cq)
 
 	switch (cq->op) {
 	case MISDN_CTRL_GETOP:
-		cq->op = MISDN_CTRL_SETPEER | MISDN_CTRL_UNSETPEER;
+		cq->op = MISDN_CTRL_SETPEER | MISDN_CTRL_UNSETPEER
+			| MISDN_CTRL_GETPEER;
 		break;
 	case MISDN_CTRL_SETPEER:
 		hc->remoteip = (u32)cq->p1;
@@ -964,6 +977,13 @@ channel_dctrl(struct dchannel *dch, struct mISDN_ctrl_req *cq)
 		hc->remoteip = 0;
 		l1oip_socket_open(hc);
 		break;
+	case MISDN_CTRL_GETPEER:
+		if (debug & DEBUG_L1OIP_SOCKET)
+			printk(KERN_DEBUG "%s: getting ip address.\n",
+				__func__);
+		(u32)cq->p1 = hc->remoteip;
+		cq->p2 = hc->remoteport | (hc->localport << 16);
+		break;
 	default:
 		printk(KERN_WARNING "%s: unknown Op %x\n",
 		    __func__, cq->op);
diff --git a/drivers/isdn/mISDN/socket.c b/drivers/isdn/mISDN/socket.c
index 2f6d6e88ff2c..916569ca156d 100644
--- a/drivers/isdn/mISDN/socket.c
+++ b/drivers/isdn/mISDN/socket.c
@@ -460,6 +460,8 @@ data_sock_bind(struct socket *sock, struct sockaddr *addr, int addr_len)
 {
 	struct sockaddr_mISDN *maddr = (struct sockaddr_mISDN *) addr;
 	struct sock *sk = sock->sk;
+	struct hlist_node *node;
+	struct sock *csk;
 	int err = 0;
 
 	if (*debug & DEBUG_SOCKET)
@@ -480,6 +482,24 @@ data_sock_bind(struct socket *sock, struct sockaddr *addr, int addr_len)
 		err = -ENODEV;
 		goto done;
 	}
+
+	read_lock_bh(&data_sockets.lock);
+	sk_for_each(csk, node, &data_sockets.head) {
+		if (sk == csk)
+			continue;
+		if (_pms(csk)->dev != _pms(sk)->dev)
+			continue;
+		if (csk->sk_protocol >= ISDN_P_B_START)
+			continue;
+		if (IS_ISDN_P_TE(csk->sk_protocol)
+				== IS_ISDN_P_TE(sk->sk_protocol))
+			continue;
+		read_unlock_bh(&data_sockets.lock);
+		err = -EBUSY;
+		goto done;
+	}
+	read_unlock_bh(&data_sockets.lock);
+
 	_pms(sk)->ch.send = mISDN_send;
 	_pms(sk)->ch.ctrl = mISDN_ctrl;
 
diff --git a/drivers/isdn/mISDN/stack.c b/drivers/isdn/mISDN/stack.c
index 8cff570bb8df..63afa8cf9e07 100644
--- a/drivers/isdn/mISDN/stack.c
+++ b/drivers/isdn/mISDN/stack.c
@@ -440,15 +440,6 @@ connect_layer1(struct mISDNdevice *dev, struct mISDNchannel *ch,
 	case ISDN_P_NT_E1:
 	case ISDN_P_TE_S0:
 	case ISDN_P_TE_E1:
-#ifdef PROTOCOL_CHECK
-		/* this should be enhanced */
-		if (!list_empty(&dev->D.st->layer2)
-			&& dev->D.protocol != protocol)
-			return -EBUSY;
-		if (!hlist_empty(&dev->D.st->l1sock.head)
-			&& dev->D.protocol != protocol)
-			return -EBUSY;
-#endif
 		ch->recv = mISDN_queue_message;
 		ch->peer = &dev->D.st->own;
 		ch->st = dev->D.st;
@@ -546,15 +537,6 @@ create_l2entity(struct mISDNdevice *dev, struct mISDNchannel *ch,
 		if (dev->Dprotocols & (1 << ISDN_P_NT_E1))
 			rq.protocol = ISDN_P_NT_E1;
 	case ISDN_P_LAPD_TE:
-#ifdef PROTOCOL_CHECK
-		/* this should be enhanced */
-		if (!list_empty(&dev->D.st->layer2)
-			&& dev->D.protocol != protocol)
-			return -EBUSY;
-		if (!hlist_empty(&dev->D.st->l1sock.head)
-			&& dev->D.protocol != protocol)
-			return -EBUSY;
-#endif
 		ch->recv = mISDN_queue_message;
 		ch->peer = &dev->D.st->own;
 		ch->st = dev->D.st;
diff --git a/include/linux/mISDNif.h b/include/linux/mISDNif.h
index 7f65aa0c1cc5..3f9988849f32 100644
--- a/include/linux/mISDNif.h
+++ b/include/linux/mISDNif.h
@@ -204,9 +204,9 @@
 #define ISDN_P_NT_UP0		0x06
 
 #define IS_ISDN_P_TE(p) ((p == ISDN_P_TE_S0) || (p == ISDN_P_TE_E1) || \
-				(p == ISDN_P_TE_UP0))
+				(p == ISDN_P_TE_UP0) || (p == ISDN_P_LAPD_TE))
 #define IS_ISDN_P_NT(p) ((p == ISDN_P_NT_S0) || (p == ISDN_P_NT_E1) || \
-				(p == ISDN_P_NT_UP0))
+				(p == ISDN_P_NT_UP0) || (p == ISDN_P_LAPD_NT))
 #define IS_ISDN_P_S0(p) ((p == ISDN_P_TE_S0) || (p == ISDN_P_NT_S0))
 #define IS_ISDN_P_E1(p) ((p == ISDN_P_TE_E1) || (p == ISDN_P_NT_E1))
 #define IS_ISDN_P_UP0(p) ((p == ISDN_P_TE_UP0) || (p == ISDN_P_NT_UP0))
@@ -333,6 +333,7 @@ clear_channelmap(u_int nr, u_char *map)
 #define MISDN_CTRL_UNSETPEER		0x0080
 #define MISDN_CTRL_RX_OFF		0x0100
 #define MISDN_CTRL_FILL_EMPTY		0x0200
+#define MISDN_CTRL_GETPEER		0x0400
 #define MISDN_CTRL_HW_FEATURES_OP	0x2000
 #define MISDN_CTRL_HW_FEATURES		0x2001
 #define MISDN_CTRL_HFC_OP		0x4000
-- 
cgit v1.2.3


From b36b654a7e82308cea063cdf909a7f246105c2a3 Mon Sep 17 00:00:00 2001
From: Matthias Urlichs <smurf@smurf.noris.de>
Date: Sat, 16 Aug 2008 00:09:24 +0200
Subject: mISDN: Create /sys/class/mISDN

Create /sys/class/mISDN and implement functions to handle
device renames.

Signed-Off-By: Matthias Urlichs <matthias@urlichs.de>
Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 drivers/isdn/hardware/mISDN/hfcmulti.c |   6 +-
 drivers/isdn/hardware/mISDN/hfcpci.c   | 100 ++++++------
 drivers/isdn/mISDN/core.c              | 271 ++++++++++++++++++++++++++-------
 drivers/isdn/mISDN/l1oip_core.c        |   3 +-
 include/linux/mISDNif.h                |   8 +-
 5 files changed, 273 insertions(+), 115 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/isdn/hardware/mISDN/hfcmulti.c b/drivers/isdn/hardware/mISDN/hfcmulti.c
index cd3f3994b7f8..97f4708b3879 100644
--- a/drivers/isdn/hardware/mISDN/hfcmulti.c
+++ b/drivers/isdn/hardware/mISDN/hfcmulti.c
@@ -4718,7 +4718,7 @@ init_e1_port(struct hfc_multi *hc, struct hm_map *m)
 	} else
 		hc->chan[hc->dslot].jitter = 2; /* default */
 	snprintf(name, MISDN_MAX_IDLEN - 1, "hfc-e1.%d", HFC_cnt + 1);
-	ret = mISDN_register_device(&dch->dev, name);
+	ret = mISDN_register_device(&dch->dev, &hc->pci_dev->dev, name);
 	if (ret)
 		goto free_chan;
 	hc->created[0] = 1;
@@ -4826,9 +4826,9 @@ init_multi_port(struct hfc_multi *hc, int pt)
 		test_and_set_bit(HFC_CFG_DIS_ECHANNEL,
 		    &hc->chan[i + 2].cfg);
 	}
-	snprintf(name, MISDN_MAX_IDLEN - 1, "hfc-%ds.%d/%d",
+	snprintf(name, MISDN_MAX_IDLEN - 1, "hfc-%ds.%d-%d",
 		hc->type, HFC_cnt + 1, pt + 1);
-	ret = mISDN_register_device(&dch->dev, name);
+	ret = mISDN_register_device(&dch->dev, &hc->pci_dev->dev, name);
 	if (ret)
 		goto free_chan;
 	hc->created[pt] = 1;
diff --git a/drivers/isdn/hardware/mISDN/hfcpci.c b/drivers/isdn/hardware/mISDN/hfcpci.c
index cd5d26c9909f..8df12bf02af3 100644
--- a/drivers/isdn/hardware/mISDN/hfcpci.c
+++ b/drivers/isdn/hardware/mISDN/hfcpci.c
@@ -64,9 +64,6 @@ MODULE_LICENSE("GPL");
 module_param(debug, uint, 0);
 module_param(poll, uint, S_IRUGO | S_IWUSR);
 
-static LIST_HEAD(HFClist);
-static DEFINE_RWLOCK(HFClock);
-
 enum {
 	HFC_CCD_2BD0,
 	HFC_CCD_B000,
@@ -136,7 +133,6 @@ struct hfcPCI_hw {
 
 
 struct hfc_pci {
-	struct list_head	list;
 	u_char			subtype;
 	u_char			chanlimit;
 	u_char			initdone;
@@ -1227,41 +1223,6 @@ hfcpci_int(int intno, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-static void
-hfcpci_softirq(void *arg)
-{
-	u_long		flags;
-	struct bchannel	*bch;
-	struct hfc_pci	*hc;
-
-	write_lock_irqsave(&HFClock, flags);
-	list_for_each_entry(hc, &HFClist, list) {
-		if (hc->hw.int_m2 & HFCPCI_IRQ_ENABLE) {
-			spin_lock(&hc->lock);
-			bch = Sel_BCS(hc, hc->hw.bswapped ? 2 : 1);
-			if (bch && bch->state == ISDN_P_B_RAW) { /* B1 rx&tx */
-				main_rec_hfcpci(bch);
-				tx_birq(bch);
-			}
-			bch = Sel_BCS(hc, hc->hw.bswapped ? 1 : 2);
-			if (bch && bch->state == ISDN_P_B_RAW) { /* B2 rx&tx */
-				main_rec_hfcpci(bch);
-				tx_birq(bch);
-			}
-			spin_unlock(&hc->lock);
-		}
-	}
-	write_unlock_irqrestore(&HFClock, flags);
-
-	/* if next event would be in the past ... */
-	if ((s32)(hfc_jiffies + tics - jiffies) <= 0)
-		hfc_jiffies = jiffies + 1;
-	else
-		hfc_jiffies += tics;
-	hfc_tl.expires = hfc_jiffies;
-	add_timer(&hfc_tl);
-}
-
 /*
  * timer callback for D-chan busy resolution. Currently no function
  */
@@ -2131,7 +2092,6 @@ release_card(struct hfc_pci *hc) {
 	mISDN_freebchannel(&hc->bch[1]);
 	mISDN_freebchannel(&hc->bch[0]);
 	mISDN_freedchannel(&hc->dch);
-	list_del(&hc->list);
 	pci_set_drvdata(hc->pdev, NULL);
 	kfree(hc);
 }
@@ -2141,7 +2101,6 @@ setup_card(struct hfc_pci *card)
 {
 	int		err = -EINVAL;
 	u_int		i;
-	u_long		flags;
 	char		name[MISDN_MAX_IDLEN];
 
 	card->dch.debug = debug;
@@ -2169,13 +2128,10 @@ setup_card(struct hfc_pci *card)
 	if (err)
 		goto error;
 	snprintf(name, MISDN_MAX_IDLEN - 1, "hfc-pci.%d", HFC_cnt + 1);
-	err = mISDN_register_device(&card->dch.dev, name);
+	err = mISDN_register_device(&card->dch.dev, &card->pdev->dev, name);
 	if (err)
 		goto error;
 	HFC_cnt++;
-	write_lock_irqsave(&HFClock, flags);
-	list_add_tail(&card->list, &HFClist);
-	write_unlock_irqrestore(&HFClock, flags);
 	printk(KERN_INFO "HFC %d cards installed\n", HFC_cnt);
 	return 0;
 error:
@@ -2311,15 +2267,12 @@ static void __devexit
 hfc_remove_pci(struct pci_dev *pdev)
 {
 	struct hfc_pci	*card = pci_get_drvdata(pdev);
-	u_long		flags;
 
-	if (card) {
-		write_lock_irqsave(&HFClock, flags);
+	if (card)
 		release_card(card);
-		write_unlock_irqrestore(&HFClock, flags);
-	} else
+	else
 		if (debug)
-			printk(KERN_WARNING "%s: drvdata allready removed\n",
+			printk(KERN_WARNING "%s: drvdata already removed\n",
 			    __func__);
 }
 
@@ -2331,6 +2284,46 @@ static struct pci_driver hfc_driver = {
 	.id_table = hfc_ids,
 };
 
+static int
+_hfcpci_softirq(struct device *dev, void *arg)
+{
+	struct hfc_pci  *hc = dev_get_drvdata(dev);
+	struct bchannel *bch;
+	if (hc == NULL)
+		return 0;
+
+	if (hc->hw.int_m2 & HFCPCI_IRQ_ENABLE) {
+		spin_lock(&hc->lock);
+		bch = Sel_BCS(hc, hc->hw.bswapped ? 2 : 1);
+		if (bch && bch->state == ISDN_P_B_RAW) { /* B1 rx&tx */
+			main_rec_hfcpci(bch);
+			tx_birq(bch);
+		}
+		bch = Sel_BCS(hc, hc->hw.bswapped ? 1 : 2);
+		if (bch && bch->state == ISDN_P_B_RAW) { /* B2 rx&tx */
+			main_rec_hfcpci(bch);
+			tx_birq(bch);
+		}
+		spin_unlock(&hc->lock);
+	}
+	return 0;
+}
+
+static void
+hfcpci_softirq(void *arg)
+{
+	(void) driver_for_each_device(&hfc_driver.driver, NULL, arg,
+					_hfcpci_softirq);
+
+	/* if next event would be in the past ... */
+	if ((s32)(hfc_jiffies + tics - jiffies) <= 0)
+		hfc_jiffies = jiffies + 1;
+	else
+		hfc_jiffies += tics;
+	hfc_tl.expires = hfc_jiffies;
+	add_timer(&hfc_tl);
+}
+
 static int __init
 HFC_init(void)
 {
@@ -2375,14 +2368,9 @@ HFC_init(void)
 static void __exit
 HFC_cleanup(void)
 {
-	struct hfc_pci	*card, *next;
-
 	if (timer_pending(&hfc_tl))
 		del_timer(&hfc_tl);
 
-	list_for_each_entry_safe(card, next, &HFClist, list) {
-		release_card(card);
-	}
 	pci_unregister_driver(&hfc_driver);
 }
 
diff --git a/drivers/isdn/mISDN/core.c b/drivers/isdn/mISDN/core.c
index 9116f54def2c..9426c9827e47 100644
--- a/drivers/isdn/mISDN/core.c
+++ b/drivers/isdn/mISDN/core.c
@@ -25,39 +25,183 @@ MODULE_AUTHOR("Karsten Keil");
 MODULE_LICENSE("GPL");
 module_param(debug, uint, S_IRUGO | S_IWUSR);
 
-static LIST_HEAD(devices);
-static DEFINE_RWLOCK(device_lock);
 static u64		device_ids;
 #define MAX_DEVICE_ID	63
 
 static LIST_HEAD(Bprotocols);
 static DEFINE_RWLOCK(bp_lock);
 
+static void mISDN_dev_release(struct device *dev)
+{
+	/* nothing to do: the device is part of its parent's data structure */
+}
+
+static ssize_t _show_id(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct mISDNdevice *mdev = dev_to_mISDN(dev);
+
+	if (!mdev)
+		return -ENODEV;
+	return sprintf(buf, "%d\n", mdev->id);
+}
+
+static ssize_t _show_nrbchan(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct mISDNdevice *mdev = dev_to_mISDN(dev);
+
+	if (!mdev)
+		return -ENODEV;
+	return sprintf(buf, "%d\n", mdev->nrbchan);
+}
+
+static ssize_t _show_d_protocols(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct mISDNdevice *mdev = dev_to_mISDN(dev);
+
+	if (!mdev)
+		return -ENODEV;
+	return sprintf(buf, "%d\n", mdev->Dprotocols);
+}
+
+static ssize_t _show_b_protocols(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct mISDNdevice *mdev = dev_to_mISDN(dev);
+
+	if (!mdev)
+		return -ENODEV;
+	return sprintf(buf, "%d\n", mdev->Bprotocols | get_all_Bprotocols());
+}
+
+static ssize_t _show_protocol(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct mISDNdevice *mdev = dev_to_mISDN(dev);
+
+	if (!mdev)
+		return -ENODEV;
+	return sprintf(buf, "%d\n", mdev->D.protocol);
+}
+
+static ssize_t _show_name(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	strcpy(buf, dev_name(dev));
+	return strlen(buf);
+}
+
+#if 0 /* hangs */
+static ssize_t _set_name(struct device *dev, struct device_attribute *attr,
+				const char *buf, size_t count)
+{
+	int err = 0;
+	char *out = kmalloc(count + 1, GFP_KERNEL);
+
+	if (!out)
+		return -ENOMEM;
+
+	memcpy(out, buf, count);
+	if (count && out[count - 1] == '\n')
+		out[--count] = 0;
+	if (count)
+		err = device_rename(dev, out);
+	kfree(out);
+
+	return (err < 0) ? err : count;
+}
+#endif
+
+static ssize_t _show_channelmap(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct mISDNdevice *mdev = dev_to_mISDN(dev);
+	char *bp = buf;
+	int i;
+
+	for (i = 0; i <= mdev->nrbchan; i++)
+		*bp++ = test_channelmap(i, mdev->channelmap) ? '1' : '0';
+
+	return bp - buf;
+}
+
+static struct device_attribute mISDN_dev_attrs[] = {
+	__ATTR(id,          S_IRUGO,         _show_id,          NULL),
+	__ATTR(d_protocols, S_IRUGO,         _show_d_protocols, NULL),
+	__ATTR(b_protocols, S_IRUGO,         _show_b_protocols, NULL),
+	__ATTR(protocol,    S_IRUGO,         _show_protocol,    NULL),
+	__ATTR(channelmap,  S_IRUGO,         _show_channelmap,  NULL),
+	__ATTR(nrbchan,     S_IRUGO,         _show_nrbchan,     NULL),
+	__ATTR(name,        S_IRUGO,         _show_name,        NULL),
+/*	__ATTR(name,        S_IRUGO|S_IWUSR, _show_name,       _set_name), */
+	{}
+};
+
+#ifdef CONFIG_HOTPLUG
+static int mISDN_uevent(struct device *dev, struct kobj_uevent_env *env)
+{
+	struct mISDNdevice *mdev = dev_to_mISDN(dev);
+
+	if (!mdev)
+		return 0;
+
+	if (add_uevent_var(env, "nchans=%d", mdev->nrbchan))
+		return -ENOMEM;
+
+	return 0;
+}
+#endif
+
+static void mISDN_class_release(struct class *cls)
+{
+	/* do nothing, it's static */
+}
+
+static struct class mISDN_class = {
+	.name = "mISDN",
+	.owner = THIS_MODULE,
+#ifdef CONFIG_HOTPLUG
+	.dev_uevent = mISDN_uevent,
+#endif
+	.dev_attrs = mISDN_dev_attrs,
+	.dev_release = mISDN_dev_release,
+	.class_release = mISDN_class_release,
+};
+
+static int
+_get_mdevice(struct device *dev, void *id)
+{
+	struct mISDNdevice *mdev = dev_to_mISDN(dev);
+
+	if (!mdev)
+		return 0;
+	if (mdev->id != *(u_int *)id)
+		return 0;
+	return 1;
+}
+
 struct mISDNdevice
 *get_mdevice(u_int id)
 {
-	struct mISDNdevice	*dev;
+	return dev_to_mISDN(class_find_device(&mISDN_class, NULL, &id,
+		_get_mdevice));
+}
 
-	read_lock(&device_lock);
-	list_for_each_entry(dev, &devices, D.list)
-		if (dev->id == id) {
-			read_unlock(&device_lock);
-			return dev;
-		}
-	read_unlock(&device_lock);
-	return NULL;
+static int
+_get_mdevice_count(struct device *dev, void *cnt)
+{
+	*(int *)cnt += 1;
+	return 0;
 }
 
 int
 get_mdevice_count(void)
 {
-	struct mISDNdevice	*dev;
-	int			cnt = 0;
+	int cnt = 0;
 
-	read_lock(&device_lock);
-	list_for_each_entry(dev, &devices, D.list)
-		cnt++;
-	read_unlock(&device_lock);
+	class_for_each_device(&mISDN_class, NULL, &cnt, _get_mdevice_count);
 	return cnt;
 }
 
@@ -68,19 +212,24 @@ get_free_devid(void)
 
 	for (i = 0; i <= MAX_DEVICE_ID; i++)
 		if (!test_and_set_bit(i, (u_long *)&device_ids))
-			return i;
-	return -1;
+			break;
+	if (i > MAX_DEVICE_ID)
+		return -1;
+	return i;
 }
 
 int
-mISDN_register_device(struct mISDNdevice *dev, char *name)
+mISDN_register_device(struct mISDNdevice *dev,
+			struct device *parent, char *name)
 {
-	u_long	flags;
 	int	err;
 
 	dev->id = get_free_devid();
+	err = -EBUSY;
 	if (dev->id < 0)
-		return -EBUSY;
+		goto error1;
+
+	device_initialize(&dev->dev);
 	if (name && name[0])
 		dev_set_name(&dev->dev, "%s", name);
 	else
@@ -90,26 +239,39 @@ mISDN_register_device(struct mISDNdevice *dev, char *name)
 			dev_name(&dev->dev), dev->id);
 	err = create_stack(dev);
 	if (err)
-		return err;
-	write_lock_irqsave(&device_lock, flags);
-	list_add_tail(&dev->D.list, &devices);
-	write_unlock_irqrestore(&device_lock, flags);
+		goto error1;
+
+	dev->dev.class = &mISDN_class;
+	dev->dev.platform_data = dev;
+	dev->dev.parent = parent;
+	dev_set_drvdata(&dev->dev, dev);
+
+	err = device_add(&dev->dev);
+	if (err)
+		goto error3;
 	return 0;
+
+error3:
+	delete_stack(dev);
+	return err;
+error1:
+	return err;
+
 }
 EXPORT_SYMBOL(mISDN_register_device);
 
 void
 mISDN_unregister_device(struct mISDNdevice *dev) {
-	u_long	flags;
-
 	if (debug & DEBUG_CORE)
 		printk(KERN_DEBUG "mISDN_unregister %s %d\n",
 			dev_name(&dev->dev), dev->id);
-	write_lock_irqsave(&device_lock, flags);
-	list_del(&dev->D.list);
-	write_unlock_irqrestore(&device_lock, flags);
+	/* sysfs_remove_link(&dev->dev.kobj, "device"); */
+	device_del(&dev->dev);
+	dev_set_drvdata(&dev->dev, NULL);
+
 	test_and_clear_bit(dev->id, (u_long *)&device_ids);
 	delete_stack(dev);
+	put_device(&dev->dev);
 }
 EXPORT_SYMBOL(mISDN_unregister_device);
 
@@ -201,42 +363,43 @@ mISDNInit(void)
 		MISDN_MAJOR_VERSION, MISDN_MINOR_VERSION, MISDN_RELEASE);
 	mISDN_init_clock(&debug);
 	mISDN_initstack(&debug);
+	err = class_register(&mISDN_class);
+	if (err)
+		goto error1;
 	err = mISDN_inittimer(&debug);
 	if (err)
-		goto error;
+		goto error2;
 	err = l1_init(&debug);
-	if (err) {
-		mISDN_timer_cleanup();
-		goto error;
-	}
+	if (err)
+		goto error3;
 	err = Isdnl2_Init(&debug);
-	if (err) {
-		mISDN_timer_cleanup();
-		l1_cleanup();
-		goto error;
-	}
+	if (err)
+		goto error4;
 	err = misdn_sock_init(&debug);
-	if (err) {
-		mISDN_timer_cleanup();
-		l1_cleanup();
-		Isdnl2_cleanup();
-	}
-error:
+	if (err)
+		goto error5;
+	return 0;
+
+error5:
+	Isdnl2_cleanup();
+error4:
+	l1_cleanup();
+error3:
+	mISDN_timer_cleanup();
+error2:
+	class_unregister(&mISDN_class);
+error1:
 	return err;
 }
 
 static void mISDN_cleanup(void)
 {
 	misdn_sock_cleanup();
-	mISDN_timer_cleanup();
-	l1_cleanup();
 	Isdnl2_cleanup();
+	l1_cleanup();
+	mISDN_timer_cleanup();
+	class_unregister(&mISDN_class);
 
-	if (!list_empty(&devices))
-		printk(KERN_ERR "%s devices still registered\n", __func__);
-
-	if (!list_empty(&Bprotocols))
-		printk(KERN_ERR "%s Bprotocols still registered\n", __func__);
 	printk(KERN_DEBUG "mISDNcore unloaded\n");
 }
 
diff --git a/drivers/isdn/mISDN/l1oip_core.c b/drivers/isdn/mISDN/l1oip_core.c
index a6a9e1af2ad6..abe574989572 100644
--- a/drivers/isdn/mISDN/l1oip_core.c
+++ b/drivers/isdn/mISDN/l1oip_core.c
@@ -1433,7 +1433,8 @@ init_card(struct l1oip *hc, int pri, int bundle)
 		hc->chan[i + ch].bch = bch;
 		set_channelmap(bch->nr, dch->dev.channelmap);
 	}
-	ret = mISDN_register_device(&dch->dev, hc->name);
+	/* TODO: create a parent device for this driver */
+	ret = mISDN_register_device(&dch->dev, NULL, hc->name);
 	if (ret)
 		return ret;
 	hc->registered = 1;
diff --git a/include/linux/mISDNif.h b/include/linux/mISDNif.h
index 3f9988849f32..d4229aebf648 100644
--- a/include/linux/mISDNif.h
+++ b/include/linux/mISDNif.h
@@ -531,7 +531,8 @@ _queue_data(struct mISDNchannel *ch, u_int prim,
 
 /* global register/unregister functions */
 
-extern int	mISDN_register_device(struct mISDNdevice *, char *name);
+extern int	mISDN_register_device(struct mISDNdevice *,
+					struct device *parent, char *name);
 extern void	mISDN_unregister_device(struct mISDNdevice *);
 extern int	mISDN_register_Bprotocol(struct Bprotocol *);
 extern void	mISDN_unregister_Bprotocol(struct Bprotocol *);
@@ -539,6 +540,11 @@ extern struct mISDNclock *mISDN_register_clock(char *, int, clockctl_func_t *,
 						void *);
 extern void	mISDN_unregister_clock(struct mISDNclock *);
 
+static inline struct mISDNdevice *dev_to_mISDN(struct device *dev)
+{
+	return dev_get_drvdata(dev);
+}
+
 extern void	set_channel_address(struct mISDNchannel *, u_int, u_int);
 extern void	mISDN_clock_update(struct mISDNclock *, int, struct timeval *);
 extern unsigned short mISDN_clock_get(void);
-- 
cgit v1.2.3


From 3f75e84a6a697c5cffb78ee15e79498a35473e05 Mon Sep 17 00:00:00 2001
From: Martin Bachem <m.bachem@gmx.de>
Date: Tue, 4 Nov 2008 14:11:22 +0100
Subject: mISDN: Add layer1 prim MPH_INFORMATION_REQ

MPH_INFORMATION provides full D- and B-Channel status overview

- new layer1 primitive: MPF_INFORMATON_REQ
- layer1 replies with MPH_INFORMATION_IND containing
   - dch->[state,Flags,nrbchan]
   - bch[]->[protocol,Flags]
- hardware driver should send MPH_INFORMATION_IND
  on all ph state changes and BChannel state changes to MISDN_ID_ANY

Signed-off-by: Martin Bachem <m.bachem@gmx.de>
Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 include/linux/mISDNif.h | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/mISDNif.h b/include/linux/mISDNif.h
index d4229aebf648..557477ac3d5b 100644
--- a/include/linux/mISDNif.h
+++ b/include/linux/mISDNif.h
@@ -289,6 +289,23 @@ struct mISDN_devrename {
 	char			name[MISDN_MAX_IDLEN]; /* new name */
 };
 
+/* MPH_INFORMATION_REQ payload */
+struct ph_info_ch {
+        __u32 protocol;
+        __u64 Flags;
+};
+
+struct ph_info_dch {
+        struct ph_info_ch ch;
+        __u16 state;
+        __u16 num_bch;
+};
+
+struct ph_info {
+        struct ph_info_dch dch;
+        struct ph_info_ch  bch[];
+};
+
 /* timer device ioctl */
 #define IMADDTIMER	_IOR('I', 64, int)
 #define IMDELTIMER	_IOR('I', 65, int)
-- 
cgit v1.2.3


From 4db8e282f2d1dfa43d51ce2a4817901312c9134d Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Fri, 9 Jan 2009 14:32:46 -0800
Subject: Revert "driver core: move knode_bus into private structure"

This reverts commit b9daa99ee533578e3f88231e7a16784dcb44ec42.

Turns out that device_initialize shouldn't fail silently.
This series needs to be reworked in order to get into proper
shape.

Reported-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/base.h    |  4 ----
 drivers/base/bus.c     | 40 +++++++++++++---------------------------
 include/linux/device.h |  1 +
 3 files changed, 14 insertions(+), 31 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/base.h b/drivers/base/base.h
index b676f8f801f8..8af0bb2c0aa8 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -69,7 +69,6 @@ struct class_private {
  * @klist_children - klist containing all children of this device
  * @knode_parent - node in sibling list
  * @knode_driver - node in driver list
- * @knode_bus - node in bus list
  * @device - pointer back to the struct class that this structure is
  * associated with.
  *
@@ -79,15 +78,12 @@ struct device_private {
 	struct klist klist_children;
 	struct klist_node knode_parent;
 	struct klist_node knode_driver;
-	struct klist_node knode_bus;
 	struct device *device;
 };
 #define to_device_private_parent(obj)	\
 	container_of(obj, struct device_private, knode_parent)
 #define to_device_private_driver(obj)	\
 	container_of(obj, struct device_private, knode_driver)
-#define to_device_private_bus(obj)	\
-	container_of(obj, struct device_private, knode_bus)
 
 /* initialisation functions */
 extern int devices_init(void);
diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 0f0a50444672..83f32b891fa9 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -253,14 +253,7 @@ static ssize_t store_drivers_probe(struct bus_type *bus,
 static struct device *next_device(struct klist_iter *i)
 {
 	struct klist_node *n = klist_next(i);
-	struct device *dev = NULL;
-	struct device_private *dev_prv;
-
-	if (n) {
-		dev_prv = to_device_private_bus(n);
-		dev = dev_prv->device;
-	}
-	return dev;
+	return n ? container_of(n, struct device, knode_bus) : NULL;
 }
 
 /**
@@ -293,7 +286,7 @@ int bus_for_each_dev(struct bus_type *bus, struct device *start,
 		return -EINVAL;
 
 	klist_iter_init_node(&bus->p->klist_devices, &i,
-			     (start ? &start->p->knode_bus : NULL));
+			     (start ? &start->knode_bus : NULL));
 	while ((dev = next_device(&i)) && !error)
 		error = fn(dev, data);
 	klist_iter_exit(&i);
@@ -327,7 +320,7 @@ struct device *bus_find_device(struct bus_type *bus,
 		return NULL;
 
 	klist_iter_init_node(&bus->p->klist_devices, &i,
-			     (start ? &start->p->knode_bus : NULL));
+			     (start ? &start->knode_bus : NULL));
 	while ((dev = next_device(&i)))
 		if (match(dev, data) && get_device(dev))
 			break;
@@ -514,8 +507,7 @@ void bus_attach_device(struct device *dev)
 			ret = device_attach(dev);
 		WARN_ON(ret < 0);
 		if (ret >= 0)
-			klist_add_tail(&dev->p->knode_bus,
-				       &bus->p->klist_devices);
+			klist_add_tail(&dev->knode_bus, &bus->p->klist_devices);
 	}
 }
 
@@ -536,8 +528,8 @@ void bus_remove_device(struct device *dev)
 		sysfs_remove_link(&dev->bus->p->devices_kset->kobj,
 				  dev_name(dev));
 		device_remove_attrs(dev->bus, dev);
-		if (klist_node_attached(&dev->p->knode_bus))
-			klist_del(&dev->p->knode_bus);
+		if (klist_node_attached(&dev->knode_bus))
+			klist_del(&dev->knode_bus);
 
 		pr_debug("bus: '%s': remove device %s\n",
 			 dev->bus->name, dev_name(dev));
@@ -839,16 +831,14 @@ static void bus_remove_attrs(struct bus_type *bus)
 
 static void klist_devices_get(struct klist_node *n)
 {
-	struct device_private *dev_prv = to_device_private_bus(n);
-	struct device *dev = dev_prv->device;
+	struct device *dev = container_of(n, struct device, knode_bus);
 
 	get_device(dev);
 }
 
 static void klist_devices_put(struct klist_node *n)
 {
-	struct device_private *dev_prv = to_device_private_bus(n);
-	struct device *dev = dev_prv->device;
+	struct device *dev = container_of(n, struct device, knode_bus);
 
 	put_device(dev);
 }
@@ -1003,20 +993,18 @@ static void device_insertion_sort_klist(struct device *a, struct list_head *list
 {
 	struct list_head *pos;
 	struct klist_node *n;
-	struct device_private *dev_prv;
 	struct device *b;
 
 	list_for_each(pos, list) {
 		n = container_of(pos, struct klist_node, n_node);
-		dev_prv = to_device_private_bus(n);
-		b = dev_prv->device;
+		b = container_of(n, struct device, knode_bus);
 		if (compare(a, b) <= 0) {
-			list_move_tail(&a->p->knode_bus.n_node,
-				       &b->p->knode_bus.n_node);
+			list_move_tail(&a->knode_bus.n_node,
+				       &b->knode_bus.n_node);
 			return;
 		}
 	}
-	list_move_tail(&a->p->knode_bus.n_node, list);
+	list_move_tail(&a->knode_bus.n_node, list);
 }
 
 void bus_sort_breadthfirst(struct bus_type *bus,
@@ -1026,7 +1014,6 @@ void bus_sort_breadthfirst(struct bus_type *bus,
 	LIST_HEAD(sorted_devices);
 	struct list_head *pos, *tmp;
 	struct klist_node *n;
-	struct device_private *dev_prv;
 	struct device *dev;
 	struct klist *device_klist;
 
@@ -1035,8 +1022,7 @@ void bus_sort_breadthfirst(struct bus_type *bus,
 	spin_lock(&device_klist->k_lock);
 	list_for_each_safe(pos, tmp, &device_klist->k_list) {
 		n = container_of(pos, struct klist_node, n_node);
-		dev_prv = to_device_private_bus(n);
-		dev = dev_prv->device;
+		dev = container_of(n, struct device, knode_bus);
 		device_insertion_sort_klist(dev, &sorted_devices, compare);
 	}
 	list_splice(&sorted_devices, &device_klist->k_list);
diff --git a/include/linux/device.h b/include/linux/device.h
index 7d9da4b4993f..8987f4776064 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -366,6 +366,7 @@ struct device_dma_parameters {
 };
 
 struct device {
+	struct klist_node	knode_bus;
 	struct device		*parent;
 
 	struct device_private	*p;
-- 
cgit v1.2.3


From cda5e83fdea476dce9c0a9b1152cd6ca46832cc4 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Fri, 9 Jan 2009 14:44:18 -0800
Subject: Revert "driver core: move knode_driver into private structure"

This reverts commit 93e746db183b3bdbbda67900f79b5835f9cb388f.

Turns out that device_initialize shouldn't fail silently.
This series needs to be reworked in order to get into proper
shape.

Reported-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/base.h    |  4 ----
 drivers/base/dd.c      | 13 +++++--------
 drivers/base/driver.c  | 13 +++----------
 include/linux/device.h |  1 +
 4 files changed, 9 insertions(+), 22 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/base.h b/drivers/base/base.h
index 8af0bb2c0aa8..f5cf31c664d7 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -68,7 +68,6 @@ struct class_private {
  *
  * @klist_children - klist containing all children of this device
  * @knode_parent - node in sibling list
- * @knode_driver - node in driver list
  * @device - pointer back to the struct class that this structure is
  * associated with.
  *
@@ -77,13 +76,10 @@ struct class_private {
 struct device_private {
 	struct klist klist_children;
 	struct klist_node knode_parent;
-	struct klist_node knode_driver;
 	struct device *device;
 };
 #define to_device_private_parent(obj)	\
 	container_of(obj, struct device_private, knode_parent)
-#define to_device_private_driver(obj)	\
-	container_of(obj, struct device_private, knode_driver)
 
 /* initialisation functions */
 extern int devices_init(void);
diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 6fdaf76f033f..315bed8d5e7f 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -28,7 +28,7 @@
 
 static void driver_bound(struct device *dev)
 {
-	if (klist_node_attached(&dev->p->knode_driver)) {
+	if (klist_node_attached(&dev->knode_driver)) {
 		printk(KERN_WARNING "%s: device %s already bound\n",
 			__func__, kobject_name(&dev->kobj));
 		return;
@@ -41,7 +41,7 @@ static void driver_bound(struct device *dev)
 		blocking_notifier_call_chain(&dev->bus->p->bus_notifier,
 					     BUS_NOTIFY_BOUND_DRIVER, dev);
 
-	klist_add_tail(&dev->p->knode_driver, &dev->driver->p->klist_devices);
+	klist_add_tail(&dev->knode_driver, &dev->driver->p->klist_devices);
 }
 
 static int driver_sysfs_add(struct device *dev)
@@ -310,7 +310,7 @@ static void __device_release_driver(struct device *dev)
 			drv->remove(dev);
 		devres_release_all(dev);
 		dev->driver = NULL;
-		klist_remove(&dev->p->knode_driver);
+		klist_remove(&dev->knode_driver);
 	}
 }
 
@@ -340,7 +340,6 @@ EXPORT_SYMBOL_GPL(device_release_driver);
  */
 void driver_detach(struct device_driver *drv)
 {
-	struct device_private *dev_prv;
 	struct device *dev;
 
 	for (;;) {
@@ -349,10 +348,8 @@ void driver_detach(struct device_driver *drv)
 			spin_unlock(&drv->p->klist_devices.k_lock);
 			break;
 		}
-		dev_prv = list_entry(drv->p->klist_devices.k_list.prev,
-				     struct device_private,
-				     knode_driver.n_node);
-		dev = dev_prv->device;
+		dev = list_entry(drv->p->klist_devices.k_list.prev,
+				struct device, knode_driver.n_node);
 		get_device(dev);
 		spin_unlock(&drv->p->klist_devices.k_lock);
 
diff --git a/drivers/base/driver.c b/drivers/base/driver.c
index b76cc69f1106..1e2bda780e48 100644
--- a/drivers/base/driver.c
+++ b/drivers/base/driver.c
@@ -19,14 +19,7 @@
 static struct device *next_device(struct klist_iter *i)
 {
 	struct klist_node *n = klist_next(i);
-	struct device *dev = NULL;
-	struct device_private *dev_prv;
-
-	if (n) {
-		dev_prv = to_device_private_driver(n);
-		dev = dev_prv->device;
-	}
-	return dev;
+	return n ? container_of(n, struct device, knode_driver) : NULL;
 }
 
 /**
@@ -49,7 +42,7 @@ int driver_for_each_device(struct device_driver *drv, struct device *start,
 		return -EINVAL;
 
 	klist_iter_init_node(&drv->p->klist_devices, &i,
-			     start ? &start->p->knode_driver : NULL);
+			     start ? &start->knode_driver : NULL);
 	while ((dev = next_device(&i)) && !error)
 		error = fn(dev, data);
 	klist_iter_exit(&i);
@@ -83,7 +76,7 @@ struct device *driver_find_device(struct device_driver *drv,
 		return NULL;
 
 	klist_iter_init_node(&drv->p->klist_devices, &i,
-			     (start ? &start->p->knode_driver : NULL));
+			     (start ? &start->knode_driver : NULL));
 	while ((dev = next_device(&i)))
 		if (match(dev, data) && get_device(dev))
 			break;
diff --git a/include/linux/device.h b/include/linux/device.h
index 8987f4776064..c66ceb15acd8 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -366,6 +366,7 @@ struct device_dma_parameters {
 };
 
 struct device {
+	struct klist_node	knode_driver;
 	struct klist_node	knode_bus;
 	struct device		*parent;
 
-- 
cgit v1.2.3


From 2526c151c31358aec66b63921dd712bbec5ee0cb Mon Sep 17 00:00:00 2001
From: Jon Smirl <jonsmirl@gmail.com>
Date: Fri, 9 Jan 2009 15:49:06 -0700
Subject: drivers/of: Add the of_find_i2c_device_by_node function.

The of_find_i2c_device_by_node function allows you to follow a
reference in the device tree to an i2c device node and then locate
the linux device instantiated by the device tree. Example use: an I2S
bus driver finding the i2c_device instance for a codec described by
a device tree node.

This was waiting for Anton's i2c patches that were just added.

Signed-off-by: Jon Smirl <jonsmirl@gmail.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
---
 drivers/of/of_i2c.c    | 19 +++++++++++++++++++
 include/linux/of_i2c.h |  3 +++
 2 files changed, 22 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/of/of_i2c.c b/drivers/of/of_i2c.c
index e1b0ad6e918f..fa65a2b2ae2e 100644
--- a/drivers/of/of_i2c.c
+++ b/drivers/of/of_i2c.c
@@ -66,4 +66,23 @@ void of_register_i2c_devices(struct i2c_adapter *adap,
 }
 EXPORT_SYMBOL(of_register_i2c_devices);
 
+static int of_dev_node_match(struct device *dev, void *data)
+{
+        return dev_archdata_get_node(&dev->archdata) == data;
+}
+
+/* must call put_device() when done with returned i2c_client device */
+struct i2c_client *of_find_i2c_device_by_node(struct device_node *node)
+{
+	struct device *dev;
+
+	dev = bus_find_device(&i2c_bus_type, NULL, node,
+					 of_dev_node_match);
+	if (!dev)
+		return NULL;
+
+	return to_i2c_client(dev);
+}
+EXPORT_SYMBOL(of_find_i2c_device_by_node);
+
 MODULE_LICENSE("GPL");
diff --git a/include/linux/of_i2c.h b/include/linux/of_i2c.h
index bd2a870ec296..34974b5a76f7 100644
--- a/include/linux/of_i2c.h
+++ b/include/linux/of_i2c.h
@@ -17,4 +17,7 @@
 void of_register_i2c_devices(struct i2c_adapter *adap,
 			     struct device_node *adap_node);
 
+/* must call put_device() when done with returned i2c_client device */
+struct i2c_client *of_find_i2c_device_by_node(struct device_node *node);
+
 #endif /* __LINUX_OF_I2C_H */
-- 
cgit v1.2.3


From e2d4077678c7ec7661003c268120582adc544897 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Fri, 9 Jan 2009 14:55:37 -0800
Subject: Revert "driver core: move klist_children into private structure"

This reverts commit 11c3b5c3e08f4d855cbef52883c266b9ab9df879.

Turns out that device_initialize shouldn't fail silently.
This series needs to be reworked in order to get into proper
shape.

Reported-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/base.h    |  6 ------
 drivers/base/core.c    | 37 +++++++++++++------------------------
 include/linux/device.h |  2 ++
 3 files changed, 15 insertions(+), 30 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/base.h b/drivers/base/base.h
index f5cf31c664d7..6b20809b5fd4 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -66,20 +66,14 @@ struct class_private {
 /**
  * struct device_private - structure to hold the private to the driver core portions of the device structure.
  *
- * @klist_children - klist containing all children of this device
- * @knode_parent - node in sibling list
  * @device - pointer back to the struct class that this structure is
  * associated with.
  *
  * Nothing outside of the driver core should ever touch these fields.
  */
 struct device_private {
-	struct klist klist_children;
-	struct klist_node knode_parent;
 	struct device *device;
 };
-#define to_device_private_parent(obj)	\
-	container_of(obj, struct device_private, knode_parent)
 
 /* initialisation functions */
 extern int devices_init(void);
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 61df508fa62b..5c2acf7116d7 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -509,16 +509,14 @@ EXPORT_SYMBOL_GPL(device_schedule_callback_owner);
 
 static void klist_children_get(struct klist_node *n)
 {
-	struct device_private *p = to_device_private_parent(n);
-	struct device *dev = p->device;
+	struct device *dev = container_of(n, struct device, knode_parent);
 
 	get_device(dev);
 }
 
 static void klist_children_put(struct klist_node *n)
 {
-	struct device_private *p = to_device_private_parent(n);
-	struct device *dev = p->device;
+	struct device *dev = container_of(n, struct device, knode_parent);
 
 	put_device(dev);
 }
@@ -548,7 +546,7 @@ void device_initialize(struct device *dev)
 	dev->p->device = dev;
 	dev->kobj.kset = devices_kset;
 	kobject_init(&dev->kobj, &device_ktype);
-	klist_init(&dev->p->klist_children, klist_children_get,
+	klist_init(&dev->klist_children, klist_children_get,
 		   klist_children_put);
 	INIT_LIST_HEAD(&dev->dma_pools);
 	init_MUTEX(&dev->sem);
@@ -932,8 +930,7 @@ int device_add(struct device *dev)
 	kobject_uevent(&dev->kobj, KOBJ_ADD);
 	bus_attach_device(dev);
 	if (parent)
-		klist_add_tail(&dev->p->knode_parent,
-			       &parent->p->klist_children);
+		klist_add_tail(&dev->knode_parent, &parent->klist_children);
 
 	if (dev->class) {
 		mutex_lock(&dev->class->p->class_mutex);
@@ -1047,7 +1044,7 @@ void device_del(struct device *dev)
 	device_pm_remove(dev);
 	dpm_sysfs_remove(dev);
 	if (parent)
-		klist_del(&dev->p->knode_parent);
+		klist_del(&dev->knode_parent);
 	if (MAJOR(dev->devt)) {
 		device_remove_sys_dev_entry(dev);
 		device_remove_file(dev, &devt_attr);
@@ -1108,14 +1105,7 @@ void device_unregister(struct device *dev)
 static struct device *next_device(struct klist_iter *i)
 {
 	struct klist_node *n = klist_next(i);
-	struct device *dev = NULL;
-	struct device_private *p;
-
-	if (n) {
-		p = to_device_private_parent(n);
-		dev = p->device;
-	}
-	return dev;
+	return n ? container_of(n, struct device, knode_parent) : NULL;
 }
 
 /**
@@ -1137,7 +1127,7 @@ int device_for_each_child(struct device *parent, void *data,
 	struct device *child;
 	int error = 0;
 
-	klist_iter_init(&parent->p->klist_children, &i);
+	klist_iter_init(&parent->klist_children, &i);
 	while ((child = next_device(&i)) && !error)
 		error = fn(child, data);
 	klist_iter_exit(&i);
@@ -1168,7 +1158,7 @@ struct device *device_find_child(struct device *parent, void *data,
 	if (!parent)
 		return NULL;
 
-	klist_iter_init(&parent->p->klist_children, &i);
+	klist_iter_init(&parent->klist_children, &i);
 	while ((child = next_device(&i)))
 		if (match(child, data) && get_device(child))
 			break;
@@ -1582,10 +1572,9 @@ int device_move(struct device *dev, struct device *new_parent)
 	old_parent = dev->parent;
 	dev->parent = new_parent;
 	if (old_parent)
-		klist_remove(&dev->p->knode_parent);
+		klist_remove(&dev->knode_parent);
 	if (new_parent) {
-		klist_add_tail(&dev->p->knode_parent,
-			       &new_parent->p->klist_children);
+		klist_add_tail(&dev->knode_parent, &new_parent->klist_children);
 		set_dev_node(dev, dev_to_node(new_parent));
 	}
 
@@ -1597,11 +1586,11 @@ int device_move(struct device *dev, struct device *new_parent)
 		device_move_class_links(dev, new_parent, old_parent);
 		if (!kobject_move(&dev->kobj, &old_parent->kobj)) {
 			if (new_parent)
-				klist_remove(&dev->p->knode_parent);
+				klist_remove(&dev->knode_parent);
 			dev->parent = old_parent;
 			if (old_parent) {
-				klist_add_tail(&dev->p->knode_parent,
-					       &old_parent->p->klist_children);
+				klist_add_tail(&dev->knode_parent,
+					       &old_parent->klist_children);
 				set_dev_node(dev, dev_to_node(old_parent));
 			}
 		}
diff --git a/include/linux/device.h b/include/linux/device.h
index c66ceb15acd8..2975351635d3 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -366,6 +366,8 @@ struct device_dma_parameters {
 };
 
 struct device {
+	struct klist		klist_children;
+	struct klist_node	knode_parent;	/* node in sibling list */
 	struct klist_node	knode_driver;
 	struct klist_node	knode_bus;
 	struct device		*parent;
-- 
cgit v1.2.3


From 926beadb3dfaddccb3348a5b9e6c2a1f8290a220 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Fri, 9 Jan 2009 15:06:12 -0800
Subject: Revert "driver core: create a private portion of struct device"

This reverts commit 2831fe6f9cc4e16c103504ee09a47a084297c0f3.

Turns out that device_initialize shouldn't fail silently.
This series needs to be reworked in order to get into proper
shape.

Reported-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/base.h    | 12 ------------
 drivers/base/core.c    |  8 --------
 include/linux/device.h |  3 ---
 3 files changed, 23 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/base/base.h b/drivers/base/base.h
index 6b20809b5fd4..0a5f055dffba 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -63,18 +63,6 @@ struct class_private {
 #define to_class(obj)	\
 	container_of(obj, struct class_private, class_subsys.kobj)
 
-/**
- * struct device_private - structure to hold the private to the driver core portions of the device structure.
- *
- * @device - pointer back to the struct class that this structure is
- * associated with.
- *
- * Nothing outside of the driver core should ever touch these fields.
- */
-struct device_private {
-	struct device *device;
-};
-
 /* initialisation functions */
 extern int devices_init(void);
 extern int buses_init(void);
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 5c2acf7116d7..8079afca4972 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -109,7 +109,6 @@ static struct sysfs_ops dev_sysfs_ops = {
 static void device_release(struct kobject *kobj)
 {
 	struct device *dev = to_dev(kobj);
-	struct device_private *p = dev->p;
 
 	if (dev->release)
 		dev->release(dev);
@@ -121,7 +120,6 @@ static void device_release(struct kobject *kobj)
 		WARN(1, KERN_ERR "Device '%s' does not have a release() "
 			"function, it is broken and must be fixed.\n",
 			dev_name(dev));
-	kfree(p);
 }
 
 static struct kobj_type device_ktype = {
@@ -538,12 +536,6 @@ static void klist_children_put(struct klist_node *n)
  */
 void device_initialize(struct device *dev)
 {
-	dev->p = kzalloc(sizeof(*dev->p), GFP_KERNEL);
-	if (!dev->p) {
-		WARN_ON(1);
-		return;
-	}
-	dev->p->device = dev;
 	dev->kobj.kset = devices_kset;
 	kobject_init(&dev->kobj, &device_ktype);
 	klist_init(&dev->klist_children, klist_children_get,
diff --git a/include/linux/device.h b/include/linux/device.h
index 2975351635d3..45e5b1921fbb 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -28,7 +28,6 @@
 #define BUS_ID_SIZE		20
 
 struct device;
-struct device_private;
 struct device_driver;
 struct driver_private;
 struct class;
@@ -372,8 +371,6 @@ struct device {
 	struct klist_node	knode_bus;
 	struct device		*parent;
 
-	struct device_private	*p;
-
 	struct kobject kobj;
 	char	bus_id[BUS_ID_SIZE];	/* position on parent bus */
 	unsigned		uevent_suppress:1;
-- 
cgit v1.2.3


From 85c210edc46d602a1562aeea0fc74919349c8cf0 Mon Sep 17 00:00:00 2001
From: Andi Kleen <andi@firstfloor.org>
Date: Fri, 9 Jan 2009 16:40:53 -0800
Subject: compiler-gcc.h: add more comments to RELOC_HIDE

Requested by C. Lameter

Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Mike Travis <travis@sgi.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Richard Henderson <rth@twiddle.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/compiler-gcc.h | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h
index af40f8eb86f0..1514d534deeb 100644
--- a/include/linux/compiler-gcc.h
+++ b/include/linux/compiler-gcc.h
@@ -11,9 +11,19 @@
 /* The "volatile" is due to gcc bugs */
 #define barrier() __asm__ __volatile__("": : :"memory")
 
-/* This macro obfuscates arithmetic on a variable address so that gcc
-   shouldn't recognize the original var, and make assumptions about it */
 /*
+ * This macro obfuscates arithmetic on a variable address so that gcc
+ * shouldn't recognize the original var, and make assumptions about it.
+ *
+ * This is needed because the C standard makes it undefined to do
+ * pointer arithmetic on "objects" outside their boundaries and the
+ * gcc optimizers assume this is the case. In particular they
+ * assume such arithmetic does not wrap.
+ *
+ * A miscompilation has been observed because of this on PPC.
+ * To work around it we hide the relationship of the pointer and the object
+ * using this macro.
+ *
  * Versions of the ppc64 compiler before 4.1 had a bug where use of
  * RELOC_HIDE could trash r30. The bug can be worked around by changing
  * the inline assembly constraint from =g to =r, in this particular
-- 
cgit v1.2.3


From 69347a236b22c3962ea812511495e502dedfd50c Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Fri, 9 Jan 2009 16:40:56 -0800
Subject: memstick: annotate endianness of attribute structs

The code was shifting the endianness appropriately everywhere, annotate
the structs to avoid the sparse warnings when assigning the endian types
to the struct members, or passing them to be[16|32]_to_cpu:

drivers/memstick/core/mspro_block.c:331:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:333:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:335:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:337:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:341:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:347:4: warning: cast to restricted __be32
drivers/memstick/core/mspro_block.c:356:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:358:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:364:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:367:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:369:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:371:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:377:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:478:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:480:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:482:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:484:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:486:4: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:689:22:    expected unsigned int [unsigned] [assigned] data_address
drivers/memstick/core/mspro_block.c:689:22:    got restricted __be32 [usertype] <noident>
drivers/memstick/core/mspro_block.c:697:3: warning: cast to restricted __be32
drivers/memstick/core/mspro_block.c:960:17: warning: incorrect type in initializer (different base types)
drivers/memstick/core/mspro_block.c:960:17:    expected unsigned short [unsigned] data_count
drivers/memstick/core/mspro_block.c:960:17:    got restricted __be16 [usertype] <noident>
drivers/memstick/core/mspro_block.c:993:6: warning: cast to restricted __be16
drivers/memstick/core/mspro_block.c:995:28: warning: cast to restricted __be16

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Alex Dubov <oakad@yahoo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/memstick/core/mspro_block.c | 43 ++++++++++++++++++-------------------
 include/linux/memstick.h            |  4 ++--
 2 files changed, 23 insertions(+), 24 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/memstick/core/mspro_block.c b/drivers/memstick/core/mspro_block.c
index 1f1e3982b1aa..de143deb06f0 100644
--- a/drivers/memstick/core/mspro_block.c
+++ b/drivers/memstick/core/mspro_block.c
@@ -52,14 +52,14 @@ struct mspro_sys_attr {
 };
 
 struct mspro_attr_entry {
-	unsigned int  address;
-	unsigned int  size;
+	__be32 address;
+	__be32 size;
 	unsigned char id;
 	unsigned char reserved[3];
 } __attribute__((packed));
 
 struct mspro_attribute {
-	unsigned short          signature;
+	__be16 signature;
 	unsigned short          version;
 	unsigned char           count;
 	unsigned char           reserved[11];
@@ -69,28 +69,28 @@ struct mspro_attribute {
 struct mspro_sys_info {
 	unsigned char  class;
 	unsigned char  reserved0;
-	unsigned short block_size;
-	unsigned short block_count;
-	unsigned short user_block_count;
-	unsigned short page_size;
+	__be16 block_size;
+	__be16 block_count;
+	__be16 user_block_count;
+	__be16 page_size;
 	unsigned char  reserved1[2];
 	unsigned char  assembly_date[8];
-	unsigned int   serial_number;
+	__be32 serial_number;
 	unsigned char  assembly_maker_code;
 	unsigned char  assembly_model_code[3];
-	unsigned short memory_maker_code;
-	unsigned short memory_model_code;
+	__be16 memory_maker_code;
+	__be16 memory_model_code;
 	unsigned char  reserved2[4];
 	unsigned char  vcc;
 	unsigned char  vpp;
-	unsigned short controller_number;
-	unsigned short controller_function;
-	unsigned short start_sector;
-	unsigned short unit_size;
+	__be16 controller_number;
+	__be16 controller_function;
+	__be16 start_sector;
+	__be16 unit_size;
 	unsigned char  ms_sub_class;
 	unsigned char  reserved3[4];
 	unsigned char  interface_type;
-	unsigned short controller_code;
+	__be16 controller_code;
 	unsigned char  format_type;
 	unsigned char  reserved4;
 	unsigned char  device_type;
@@ -124,11 +124,11 @@ struct mspro_specfile {
 } __attribute__((packed));
 
 struct mspro_devinfo {
-	unsigned short cylinders;
-	unsigned short heads;
-	unsigned short bytes_per_track;
-	unsigned short bytes_per_sector;
-	unsigned short sectors_per_track;
+	__be16 cylinders;
+	__be16 heads;
+	__be16 bytes_per_track;
+	__be16 bytes_per_sector;
+	__be16 sectors_per_track;
 	unsigned char  reserved[6];
 } __attribute__((packed));
 
@@ -338,8 +338,7 @@ static ssize_t mspro_block_attr_show_sysinfo(struct device *dev,
 	rc += scnprintf(buffer + rc, PAGE_SIZE - rc, "assembly date: "
 			"GMT%+d:%d %04u-%02u-%02u %02u:%02u:%02u\n",
 			date_tz, date_tz_f,
-			be16_to_cpu(*(unsigned short *)
-				    &x_sys->assembly_date[1]),
+			be16_to_cpup((__be16 *)&x_sys->assembly_date[1]),
 			x_sys->assembly_date[3], x_sys->assembly_date[4],
 			x_sys->assembly_date[5], x_sys->assembly_date[6],
 			x_sys->assembly_date[7]);
diff --git a/include/linux/memstick.h b/include/linux/memstick.h
index d0c37e682234..690c35a9d4cc 100644
--- a/include/linux/memstick.h
+++ b/include/linux/memstick.h
@@ -100,8 +100,8 @@ struct mspro_param_register {
 #define MEMSTICK_SYS_PAR8   0x40
 #define MEMSTICK_SYS_SERIAL 0x80
 
-	unsigned short data_count;
-	unsigned int   data_address;
+	__be16 data_count;
+	__be32 data_address;
 	unsigned char  tpc_param;
 } __attribute__((packed));
 
-- 
cgit v1.2.3


From c4be0c1dc4cdc37b175579be1460f15ac6495e9a Mon Sep 17 00:00:00 2001
From: Takashi Sato <t-sato@yk.jp.nec.com>
Date: Fri, 9 Jan 2009 16:40:58 -0800
Subject: filesystem freeze: add error handling of write_super_lockfs/unlockfs

Currently, ext3 in mainline Linux doesn't have the freeze feature which
suspends write requests.  So, we cannot take a backup which keeps the
filesystem's consistency with the storage device's features (snapshot and
replication) while it is mounted.

In many case, a commercial filesystem (e.g.  VxFS) has the freeze feature
and it would be used to get the consistent backup.

If Linux's standard filesystem ext3 has the freeze feature, we can do it
without a commercial filesystem.

So I have implemented the ioctls of the freeze feature.
I think we can take the consistent backup with the following steps.
1. Freeze the filesystem with the freeze ioctl.
2. Separate the replication volume or create the snapshot
   with the storage device's feature.
3. Unfreeze the filesystem with the unfreeze ioctl.
4. Take the backup from the separated replication volume
   or the snapshot.

This patch:

VFS:
Changed the type of write_super_lockfs and unlockfs from "void"
to "int" so that they can return an error.
Rename write_super_lockfs and unlockfs of the super block operation
freeze_fs and unfreeze_fs to avoid a confusion.

ext3, ext4, xfs, gfs2, jfs:
Changed the type of write_super_lockfs and unlockfs from "void"
to "int" so that write_super_lockfs returns an error if needed,
and unlockfs always returns 0.

reiserfs:
Changed the type of write_super_lockfs and unlockfs from "void"
to "int" so that they always return 0 (success) to keep a current behavior.

Signed-off-by: Takashi Sato <t-sato@yk.jp.nec.com>
Signed-off-by: Masayuki Hamaguchi <m-hamaguchi@ys.jp.nec.com>
Cc: <xfs-masters@oss.sgi.com>
Cc: <linux-ext4@vger.kernel.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Alasdair G Kergon <agk@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/filesystems/Locking |  8 +++----
 Documentation/filesystems/vfs.txt |  8 +++----
 fs/buffer.c                       |  8 +++----
 fs/ext3/super.c                   | 45 +++++++++++++++++++++++++--------------
 fs/ext4/super.c                   | 45 +++++++++++++++++++++++++++------------
 fs/gfs2/ops_super.c               | 16 ++++++++------
 fs/jfs/super.c                    | 10 +++++----
 fs/reiserfs/super.c               | 10 +++++----
 fs/xfs/linux-2.6/xfs_super.c      |  8 +++----
 fs/xfs/xfs_fsops.c                | 11 ++++++----
 fs/xfs/xfs_fsops.h                |  2 +-
 include/linux/fs.h                |  4 ++--
 12 files changed, 107 insertions(+), 68 deletions(-)

(limited to 'include/linux')

diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking
index cfbfa15a46ba..ec6a9392a173 100644
--- a/Documentation/filesystems/Locking
+++ b/Documentation/filesystems/Locking
@@ -97,8 +97,8 @@ prototypes:
 	void (*put_super) (struct super_block *);
 	void (*write_super) (struct super_block *);
 	int (*sync_fs)(struct super_block *sb, int wait);
-	void (*write_super_lockfs) (struct super_block *);
-	void (*unlockfs) (struct super_block *);
+	int (*freeze_fs) (struct super_block *);
+	int (*unfreeze_fs) (struct super_block *);
 	int (*statfs) (struct dentry *, struct kstatfs *);
 	int (*remount_fs) (struct super_block *, int *, char *);
 	void (*clear_inode) (struct inode *);
@@ -119,8 +119,8 @@ delete_inode:		no
 put_super:		yes	yes	no
 write_super:		no	yes	read
 sync_fs:		no	no	read
-write_super_lockfs:	?
-unlockfs:		?
+freeze_fs:		?
+unfreeze_fs:		?
 statfs:			no	no	no
 remount_fs:		yes	yes	maybe		(see below)
 clear_inode:		no
diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt
index ef19afa186a9..deeeed0faa8f 100644
--- a/Documentation/filesystems/vfs.txt
+++ b/Documentation/filesystems/vfs.txt
@@ -210,8 +210,8 @@ struct super_operations {
         void (*put_super) (struct super_block *);
         void (*write_super) (struct super_block *);
         int (*sync_fs)(struct super_block *sb, int wait);
-        void (*write_super_lockfs) (struct super_block *);
-        void (*unlockfs) (struct super_block *);
+        int (*freeze_fs) (struct super_block *);
+        int (*unfreeze_fs) (struct super_block *);
         int (*statfs) (struct dentry *, struct kstatfs *);
         int (*remount_fs) (struct super_block *, int *, char *);
         void (*clear_inode) (struct inode *);
@@ -270,11 +270,11 @@ or bottom half).
   	a superblock. The second parameter indicates whether the method
 	should wait until the write out has been completed. Optional.
 
-  write_super_lockfs: called when VFS is locking a filesystem and
+  freeze_fs: called when VFS is locking a filesystem and
   	forcing it into a consistent state.  This method is currently
   	used by the Logical Volume Manager (LVM).
 
-  unlockfs: called when VFS is unlocking a filesystem and making it writable
+  unfreeze_fs: called when VFS is unlocking a filesystem and making it writable
   	again.
 
   statfs: called when the VFS needs to get filesystem statistics. This
diff --git a/fs/buffer.c b/fs/buffer.c
index c26da785938a..87f9e537b8c3 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -221,8 +221,8 @@ struct super_block *freeze_bdev(struct block_device *bdev)
 
 		sync_blockdev(sb->s_bdev);
 
-		if (sb->s_op->write_super_lockfs)
-			sb->s_op->write_super_lockfs(sb);
+		if (sb->s_op->freeze_fs)
+			sb->s_op->freeze_fs(sb);
 	}
 
 	sync_blockdev(bdev);
@@ -242,8 +242,8 @@ void thaw_bdev(struct block_device *bdev, struct super_block *sb)
 	if (sb) {
 		BUG_ON(sb->s_bdev != bdev);
 
-		if (sb->s_op->unlockfs)
-			sb->s_op->unlockfs(sb);
+		if (sb->s_op->unfreeze_fs)
+			sb->s_op->unfreeze_fs(sb);
 		sb->s_frozen = SB_UNFROZEN;
 		smp_wmb();
 		wake_up(&sb->s_wait_unfrozen);
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 5d047a030a73..b70d90e08a3c 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -48,8 +48,8 @@ static int ext3_load_journal(struct super_block *, struct ext3_super_block *,
 			     unsigned long journal_devnum);
 static int ext3_create_journal(struct super_block *, struct ext3_super_block *,
 			       unsigned int);
-static void ext3_commit_super (struct super_block * sb,
-			       struct ext3_super_block * es,
+static int ext3_commit_super(struct super_block *sb,
+			       struct ext3_super_block *es,
 			       int sync);
 static void ext3_mark_recovery_complete(struct super_block * sb,
 					struct ext3_super_block * es);
@@ -60,9 +60,9 @@ static const char *ext3_decode_error(struct super_block * sb, int errno,
 				     char nbuf[16]);
 static int ext3_remount (struct super_block * sb, int * flags, char * data);
 static int ext3_statfs (struct dentry * dentry, struct kstatfs * buf);
-static void ext3_unlockfs(struct super_block *sb);
+static int ext3_unfreeze(struct super_block *sb);
 static void ext3_write_super (struct super_block * sb);
-static void ext3_write_super_lockfs(struct super_block *sb);
+static int ext3_freeze(struct super_block *sb);
 
 /*
  * Wrappers for journal_start/end.
@@ -759,8 +759,8 @@ static const struct super_operations ext3_sops = {
 	.put_super	= ext3_put_super,
 	.write_super	= ext3_write_super,
 	.sync_fs	= ext3_sync_fs,
-	.write_super_lockfs = ext3_write_super_lockfs,
-	.unlockfs	= ext3_unlockfs,
+	.freeze_fs	= ext3_freeze,
+	.unfreeze_fs	= ext3_unfreeze,
 	.statfs		= ext3_statfs,
 	.remount_fs	= ext3_remount,
 	.clear_inode	= ext3_clear_inode,
@@ -2311,21 +2311,23 @@ static int ext3_create_journal(struct super_block * sb,
 	return 0;
 }
 
-static void ext3_commit_super (struct super_block * sb,
-			       struct ext3_super_block * es,
+static int ext3_commit_super(struct super_block *sb,
+			       struct ext3_super_block *es,
 			       int sync)
 {
 	struct buffer_head *sbh = EXT3_SB(sb)->s_sbh;
+	int error = 0;
 
 	if (!sbh)
-		return;
+		return error;
 	es->s_wtime = cpu_to_le32(get_seconds());
 	es->s_free_blocks_count = cpu_to_le32(ext3_count_free_blocks(sb));
 	es->s_free_inodes_count = cpu_to_le32(ext3_count_free_inodes(sb));
 	BUFFER_TRACE(sbh, "marking dirty");
 	mark_buffer_dirty(sbh);
 	if (sync)
-		sync_dirty_buffer(sbh);
+		error = sync_dirty_buffer(sbh);
+	return error;
 }
 
 
@@ -2439,12 +2441,14 @@ static int ext3_sync_fs(struct super_block *sb, int wait)
  * LVM calls this function before a (read-only) snapshot is created.  This
  * gives us a chance to flush the journal completely and mark the fs clean.
  */
-static void ext3_write_super_lockfs(struct super_block *sb)
+static int ext3_freeze(struct super_block *sb)
 {
+	int error = 0;
+	journal_t *journal;
 	sb->s_dirt = 0;
 
 	if (!(sb->s_flags & MS_RDONLY)) {
-		journal_t *journal = EXT3_SB(sb)->s_journal;
+		journal = EXT3_SB(sb)->s_journal;
 
 		/* Now we set up the journal barrier. */
 		journal_lock_updates(journal);
@@ -2453,20 +2457,28 @@ static void ext3_write_super_lockfs(struct super_block *sb)
 		 * We don't want to clear needs_recovery flag when we failed
 		 * to flush the journal.
 		 */
-		if (journal_flush(journal) < 0)
-			return;
+		error = journal_flush(journal);
+		if (error < 0)
+			goto out;
 
 		/* Journal blocked and flushed, clear needs_recovery flag. */
 		EXT3_CLEAR_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER);
-		ext3_commit_super(sb, EXT3_SB(sb)->s_es, 1);
+		error = ext3_commit_super(sb, EXT3_SB(sb)->s_es, 1);
+		if (error)
+			goto out;
 	}
+	return 0;
+
+out:
+	journal_unlock_updates(journal);
+	return error;
 }
 
 /*
  * Called by LVM after the snapshot is done.  We need to reset the RECOVER
  * flag here, even though the filesystem is not technically dirty yet.
  */
-static void ext3_unlockfs(struct super_block *sb)
+static int ext3_unfreeze(struct super_block *sb)
 {
 	if (!(sb->s_flags & MS_RDONLY)) {
 		lock_super(sb);
@@ -2476,6 +2488,7 @@ static void ext3_unlockfs(struct super_block *sb)
 		unlock_super(sb);
 		journal_unlock_updates(EXT3_SB(sb)->s_journal);
 	}
+	return 0;
 }
 
 static int ext3_remount (struct super_block * sb, int * flags, char * data)
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 8f7e0be8ab1b..e5f06a5f045e 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -51,7 +51,7 @@ struct proc_dir_entry *ext4_proc_root;
 
 static int ext4_load_journal(struct super_block *, struct ext4_super_block *,
 			     unsigned long journal_devnum);
-static void ext4_commit_super(struct super_block *sb,
+static int ext4_commit_super(struct super_block *sb,
 			      struct ext4_super_block *es, int sync);
 static void ext4_mark_recovery_complete(struct super_block *sb,
 					struct ext4_super_block *es);
@@ -62,9 +62,9 @@ static const char *ext4_decode_error(struct super_block *sb, int errno,
 				     char nbuf[16]);
 static int ext4_remount(struct super_block *sb, int *flags, char *data);
 static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf);
-static void ext4_unlockfs(struct super_block *sb);
+static int ext4_unfreeze(struct super_block *sb);
 static void ext4_write_super(struct super_block *sb);
-static void ext4_write_super_lockfs(struct super_block *sb);
+static int ext4_freeze(struct super_block *sb);
 
 
 ext4_fsblk_t ext4_block_bitmap(struct super_block *sb,
@@ -978,8 +978,8 @@ static const struct super_operations ext4_sops = {
 	.put_super	= ext4_put_super,
 	.write_super	= ext4_write_super,
 	.sync_fs	= ext4_sync_fs,
-	.write_super_lockfs = ext4_write_super_lockfs,
-	.unlockfs	= ext4_unlockfs,
+	.freeze_fs	= ext4_freeze,
+	.unfreeze_fs	= ext4_unfreeze,
 	.statfs		= ext4_statfs,
 	.remount_fs	= ext4_remount,
 	.clear_inode	= ext4_clear_inode,
@@ -2888,13 +2888,14 @@ static int ext4_load_journal(struct super_block *sb,
 	return 0;
 }
 
-static void ext4_commit_super(struct super_block *sb,
+static int ext4_commit_super(struct super_block *sb,
 			      struct ext4_super_block *es, int sync)
 {
 	struct buffer_head *sbh = EXT4_SB(sb)->s_sbh;
+	int error = 0;
 
 	if (!sbh)
-		return;
+		return error;
 	if (buffer_write_io_error(sbh)) {
 		/*
 		 * Oh, dear.  A previous attempt to write the
@@ -2918,14 +2919,19 @@ static void ext4_commit_super(struct super_block *sb,
 	BUFFER_TRACE(sbh, "marking dirty");
 	mark_buffer_dirty(sbh);
 	if (sync) {
-		sync_dirty_buffer(sbh);
-		if (buffer_write_io_error(sbh)) {
+		error = sync_dirty_buffer(sbh);
+		if (error)
+			return error;
+
+		error = buffer_write_io_error(sbh);
+		if (error) {
 			printk(KERN_ERR "EXT4-fs: I/O error while writing "
 			       "superblock for %s.\n", sb->s_id);
 			clear_buffer_write_io_error(sbh);
 			set_buffer_uptodate(sbh);
 		}
 	}
+	return error;
 }
 
 
@@ -3058,12 +3064,14 @@ static int ext4_sync_fs(struct super_block *sb, int wait)
  * LVM calls this function before a (read-only) snapshot is created.  This
  * gives us a chance to flush the journal completely and mark the fs clean.
  */
-static void ext4_write_super_lockfs(struct super_block *sb)
+static int ext4_freeze(struct super_block *sb)
 {
+	int error = 0;
+	journal_t *journal;
 	sb->s_dirt = 0;
 
 	if (!(sb->s_flags & MS_RDONLY)) {
-		journal_t *journal = EXT4_SB(sb)->s_journal;
+		journal = EXT4_SB(sb)->s_journal;
 
 		if (journal) {
 			/* Now we set up the journal barrier. */
@@ -3073,21 +3081,29 @@ static void ext4_write_super_lockfs(struct super_block *sb)
 			 * We don't want to clear needs_recovery flag when we
 			 * failed to flush the journal.
 			 */
-			if (jbd2_journal_flush(journal) < 0)
-				return;
+			error = jbd2_journal_flush(journal);
+			if (error < 0)
+				goto out;
 		}
 
 		/* Journal blocked and flushed, clear needs_recovery flag. */
 		EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
 		ext4_commit_super(sb, EXT4_SB(sb)->s_es, 1);
+		error = ext4_commit_super(sb, EXT4_SB(sb)->s_es, 1);
+		if (error)
+			goto out;
 	}
+	return 0;
+out:
+	jbd2_journal_unlock_updates(journal);
+	return error;
 }
 
 /*
  * Called by LVM after the snapshot is done.  We need to reset the RECOVER
  * flag here, even though the filesystem is not technically dirty yet.
  */
-static void ext4_unlockfs(struct super_block *sb)
+static int ext4_unfreeze(struct super_block *sb)
 {
 	if (EXT4_SB(sb)->s_journal && !(sb->s_flags & MS_RDONLY)) {
 		lock_super(sb);
@@ -3097,6 +3113,7 @@ static void ext4_unlockfs(struct super_block *sb)
 		unlock_super(sb);
 		jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
 	}
+	return 0;
 }
 
 static int ext4_remount(struct super_block *sb, int *flags, char *data)
diff --git a/fs/gfs2/ops_super.c b/fs/gfs2/ops_super.c
index 777783deddcb..320323d03479 100644
--- a/fs/gfs2/ops_super.c
+++ b/fs/gfs2/ops_super.c
@@ -211,18 +211,18 @@ static int gfs2_sync_fs(struct super_block *sb, int wait)
 }
 
 /**
- * gfs2_write_super_lockfs - prevent further writes to the filesystem
+ * gfs2_freeze - prevent further writes to the filesystem
  * @sb: the VFS structure for the filesystem
  *
  */
 
-static void gfs2_write_super_lockfs(struct super_block *sb)
+static int gfs2_freeze(struct super_block *sb)
 {
 	struct gfs2_sbd *sdp = sb->s_fs_info;
 	int error;
 
 	if (test_bit(SDF_SHUTDOWN, &sdp->sd_flags))
-		return;
+		return -EINVAL;
 
 	for (;;) {
 		error = gfs2_freeze_fs(sdp);
@@ -242,17 +242,19 @@ static void gfs2_write_super_lockfs(struct super_block *sb)
 		fs_err(sdp, "retrying...\n");
 		msleep(1000);
 	}
+	return 0;
 }
 
 /**
- * gfs2_unlockfs - reallow writes to the filesystem
+ * gfs2_unfreeze - reallow writes to the filesystem
  * @sb: the VFS structure for the filesystem
  *
  */
 
-static void gfs2_unlockfs(struct super_block *sb)
+static int gfs2_unfreeze(struct super_block *sb)
 {
 	gfs2_unfreeze_fs(sb->s_fs_info);
+	return 0;
 }
 
 /**
@@ -688,8 +690,8 @@ const struct super_operations gfs2_super_ops = {
 	.put_super		= gfs2_put_super,
 	.write_super		= gfs2_write_super,
 	.sync_fs		= gfs2_sync_fs,
-	.write_super_lockfs 	= gfs2_write_super_lockfs,
-	.unlockfs		= gfs2_unlockfs,
+	.freeze_fs 		= gfs2_freeze,
+	.unfreeze_fs		= gfs2_unfreeze,
 	.statfs			= gfs2_statfs,
 	.remount_fs		= gfs2_remount_fs,
 	.clear_inode		= gfs2_clear_inode,
diff --git a/fs/jfs/super.c b/fs/jfs/super.c
index 0dae345e481b..b37d1f78b854 100644
--- a/fs/jfs/super.c
+++ b/fs/jfs/super.c
@@ -543,7 +543,7 @@ out_kfree:
 	return ret;
 }
 
-static void jfs_write_super_lockfs(struct super_block *sb)
+static int jfs_freeze(struct super_block *sb)
 {
 	struct jfs_sb_info *sbi = JFS_SBI(sb);
 	struct jfs_log *log = sbi->log;
@@ -553,9 +553,10 @@ static void jfs_write_super_lockfs(struct super_block *sb)
 		lmLogShutdown(log);
 		updateSuper(sb, FM_CLEAN);
 	}
+	return 0;
 }
 
-static void jfs_unlockfs(struct super_block *sb)
+static int jfs_unfreeze(struct super_block *sb)
 {
 	struct jfs_sb_info *sbi = JFS_SBI(sb);
 	struct jfs_log *log = sbi->log;
@@ -568,6 +569,7 @@ static void jfs_unlockfs(struct super_block *sb)
 		else
 			txResume(sb);
 	}
+	return 0;
 }
 
 static int jfs_get_sb(struct file_system_type *fs_type,
@@ -735,8 +737,8 @@ static const struct super_operations jfs_super_operations = {
 	.delete_inode	= jfs_delete_inode,
 	.put_super	= jfs_put_super,
 	.sync_fs	= jfs_sync_fs,
-	.write_super_lockfs = jfs_write_super_lockfs,
-	.unlockfs       = jfs_unlockfs,
+	.freeze_fs	= jfs_freeze,
+	.unfreeze_fs	= jfs_unfreeze,
 	.statfs		= jfs_statfs,
 	.remount_fs	= jfs_remount,
 	.show_options	= jfs_show_options,
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index c55651f1407c..f3c820b75829 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -83,7 +83,7 @@ static void reiserfs_write_super(struct super_block *s)
 	reiserfs_sync_fs(s, 1);
 }
 
-static void reiserfs_write_super_lockfs(struct super_block *s)
+static int reiserfs_freeze(struct super_block *s)
 {
 	struct reiserfs_transaction_handle th;
 	reiserfs_write_lock(s);
@@ -101,11 +101,13 @@ static void reiserfs_write_super_lockfs(struct super_block *s)
 	}
 	s->s_dirt = 0;
 	reiserfs_write_unlock(s);
+	return 0;
 }
 
-static void reiserfs_unlockfs(struct super_block *s)
+static int reiserfs_unfreeze(struct super_block *s)
 {
 	reiserfs_allow_writes(s);
+	return 0;
 }
 
 extern const struct in_core_key MAX_IN_CORE_KEY;
@@ -613,8 +615,8 @@ static const struct super_operations reiserfs_sops = {
 	.put_super = reiserfs_put_super,
 	.write_super = reiserfs_write_super,
 	.sync_fs = reiserfs_sync_fs,
-	.write_super_lockfs = reiserfs_write_super_lockfs,
-	.unlockfs = reiserfs_unlockfs,
+	.freeze_fs = reiserfs_freeze,
+	.unfreeze_fs = reiserfs_unfreeze,
 	.statfs = reiserfs_statfs,
 	.remount_fs = reiserfs_remount,
 	.show_options = generic_show_options,
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index be846d606ae8..95a971080368 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -1269,14 +1269,14 @@ xfs_fs_remount(
  * need to take care of the metadata. Once that's done write a dummy
  * record to dirty the log in case of a crash while frozen.
  */
-STATIC void
-xfs_fs_lockfs(
+STATIC int
+xfs_fs_freeze(
 	struct super_block	*sb)
 {
 	struct xfs_mount	*mp = XFS_M(sb);
 
 	xfs_quiesce_attr(mp);
-	xfs_fs_log_dummy(mp);
+	return -xfs_fs_log_dummy(mp);
 }
 
 STATIC int
@@ -1557,7 +1557,7 @@ static struct super_operations xfs_super_operations = {
 	.put_super		= xfs_fs_put_super,
 	.write_super		= xfs_fs_write_super,
 	.sync_fs		= xfs_fs_sync_super,
-	.write_super_lockfs	= xfs_fs_lockfs,
+	.freeze_fs		= xfs_fs_freeze,
 	.statfs			= xfs_fs_statfs,
 	.remount_fs		= xfs_fs_remount,
 	.show_options		= xfs_fs_show_options,
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index 852b6d32e8d0..680d0e0ec932 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -595,17 +595,19 @@ out:
 	return 0;
 }
 
-void
+int
 xfs_fs_log_dummy(
 	xfs_mount_t	*mp)
 {
 	xfs_trans_t	*tp;
 	xfs_inode_t	*ip;
+	int		error;
 
 	tp = _xfs_trans_alloc(mp, XFS_TRANS_DUMMY1);
-	if (xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0)) {
+	error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0);
+	if (error) {
 		xfs_trans_cancel(tp, 0);
-		return;
+		return error;
 	}
 
 	ip = mp->m_rootip;
@@ -615,9 +617,10 @@ xfs_fs_log_dummy(
 	xfs_trans_ihold(tp, ip);
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 	xfs_trans_set_sync(tp);
-	xfs_trans_commit(tp, 0);
+	error = xfs_trans_commit(tp, 0);
 
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	return error;
 }
 
 int
diff --git a/fs/xfs/xfs_fsops.h b/fs/xfs/xfs_fsops.h
index 300d0c9d61ad..88435e0a77c9 100644
--- a/fs/xfs/xfs_fsops.h
+++ b/fs/xfs/xfs_fsops.h
@@ -25,6 +25,6 @@ extern int xfs_fs_counts(xfs_mount_t *mp, xfs_fsop_counts_t *cnt);
 extern int xfs_reserve_blocks(xfs_mount_t *mp, __uint64_t *inval,
 				xfs_fsop_resblks_t *outval);
 extern int xfs_fs_goingdown(xfs_mount_t *mp, __uint32_t inflags);
-extern void xfs_fs_log_dummy(xfs_mount_t *mp);
+extern int xfs_fs_log_dummy(xfs_mount_t *mp);
 
 #endif	/* __XFS_FSOPS_H__ */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 0b87b29f4797..3e59182de9df 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1377,8 +1377,8 @@ struct super_operations {
 	void (*put_super) (struct super_block *);
 	void (*write_super) (struct super_block *);
 	int (*sync_fs)(struct super_block *sb, int wait);
-	void (*write_super_lockfs) (struct super_block *);
-	void (*unlockfs) (struct super_block *);
+	int (*freeze_fs) (struct super_block *);
+	int (*unfreeze_fs) (struct super_block *);
 	int (*statfs) (struct dentry *, struct kstatfs *);
 	int (*remount_fs) (struct super_block *, int *, char *);
 	void (*clear_inode) (struct inode *);
-- 
cgit v1.2.3


From fcccf502540e3d752d33b2d8e976034dee81f9f7 Mon Sep 17 00:00:00 2001
From: Takashi Sato <t-sato@yk.jp.nec.com>
Date: Fri, 9 Jan 2009 16:40:59 -0800
Subject: filesystem freeze: implement generic freeze feature

The ioctls for the generic freeze feature are below.
o Freeze the filesystem
  int ioctl(int fd, int FIFREEZE, arg)
    fd: The file descriptor of the mountpoint
    FIFREEZE: request code for the freeze
    arg: Ignored
    Return value: 0 if the operation succeeds. Otherwise, -1

o Unfreeze the filesystem
  int ioctl(int fd, int FITHAW, arg)
    fd: The file descriptor of the mountpoint
    FITHAW: request code for unfreeze
    arg: Ignored
    Return value: 0 if the operation succeeds. Otherwise, -1
    Error number: If the filesystem has already been unfrozen,
                  errno is set to EINVAL.

[akpm@linux-foundation.org: fix CONFIG_BLOCK=n]
Signed-off-by: Takashi Sato <t-sato@yk.jp.nec.com>
Signed-off-by: Masayuki Hamaguchi <m-hamaguchi@ys.jp.nec.com>
Cc: <xfs-masters@oss.sgi.com>
Cc: <linux-ext4@vger.kernel.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Alasdair G Kergon <agk@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/block_dev.c              |  2 ++
 fs/buffer.c                 | 74 +++++++++++++++++++++++++++++++++++++++------
 fs/ioctl.c                  | 46 ++++++++++++++++++++++++++++
 include/linux/buffer_head.h | 11 ++++++-
 include/linux/fs.h          |  7 +++++
 5 files changed, 130 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/fs/block_dev.c b/fs/block_dev.c
index ac7031f12ea5..b3c1efff5e1d 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -285,6 +285,8 @@ static void init_once(void *foo)
 	INIT_LIST_HEAD(&bdev->bd_holder_list);
 #endif
 	inode_init_once(&ei->vfs_inode);
+	/* Initialize mutex for freeze. */
+	mutex_init(&bdev->bd_fsfreeze_mutex);
 }
 
 static inline void __bd_forget(struct inode *inode)
diff --git a/fs/buffer.c b/fs/buffer.c
index 87f9e537b8c3..b6e8b8632e2f 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -203,10 +203,25 @@ int fsync_bdev(struct block_device *bdev)
  * happen on bdev until thaw_bdev() is called.
  * If a superblock is found on this device, we take the s_umount semaphore
  * on it to make sure nobody unmounts until the snapshot creation is done.
+ * The reference counter (bd_fsfreeze_count) guarantees that only the last
+ * unfreeze process can unfreeze the frozen filesystem actually when multiple
+ * freeze requests arrive simultaneously. It counts up in freeze_bdev() and
+ * count down in thaw_bdev(). When it becomes 0, thaw_bdev() will unfreeze
+ * actually.
  */
 struct super_block *freeze_bdev(struct block_device *bdev)
 {
 	struct super_block *sb;
+	int error = 0;
+
+	mutex_lock(&bdev->bd_fsfreeze_mutex);
+	if (bdev->bd_fsfreeze_count > 0) {
+		bdev->bd_fsfreeze_count++;
+		sb = get_super(bdev);
+		mutex_unlock(&bdev->bd_fsfreeze_mutex);
+		return sb;
+	}
+	bdev->bd_fsfreeze_count++;
 
 	down(&bdev->bd_mount_sem);
 	sb = get_super(bdev);
@@ -221,11 +236,24 @@ struct super_block *freeze_bdev(struct block_device *bdev)
 
 		sync_blockdev(sb->s_bdev);
 
-		if (sb->s_op->freeze_fs)
-			sb->s_op->freeze_fs(sb);
+		if (sb->s_op->freeze_fs) {
+			error = sb->s_op->freeze_fs(sb);
+			if (error) {
+				printk(KERN_ERR
+					"VFS:Filesystem freeze failed\n");
+				sb->s_frozen = SB_UNFROZEN;
+				drop_super(sb);
+				up(&bdev->bd_mount_sem);
+				bdev->bd_fsfreeze_count--;
+				mutex_unlock(&bdev->bd_fsfreeze_mutex);
+				return ERR_PTR(error);
+			}
+		}
 	}
 
 	sync_blockdev(bdev);
+	mutex_unlock(&bdev->bd_fsfreeze_mutex);
+
 	return sb;	/* thaw_bdev releases s->s_umount and bd_mount_sem */
 }
 EXPORT_SYMBOL(freeze_bdev);
@@ -237,20 +265,48 @@ EXPORT_SYMBOL(freeze_bdev);
  *
  * Unlocks the filesystem and marks it writeable again after freeze_bdev().
  */
-void thaw_bdev(struct block_device *bdev, struct super_block *sb)
+int thaw_bdev(struct block_device *bdev, struct super_block *sb)
 {
+	int error = 0;
+
+	mutex_lock(&bdev->bd_fsfreeze_mutex);
+	if (!bdev->bd_fsfreeze_count) {
+		mutex_unlock(&bdev->bd_fsfreeze_mutex);
+		return -EINVAL;
+	}
+
+	bdev->bd_fsfreeze_count--;
+	if (bdev->bd_fsfreeze_count > 0) {
+		if (sb)
+			drop_super(sb);
+		mutex_unlock(&bdev->bd_fsfreeze_mutex);
+		return 0;
+	}
+
 	if (sb) {
 		BUG_ON(sb->s_bdev != bdev);
-
-		if (sb->s_op->unfreeze_fs)
-			sb->s_op->unfreeze_fs(sb);
-		sb->s_frozen = SB_UNFROZEN;
-		smp_wmb();
-		wake_up(&sb->s_wait_unfrozen);
+		if (!(sb->s_flags & MS_RDONLY)) {
+			if (sb->s_op->unfreeze_fs) {
+				error = sb->s_op->unfreeze_fs(sb);
+				if (error) {
+					printk(KERN_ERR
+						"VFS:Filesystem thaw failed\n");
+					sb->s_frozen = SB_FREEZE_TRANS;
+					bdev->bd_fsfreeze_count++;
+					mutex_unlock(&bdev->bd_fsfreeze_mutex);
+					return error;
+				}
+			}
+			sb->s_frozen = SB_UNFROZEN;
+			smp_wmb();
+			wake_up(&sb->s_wait_unfrozen);
+		}
 		drop_super(sb);
 	}
 
 	up(&bdev->bd_mount_sem);
+	mutex_unlock(&bdev->bd_fsfreeze_mutex);
+	return 0;
 }
 EXPORT_SYMBOL(thaw_bdev);
 
diff --git a/fs/ioctl.c b/fs/ioctl.c
index cc3f1aa1cf7b..20b0a8a24c6b 100644
--- a/fs/ioctl.c
+++ b/fs/ioctl.c
@@ -439,6 +439,43 @@ static int ioctl_fioasync(unsigned int fd, struct file *filp,
 	return error;
 }
 
+static int ioctl_fsfreeze(struct file *filp)
+{
+	struct super_block *sb = filp->f_path.dentry->d_inode->i_sb;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	/* If filesystem doesn't support freeze feature, return. */
+	if (sb->s_op->freeze_fs == NULL)
+		return -EOPNOTSUPP;
+
+	/* If a blockdevice-backed filesystem isn't specified, return. */
+	if (sb->s_bdev == NULL)
+		return -EINVAL;
+
+	/* Freeze */
+	sb = freeze_bdev(sb->s_bdev);
+	if (IS_ERR(sb))
+		return PTR_ERR(sb);
+	return 0;
+}
+
+static int ioctl_fsthaw(struct file *filp)
+{
+	struct super_block *sb = filp->f_path.dentry->d_inode->i_sb;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	/* If a blockdevice-backed filesystem isn't specified, return EINVAL. */
+	if (sb->s_bdev == NULL)
+		return -EINVAL;
+
+	/* Thaw */
+	return thaw_bdev(sb->s_bdev, sb);
+}
+
 /*
  * When you add any new common ioctls to the switches above and below
  * please update compat_sys_ioctl() too.
@@ -486,6 +523,15 @@ int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd,
 		} else
 			error = -ENOTTY;
 		break;
+
+	case FIFREEZE:
+		error = ioctl_fsfreeze(filp);
+		break;
+
+	case FITHAW:
+		error = ioctl_fsthaw(filp);
+		break;
+
 	default:
 		if (S_ISREG(filp->f_path.dentry->d_inode->i_mode))
 			error = file_ioctl(filp, cmd, arg);
diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h
index 8605f8a74df9..bd7ac793be19 100644
--- a/include/linux/buffer_head.h
+++ b/include/linux/buffer_head.h
@@ -171,7 +171,7 @@ void __wait_on_buffer(struct buffer_head *);
 wait_queue_head_t *bh_waitq_head(struct buffer_head *bh);
 int fsync_bdev(struct block_device *);
 struct super_block *freeze_bdev(struct block_device *);
-void thaw_bdev(struct block_device *, struct super_block *);
+int thaw_bdev(struct block_device *, struct super_block *);
 int fsync_super(struct super_block *);
 int fsync_no_super(struct block_device *);
 struct buffer_head *__find_get_block(struct block_device *bdev, sector_t block,
@@ -346,6 +346,15 @@ static inline int remove_inode_buffers(struct inode *inode) { return 1; }
 static inline int sync_mapping_buffers(struct address_space *mapping) { return 0; }
 static inline void invalidate_bdev(struct block_device *bdev) {}
 
+static inline struct super_block *freeze_bdev(struct block_device *sb)
+{
+	return NULL;
+}
+
+static inline int thaw_bdev(struct block_device *bdev, struct super_block *sb)
+{
+	return 0;
+}
 
 #endif /* CONFIG_BLOCK */
 #endif /* _LINUX_BUFFER_HEAD_H */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 3e59182de9df..6022f44043f2 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -234,6 +234,8 @@ struct inodes_stat_t {
 #define BMAP_IOCTL 1		/* obsolete - kept for compatibility */
 #define FIBMAP	   _IO(0x00,1)	/* bmap access */
 #define FIGETBSZ   _IO(0x00,2)	/* get the block size used for bmap */
+#define FIFREEZE	_IOWR('X', 119, int)	/* Freeze */
+#define FITHAW		_IOWR('X', 120, int)	/* Thaw */
 
 #define	FS_IOC_GETFLAGS			_IOR('f', 1, long)
 #define	FS_IOC_SETFLAGS			_IOW('f', 2, long)
@@ -591,6 +593,11 @@ struct block_device {
 	 * care to not mess up bd_private for that case.
 	 */
 	unsigned long		bd_private;
+
+	/* The counter of freeze processes */
+	int			bd_fsfreeze_count;
+	/* Mutex for freeze */
+	struct mutex		bd_fsfreeze_mutex;
 };
 
 /*
-- 
cgit v1.2.3


From f4b477c47332367d35686bd2b808c2156b96d7c7 Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Sat, 10 Jan 2009 11:12:09 +0000
Subject: rbtree: add const qualifier to some functions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The 'rb_first()', 'rb_last()', 'rb_next()' and 'rb_prev()' calls
take a pointer to an RB node or RB root. They do not change the
pointed objects, so add a 'const' qualifier in order to make life
of the users of these functions easier.

Indeed, if I have my own constant pointer &const struct my_type *p,
and I call 'rb_next(&p->rb)', I get a GCC warning:

warning: passing argument 1 of ‘rb_next’ discards qualifiers from pointer target type

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/rbtree.h |  8 ++++----
 lib/rbtree.c           | 12 ++++++------
 2 files changed, 10 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/rbtree.h b/include/linux/rbtree.h
index 344bc3495ddb..9c295411d01f 100644
--- a/include/linux/rbtree.h
+++ b/include/linux/rbtree.h
@@ -140,10 +140,10 @@ extern void rb_insert_color(struct rb_node *, struct rb_root *);
 extern void rb_erase(struct rb_node *, struct rb_root *);
 
 /* Find logical next and previous nodes in a tree */
-extern struct rb_node *rb_next(struct rb_node *);
-extern struct rb_node *rb_prev(struct rb_node *);
-extern struct rb_node *rb_first(struct rb_root *);
-extern struct rb_node *rb_last(struct rb_root *);
+extern struct rb_node *rb_next(const struct rb_node *);
+extern struct rb_node *rb_prev(const struct rb_node *);
+extern struct rb_node *rb_first(const struct rb_root *);
+extern struct rb_node *rb_last(const struct rb_root *);
 
 /* Fast replacement of a single node without remove/rebalance/add/rebalance */
 extern void rb_replace_node(struct rb_node *victim, struct rb_node *new, 
diff --git a/lib/rbtree.c b/lib/rbtree.c
index 48499c2d88cc..9956b99649f0 100644
--- a/lib/rbtree.c
+++ b/lib/rbtree.c
@@ -292,7 +292,7 @@ EXPORT_SYMBOL(rb_erase);
 /*
  * This function returns the first node (in sort order) of the tree.
  */
-struct rb_node *rb_first(struct rb_root *root)
+struct rb_node *rb_first(const struct rb_root *root)
 {
 	struct rb_node	*n;
 
@@ -305,7 +305,7 @@ struct rb_node *rb_first(struct rb_root *root)
 }
 EXPORT_SYMBOL(rb_first);
 
-struct rb_node *rb_last(struct rb_root *root)
+struct rb_node *rb_last(const struct rb_root *root)
 {
 	struct rb_node	*n;
 
@@ -318,7 +318,7 @@ struct rb_node *rb_last(struct rb_root *root)
 }
 EXPORT_SYMBOL(rb_last);
 
-struct rb_node *rb_next(struct rb_node *node)
+struct rb_node *rb_next(const struct rb_node *node)
 {
 	struct rb_node *parent;
 
@@ -331,7 +331,7 @@ struct rb_node *rb_next(struct rb_node *node)
 		node = node->rb_right; 
 		while (node->rb_left)
 			node=node->rb_left;
-		return node;
+		return (struct rb_node *)node;
 	}
 
 	/* No right-hand children.  Everything down and left is
@@ -347,7 +347,7 @@ struct rb_node *rb_next(struct rb_node *node)
 }
 EXPORT_SYMBOL(rb_next);
 
-struct rb_node *rb_prev(struct rb_node *node)
+struct rb_node *rb_prev(const struct rb_node *node)
 {
 	struct rb_node *parent;
 
@@ -360,7 +360,7 @@ struct rb_node *rb_prev(struct rb_node *node)
 		node = node->rb_left; 
 		while (node->rb_right)
 			node=node->rb_right;
-		return node;
+		return (struct rb_node *)node;
 	}
 
 	/* No left-hand children. Go up till we find an ancestor which
-- 
cgit v1.2.3


From 886ad09fc83342aa1c5a02a0b6d3298b78a8067f Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 9 Jan 2009 15:54:07 -0800
Subject: libata: Add a per-host flag to opt-in into parallel port probes

This patch adds a per host flag that allows drivers to opt in into
having its busses scanned in parallel.

Drivers that do not set this flag get their ports scanned in
the "original" sequence.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/ata/ahci.c        | 3 +++
 drivers/ata/libata-core.c | 9 +++++++++
 include/linux/libata.h    | 1 +
 3 files changed, 13 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 7f701cbe14ab..96039671e3b9 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -2660,6 +2660,9 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	host->iomap = pcim_iomap_table(pdev);
 	host->private_data = hpriv;
 
+	if (!(hpriv->cap & HOST_CAP_SSS))
+		host->flags |= ATA_HOST_PARALLEL_SCAN;
+
 	if (pi.flags & ATA_FLAG_EM)
 		ahci_reset_em(host);
 
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index c507a9ac78f4..f810078fafcc 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -5920,6 +5920,15 @@ static void async_port_probe(void *data, async_cookie_t cookie)
 {
 	int rc;
 	struct ata_port *ap = data;
+
+	/*
+	 * If we're not allowed to scan this host in parallel,
+	 * we need to wait until all previous scans have completed
+	 * before going further.
+	 */
+	if (!(ap->host->flags & ATA_HOST_PARALLEL_SCAN))
+		async_synchronize_cookie(cookie);
+
 	/* probe */
 	if (ap->ops->error_handler) {
 		struct ata_eh_info *ehi = &ap->link.eh_info;
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 4f7c8fb4d3fe..b6b8a7f3ec66 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -239,6 +239,7 @@ enum {
 	/* host set flags */
 	ATA_HOST_SIMPLEX	= (1 << 0),	/* Host is simplex, one DMA channel per host only */
 	ATA_HOST_STARTED	= (1 << 1),	/* Host started */
+	ATA_HOST_PARALLEL_SCAN	= (1 << 2),	/* Ports on this host can be scanned in parallel */
 
 	/* bits 24:31 of host->flags are reserved for LLD specific flags */
 
-- 
cgit v1.2.3


From f52046b14b1e1a8a02ae48d0c69d39c5e204644f Mon Sep 17 00:00:00 2001
From: Balaji Rao <balajirrao@openmoko.org>
Date: Fri, 9 Jan 2009 01:49:01 +0100
Subject: mfd: PCF50633 core driver

This patch implements the core of the PCF50633 driver. This core driver has
generic register read/write functions and does interrupt management for its
sub devices.

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
Cc: Andy Green <andy@openmoko.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/Kconfig               |   9 +
 drivers/mfd/Makefile              |   2 +
 drivers/mfd/pcf50633-core.c       | 710 ++++++++++++++++++++++++++++++++++++++
 include/linux/mfd/pcf50633/core.h | 218 ++++++++++++
 4 files changed, 939 insertions(+)
 create mode 100644 drivers/mfd/pcf50633-core.c
 create mode 100644 include/linux/mfd/pcf50633/core.h

(limited to 'include/linux')

diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index 416f9e7286ba..3ac32e45b03b 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -217,6 +217,15 @@ config MFD_WM8350_I2C
 	  I2C as the control interface.  Additional options must be
 	  selected to enable support for the functionality of the chip.
 
+config MFD_PCF50633
+	tristate "Support for NXP PCF50633"
+	depends on I2C
+	help
+	  Say yes here if you have NXP PCF50633 chip on your board.
+	  This core driver provides register access and IRQ handling
+	  facilities, and registers devices for the various functions
+	  so that function-specific drivers can bind to them.
+
 endmenu
 
 menu "Multimedia Capabilities Port drivers"
diff --git a/drivers/mfd/Makefile b/drivers/mfd/Makefile
index 0c9418b36c26..23d4d10981b3 100644
--- a/drivers/mfd/Makefile
+++ b/drivers/mfd/Makefile
@@ -37,3 +37,5 @@ endif
 obj-$(CONFIG_UCB1400_CORE)	+= ucb1400_core.o
 
 obj-$(CONFIG_PMIC_DA903X)	+= da903x.o
+
+obj-$(CONFIG_MFD_PCF50633)	+= pcf50633-core.o
\ No newline at end of file
diff --git a/drivers/mfd/pcf50633-core.c b/drivers/mfd/pcf50633-core.c
new file mode 100644
index 000000000000..24508e28e3fb
--- /dev/null
+++ b/drivers/mfd/pcf50633-core.c
@@ -0,0 +1,710 @@
+/* NXP PCF50633 Power Management Unit (PMU) driver
+ *
+ * (C) 2006-2008 by Openmoko, Inc.
+ * Author: Harald Welte <laforge@openmoko.org>
+ * 	   Balaji Rao <balajirrao@openmoko.org>
+ * All rights reserved.
+ *
+ *  This program is free software; you can redistribute  it and/or modify it
+ *  under  the terms of  the GNU General  Public License as published by the
+ *  Free Software Foundation;  either version 2 of the  License, or (at your
+ *  option) any later version.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/device.h>
+#include <linux/sysfs.h>
+#include <linux/device.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/interrupt.h>
+#include <linux/workqueue.h>
+#include <linux/platform_device.h>
+#include <linux/i2c.h>
+#include <linux/irq.h>
+
+#include <linux/mfd/pcf50633/core.h>
+
+/* Two MBCS registers used during cold start */
+#define PCF50633_REG_MBCS1		0x4b
+#define PCF50633_REG_MBCS2		0x4c
+#define PCF50633_MBCS1_USBPRES 		0x01
+#define PCF50633_MBCS1_ADAPTPRES	0x01
+
+static int __pcf50633_read(struct pcf50633 *pcf, u8 reg, int num, u8 *data)
+{
+	int ret;
+
+	ret = i2c_smbus_read_i2c_block_data(pcf->i2c_client, reg,
+				num, data);
+	if (ret < 0)
+		dev_err(pcf->dev, "Error reading %d regs at %d\n", num, reg);
+
+	return ret;
+}
+
+static int __pcf50633_write(struct pcf50633 *pcf, u8 reg, int num, u8 *data)
+{
+	int ret;
+
+	ret = i2c_smbus_write_i2c_block_data(pcf->i2c_client, reg,
+				num, data);
+	if (ret < 0)
+		dev_err(pcf->dev, "Error writing %d regs at %d\n", num, reg);
+
+	return ret;
+
+}
+
+/* Read a block of upto 32 regs  */
+int pcf50633_read_block(struct pcf50633 *pcf, u8 reg,
+					int nr_regs, u8 *data)
+{
+	int ret;
+
+	mutex_lock(&pcf->lock);
+	ret = __pcf50633_read(pcf, reg, nr_regs, data);
+	mutex_unlock(&pcf->lock);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(pcf50633_read_block);
+
+/* Write a block of upto 32 regs  */
+int pcf50633_write_block(struct pcf50633 *pcf , u8 reg,
+					int nr_regs, u8 *data)
+{
+	int ret;
+
+	mutex_lock(&pcf->lock);
+	ret = __pcf50633_write(pcf, reg, nr_regs, data);
+	mutex_unlock(&pcf->lock);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(pcf50633_write_block);
+
+u8 pcf50633_reg_read(struct pcf50633 *pcf, u8 reg)
+{
+	u8 val;
+
+	mutex_lock(&pcf->lock);
+	__pcf50633_read(pcf, reg, 1, &val);
+	mutex_unlock(&pcf->lock);
+
+	return val;
+}
+EXPORT_SYMBOL_GPL(pcf50633_reg_read);
+
+int pcf50633_reg_write(struct pcf50633 *pcf, u8 reg, u8 val)
+{
+	int ret;
+
+	mutex_lock(&pcf->lock);
+	ret = __pcf50633_write(pcf, reg, 1, &val);
+	mutex_unlock(&pcf->lock);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(pcf50633_reg_write);
+
+int pcf50633_reg_set_bit_mask(struct pcf50633 *pcf, u8 reg, u8 mask, u8 val)
+{
+	int ret;
+	u8 tmp;
+
+	val &= mask;
+
+	mutex_lock(&pcf->lock);
+	ret = __pcf50633_read(pcf, reg, 1, &tmp);
+	if (ret < 0)
+		goto out;
+
+	tmp &= ~mask;
+	tmp |= val;
+	ret = __pcf50633_write(pcf, reg, 1, &tmp);
+
+out:
+	mutex_unlock(&pcf->lock);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(pcf50633_reg_set_bit_mask);
+
+int pcf50633_reg_clear_bits(struct pcf50633 *pcf, u8 reg, u8 val)
+{
+	int ret;
+	u8 tmp;
+
+	mutex_lock(&pcf->lock);
+	ret = __pcf50633_read(pcf, reg, 1, &tmp);
+	if (ret < 0)
+		goto out;
+
+	tmp &= ~val;
+	ret = __pcf50633_write(pcf, reg, 1, &tmp);
+
+out:
+	mutex_unlock(&pcf->lock);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(pcf50633_reg_clear_bits);
+
+/* sysfs attributes */
+static ssize_t show_dump_regs(struct device *dev, struct device_attribute *attr,
+			    char *buf)
+{
+	struct pcf50633 *pcf = dev_get_drvdata(dev);
+	u8 dump[16];
+	int n, n1, idx = 0;
+	char *buf1 = buf;
+	static u8 address_no_read[] = { /* must be ascending */
+		PCF50633_REG_INT1,
+		PCF50633_REG_INT2,
+		PCF50633_REG_INT3,
+		PCF50633_REG_INT4,
+		PCF50633_REG_INT5,
+		0 /* terminator */
+	};
+
+	for (n = 0; n < 256; n += sizeof(dump)) {
+		for (n1 = 0; n1 < sizeof(dump); n1++)
+			if (n == address_no_read[idx]) {
+				idx++;
+				dump[n1] = 0x00;
+			} else
+				dump[n1] = pcf50633_reg_read(pcf, n + n1);
+
+		hex_dump_to_buffer(dump, sizeof(dump), 16, 1, buf1, 128, 0);
+		buf1 += strlen(buf1);
+		*buf1++ = '\n';
+		*buf1 = '\0';
+	}
+
+	return buf1 - buf;
+}
+static DEVICE_ATTR(dump_regs, 0400, show_dump_regs, NULL);
+
+static ssize_t show_resume_reason(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct pcf50633 *pcf = dev_get_drvdata(dev);
+	int n;
+
+	n = sprintf(buf, "%02x%02x%02x%02x%02x\n",
+				pcf->resume_reason[0],
+				pcf->resume_reason[1],
+				pcf->resume_reason[2],
+				pcf->resume_reason[3],
+				pcf->resume_reason[4]);
+
+	return n;
+}
+static DEVICE_ATTR(resume_reason, 0400, show_resume_reason, NULL);
+
+static struct attribute *pcf_sysfs_entries[] = {
+	&dev_attr_dump_regs.attr,
+	&dev_attr_resume_reason.attr,
+	NULL,
+};
+
+static struct attribute_group pcf_attr_group = {
+	.name	= NULL,			/* put in device directory */
+	.attrs	= pcf_sysfs_entries,
+};
+
+int pcf50633_register_irq(struct pcf50633 *pcf, int irq,
+			void (*handler) (int, void *), void *data)
+{
+	if (irq < 0 || irq > PCF50633_NUM_IRQ || !handler)
+		return -EINVAL;
+
+	if (WARN_ON(pcf->irq_handler[irq].handler))
+		return -EBUSY;
+
+	mutex_lock(&pcf->lock);
+	pcf->irq_handler[irq].handler = handler;
+	pcf->irq_handler[irq].data = data;
+	mutex_unlock(&pcf->lock);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(pcf50633_register_irq);
+
+int pcf50633_free_irq(struct pcf50633 *pcf, int irq)
+{
+	if (irq < 0 || irq > PCF50633_NUM_IRQ)
+		return -EINVAL;
+
+	mutex_lock(&pcf->lock);
+	pcf->irq_handler[irq].handler = NULL;
+	mutex_unlock(&pcf->lock);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(pcf50633_free_irq);
+
+static int __pcf50633_irq_mask_set(struct pcf50633 *pcf, int irq, u8 mask)
+{
+	u8 reg, bits, tmp;
+	int ret = 0, idx;
+
+	idx = irq >> 3;
+	reg =  PCF50633_REG_INT1M + idx;
+	bits = 1 << (irq & 0x07);
+
+	mutex_lock(&pcf->lock);
+
+	if (mask) {
+		ret = __pcf50633_read(pcf, reg, 1, &tmp);
+		if (ret < 0)
+			goto out;
+
+		tmp |= bits;
+
+		ret = __pcf50633_write(pcf, reg, 1, &tmp);
+		if (ret < 0)
+			goto out;
+
+		pcf->mask_regs[idx] &= ~bits;
+		pcf->mask_regs[idx] |= bits;
+	} else {
+		ret = __pcf50633_read(pcf, reg, 1, &tmp);
+		if (ret < 0)
+			goto out;
+
+		tmp &= ~bits;
+
+		ret = __pcf50633_write(pcf, reg, 1, &tmp);
+		if (ret < 0)
+			goto out;
+
+		pcf->mask_regs[idx] &= ~bits;
+	}
+out:
+	mutex_unlock(&pcf->lock);
+
+	return ret;
+}
+
+int pcf50633_irq_mask(struct pcf50633 *pcf, int irq)
+{
+	dev_info(pcf->dev, "Masking IRQ %d\n", irq);
+
+	return __pcf50633_irq_mask_set(pcf, irq, 1);
+}
+EXPORT_SYMBOL_GPL(pcf50633_irq_mask);
+
+int pcf50633_irq_unmask(struct pcf50633 *pcf, int irq)
+{
+	dev_info(pcf->dev, "Unmasking IRQ %d\n", irq);
+
+	return __pcf50633_irq_mask_set(pcf, irq, 0);
+}
+EXPORT_SYMBOL_GPL(pcf50633_irq_unmask);
+
+int pcf50633_irq_mask_get(struct pcf50633 *pcf, int irq)
+{
+	u8 reg, bits;
+
+	reg =  irq >> 3;
+	bits = 1 << (irq & 0x07);
+
+	return pcf->mask_regs[reg] & bits;
+}
+EXPORT_SYMBOL_GPL(pcf50633_irq_mask_get);
+
+static void pcf50633_irq_call_handler(struct pcf50633 *pcf, int irq)
+{
+	if (pcf->irq_handler[irq].handler)
+		pcf->irq_handler[irq].handler(irq, pcf->irq_handler[irq].data);
+}
+
+/* Maximum amount of time ONKEY is held before emergency action is taken */
+#define PCF50633_ONKEY1S_TIMEOUT 8
+
+static void pcf50633_irq_worker(struct work_struct *work)
+{
+	struct pcf50633 *pcf;
+	int ret, i, j;
+	u8 pcf_int[5], chgstat;
+
+	pcf = container_of(work, struct pcf50633, irq_work);
+
+	/* Read the 5 INT regs in one transaction */
+	ret = pcf50633_read_block(pcf, PCF50633_REG_INT1,
+						ARRAY_SIZE(pcf_int), pcf_int);
+	if (ret != ARRAY_SIZE(pcf_int)) {
+		dev_err(pcf->dev, "Error reading INT registers\n");
+
+		/*
+		 * If this doesn't ACK the interrupt to the chip, we'll be
+		 * called once again as we're level triggered.
+		 */
+		goto out;
+	}
+
+	/* We immediately read the usb and adapter status. We thus make sure
+	 * only of USBINS/USBREM IRQ handlers are called */
+	if (pcf_int[0] & (PCF50633_INT1_USBINS | PCF50633_INT1_USBREM)) {
+		chgstat = pcf50633_reg_read(pcf, PCF50633_REG_MBCS2);
+		if (chgstat & (0x3 << 4))
+			pcf_int[0] &= ~(1 << PCF50633_INT1_USBREM);
+		else
+			pcf_int[0] &= ~(1 << PCF50633_INT1_USBINS);
+	}
+
+	/* Make sure only one of ADPINS or ADPREM is set */
+	if (pcf_int[0] & (PCF50633_INT1_ADPINS | PCF50633_INT1_ADPREM)) {
+		chgstat = pcf50633_reg_read(pcf, PCF50633_REG_MBCS2);
+		if (chgstat & (0x3 << 4))
+			pcf_int[0] &= ~(1 << PCF50633_INT1_ADPREM);
+		else
+			pcf_int[0] &= ~(1 << PCF50633_INT1_ADPINS);
+	}
+
+	dev_dbg(pcf->dev, "INT1=0x%02x INT2=0x%02x INT3=0x%02x "
+			"INT4=0x%02x INT5=0x%02x\n", pcf_int[0],
+			pcf_int[1], pcf_int[2], pcf_int[3], pcf_int[4]);
+
+	/* Some revisions of the chip don't have a 8s standby mode on
+	 * ONKEY1S press. We try to manually do it in such cases. */
+	if ((pcf_int[0] & PCF50633_INT1_SECOND) && pcf->onkey1s_held) {
+		dev_info(pcf->dev, "ONKEY1S held for %d secs\n",
+							pcf->onkey1s_held);
+		if (pcf->onkey1s_held++ == PCF50633_ONKEY1S_TIMEOUT)
+			if (pcf->pdata->force_shutdown)
+				pcf->pdata->force_shutdown(pcf);
+	}
+
+	if (pcf_int[2] & PCF50633_INT3_ONKEY1S) {
+		dev_info(pcf->dev, "ONKEY1S held\n");
+		pcf->onkey1s_held = 1 ;
+
+		/* Unmask IRQ_SECOND */
+		pcf50633_reg_clear_bits(pcf, PCF50633_REG_INT1M,
+						PCF50633_INT1_SECOND);
+
+		/* Unmask IRQ_ONKEYR */
+		pcf50633_reg_clear_bits(pcf, PCF50633_REG_INT2M,
+						PCF50633_INT2_ONKEYR);
+	}
+
+	if ((pcf_int[1] & PCF50633_INT2_ONKEYR) && pcf->onkey1s_held) {
+		pcf->onkey1s_held = 0;
+
+		/* Mask SECOND and ONKEYR interrupts */
+		if (pcf->mask_regs[0] & PCF50633_INT1_SECOND)
+			pcf50633_reg_set_bit_mask(pcf,
+					PCF50633_REG_INT1M,
+					PCF50633_INT1_SECOND,
+					PCF50633_INT1_SECOND);
+
+		if (pcf->mask_regs[1] & PCF50633_INT2_ONKEYR)
+			pcf50633_reg_set_bit_mask(pcf,
+					PCF50633_REG_INT2M,
+					PCF50633_INT2_ONKEYR,
+					PCF50633_INT2_ONKEYR);
+	}
+
+	/* Have we just resumed ? */
+	if (pcf->is_suspended) {
+		pcf->is_suspended = 0;
+
+		/* Set the resume reason filtering out non resumers */
+		for (i = 0; i < ARRAY_SIZE(pcf_int); i++)
+			pcf->resume_reason[i] = pcf_int[i] &
+						pcf->pdata->resumers[i];
+
+		/* Make sure we don't pass on any ONKEY events to
+		 * userspace now */
+		pcf_int[1] &= ~(PCF50633_INT2_ONKEYR | PCF50633_INT2_ONKEYF);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(pcf_int); i++) {
+		/* Unset masked interrupts */
+		pcf_int[i] &= ~pcf->mask_regs[i];
+
+		for (j = 0; j < 8 ; j++)
+			if (pcf_int[i] & (1 << j))
+				pcf50633_irq_call_handler(pcf, (i * 8) + j);
+	}
+
+out:
+	put_device(pcf->dev);
+	enable_irq(pcf->irq);
+}
+
+static irqreturn_t pcf50633_irq(int irq, void *data)
+{
+	struct pcf50633 *pcf = data;
+
+	dev_dbg(pcf->dev, "pcf50633_irq\n");
+
+	get_device(pcf->dev);
+	disable_irq(pcf->irq);
+	schedule_work(&pcf->irq_work);
+
+	return IRQ_HANDLED;
+}
+
+static void
+pcf50633_client_dev_register(struct pcf50633 *pcf, const char *name,
+						struct platform_device **pdev)
+{
+	struct pcf50633_subdev_pdata *subdev_pdata;
+	int ret;
+
+	*pdev = platform_device_alloc(name, -1);
+	if (!*pdev) {
+		dev_err(pcf->dev, "Falied to allocate %s\n", name);
+		return;
+	}
+
+	subdev_pdata = kmalloc(sizeof(*subdev_pdata), GFP_KERNEL);
+	if (!subdev_pdata) {
+		dev_err(pcf->dev, "Error allocating subdev pdata\n");
+		platform_device_put(*pdev);
+	}
+
+	subdev_pdata->pcf = pcf;
+	platform_device_add_data(*pdev, subdev_pdata, sizeof(*subdev_pdata));
+
+	(*pdev)->dev.parent = pcf->dev;
+
+	ret = platform_device_add(*pdev);
+	if (ret) {
+		dev_err(pcf->dev, "Failed to register %s: %d\n", name, ret);
+		platform_device_put(*pdev);
+		*pdev = NULL;
+	}
+}
+
+#ifdef CONFIG_PM
+static int pcf50633_suspend(struct device *dev, pm_message_t state)
+{
+	struct pcf50633 *pcf;
+	int ret = 0, i;
+	u8 res[5];
+
+	pcf = dev_get_drvdata(dev);
+
+	/* Make sure our interrupt handlers are not called
+	 * henceforth */
+	disable_irq(pcf->irq);
+
+	/* Make sure that any running IRQ worker has quit */
+	cancel_work_sync(&pcf->irq_work);
+
+	/* Save the masks */
+	ret = pcf50633_read_block(pcf, PCF50633_REG_INT1M,
+				ARRAY_SIZE(pcf->suspend_irq_masks),
+					pcf->suspend_irq_masks);
+	if (ret < 0) {
+		dev_err(pcf->dev, "error saving irq masks\n");
+		goto out;
+	}
+
+	/* Write wakeup irq masks */
+	for (i = 0; i < ARRAY_SIZE(res); i++)
+		res[i] = ~pcf->pdata->resumers[i];
+
+	ret = pcf50633_write_block(pcf, PCF50633_REG_INT1M,
+					ARRAY_SIZE(res), &res[0]);
+	if (ret < 0) {
+		dev_err(pcf->dev, "error writing wakeup irq masks\n");
+		goto out;
+	}
+
+	pcf->is_suspended = 1;
+
+out:
+	return ret;
+}
+
+static int pcf50633_resume(struct device *dev)
+{
+	struct pcf50633 *pcf;
+	int ret;
+
+	pcf = dev_get_drvdata(dev);
+
+	/* Write the saved mask registers */
+	ret = pcf50633_write_block(pcf, PCF50633_REG_INT1M,
+				ARRAY_SIZE(pcf->suspend_irq_masks),
+					pcf->suspend_irq_masks);
+	if (ret < 0)
+		dev_err(pcf->dev, "Error restoring saved suspend masks\n");
+
+	/* Restore regulators' state */
+
+
+	get_device(pcf->dev);
+
+	/*
+	 * Clear any pending interrupts and set resume reason if any.
+	 * This will leave with enable_irq()
+	 */
+	pcf50633_irq_worker(&pcf->irq_work);
+
+	return 0;
+}
+#else
+#define pcf50633_suspend NULL
+#define pcf50633_resume NULL
+#endif
+
+static int __devinit pcf50633_probe(struct i2c_client *client,
+				const struct i2c_device_id *ids)
+{
+	struct pcf50633 *pcf;
+	struct pcf50633_platform_data *pdata = client->dev.platform_data;
+	int i, ret = 0;
+	int version, variant;
+
+	pcf = kzalloc(sizeof(*pcf), GFP_KERNEL);
+	if (!pcf)
+		return -ENOMEM;
+
+	pcf->pdata = pdata;
+
+	mutex_init(&pcf->lock);
+
+	i2c_set_clientdata(client, pcf);
+	pcf->dev = &client->dev;
+	pcf->i2c_client = client;
+	pcf->irq = client->irq;
+
+	INIT_WORK(&pcf->irq_work, pcf50633_irq_worker);
+
+	version = pcf50633_reg_read(pcf, 0);
+	variant = pcf50633_reg_read(pcf, 1);
+	if (version < 0 || variant < 0) {
+		dev_err(pcf->dev, "Unable to probe pcf50633\n");
+		ret = -ENODEV;
+		goto err;
+	}
+
+	dev_info(pcf->dev, "Probed device version %d variant %d\n",
+							version, variant);
+
+	/* Enable all interrupts except RTC SECOND */
+	pcf->mask_regs[0] = 0x80;
+	pcf50633_reg_write(pcf, PCF50633_REG_INT1M, pcf->mask_regs[0]);
+	pcf50633_reg_write(pcf, PCF50633_REG_INT2M, 0x00);
+	pcf50633_reg_write(pcf, PCF50633_REG_INT3M, 0x00);
+	pcf50633_reg_write(pcf, PCF50633_REG_INT4M, 0x00);
+	pcf50633_reg_write(pcf, PCF50633_REG_INT5M, 0x00);
+
+	/* Create sub devices */
+	pcf50633_client_dev_register(pcf, "pcf50633-input",
+						&pcf->input_pdev);
+	pcf50633_client_dev_register(pcf, "pcf50633-rtc",
+						&pcf->rtc_pdev);
+	pcf50633_client_dev_register(pcf, "pcf50633-mbc",
+						&pcf->mbc_pdev);
+	pcf50633_client_dev_register(pcf, "pcf50633-adc",
+						&pcf->adc_pdev);
+
+	for (i = 0; i < PCF50633_NUM_REGULATORS; i++) {
+		struct platform_device *pdev;
+
+		pdev = platform_device_alloc("pcf50633-regltr", i);
+		if (!pdev) {
+			dev_err(pcf->dev, "Cannot create regulator\n");
+			continue;
+		}
+
+		pdev->dev.parent = pcf->dev;
+		pdev->dev.platform_data = &pdata->reg_init_data[i];
+		pdev->dev.driver_data = pcf;
+		pcf->regulator_pdev[i] = pdev;
+
+		platform_device_add(pdev);
+	}
+
+	if (client->irq) {
+		set_irq_handler(client->irq, handle_level_irq);
+		ret = request_irq(client->irq, pcf50633_irq,
+				IRQF_TRIGGER_LOW, "pcf50633", pcf);
+
+		if (ret) {
+			dev_err(pcf->dev, "Failed to request IRQ %d\n", ret);
+			goto err;
+		}
+	} else {
+		dev_err(pcf->dev, "No IRQ configured\n");
+		goto err;
+	}
+
+	if (enable_irq_wake(client->irq) < 0)
+		dev_err(pcf->dev, "IRQ %u cannot be enabled as wake-up source"
+			"in this hardware revision", client->irq);
+
+	ret = sysfs_create_group(&client->dev.kobj, &pcf_attr_group);
+	if (ret)
+		dev_err(pcf->dev, "error creating sysfs entries\n");
+
+	if (pdata->probe_done)
+		pdata->probe_done(pcf);
+
+	return 0;
+
+err:
+	kfree(pcf);
+	return ret;
+}
+
+static int __devexit pcf50633_remove(struct i2c_client *client)
+{
+	struct pcf50633 *pcf = i2c_get_clientdata(client);
+	int i;
+
+	free_irq(pcf->irq, pcf);
+
+	platform_device_unregister(pcf->input_pdev);
+	platform_device_unregister(pcf->rtc_pdev);
+	platform_device_unregister(pcf->mbc_pdev);
+	platform_device_unregister(pcf->adc_pdev);
+
+	for (i = 0; i < PCF50633_NUM_REGULATORS; i++)
+		platform_device_unregister(pcf->regulator_pdev[i]);
+
+	kfree(pcf);
+
+	return 0;
+}
+
+static struct i2c_device_id pcf50633_id_table[] = {
+	{"pcf50633", 0x73},
+};
+
+static struct i2c_driver pcf50633_driver = {
+	.driver = {
+		.name	= "pcf50633",
+		.suspend = pcf50633_suspend,
+		.resume	= pcf50633_resume,
+	},
+	.id_table = pcf50633_id_table,
+	.probe = pcf50633_probe,
+	.remove = __devexit_p(pcf50633_remove),
+};
+
+static int __init pcf50633_init(void)
+{
+	return i2c_add_driver(&pcf50633_driver);
+}
+
+static void __exit pcf50633_exit(void)
+{
+	i2c_del_driver(&pcf50633_driver);
+}
+
+MODULE_DESCRIPTION("I2C chip driver for NXP PCF50633 PMU");
+MODULE_AUTHOR("Harald Welte <laforge@openmoko.org>");
+MODULE_LICENSE("GPL");
+
+module_init(pcf50633_init);
+module_exit(pcf50633_exit);
diff --git a/include/linux/mfd/pcf50633/core.h b/include/linux/mfd/pcf50633/core.h
new file mode 100644
index 000000000000..4455b212d75a
--- /dev/null
+++ b/include/linux/mfd/pcf50633/core.h
@@ -0,0 +1,218 @@
+/*
+ * core.h  -- Core driver for NXP PCF50633
+ *
+ * (C) 2006-2008 by Openmoko, Inc.
+ * All rights reserved.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+#ifndef __LINUX_MFD_PCF50633_CORE_H
+#define __LINUX_MFD_PCF50633_CORE_H
+
+#include <linux/i2c.h>
+#include <linux/workqueue.h>
+#include <linux/regulator/driver.h>
+#include <linux/regulator/machine.h>
+#include <linux/power_supply.h>
+
+struct pcf50633;
+
+#define PCF50633_NUM_REGULATORS	11
+
+struct pcf50633_platform_data {
+	struct regulator_init_data reg_init_data[PCF50633_NUM_REGULATORS];
+
+	char **batteries;
+	int num_batteries;
+
+	/* Callbacks */
+	void (*probe_done)(struct pcf50633 *);
+	void (*mbc_event_callback)(struct pcf50633 *, int);
+	void (*regulator_registered)(struct pcf50633 *, int);
+	void (*force_shutdown)(struct pcf50633 *);
+
+	u8 resumers[5];
+};
+
+struct pcf50633_subdev_pdata {
+	struct pcf50633 *pcf;
+};
+
+struct pcf50633_irq {
+	void (*handler) (int, void *);
+	void *data;
+};
+
+int pcf50633_register_irq(struct pcf50633 *pcf, int irq,
+			void (*handler) (int, void *), void *data);
+int pcf50633_free_irq(struct pcf50633 *pcf, int irq);
+
+int pcf50633_irq_mask(struct pcf50633 *pcf, int irq);
+int pcf50633_irq_unmask(struct pcf50633 *pcf, int irq);
+int pcf50633_irq_mask_get(struct pcf50633 *pcf, int irq);
+
+int pcf50633_read_block(struct pcf50633 *, u8 reg,
+					int nr_regs, u8 *data);
+int pcf50633_write_block(struct pcf50633 *pcf, u8 reg,
+					int nr_regs, u8 *data);
+u8 pcf50633_reg_read(struct pcf50633 *, u8 reg);
+int pcf50633_reg_write(struct pcf50633 *pcf, u8 reg, u8 val);
+
+int pcf50633_reg_set_bit_mask(struct pcf50633 *pcf, u8 reg, u8 mask, u8 val);
+int pcf50633_reg_clear_bits(struct pcf50633 *pcf, u8 reg, u8 bits);
+
+/* Interrupt registers */
+
+#define PCF50633_REG_INT1	0x02
+#define PCF50633_REG_INT2	0x03
+#define PCF50633_REG_INT3	0x04
+#define PCF50633_REG_INT4	0x05
+#define PCF50633_REG_INT5	0x06
+
+#define PCF50633_REG_INT1M	0x07
+#define PCF50633_REG_INT2M	0x08
+#define PCF50633_REG_INT3M	0x09
+#define PCF50633_REG_INT4M	0x0a
+#define PCF50633_REG_INT5M	0x0b
+
+enum {
+	/* Chip IRQs */
+	PCF50633_IRQ_ADPINS,
+	PCF50633_IRQ_ADPREM,
+	PCF50633_IRQ_USBINS,
+	PCF50633_IRQ_USBREM,
+	PCF50633_IRQ_RESERVED1,
+	PCF50633_IRQ_RESERVED2,
+	PCF50633_IRQ_ALARM,
+	PCF50633_IRQ_SECOND,
+	PCF50633_IRQ_ONKEYR,
+	PCF50633_IRQ_ONKEYF,
+	PCF50633_IRQ_EXTON1R,
+	PCF50633_IRQ_EXTON1F,
+	PCF50633_IRQ_EXTON2R,
+	PCF50633_IRQ_EXTON2F,
+	PCF50633_IRQ_EXTON3R,
+	PCF50633_IRQ_EXTON3F,
+	PCF50633_IRQ_BATFULL,
+	PCF50633_IRQ_CHGHALT,
+	PCF50633_IRQ_THLIMON,
+	PCF50633_IRQ_THLIMOFF,
+	PCF50633_IRQ_USBLIMON,
+	PCF50633_IRQ_USBLIMOFF,
+	PCF50633_IRQ_ADCRDY,
+	PCF50633_IRQ_ONKEY1S,
+	PCF50633_IRQ_LOWSYS,
+	PCF50633_IRQ_LOWBAT,
+	PCF50633_IRQ_HIGHTMP,
+	PCF50633_IRQ_AUTOPWRFAIL,
+	PCF50633_IRQ_DWN1PWRFAIL,
+	PCF50633_IRQ_DWN2PWRFAIL,
+	PCF50633_IRQ_LEDPWRFAIL,
+	PCF50633_IRQ_LEDOVP,
+	PCF50633_IRQ_LDO1PWRFAIL,
+	PCF50633_IRQ_LDO2PWRFAIL,
+	PCF50633_IRQ_LDO3PWRFAIL,
+	PCF50633_IRQ_LDO4PWRFAIL,
+	PCF50633_IRQ_LDO5PWRFAIL,
+	PCF50633_IRQ_LDO6PWRFAIL,
+	PCF50633_IRQ_HCLDOPWRFAIL,
+	PCF50633_IRQ_HCLDOOVL,
+
+	/* Always last */
+	PCF50633_NUM_IRQ,
+};
+
+struct pcf50633 {
+	struct device *dev;
+	struct i2c_client *i2c_client;
+
+	struct pcf50633_platform_data *pdata;
+	int irq;
+	struct pcf50633_irq irq_handler[PCF50633_NUM_IRQ];
+	struct work_struct irq_work;
+	struct mutex lock;
+
+	u8 mask_regs[5];
+
+	u8 suspend_irq_masks[5];
+	u8 resume_reason[5];
+	int is_suspended;
+
+	int onkey1s_held;
+
+	struct platform_device *rtc_pdev;
+	struct platform_device *mbc_pdev;
+	struct platform_device *adc_pdev;
+	struct platform_device *input_pdev;
+	struct platform_device *regulator_pdev[PCF50633_NUM_REGULATORS];
+};
+
+enum pcf50633_reg_int1 {
+	PCF50633_INT1_ADPINS	= 0x01,	/* Adapter inserted */
+	PCF50633_INT1_ADPREM	= 0x02,	/* Adapter removed */
+	PCF50633_INT1_USBINS	= 0x04,	/* USB inserted */
+	PCF50633_INT1_USBREM	= 0x08,	/* USB removed */
+	/* reserved */
+	PCF50633_INT1_ALARM	= 0x40, /* RTC alarm time is reached */
+	PCF50633_INT1_SECOND	= 0x80,	/* RTC periodic second interrupt */
+};
+
+enum pcf50633_reg_int2 {
+	PCF50633_INT2_ONKEYR	= 0x01, /* ONKEY rising edge */
+	PCF50633_INT2_ONKEYF	= 0x02, /* ONKEY falling edge */
+	PCF50633_INT2_EXTON1R	= 0x04, /* EXTON1 rising edge */
+	PCF50633_INT2_EXTON1F	= 0x08, /* EXTON1 falling edge */
+	PCF50633_INT2_EXTON2R	= 0x10, /* EXTON2 rising edge */
+	PCF50633_INT2_EXTON2F	= 0x20, /* EXTON2 falling edge */
+	PCF50633_INT2_EXTON3R	= 0x40, /* EXTON3 rising edge */
+	PCF50633_INT2_EXTON3F	= 0x80, /* EXTON3 falling edge */
+};
+
+enum pcf50633_reg_int3 {
+	PCF50633_INT3_BATFULL	= 0x01, /* Battery full */
+	PCF50633_INT3_CHGHALT	= 0x02,	/* Charger halt */
+	PCF50633_INT3_THLIMON	= 0x04,
+	PCF50633_INT3_THLIMOFF	= 0x08,
+	PCF50633_INT3_USBLIMON	= 0x10,
+	PCF50633_INT3_USBLIMOFF	= 0x20,
+	PCF50633_INT3_ADCRDY	= 0x40, /* ADC result ready */
+	PCF50633_INT3_ONKEY1S	= 0x80,	/* ONKEY pressed 1 second */
+};
+
+enum pcf50633_reg_int4 {
+	PCF50633_INT4_LOWSYS		= 0x01,
+	PCF50633_INT4_LOWBAT		= 0x02,
+	PCF50633_INT4_HIGHTMP		= 0x04,
+	PCF50633_INT4_AUTOPWRFAIL	= 0x08,
+	PCF50633_INT4_DWN1PWRFAIL	= 0x10,
+	PCF50633_INT4_DWN2PWRFAIL	= 0x20,
+	PCF50633_INT4_LEDPWRFAIL	= 0x40,
+	PCF50633_INT4_LEDOVP		= 0x80,
+};
+
+enum pcf50633_reg_int5 {
+	PCF50633_INT5_LDO1PWRFAIL	= 0x01,
+	PCF50633_INT5_LDO2PWRFAIL	= 0x02,
+	PCF50633_INT5_LDO3PWRFAIL	= 0x04,
+	PCF50633_INT5_LDO4PWRFAIL	= 0x08,
+	PCF50633_INT5_LDO5PWRFAIL	= 0x10,
+	PCF50633_INT5_LDO6PWRFAIL	= 0x20,
+	PCF50633_INT5_HCLDOPWRFAIL	= 0x40,
+	PCF50633_INT5_HCLDOOVL		= 0x80,
+};
+
+/* misc. registers */
+#define PCF50633_REG_OOCSHDWN	0x0c
+
+/* LED registers */
+#define PCF50633_REG_LEDOUT 0x28
+#define PCF50633_REG_LEDENA 0x29
+#define PCF50633_REG_LEDCTL 0x2a
+#define PCF50633_REG_LEDDIM 0x2b
+
+#endif
+
-- 
cgit v1.2.3


From 08c3e06a5eb27d43b712adef18379f8464425e71 Mon Sep 17 00:00:00 2001
From: Balaji Rao <balajirrao@openmoko.org>
Date: Fri, 9 Jan 2009 01:49:26 +0100
Subject: mfd: PCF50633 adc driver

This patch adds basic support for the PCF50633 ADC. The subtractive mode
is not supported yet.

Since we don't have adc subsystem, it currently lives in drivers/mfd.

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
Cc: Andy Green <andy@openmoko.com>
Acked-by: Jonathan Cameron <jonathan.cameron@gmail.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/Kconfig              |   7 +
 drivers/mfd/Makefile             |   3 +-
 drivers/mfd/pcf50633-adc.c       | 277 +++++++++++++++++++++++++++++++++++++++
 include/linux/mfd/pcf50633/adc.h |  72 ++++++++++
 4 files changed, 358 insertions(+), 1 deletion(-)
 create mode 100644 drivers/mfd/pcf50633-adc.c
 create mode 100644 include/linux/mfd/pcf50633/adc.h

(limited to 'include/linux')

diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index 3ac32e45b03b..2fa3504e165a 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -226,6 +226,13 @@ config MFD_PCF50633
 	  facilities, and registers devices for the various functions
 	  so that function-specific drivers can bind to them.
 
+config PCF50633_ADC
+	tristate "Support for NXP PCF50633 ADC"
+	depends on MFD_PCF50633
+	help
+	 Say yes here if you want to include support for ADC in the
+	 NXP PCF50633 chip.
+
 endmenu
 
 menu "Multimedia Capabilities Port drivers"
diff --git a/drivers/mfd/Makefile b/drivers/mfd/Makefile
index 23d4d10981b3..138f9c4d3d7b 100644
--- a/drivers/mfd/Makefile
+++ b/drivers/mfd/Makefile
@@ -38,4 +38,5 @@ obj-$(CONFIG_UCB1400_CORE)	+= ucb1400_core.o
 
 obj-$(CONFIG_PMIC_DA903X)	+= da903x.o
 
-obj-$(CONFIG_MFD_PCF50633)	+= pcf50633-core.o
\ No newline at end of file
+obj-$(CONFIG_MFD_PCF50633)	+= pcf50633-core.o
+obj-$(CONFIG_PCF50633_ADC)	+= pcf50633-adc.o
\ No newline at end of file
diff --git a/drivers/mfd/pcf50633-adc.c b/drivers/mfd/pcf50633-adc.c
new file mode 100644
index 000000000000..c2d05becfa97
--- /dev/null
+++ b/drivers/mfd/pcf50633-adc.c
@@ -0,0 +1,277 @@
+/* NXP PCF50633 ADC Driver
+ *
+ * (C) 2006-2008 by Openmoko, Inc.
+ * Author: Balaji Rao <balajirrao@openmoko.org>
+ * All rights reserved.
+ *
+ * Broken down from monstrous PCF50633 driver mainly by
+ * Harald Welte, Andy Green and Werner Almesberger
+ *
+ *  This program is free software; you can redistribute  it and/or modify it
+ *  under  the terms of  the GNU General  Public License as published by the
+ *  Free Software Foundation;  either version 2 of the  License, or (at your
+ *  option) any later version.
+ *
+ *  NOTE: This driver does not yet support subtractive ADC mode, which means
+ *  you can do only one measurement per read request.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/device.h>
+#include <linux/platform_device.h>
+#include <linux/completion.h>
+
+#include <linux/mfd/pcf50633/core.h>
+#include <linux/mfd/pcf50633/adc.h>
+
+struct pcf50633_adc_request {
+	int mux;
+	int avg;
+	int result;
+	void (*callback)(struct pcf50633 *, void *, int);
+	void *callback_param;
+
+	/* Used in case of sync requests */
+	struct completion completion;
+
+};
+
+#define PCF50633_MAX_ADC_FIFO_DEPTH 8
+
+struct pcf50633_adc {
+	struct pcf50633 *pcf;
+
+	/* Private stuff */
+	struct pcf50633_adc_request *queue[PCF50633_MAX_ADC_FIFO_DEPTH];
+	int queue_head;
+	int queue_tail;
+	struct mutex queue_mutex;
+};
+
+static inline struct pcf50633_adc *__to_adc(struct pcf50633 *pcf)
+{
+	return platform_get_drvdata(pcf->adc_pdev);
+}
+
+static void adc_setup(struct pcf50633 *pcf, int channel, int avg)
+{
+	channel &= PCF50633_ADCC1_ADCMUX_MASK;
+
+	/* kill ratiometric, but enable ACCSW biasing */
+	pcf50633_reg_write(pcf, PCF50633_REG_ADCC2, 0x00);
+	pcf50633_reg_write(pcf, PCF50633_REG_ADCC3, 0x01);
+
+	/* start ADC conversion on selected channel */
+	pcf50633_reg_write(pcf, PCF50633_REG_ADCC1, channel | avg |
+		    PCF50633_ADCC1_ADCSTART | PCF50633_ADCC1_RES_10BIT);
+}
+
+static void trigger_next_adc_job_if_any(struct pcf50633 *pcf)
+{
+	struct pcf50633_adc *adc = __to_adc(pcf);
+	int head;
+
+	mutex_lock(&adc->queue_mutex);
+
+	head = adc->queue_head;
+
+	if (!adc->queue[head]) {
+		mutex_unlock(&adc->queue_mutex);
+		return;
+	}
+	mutex_unlock(&adc->queue_mutex);
+
+	adc_setup(pcf, adc->queue[head]->mux, adc->queue[head]->avg);
+}
+
+static int
+adc_enqueue_request(struct pcf50633 *pcf, struct pcf50633_adc_request *req)
+{
+	struct pcf50633_adc *adc = __to_adc(pcf);
+	int head, tail;
+
+	mutex_lock(&adc->queue_mutex);
+
+	head = adc->queue_head;
+	tail = adc->queue_tail;
+
+	if (adc->queue[tail]) {
+		mutex_unlock(&adc->queue_mutex);
+		return -EBUSY;
+	}
+
+	adc->queue[tail] = req;
+	adc->queue_tail = (tail + 1) & (PCF50633_MAX_ADC_FIFO_DEPTH - 1);
+
+	mutex_unlock(&adc->queue_mutex);
+
+	trigger_next_adc_job_if_any(pcf);
+
+	return 0;
+}
+
+static void
+pcf50633_adc_sync_read_callback(struct pcf50633 *pcf, void *param, int result)
+{
+	struct pcf50633_adc_request *req = param;
+
+	req->result = result;
+	complete(&req->completion);
+}
+
+int pcf50633_adc_sync_read(struct pcf50633 *pcf, int mux, int avg)
+{
+	struct pcf50633_adc_request *req;
+
+	/* req is freed when the result is ready, in interrupt handler */
+	req = kzalloc(sizeof(*req), GFP_KERNEL);
+	if (!req)
+		return -ENOMEM;
+
+	req->mux = mux;
+	req->avg = avg;
+	req->callback =  pcf50633_adc_sync_read_callback;
+	req->callback_param = req;
+
+	init_completion(&req->completion);
+	adc_enqueue_request(pcf, req);
+	wait_for_completion(&req->completion);
+
+	return req->result;
+}
+EXPORT_SYMBOL_GPL(pcf50633_adc_sync_read);
+
+int pcf50633_adc_async_read(struct pcf50633 *pcf, int mux, int avg,
+			     void (*callback)(struct pcf50633 *, void *, int),
+			     void *callback_param)
+{
+	struct pcf50633_adc_request *req;
+
+	/* req is freed when the result is ready, in interrupt handler */
+	req = kmalloc(sizeof(*req), GFP_KERNEL);
+	if (!req)
+		return -ENOMEM;
+
+	req->mux = mux;
+	req->avg = avg;
+	req->callback = callback;
+	req->callback_param = callback_param;
+
+	adc_enqueue_request(pcf, req);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(pcf50633_adc_async_read);
+
+static int adc_result(struct pcf50633 *pcf)
+{
+	u8 adcs1, adcs3;
+	u16 result;
+
+	adcs1 = pcf50633_reg_read(pcf, PCF50633_REG_ADCS1);
+	adcs3 = pcf50633_reg_read(pcf, PCF50633_REG_ADCS3);
+	result = (adcs1 << 2) | (adcs3 & PCF50633_ADCS3_ADCDAT1L_MASK);
+
+	dev_dbg(pcf->dev, "adc result = %d\n", result);
+
+	return result;
+}
+
+static void pcf50633_adc_irq(int irq, void *data)
+{
+	struct pcf50633_adc *adc = data;
+	struct pcf50633 *pcf = adc->pcf;
+	struct pcf50633_adc_request *req;
+	int head;
+
+	mutex_lock(&adc->queue_mutex);
+	head = adc->queue_head;
+
+	req = adc->queue[head];
+	if (WARN_ON(!req)) {
+		dev_err(pcf->dev, "pcf50633-adc irq: ADC queue empty!\n");
+		mutex_unlock(&adc->queue_mutex);
+		return;
+	}
+	adc->queue[head] = NULL;
+	adc->queue_head = (head + 1) &
+				      (PCF50633_MAX_ADC_FIFO_DEPTH - 1);
+
+	mutex_unlock(&adc->queue_mutex);
+
+	req->callback(pcf, req->callback_param, adc_result(pcf));
+	kfree(req);
+
+	trigger_next_adc_job_if_any(pcf);
+}
+
+static int __devinit pcf50633_adc_probe(struct platform_device *pdev)
+{
+	struct pcf50633_subdev_pdata *pdata = pdev->dev.platform_data;
+	struct pcf50633_adc *adc;
+
+	adc = kzalloc(sizeof(*adc), GFP_KERNEL);
+	if (!adc)
+		return -ENOMEM;
+
+	adc->pcf = pdata->pcf;
+	platform_set_drvdata(pdev, adc);
+
+	pcf50633_register_irq(pdata->pcf, PCF50633_IRQ_ADCRDY,
+					pcf50633_adc_irq, adc);
+
+	mutex_init(&adc->queue_mutex);
+
+	return 0;
+}
+
+static int __devexit pcf50633_adc_remove(struct platform_device *pdev)
+{
+	struct pcf50633_adc *adc = platform_get_drvdata(pdev);
+	int i, head;
+
+	pcf50633_free_irq(adc->pcf, PCF50633_IRQ_ADCRDY);
+
+	mutex_lock(&adc->queue_mutex);
+	head = adc->queue_head;
+
+	if (WARN_ON(adc->queue[head]))
+		dev_err(adc->pcf->dev,
+			"adc driver removed with request pending\n");
+
+	for (i = 0; i < PCF50633_MAX_ADC_FIFO_DEPTH; i++)
+		kfree(adc->queue[i]);
+
+	mutex_unlock(&adc->queue_mutex);
+	kfree(adc);
+
+	return 0;
+}
+
+static struct platform_driver pcf50633_adc_driver = {
+	.driver = {
+		.name = "pcf50633-adc",
+	},
+	.probe = pcf50633_adc_probe,
+	.remove = __devexit_p(pcf50633_adc_remove),
+};
+
+static int __init pcf50633_adc_init(void)
+{
+	return platform_driver_register(&pcf50633_adc_driver);
+}
+module_init(pcf50633_adc_init);
+
+static void __exit pcf50633_adc_exit(void)
+{
+	platform_driver_unregister(&pcf50633_adc_driver);
+}
+module_exit(pcf50633_adc_exit);
+
+MODULE_AUTHOR("Balaji Rao <balajirrao@openmoko.org>");
+MODULE_DESCRIPTION("PCF50633 adc driver");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:pcf50633-adc");
+
diff --git a/include/linux/mfd/pcf50633/adc.h b/include/linux/mfd/pcf50633/adc.h
new file mode 100644
index 000000000000..56669b4183ad
--- /dev/null
+++ b/include/linux/mfd/pcf50633/adc.h
@@ -0,0 +1,72 @@
+/*
+ * adc.h  -- Driver for NXP PCF50633 ADC
+ *
+ * (C) 2006-2008 by Openmoko, Inc.
+ * All rights reserved.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+#ifndef __LINUX_MFD_PCF50633_ADC_H
+#define __LINUX_MFD_PCF50633_ADC_H
+
+#include <linux/mfd/pcf50633/core.h>
+#include <linux/platform_device.h>
+
+/* ADC Registers */
+#define PCF50633_REG_ADCC3		0x52
+#define PCF50633_REG_ADCC2		0x53
+#define PCF50633_REG_ADCC1		0x54
+#define PCF50633_REG_ADCS1		0x55
+#define PCF50633_REG_ADCS2		0x56
+#define PCF50633_REG_ADCS3		0x57
+
+#define PCF50633_ADCC1_ADCSTART		0x01
+#define PCF50633_ADCC1_RES_10BIT	0x02
+#define PCF50633_ADCC1_AVERAGE_NO	0x00
+#define PCF50633_ADCC1_AVERAGE_4	0x04
+#define PCF50633_ADCC1_AVERAGE_8	0x08
+#define PCF50633_ADCC1_AVERAGE_16	0x0c
+#define PCF50633_ADCC1_MUX_BATSNS_RES	0x00
+#define PCF50633_ADCC1_MUX_BATSNS_SUBTR	0x10
+#define PCF50633_ADCC1_MUX_ADCIN2_RES	0x20
+#define PCF50633_ADCC1_MUX_ADCIN2_SUBTR	0x30
+#define PCF50633_ADCC1_MUX_BATTEMP	0x60
+#define PCF50633_ADCC1_MUX_ADCIN1	0x70
+#define PCF50633_ADCC1_AVERAGE_MASK	0x0c
+#define PCF50633_ADCC1_ADCMUX_MASK	0xf0
+
+#define PCF50633_ADCC2_RATIO_NONE	0x00
+#define PCF50633_ADCC2_RATIO_BATTEMP	0x01
+#define PCF50633_ADCC2_RATIO_ADCIN1	0x02
+#define PCF50633_ADCC2_RATIO_BOTH	0x03
+#define PCF50633_ADCC2_RATIOSETTL_100US 0x04
+
+#define PCF50633_ADCC3_ACCSW_EN		0x01
+#define PCF50633_ADCC3_NTCSW_EN		0x04
+#define PCF50633_ADCC3_RES_DIV_TWO	0x10
+#define PCF50633_ADCC3_RES_DIV_THREE	0x00
+
+#define PCF50633_ADCS3_REF_NTCSW	0x00
+#define PCF50633_ADCS3_REF_ACCSW	0x10
+#define PCF50633_ADCS3_REF_2V0		0x20
+#define PCF50633_ADCS3_REF_VISA		0x30
+#define PCF50633_ADCS3_REF_2V0_2	0x70
+#define PCF50633_ADCS3_ADCRDY		0x80
+
+#define PCF50633_ADCS3_ADCDAT1L_MASK	0x03
+#define PCF50633_ADCS3_ADCDAT2L_MASK	0x0c
+#define PCF50633_ADCS3_ADCDAT2L_SHIFT	2
+#define PCF50633_ASCS3_REF_MASK		0x70
+
+extern int
+pcf50633_adc_async_read(struct pcf50633 *pcf, int mux, int avg,
+		void (*callback)(struct pcf50633 *, void *, int),
+		void *callback_param);
+extern int
+pcf50633_adc_sync_read(struct pcf50633 *pcf, int mux, int avg);
+
+#endif /* __LINUX_PCF50633_ADC_H */
-- 
cgit v1.2.3


From 6a3d119b4ce29cf32bfe91eb61d46e9dbd8ce38a Mon Sep 17 00:00:00 2001
From: Balaji Rao <balajirrao@openmoko.org>
Date: Fri, 9 Jan 2009 01:49:37 +0100
Subject: mfd: PCF50633 gpio support

What the PCF05633 calls as a 'GPIO' is much more than the GPIO in the linux
sense and there are only 4 of them - which means, the gpiolib is not used
here.

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
Cc: Andy Green <andy@openmoko.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/mfd/Kconfig               |   7 +++
 drivers/mfd/Makefile              |   3 +-
 drivers/mfd/pcf50633-gpio.c       | 118 ++++++++++++++++++++++++++++++++++++++
 include/linux/mfd/pcf50633/gpio.h |  52 +++++++++++++++++
 4 files changed, 179 insertions(+), 1 deletion(-)
 create mode 100644 drivers/mfd/pcf50633-gpio.c
 create mode 100644 include/linux/mfd/pcf50633/gpio.h

(limited to 'include/linux')

diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index 2fa3504e165a..06a2b0f7737c 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -233,6 +233,13 @@ config PCF50633_ADC
 	 Say yes here if you want to include support for ADC in the
 	 NXP PCF50633 chip.
 
+config PCF50633_GPIO
+	tristate "Support for NXP PCF50633 GPIO"
+	depends on MFD_PCF50633
+	help
+	 Say yes here if you want to include support GPIO for pins on
+	 the PCF50633 chip.
+
 endmenu
 
 menu "Multimedia Capabilities Port drivers"
diff --git a/drivers/mfd/Makefile b/drivers/mfd/Makefile
index 138f9c4d3d7b..3afb5192e4da 100644
--- a/drivers/mfd/Makefile
+++ b/drivers/mfd/Makefile
@@ -39,4 +39,5 @@ obj-$(CONFIG_UCB1400_CORE)	+= ucb1400_core.o
 obj-$(CONFIG_PMIC_DA903X)	+= da903x.o
 
 obj-$(CONFIG_MFD_PCF50633)	+= pcf50633-core.o
-obj-$(CONFIG_PCF50633_ADC)	+= pcf50633-adc.o
\ No newline at end of file
+obj-$(CONFIG_PCF50633_ADC)	+= pcf50633-adc.o
+obj-$(CONFIG_PCF50633_GPIO)	+= pcf50633-gpio.o
\ No newline at end of file
diff --git a/drivers/mfd/pcf50633-gpio.c b/drivers/mfd/pcf50633-gpio.c
new file mode 100644
index 000000000000..2fa2eca5c9cc
--- /dev/null
+++ b/drivers/mfd/pcf50633-gpio.c
@@ -0,0 +1,118 @@
+/* NXP PCF50633 GPIO Driver
+ *
+ * (C) 2006-2008 by Openmoko, Inc.
+ * Author: Balaji Rao <balajirrao@openmoko.org>
+ * All rights reserved.
+ *
+ * Broken down from monstrous PCF50633 driver mainly by
+ * Harald Welte, Andy Green and Werner Almesberger
+ *
+ *  This program is free software; you can redistribute  it and/or modify it
+ *  under  the terms of  the GNU General  Public License as published by the
+ *  Free Software Foundation;  either version 2 of the  License, or (at your
+ *  option) any later version.
+ *
+ */
+
+#include <linux/kernel.h>
+
+#include <linux/mfd/pcf50633/core.h>
+#include <linux/mfd/pcf50633/gpio.h>
+
+enum pcf50633_regulator_id {
+	PCF50633_REGULATOR_AUTO,
+	PCF50633_REGULATOR_DOWN1,
+	PCF50633_REGULATOR_DOWN2,
+	PCF50633_REGULATOR_LDO1,
+	PCF50633_REGULATOR_LDO2,
+	PCF50633_REGULATOR_LDO3,
+	PCF50633_REGULATOR_LDO4,
+	PCF50633_REGULATOR_LDO5,
+	PCF50633_REGULATOR_LDO6,
+	PCF50633_REGULATOR_HCLDO,
+	PCF50633_REGULATOR_MEMLDO,
+};
+
+#define PCF50633_REG_AUTOOUT	0x1a
+#define PCF50633_REG_DOWN1OUT	0x1e
+#define PCF50633_REG_DOWN2OUT	0x22
+#define PCF50633_REG_MEMLDOOUT	0x26
+#define PCF50633_REG_LDO1OUT	0x2d
+#define PCF50633_REG_LDO2OUT	0x2f
+#define PCF50633_REG_LDO3OUT	0x31
+#define PCF50633_REG_LDO4OUT	0x33
+#define PCF50633_REG_LDO5OUT	0x35
+#define PCF50633_REG_LDO6OUT	0x37
+#define PCF50633_REG_HCLDOOUT	0x39
+
+static const u8 pcf50633_regulator_registers[PCF50633_NUM_REGULATORS] = {
+	[PCF50633_REGULATOR_AUTO]	= PCF50633_REG_AUTOOUT,
+	[PCF50633_REGULATOR_DOWN1]	= PCF50633_REG_DOWN1OUT,
+	[PCF50633_REGULATOR_DOWN2]	= PCF50633_REG_DOWN2OUT,
+	[PCF50633_REGULATOR_MEMLDO]	= PCF50633_REG_MEMLDOOUT,
+	[PCF50633_REGULATOR_LDO1]	= PCF50633_REG_LDO1OUT,
+	[PCF50633_REGULATOR_LDO2]	= PCF50633_REG_LDO2OUT,
+	[PCF50633_REGULATOR_LDO3]	= PCF50633_REG_LDO3OUT,
+	[PCF50633_REGULATOR_LDO4]	= PCF50633_REG_LDO4OUT,
+	[PCF50633_REGULATOR_LDO5]	= PCF50633_REG_LDO5OUT,
+	[PCF50633_REGULATOR_LDO6]	= PCF50633_REG_LDO6OUT,
+	[PCF50633_REGULATOR_HCLDO]	= PCF50633_REG_HCLDOOUT,
+};
+
+int pcf50633_gpio_set(struct pcf50633 *pcf, int gpio, u8 val)
+{
+	u8 reg;
+
+	reg = gpio - PCF50633_GPIO1 + PCF50633_REG_GPIO1CFG;
+
+	return pcf50633_reg_set_bit_mask(pcf, reg, 0x07, val);
+}
+EXPORT_SYMBOL_GPL(pcf50633_gpio_set);
+
+u8 pcf50633_gpio_get(struct pcf50633 *pcf, int gpio)
+{
+	u8 reg, val;
+
+	reg = gpio - PCF50633_GPIO1 + PCF50633_REG_GPIO1CFG;
+	val = pcf50633_reg_read(pcf, reg) & 0x07;
+
+	return val;
+}
+EXPORT_SYMBOL_GPL(pcf50633_gpio_get);
+
+int pcf50633_gpio_invert_set(struct pcf50633 *pcf, int gpio, int invert)
+{
+	u8 val, reg;
+
+	reg = gpio - PCF50633_GPIO1 + PCF50633_REG_GPIO1CFG;
+	val = !!invert << 3;
+
+	return pcf50633_reg_set_bit_mask(pcf, reg, 1 << 3, val);
+}
+EXPORT_SYMBOL_GPL(pcf50633_gpio_invert_set);
+
+int pcf50633_gpio_invert_get(struct pcf50633 *pcf, int gpio)
+{
+	u8 reg, val;
+
+	reg = gpio - PCF50633_GPIO1 + PCF50633_REG_GPIO1CFG;
+	val = pcf50633_reg_read(pcf, reg);
+
+	return val & (1 << 3);
+}
+EXPORT_SYMBOL_GPL(pcf50633_gpio_invert_get);
+
+int pcf50633_gpio_power_supply_set(struct pcf50633 *pcf,
+					int gpio, int regulator, int on)
+{
+	u8 reg, val, mask;
+
+	/* the *ENA register is always one after the *OUT register */
+	reg = pcf50633_regulator_registers[regulator] + 1;
+
+	val = !!on << (gpio - PCF50633_GPIO1);
+	mask = 1 << (gpio - PCF50633_GPIO1);
+
+	return pcf50633_reg_set_bit_mask(pcf, reg, mask, val);
+}
+EXPORT_SYMBOL_GPL(pcf50633_gpio_power_supply_set);
diff --git a/include/linux/mfd/pcf50633/gpio.h b/include/linux/mfd/pcf50633/gpio.h
new file mode 100644
index 000000000000..a42b845efc54
--- /dev/null
+++ b/include/linux/mfd/pcf50633/gpio.h
@@ -0,0 +1,52 @@
+/*
+ * gpio.h -- GPIO driver for NXP PCF50633
+ *
+ * (C) 2006-2008 by Openmoko, Inc.
+ * All rights reserved.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+#ifndef __LINUX_MFD_PCF50633_GPIO_H
+#define __LINUX_MFD_PCF50633_GPIO_H
+
+#include <linux/mfd/pcf50633/core.h>
+
+#define PCF50633_GPIO1		1
+#define PCF50633_GPIO2		2
+#define PCF50633_GPIO3		3
+#define PCF50633_GPO		4
+
+#define PCF50633_REG_GPIO1CFG	0x14
+#define PCF50633_REG_GPIO2CFG	0x15
+#define PCF50633_REG_GPIO3CFG	0x16
+#define PCF50633_REG_GPOCFG 	0x17
+
+#define PCF50633_GPOCFG_GPOSEL_MASK	0x07
+
+enum pcf50633_reg_gpocfg {
+	PCF50633_GPOCFG_GPOSEL_0	= 0x00,
+	PCF50633_GPOCFG_GPOSEL_LED_NFET	= 0x01,
+	PCF50633_GPOCFG_GPOSEL_SYSxOK	= 0x02,
+	PCF50633_GPOCFG_GPOSEL_CLK32K	= 0x03,
+	PCF50633_GPOCFG_GPOSEL_ADAPUSB	= 0x04,
+	PCF50633_GPOCFG_GPOSEL_USBxOK	= 0x05,
+	PCF50633_GPOCFG_GPOSEL_ACTPH4	= 0x06,
+	PCF50633_GPOCFG_GPOSEL_1	= 0x07,
+	PCF50633_GPOCFG_GPOSEL_INVERSE	= 0x08,
+};
+
+int pcf50633_gpio_set(struct pcf50633 *pcf, int gpio, u8 val);
+u8 pcf50633_gpio_get(struct pcf50633 *pcf, int gpio);
+
+int pcf50633_gpio_invert_set(struct pcf50633 *, int gpio, int invert);
+int pcf50633_gpio_invert_get(struct pcf50633 *pcf, int gpio);
+
+int pcf50633_gpio_power_supply_set(struct pcf50633 *,
+					int gpio, int regulator, int on);
+#endif /* __LINUX_MFD_PCF50633_GPIO_H */
+
+
-- 
cgit v1.2.3


From f5714dc97d63cc0dd1219bd0eb2e1f8df1e4347a Mon Sep 17 00:00:00 2001
From: Balaji Rao <balajirrao@openmoko.org>
Date: Fri, 9 Jan 2009 01:50:55 +0100
Subject: power_supply: PCF50633 battery charger driver

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
Cc: Andy Green <andy@openmoko.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Acked-by: Anton Vorontsov <cbouatmailru@gmail.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/power/Kconfig            |   6 +
 drivers/power/Makefile           |   1 +
 drivers/power/pcf50633-charger.c | 358 +++++++++++++++++++++++++++++++++++++++
 include/linux/mfd/pcf50633/mbc.h | 134 +++++++++++++++
 4 files changed, 499 insertions(+)
 create mode 100644 drivers/power/pcf50633-charger.c
 create mode 100644 include/linux/mfd/pcf50633/mbc.h

(limited to 'include/linux')

diff --git a/drivers/power/Kconfig b/drivers/power/Kconfig
index 668472405a57..33da1127992a 100644
--- a/drivers/power/Kconfig
+++ b/drivers/power/Kconfig
@@ -82,4 +82,10 @@ config BATTERY_DA9030
 	  Say Y here to enable support for batteries charger integrated into
 	  DA9030 PMIC.
 
+config CHARGER_PCF50633
+	tristate "NXP PCF50633 MBC"
+	depends on MFD_PCF50633
+	help
+	 Say Y to include support for NXP PCF50633 Main Battery Charger.
+
 endif # POWER_SUPPLY
diff --git a/drivers/power/Makefile b/drivers/power/Makefile
index eebb15505a40..2fcf41d13e5c 100644
--- a/drivers/power/Makefile
+++ b/drivers/power/Makefile
@@ -25,3 +25,4 @@ obj-$(CONFIG_BATTERY_TOSA)	+= tosa_battery.o
 obj-$(CONFIG_BATTERY_WM97XX)	+= wm97xx_battery.o
 obj-$(CONFIG_BATTERY_BQ27x00)	+= bq27x00_battery.o
 obj-$(CONFIG_BATTERY_DA9030)	+= da9030_battery.o
+obj-$(CONFIG_CHARGER_PCF50633)	+= pcf50633-charger.o
\ No newline at end of file
diff --git a/drivers/power/pcf50633-charger.c b/drivers/power/pcf50633-charger.c
new file mode 100644
index 000000000000..e988ec130fcd
--- /dev/null
+++ b/drivers/power/pcf50633-charger.c
@@ -0,0 +1,358 @@
+/* NXP PCF50633 Main Battery Charger Driver
+ *
+ * (C) 2006-2008 by Openmoko, Inc.
+ * Author: Balaji Rao <balajirrao@openmoko.org>
+ * All rights reserved.
+ *
+ * Broken down from monstrous PCF50633 driver mainly by
+ * Harald Welte, Andy Green and Werner Almesberger
+ *
+ *  This program is free software; you can redistribute  it and/or modify it
+ *  under  the terms of  the GNU General  Public License as published by the
+ *  Free Software Foundation;  either version 2 of the  License, or (at your
+ *  option) any later version.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/device.h>
+#include <linux/sysfs.h>
+#include <linux/platform_device.h>
+#include <linux/power_supply.h>
+
+#include <linux/mfd/pcf50633/core.h>
+#include <linux/mfd/pcf50633/mbc.h>
+
+struct pcf50633_mbc {
+	struct pcf50633 *pcf;
+
+	int adapter_active;
+	int adapter_online;
+	int usb_active;
+	int usb_online;
+
+	struct power_supply usb;
+	struct power_supply adapter;
+};
+
+int pcf50633_mbc_usb_curlim_set(struct pcf50633 *pcf, int ma)
+{
+	struct pcf50633_mbc *mbc = platform_get_drvdata(pcf->mbc_pdev);
+	int ret = 0;
+	u8 bits;
+
+	if (ma >= 1000)
+		bits = PCF50633_MBCC7_USB_1000mA;
+	else if (ma >= 500)
+		bits = PCF50633_MBCC7_USB_500mA;
+	else if (ma >= 100)
+		bits = PCF50633_MBCC7_USB_100mA;
+	else
+		bits = PCF50633_MBCC7_USB_SUSPEND;
+
+	ret = pcf50633_reg_set_bit_mask(pcf, PCF50633_REG_MBCC7,
+					PCF50633_MBCC7_USB_MASK, bits);
+	if (ret)
+		dev_err(pcf->dev, "error setting usb curlim to %d mA\n", ma);
+	else
+		dev_info(pcf->dev, "usb curlim to %d mA\n", ma);
+
+	power_supply_changed(&mbc->usb);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(pcf50633_mbc_usb_curlim_set);
+
+int pcf50633_mbc_get_status(struct pcf50633 *pcf)
+{
+	struct pcf50633_mbc *mbc  = platform_get_drvdata(pcf->mbc_pdev);
+	int status = 0;
+
+	if (mbc->usb_online)
+		status |= PCF50633_MBC_USB_ONLINE;
+	if (mbc->usb_active)
+		status |= PCF50633_MBC_USB_ACTIVE;
+	if (mbc->adapter_online)
+		status |= PCF50633_MBC_ADAPTER_ONLINE;
+	if (mbc->adapter_active)
+		status |= PCF50633_MBC_ADAPTER_ACTIVE;
+
+	return status;
+}
+EXPORT_SYMBOL_GPL(pcf50633_mbc_get_status);
+
+void pcf50633_mbc_set_status(struct pcf50633 *pcf, int what, int status)
+{
+	struct pcf50633_mbc *mbc = platform_get_drvdata(pcf->mbc_pdev);
+
+	if (what & PCF50633_MBC_USB_ONLINE)
+		mbc->usb_online = !!status;
+	if (what & PCF50633_MBC_USB_ACTIVE)
+		mbc->usb_active = !!status;
+	if (what & PCF50633_MBC_ADAPTER_ONLINE)
+		mbc->adapter_online = !!status;
+	if (what & PCF50633_MBC_ADAPTER_ACTIVE)
+		mbc->adapter_active = !!status;
+}
+EXPORT_SYMBOL_GPL(pcf50633_mbc_set_status);
+
+static ssize_t
+show_chgmode(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct pcf50633_mbc *mbc = dev_get_drvdata(dev);
+
+	u8 mbcs2 = pcf50633_reg_read(mbc->pcf, PCF50633_REG_MBCS2);
+	u8 chgmod = (mbcs2 & PCF50633_MBCS2_MBC_MASK);
+
+	return sprintf(buf, "%d\n", chgmod);
+}
+static DEVICE_ATTR(chgmode, S_IRUGO, show_chgmode, NULL);
+
+static ssize_t
+show_usblim(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct pcf50633_mbc *mbc = dev_get_drvdata(dev);
+	u8 usblim = pcf50633_reg_read(mbc->pcf, PCF50633_REG_MBCC7) &
+						PCF50633_MBCC7_USB_MASK;
+	unsigned int ma;
+
+	if (usblim == PCF50633_MBCC7_USB_1000mA)
+		ma = 1000;
+	else if (usblim == PCF50633_MBCC7_USB_500mA)
+		ma = 500;
+	else if (usblim == PCF50633_MBCC7_USB_100mA)
+		ma = 100;
+	else
+		ma = 0;
+
+	return sprintf(buf, "%u\n", ma);
+}
+
+static ssize_t set_usblim(struct device *dev,
+		struct device_attribute *attr, const char *buf, size_t count)
+{
+	struct pcf50633_mbc *mbc = dev_get_drvdata(dev);
+	unsigned long ma;
+	int ret;
+
+	ret = strict_strtoul(buf, 10, &ma);
+	if (ret)
+		return -EINVAL;
+
+	pcf50633_mbc_usb_curlim_set(mbc->pcf, ma);
+
+	return count;
+}
+
+static DEVICE_ATTR(usb_curlim, S_IRUGO | S_IWUSR, show_usblim, set_usblim);
+
+static struct attribute *pcf50633_mbc_sysfs_entries[] = {
+	&dev_attr_chgmode.attr,
+	&dev_attr_usb_curlim.attr,
+	NULL,
+};
+
+static struct attribute_group mbc_attr_group = {
+	.name	= NULL,			/* put in device directory */
+	.attrs	= pcf50633_mbc_sysfs_entries,
+};
+
+static void
+pcf50633_mbc_irq_handler(int irq, void *data)
+{
+	struct pcf50633_mbc *mbc = data;
+
+	/* USB */
+	if (irq == PCF50633_IRQ_USBINS) {
+		mbc->usb_online = 1;
+	} else if (irq == PCF50633_IRQ_USBREM) {
+		mbc->usb_online = 0;
+		mbc->usb_active = 0;
+		pcf50633_mbc_usb_curlim_set(mbc->pcf, 0);
+	}
+
+	/* Adapter */
+	if (irq == PCF50633_IRQ_ADPINS) {
+		mbc->adapter_online = 1;
+		mbc->adapter_active = 1;
+	} else if (irq == PCF50633_IRQ_ADPREM) {
+		mbc->adapter_online = 0;
+		mbc->adapter_active = 0;
+	}
+
+	if (irq == PCF50633_IRQ_BATFULL) {
+		mbc->usb_active = 0;
+		mbc->adapter_active = 0;
+	}
+
+	power_supply_changed(&mbc->usb);
+	power_supply_changed(&mbc->adapter);
+
+	if (mbc->pcf->pdata->mbc_event_callback)
+		mbc->pcf->pdata->mbc_event_callback(mbc->pcf, irq);
+}
+
+static int adapter_get_property(struct power_supply *psy,
+			enum power_supply_property psp,
+			union power_supply_propval *val)
+{
+	struct pcf50633_mbc *mbc = container_of(psy, struct pcf50633_mbc, usb);
+	int ret = 0;
+
+	switch (psp) {
+	case POWER_SUPPLY_PROP_ONLINE:
+		val->intval =  mbc->adapter_online;
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+	return ret;
+}
+
+static int usb_get_property(struct power_supply *psy,
+			enum power_supply_property psp,
+			union power_supply_propval *val)
+{
+	struct pcf50633_mbc *mbc = container_of(psy, struct pcf50633_mbc, usb);
+	int ret = 0;
+
+	switch (psp) {
+	case POWER_SUPPLY_PROP_ONLINE:
+		val->intval = mbc->usb_online;
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+	return ret;
+}
+
+static enum power_supply_property power_props[] = {
+	POWER_SUPPLY_PROP_ONLINE,
+};
+
+static const u8 mbc_irq_handlers[] = {
+	PCF50633_IRQ_ADPINS,
+	PCF50633_IRQ_ADPREM,
+	PCF50633_IRQ_USBINS,
+	PCF50633_IRQ_USBREM,
+	PCF50633_IRQ_BATFULL,
+	PCF50633_IRQ_CHGHALT,
+	PCF50633_IRQ_THLIMON,
+	PCF50633_IRQ_THLIMOFF,
+	PCF50633_IRQ_USBLIMON,
+	PCF50633_IRQ_USBLIMOFF,
+	PCF50633_IRQ_LOWSYS,
+	PCF50633_IRQ_LOWBAT,
+};
+
+static int __devinit pcf50633_mbc_probe(struct platform_device *pdev)
+{
+	struct pcf50633_mbc *mbc;
+	struct pcf50633_subdev_pdata *pdata = pdev->dev.platform_data;
+	int ret;
+	int i;
+	u8 mbcs1;
+
+	mbc = kzalloc(sizeof(*mbc), GFP_KERNEL);
+	if (!mbc)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, mbc);
+	mbc->pcf = pdata->pcf;
+
+	/* Set up IRQ handlers */
+	for (i = 0; i < ARRAY_SIZE(mbc_irq_handlers); i++)
+		pcf50633_register_irq(mbc->pcf, mbc_irq_handlers[i],
+					pcf50633_mbc_irq_handler, mbc);
+
+	/* Create power supplies */
+	mbc->adapter.name		= "adapter";
+	mbc->adapter.type		= POWER_SUPPLY_TYPE_MAINS;
+	mbc->adapter.properties		= power_props;
+	mbc->adapter.num_properties	= ARRAY_SIZE(power_props);
+	mbc->adapter.get_property	= &adapter_get_property;
+	mbc->adapter.supplied_to	= mbc->pcf->pdata->batteries;
+	mbc->adapter.num_supplicants	= mbc->pcf->pdata->num_batteries;
+
+	mbc->usb.name			= "usb";
+	mbc->usb.type			= POWER_SUPPLY_TYPE_USB;
+	mbc->usb.properties		= power_props;
+	mbc->usb.num_properties		= ARRAY_SIZE(power_props);
+	mbc->usb.get_property		= usb_get_property;
+	mbc->usb.supplied_to		= mbc->pcf->pdata->batteries;
+	mbc->usb.num_supplicants	= mbc->pcf->pdata->num_batteries;
+
+	ret = power_supply_register(&pdev->dev, &mbc->adapter);
+	if (ret) {
+		dev_err(mbc->pcf->dev, "failed to register adapter\n");
+		kfree(mbc);
+		return ret;
+	}
+
+	ret = power_supply_register(&pdev->dev, &mbc->usb);
+	if (ret) {
+		dev_err(mbc->pcf->dev, "failed to register usb\n");
+		power_supply_unregister(&mbc->adapter);
+		kfree(mbc);
+		return ret;
+	}
+
+	ret = sysfs_create_group(&pdev->dev.kobj, &mbc_attr_group);
+	if (ret)
+		dev_err(mbc->pcf->dev, "failed to create sysfs entries\n");
+
+	mbcs1 = pcf50633_reg_read(mbc->pcf, PCF50633_REG_MBCS1);
+	if (mbcs1 & PCF50633_MBCS1_USBPRES)
+		pcf50633_mbc_irq_handler(PCF50633_IRQ_USBINS, mbc);
+	if (mbcs1 & PCF50633_MBCS1_ADAPTPRES)
+		pcf50633_mbc_irq_handler(PCF50633_IRQ_ADPINS, mbc);
+
+	return 0;
+}
+
+static int __devexit pcf50633_mbc_remove(struct platform_device *pdev)
+{
+	struct pcf50633_mbc *mbc = platform_get_drvdata(pdev);
+	int i;
+
+	/* Remove IRQ handlers */
+	for (i = 0; i < ARRAY_SIZE(mbc_irq_handlers); i++)
+		pcf50633_free_irq(mbc->pcf, mbc_irq_handlers[i]);
+
+	power_supply_unregister(&mbc->usb);
+	power_supply_unregister(&mbc->adapter);
+
+	kfree(mbc);
+
+	return 0;
+}
+
+static struct platform_driver pcf50633_mbc_driver = {
+	.driver = {
+		.name = "pcf50633-mbc",
+	},
+	.probe = pcf50633_mbc_probe,
+	.remove = __devexit_p(pcf50633_mbc_remove),
+};
+
+static int __init pcf50633_mbc_init(void)
+{
+	return platform_driver_register(&pcf50633_mbc_driver);
+}
+module_init(pcf50633_mbc_init);
+
+static void __exit pcf50633_mbc_exit(void)
+{
+	platform_driver_unregister(&pcf50633_mbc_driver);
+}
+module_exit(pcf50633_mbc_exit);
+
+MODULE_AUTHOR("Balaji Rao <balajirrao@openmoko.org>");
+MODULE_DESCRIPTION("PCF50633 mbc driver");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:pcf50633-mbc");
diff --git a/include/linux/mfd/pcf50633/mbc.h b/include/linux/mfd/pcf50633/mbc.h
new file mode 100644
index 000000000000..6e17619b773a
--- /dev/null
+++ b/include/linux/mfd/pcf50633/mbc.h
@@ -0,0 +1,134 @@
+/*
+ * mbc.h  -- Driver for NXP PCF50633 Main Battery Charger
+ *
+ * (C) 2006-2008 by Openmoko, Inc.
+ * All rights reserved.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+#ifndef __LINUX_MFD_PCF50633_MBC_H
+#define __LINUX_MFD_PCF50633_MBC_H
+
+#include <linux/mfd/pcf50633/core.h>
+#include <linux/platform_device.h>
+
+#define PCF50633_REG_MBCC1	0x43
+#define PCF50633_REG_MBCC2	0x44
+#define PCF50633_REG_MBCC3	0x45
+#define PCF50633_REG_MBCC4	0x46
+#define PCF50633_REG_MBCC5	0x47
+#define PCF50633_REG_MBCC6	0x48
+#define PCF50633_REG_MBCC7	0x49
+#define PCF50633_REG_MBCC8	0x4a
+#define PCF50633_REG_MBCS1	0x4b
+#define PCF50633_REG_MBCS2	0x4c
+#define PCF50633_REG_MBCS3	0x4d
+
+enum pcf50633_reg_mbcc1 {
+	PCF50633_MBCC1_CHGENA		= 0x01,	/* Charger enable */
+	PCF50633_MBCC1_AUTOSTOP		= 0x02,
+	PCF50633_MBCC1_AUTORES		= 0x04, /* automatic resume */
+	PCF50633_MBCC1_RESUME		= 0x08, /* explicit resume cmd */
+	PCF50633_MBCC1_RESTART		= 0x10, /* restart charging */
+	PCF50633_MBCC1_PREWDTIME_60M	= 0x20,	/* max. precharging time */
+	PCF50633_MBCC1_WDTIME_1H	= 0x00,
+	PCF50633_MBCC1_WDTIME_2H	= 0x40,
+	PCF50633_MBCC1_WDTIME_4H	= 0x80,
+	PCF50633_MBCC1_WDTIME_6H	= 0xc0,
+};
+#define PCF50633_MBCC1_WDTIME_MASK	  0xc0
+
+enum pcf50633_reg_mbcc2 {
+	PCF50633_MBCC2_VBATCOND_2V7	= 0x00,
+	PCF50633_MBCC2_VBATCOND_2V85	= 0x01,
+	PCF50633_MBCC2_VBATCOND_3V0	= 0x02,
+	PCF50633_MBCC2_VBATCOND_3V15	= 0x03,
+	PCF50633_MBCC2_VMAX_4V		= 0x00,
+	PCF50633_MBCC2_VMAX_4V20	= 0x28,
+	PCF50633_MBCC2_VRESDEBTIME_64S	= 0x80,	/* debounce time (32/64sec) */
+};
+
+enum pcf50633_reg_mbcc7 {
+	PCF50633_MBCC7_USB_100mA	= 0x00,
+	PCF50633_MBCC7_USB_500mA	= 0x01,
+	PCF50633_MBCC7_USB_1000mA	= 0x02,
+	PCF50633_MBCC7_USB_SUSPEND	= 0x03,
+	PCF50633_MBCC7_BATTEMP_EN	= 0x04,
+	PCF50633_MBCC7_BATSYSIMAX_1A6	= 0x00,
+	PCF50633_MBCC7_BATSYSIMAX_1A8	= 0x40,
+	PCF50633_MBCC7_BATSYSIMAX_2A0	= 0x80,
+	PCF50633_MBCC7_BATSYSIMAX_2A2	= 0xc0,
+};
+#define PCF50633_MBCC7_USB_MASK 0x03
+
+enum pcf50633_reg_mbcc8 {
+	PCF50633_MBCC8_USBENASUS	= 0x10,
+};
+
+enum pcf50633_reg_mbcs1 {
+	PCF50633_MBCS1_USBPRES		= 0x01,
+	PCF50633_MBCS1_USBOK		= 0x02,
+	PCF50633_MBCS1_ADAPTPRES	= 0x04,
+	PCF50633_MBCS1_ADAPTOK		= 0x08,
+	PCF50633_MBCS1_TBAT_OK		= 0x00,
+	PCF50633_MBCS1_TBAT_ABOVE	= 0x10,
+	PCF50633_MBCS1_TBAT_BELOW	= 0x20,
+	PCF50633_MBCS1_TBAT_UNDEF	= 0x30,
+	PCF50633_MBCS1_PREWDTEXP	= 0x40,
+	PCF50633_MBCS1_WDTEXP		= 0x80,
+};
+
+enum pcf50633_reg_mbcs2_mbcmod {
+	PCF50633_MBCS2_MBC_PLAY		= 0x00,
+	PCF50633_MBCS2_MBC_USB_PRE	= 0x01,
+	PCF50633_MBCS2_MBC_USB_PRE_WAIT	= 0x02,
+	PCF50633_MBCS2_MBC_USB_FAST	= 0x03,
+	PCF50633_MBCS2_MBC_USB_FAST_WAIT = 0x04,
+	PCF50633_MBCS2_MBC_USB_SUSPEND	= 0x05,
+	PCF50633_MBCS2_MBC_ADP_PRE	= 0x06,
+	PCF50633_MBCS2_MBC_ADP_PRE_WAIT	= 0x07,
+	PCF50633_MBCS2_MBC_ADP_FAST	= 0x08,
+	PCF50633_MBCS2_MBC_ADP_FAST_WAIT = 0x09,
+	PCF50633_MBCS2_MBC_BAT_FULL	= 0x0a,
+	PCF50633_MBCS2_MBC_HALT		= 0x0b,
+};
+#define PCF50633_MBCS2_MBC_MASK		0x0f
+enum pcf50633_reg_mbcs2_chgstat {
+	PCF50633_MBCS2_CHGS_NONE	= 0x00,
+	PCF50633_MBCS2_CHGS_ADAPTER	= 0x10,
+	PCF50633_MBCS2_CHGS_USB		= 0x20,
+	PCF50633_MBCS2_CHGS_BOTH	= 0x30,
+};
+#define PCF50633_MBCS2_RESSTAT_AUTO	0x40
+
+enum pcf50633_reg_mbcs3 {
+	PCF50633_MBCS3_USBLIM_PLAY	= 0x01,
+	PCF50633_MBCS3_USBLIM_CGH	= 0x02,
+	PCF50633_MBCS3_TLIM_PLAY	= 0x04,
+	PCF50633_MBCS3_TLIM_CHG		= 0x08,
+	PCF50633_MBCS3_ILIM		= 0x10,	/* 1: Ibat > Icutoff */
+	PCF50633_MBCS3_VLIM		= 0x20,	/* 1: Vbat == Vmax */
+	PCF50633_MBCS3_VBATSTAT		= 0x40,	/* 1: Vbat > Vbatcond */
+	PCF50633_MBCS3_VRES		= 0x80, /* 1: Vbat > Vth(RES) */
+};
+
+#define PCF50633_MBCC2_VBATCOND_MASK	  0x03
+#define PCF50633_MBCC2_VMAX_MASK	  0x3c
+
+/* Charger status */
+#define PCF50633_MBC_USB_ONLINE		0x01
+#define PCF50633_MBC_USB_ACTIVE		0x02
+#define PCF50633_MBC_ADAPTER_ONLINE	0x04
+#define PCF50633_MBC_ADAPTER_ACTIVE	0x08
+
+int pcf50633_mbc_usb_curlim_set(struct pcf50633 *pcf, int ma);
+
+int pcf50633_mbc_get_status(struct pcf50633 *);
+void pcf50633_mbc_set_status(struct pcf50633 *, int what, int status);
+
+#endif
+
-- 
cgit v1.2.3


From 5ec271e745350c7df6a6ebca24b43cb7a10bfa4a Mon Sep 17 00:00:00 2001
From: Balaji Rao <balajirrao@openmoko.org>
Date: Fri, 9 Jan 2009 01:51:01 +0100
Subject: regulator: PCF50633 pmic driver

Changes from V1:
	- Removed support for suspend_enable & suspend_disable functions.

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
Cc: Andy Green <andy@openmoko.com>
Cc: Liam Girdwood <lrg@slimlogic.co.uk>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Samuel Ortiz <sameo@openedhand.com>
---
 drivers/regulator/Kconfig              |   7 +
 drivers/regulator/Makefile             |   1 +
 drivers/regulator/pcf50633-regulator.c | 329 +++++++++++++++++++++++++++++++++
 include/linux/mfd/pcf50633/pmic.h      |  67 +++++++
 4 files changed, 404 insertions(+)
 create mode 100644 drivers/regulator/pcf50633-regulator.c
 create mode 100644 include/linux/mfd/pcf50633/pmic.h

(limited to 'include/linux')

diff --git a/drivers/regulator/Kconfig b/drivers/regulator/Kconfig
index 39360e2a4540..e7e0cf102d6d 100644
--- a/drivers/regulator/Kconfig
+++ b/drivers/regulator/Kconfig
@@ -73,4 +73,11 @@ config REGULATOR_DA903X
 	  Say y here to support the BUCKs and LDOs regulators found on
 	  Dialog Semiconductor DA9030/DA9034 PMIC.
 
+config REGULATOR_PCF50633
+	tristate "PCF50633 regulator driver"
+        depends on MFD_PCF50633
+	help
+	 Say Y here to support the voltage regulators and convertors
+	 on PCF50633
+
 endif
diff --git a/drivers/regulator/Makefile b/drivers/regulator/Makefile
index 254d40c02ee8..61b30c6ddecc 100644
--- a/drivers/regulator/Makefile
+++ b/drivers/regulator/Makefile
@@ -11,5 +11,6 @@ obj-$(CONFIG_REGULATOR_BQ24022) += bq24022.o
 obj-$(CONFIG_REGULATOR_WM8350) += wm8350-regulator.o
 obj-$(CONFIG_REGULATOR_WM8400) += wm8400-regulator.o
 obj-$(CONFIG_REGULATOR_DA903X)	+= da903x.o
+obj-$(CONFIG_REGULATOR_PCF50633) += pcf50633-regulator.o
 
 ccflags-$(CONFIG_REGULATOR_DEBUG) += -DDEBUG
diff --git a/drivers/regulator/pcf50633-regulator.c b/drivers/regulator/pcf50633-regulator.c
new file mode 100644
index 000000000000..4cc85ec6e120
--- /dev/null
+++ b/drivers/regulator/pcf50633-regulator.c
@@ -0,0 +1,329 @@
+/* NXP PCF50633 PMIC Driver
+ *
+ * (C) 2006-2008 by Openmoko, Inc.
+ * Author: Balaji Rao <balajirrao@openmoko.org>
+ * All rights reserved.
+ *
+ * Broken down from monstrous PCF50633 driver mainly by
+ * Harald Welte and Andy Green and Werner Almesberger
+ *
+ *  This program is free software; you can redistribute  it and/or modify it
+ *  under  the terms of  the GNU General  Public License as published by the
+ *  Free Software Foundation;  either version 2 of the  License, or (at your
+ *  option) any later version.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/platform_device.h>
+
+#include <linux/mfd/pcf50633/core.h>
+#include <linux/mfd/pcf50633/pmic.h>
+
+#define PCF50633_REGULATOR(_name, _id) 		\
+	{					\
+		.name = _name, 			\
+		.id = _id,			\
+		.ops = &pcf50633_regulator_ops,	\
+		.type = REGULATOR_VOLTAGE, 	\
+		.owner = THIS_MODULE, 		\
+	}
+
+static const u8 pcf50633_regulator_registers[PCF50633_NUM_REGULATORS] = {
+	[PCF50633_REGULATOR_AUTO]	= PCF50633_REG_AUTOOUT,
+	[PCF50633_REGULATOR_DOWN1]	= PCF50633_REG_DOWN1OUT,
+	[PCF50633_REGULATOR_DOWN2]	= PCF50633_REG_DOWN2OUT,
+	[PCF50633_REGULATOR_MEMLDO]	= PCF50633_REG_MEMLDOOUT,
+	[PCF50633_REGULATOR_LDO1]	= PCF50633_REG_LDO1OUT,
+	[PCF50633_REGULATOR_LDO2]	= PCF50633_REG_LDO2OUT,
+	[PCF50633_REGULATOR_LDO3]	= PCF50633_REG_LDO3OUT,
+	[PCF50633_REGULATOR_LDO4]	= PCF50633_REG_LDO4OUT,
+	[PCF50633_REGULATOR_LDO5]	= PCF50633_REG_LDO5OUT,
+	[PCF50633_REGULATOR_LDO6]	= PCF50633_REG_LDO6OUT,
+	[PCF50633_REGULATOR_HCLDO]	= PCF50633_REG_HCLDOOUT,
+};
+
+/* Bits from voltage value */
+static u8 auto_voltage_bits(unsigned int millivolts)
+{
+	if (millivolts < 1800)
+		return 0;
+	if (millivolts > 3800)
+		return 0xff;
+
+	millivolts -= 625;
+
+	return millivolts / 25;
+}
+
+static u8 down_voltage_bits(unsigned int millivolts)
+{
+	if (millivolts < 625)
+		return 0;
+	else if (millivolts > 3000)
+		return 0xff;
+
+	millivolts -= 625;
+
+	return millivolts / 25;
+}
+
+static u8 ldo_voltage_bits(unsigned int millivolts)
+{
+	if (millivolts < 900)
+		return 0;
+	else if (millivolts > 3600)
+		return 0x1f;
+
+	millivolts -= 900;
+	return millivolts / 100;
+}
+
+/* Obtain voltage value from bits */
+static unsigned int auto_voltage_value(u8 bits)
+{
+	if (bits < 0x2f)
+		return 0;
+
+	return 625 + (bits * 25);
+}
+
+
+static unsigned int down_voltage_value(u8 bits)
+{
+	return 625 + (bits * 25);
+}
+
+
+static unsigned int ldo_voltage_value(u8 bits)
+{
+	bits &= 0x1f;
+
+	return 900 + (bits * 100);
+}
+
+static int pcf50633_regulator_set_voltage(struct regulator_dev *rdev,
+						int min_uV, int max_uV)
+{
+	struct pcf50633 *pcf;
+	int regulator_id, millivolts;
+	u8 volt_bits, regnr;
+
+	pcf = rdev_get_drvdata(rdev);
+
+	regulator_id = rdev_get_id(rdev);
+	if (regulator_id >= PCF50633_NUM_REGULATORS)
+		return -EINVAL;
+
+	millivolts = min_uV / 1000;
+
+	regnr = pcf50633_regulator_registers[regulator_id];
+
+	switch (regulator_id) {
+	case PCF50633_REGULATOR_AUTO:
+		volt_bits = auto_voltage_bits(millivolts);
+		break;
+	case PCF50633_REGULATOR_DOWN1:
+		volt_bits = down_voltage_bits(millivolts);
+		break;
+	case PCF50633_REGULATOR_DOWN2:
+		volt_bits = down_voltage_bits(millivolts);
+		break;
+	case PCF50633_REGULATOR_LDO1:
+	case PCF50633_REGULATOR_LDO2:
+	case PCF50633_REGULATOR_LDO3:
+	case PCF50633_REGULATOR_LDO4:
+	case PCF50633_REGULATOR_LDO5:
+	case PCF50633_REGULATOR_LDO6:
+	case PCF50633_REGULATOR_HCLDO:
+		volt_bits = ldo_voltage_bits(millivolts);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return pcf50633_reg_write(pcf, regnr, volt_bits);
+}
+
+static int pcf50633_regulator_get_voltage(struct regulator_dev *rdev)
+{
+	struct pcf50633 *pcf;
+	int regulator_id, millivolts, volt_bits;
+	u8 regnr;
+
+	pcf = rdev_get_drvdata(rdev);;
+
+	regulator_id = rdev_get_id(rdev);
+	if (regulator_id >= PCF50633_NUM_REGULATORS)
+		return -EINVAL;
+
+	regnr = pcf50633_regulator_registers[regulator_id];
+
+	volt_bits = pcf50633_reg_read(pcf, regnr);
+	if (volt_bits < 0)
+		return -1;
+
+	switch (regulator_id) {
+	case PCF50633_REGULATOR_AUTO:
+		millivolts = auto_voltage_value(volt_bits);
+		break;
+	case PCF50633_REGULATOR_DOWN1:
+		millivolts = down_voltage_value(volt_bits);
+		break;
+	case PCF50633_REGULATOR_DOWN2:
+		millivolts = down_voltage_value(volt_bits);
+		break;
+	case PCF50633_REGULATOR_LDO1:
+	case PCF50633_REGULATOR_LDO2:
+	case PCF50633_REGULATOR_LDO3:
+	case PCF50633_REGULATOR_LDO4:
+	case PCF50633_REGULATOR_LDO5:
+	case PCF50633_REGULATOR_LDO6:
+	case PCF50633_REGULATOR_HCLDO:
+		millivolts = ldo_voltage_value(volt_bits);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return millivolts * 1000;
+}
+
+static int pcf50633_regulator_enable(struct regulator_dev *rdev)
+{
+	struct pcf50633 *pcf = rdev_get_drvdata(rdev);
+	int regulator_id;
+	u8 regnr;
+
+	regulator_id = rdev_get_id(rdev);
+	if (regulator_id >= PCF50633_NUM_REGULATORS)
+		return -EINVAL;
+
+	/* The *ENA register is always one after the *OUT register */
+	regnr = pcf50633_regulator_registers[regulator_id] + 1;
+
+	return pcf50633_reg_set_bit_mask(pcf, regnr, PCF50633_REGULATOR_ON,
+						       PCF50633_REGULATOR_ON);
+}
+
+static int pcf50633_regulator_disable(struct regulator_dev *rdev)
+{
+	struct pcf50633 *pcf = rdev_get_drvdata(rdev);
+	int regulator_id;
+	u8 regnr;
+
+	regulator_id = rdev_get_id(rdev);
+	if (regulator_id >= PCF50633_NUM_REGULATORS)
+		return -EINVAL;
+
+	/* the *ENA register is always one after the *OUT register */
+	regnr = pcf50633_regulator_registers[regulator_id] + 1;
+
+	return pcf50633_reg_set_bit_mask(pcf, regnr,
+					PCF50633_REGULATOR_ON, 0);
+}
+
+static int pcf50633_regulator_is_enabled(struct regulator_dev *rdev)
+{
+	struct pcf50633 *pcf = rdev_get_drvdata(rdev);
+	int regulator_id = rdev_get_id(rdev);
+	u8 regnr;
+
+	regulator_id = rdev_get_id(rdev);
+	if (regulator_id >= PCF50633_NUM_REGULATORS)
+		return -EINVAL;
+
+	/* the *ENA register is always one after the *OUT register */
+	regnr = pcf50633_regulator_registers[regulator_id] + 1;
+
+	return pcf50633_reg_read(pcf, regnr) & PCF50633_REGULATOR_ON;
+}
+
+static struct regulator_ops pcf50633_regulator_ops = {
+	.set_voltage = pcf50633_regulator_set_voltage,
+	.get_voltage = pcf50633_regulator_get_voltage,
+	.enable = pcf50633_regulator_enable,
+	.disable = pcf50633_regulator_disable,
+	.is_enabled = pcf50633_regulator_is_enabled,
+};
+
+static struct regulator_desc regulators[] = {
+	[PCF50633_REGULATOR_AUTO] =
+		PCF50633_REGULATOR("auto", PCF50633_REGULATOR_AUTO),
+	[PCF50633_REGULATOR_DOWN1] =
+		PCF50633_REGULATOR("down1", PCF50633_REGULATOR_DOWN1),
+	[PCF50633_REGULATOR_DOWN2] =
+		PCF50633_REGULATOR("down2", PCF50633_REGULATOR_DOWN2),
+	[PCF50633_REGULATOR_LDO1] =
+		PCF50633_REGULATOR("ldo1", PCF50633_REGULATOR_LDO1),
+	[PCF50633_REGULATOR_LDO2] =
+		PCF50633_REGULATOR("ldo2", PCF50633_REGULATOR_LDO2),
+	[PCF50633_REGULATOR_LDO3] =
+		PCF50633_REGULATOR("ldo3", PCF50633_REGULATOR_LDO3),
+	[PCF50633_REGULATOR_LDO4] =
+		PCF50633_REGULATOR("ldo4", PCF50633_REGULATOR_LDO4),
+	[PCF50633_REGULATOR_LDO5] =
+		PCF50633_REGULATOR("ldo5", PCF50633_REGULATOR_LDO5),
+	[PCF50633_REGULATOR_LDO6] =
+		PCF50633_REGULATOR("ldo6", PCF50633_REGULATOR_LDO6),
+	[PCF50633_REGULATOR_HCLDO] =
+		PCF50633_REGULATOR("hcldo", PCF50633_REGULATOR_HCLDO),
+	[PCF50633_REGULATOR_MEMLDO] =
+		PCF50633_REGULATOR("memldo", PCF50633_REGULATOR_MEMLDO),
+};
+
+static int __devinit pcf50633_regulator_probe(struct platform_device *pdev)
+{
+	struct regulator_dev *rdev;
+	struct pcf50633 *pcf;
+
+	/* Already set by core driver */
+	pcf = platform_get_drvdata(pdev);
+
+	rdev = regulator_register(&regulators[pdev->id], &pdev->dev, pcf);
+	if (IS_ERR(rdev))
+		return PTR_ERR(rdev);
+
+	if (pcf->pdata->regulator_registered)
+		pcf->pdata->regulator_registered(pcf, pdev->id);
+
+	return 0;
+}
+
+static int __devexit pcf50633_regulator_remove(struct platform_device *pdev)
+{
+	struct regulator_dev *rdev = platform_get_drvdata(pdev);
+
+	regulator_unregister(rdev);
+
+	return 0;
+}
+
+static struct platform_driver pcf50633_regulator_driver = {
+	.driver = {
+		.name = "pcf50633-regltr",
+	},
+	.probe = pcf50633_regulator_probe,
+	.remove = __devexit_p(pcf50633_regulator_remove),
+};
+
+static int __init pcf50633_regulator_init(void)
+{
+	return platform_driver_register(&pcf50633_regulator_driver);
+}
+module_init(pcf50633_regulator_init);
+
+static void __exit pcf50633_regulator_exit(void)
+{
+	platform_driver_unregister(&pcf50633_regulator_driver);
+}
+module_exit(pcf50633_regulator_exit);
+
+MODULE_AUTHOR("Balaji Rao <balajirrao@openmoko.org>");
+MODULE_DESCRIPTION("PCF50633 regulator driver");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:pcf50633-regulator");
diff --git a/include/linux/mfd/pcf50633/pmic.h b/include/linux/mfd/pcf50633/pmic.h
new file mode 100644
index 000000000000..2d3dbe53b235
--- /dev/null
+++ b/include/linux/mfd/pcf50633/pmic.h
@@ -0,0 +1,67 @@
+#ifndef __LINUX_MFD_PCF50633_PMIC_H
+#define __LINUX_MFD_PCF50633_PMIC_H
+
+#include <linux/mfd/pcf50633/core.h>
+#include <linux/platform_device.h>
+
+#define PCF50633_REG_AUTOOUT	0x1a
+#define PCF50633_REG_AUTOENA	0x1b
+#define PCF50633_REG_AUTOCTL	0x1c
+#define PCF50633_REG_AUTOMXC	0x1d
+#define PCF50633_REG_DOWN1OUT	0x1e
+#define PCF50633_REG_DOWN1ENA	0x1f
+#define PCF50633_REG_DOWN1CTL	0x20
+#define PCF50633_REG_DOWN1MXC	0x21
+#define PCF50633_REG_DOWN2OUT	0x22
+#define PCF50633_REG_DOWN2ENA	0x23
+#define PCF50633_REG_DOWN2CTL	0x24
+#define PCF50633_REG_DOWN2MXC	0x25
+#define PCF50633_REG_MEMLDOOUT	0x26
+#define PCF50633_REG_MEMLDOENA	0x27
+#define PCF50633_REG_LDO1OUT	0x2d
+#define PCF50633_REG_LDO1ENA	0x2e
+#define PCF50633_REG_LDO2OUT	0x2f
+#define PCF50633_REG_LDO2ENA	0x30
+#define PCF50633_REG_LDO3OUT	0x31
+#define PCF50633_REG_LDO3ENA	0x32
+#define PCF50633_REG_LDO4OUT	0x33
+#define PCF50633_REG_LDO4ENA	0x34
+#define PCF50633_REG_LDO5OUT	0x35
+#define PCF50633_REG_LDO5ENA	0x36
+#define PCF50633_REG_LDO6OUT	0x37
+#define PCF50633_REG_LDO6ENA	0x38
+#define PCF50633_REG_HCLDOOUT	0x39
+#define PCF50633_REG_HCLDOENA	0x3a
+#define PCF50633_REG_HCLDOOVL	0x40
+
+enum pcf50633_regulator_enable {
+	PCF50633_REGULATOR_ON		= 0x01,
+	PCF50633_REGULATOR_ON_GPIO1	= 0x02,
+	PCF50633_REGULATOR_ON_GPIO2	= 0x04,
+	PCF50633_REGULATOR_ON_GPIO3	= 0x08,
+};
+#define PCF50633_REGULATOR_ON_MASK	0x0f
+
+enum pcf50633_regulator_phase {
+	PCF50633_REGULATOR_ACTPH1	= 0x00,
+	PCF50633_REGULATOR_ACTPH2	= 0x10,
+	PCF50633_REGULATOR_ACTPH3	= 0x20,
+	PCF50633_REGULATOR_ACTPH4	= 0x30,
+};
+#define PCF50633_REGULATOR_ACTPH_MASK	0x30
+
+enum pcf50633_regulator_id {
+	PCF50633_REGULATOR_AUTO,
+	PCF50633_REGULATOR_DOWN1,
+	PCF50633_REGULATOR_DOWN2,
+	PCF50633_REGULATOR_LDO1,
+	PCF50633_REGULATOR_LDO2,
+	PCF50633_REGULATOR_LDO3,
+	PCF50633_REGULATOR_LDO4,
+	PCF50633_REGULATOR_LDO5,
+	PCF50633_REGULATOR_LDO6,
+	PCF50633_REGULATOR_HCLDO,
+	PCF50633_REGULATOR_MEMLDO,
+};
+#endif
+
-- 
cgit v1.2.3


From 53ce3d9564908794ae7dd32969089b57df5fc098 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Fri, 9 Jan 2009 12:27:08 -0800
Subject: smp_call_function_single(): be slightly less stupid

If you do

	smp_call_function_single(expression-with-side-effects, ...)

then expression-with-side-effects never gets evaluated on UP builds.

As always, implementing it in C is the correct thing to do.

While we're there, uninline it for size and possible header dependency
reasons.

And create a new kernel/up.c, as a place in which to put
uniprocessor-specific code and storage.  It should mirror kernel/smp.c.

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/smp.h | 13 +++----------
 kernel/Makefile     |  6 +++++-
 kernel/up.c         | 18 ++++++++++++++++++
 3 files changed, 26 insertions(+), 11 deletions(-)
 create mode 100644 kernel/up.c

(limited to 'include/linux')

diff --git a/include/linux/smp.h b/include/linux/smp.h
index b82466968101..715196b09d67 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -24,6 +24,9 @@ struct call_single_data {
 /* total number of cpus in this system (may exceed NR_CPUS) */
 extern unsigned int total_cpus;
 
+int smp_call_function_single(int cpuid, void (*func) (void *info), void *info,
+				int wait);
+
 #ifdef CONFIG_SMP
 
 #include <linux/preempt.h>
@@ -79,8 +82,6 @@ smp_call_function_mask(cpumask_t mask, void(*func)(void *info), void *info,
 	return 0;
 }
 
-int smp_call_function_single(int cpuid, void (*func) (void *info), void *info,
-				int wait);
 void __smp_call_function_single(int cpuid, struct call_single_data *data);
 
 /*
@@ -140,14 +141,6 @@ static inline int up_smp_call_function(void (*func)(void *), void *info)
 static inline void smp_send_reschedule(int cpu) { }
 #define num_booting_cpus()			1
 #define smp_prepare_boot_cpu()			do {} while (0)
-#define smp_call_function_single(cpuid, func, info, wait) \
-({ \
-	WARN_ON(cpuid != 0);	\
-	local_irq_disable();	\
-	(func)(info);		\
-	local_irq_enable();	\
-	0;			\
-})
 #define smp_call_function_mask(mask, func, info, wait) \
 			(up_smp_call_function(func, info))
 #define smp_call_function_many(mask, func, info, wait) \
diff --git a/kernel/Makefile b/kernel/Makefile
index 2921d90ce32f..2aebc4cd7878 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -40,7 +40,11 @@ obj-$(CONFIG_RT_MUTEXES) += rtmutex.o
 obj-$(CONFIG_DEBUG_RT_MUTEXES) += rtmutex-debug.o
 obj-$(CONFIG_RT_MUTEX_TESTER) += rtmutex-tester.o
 obj-$(CONFIG_GENERIC_ISA_DMA) += dma.o
-obj-$(CONFIG_USE_GENERIC_SMP_HELPERS) += smp.o
+ifeq ($(CONFIG_USE_GENERIC_SMP_HELPERS),y)
+obj-y += smp.o
+else
+obj-y += up.o
+endif
 obj-$(CONFIG_SMP) += spinlock.o
 obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock.o
 obj-$(CONFIG_PROVE_LOCKING) += spinlock.o
diff --git a/kernel/up.c b/kernel/up.c
new file mode 100644
index 000000000000..ce62cc9e9f71
--- /dev/null
+++ b/kernel/up.c
@@ -0,0 +1,18 @@
+/*
+ * Uniprocessor-only support functions.  The counterpart to kernel/smp.c
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/smp.h>
+
+int smp_call_function_single(int cpu, void (*func) (void *info), void *info,
+				int wait)
+{
+	WARN_ON(cpuid != 0);
+	local_irq_disable();
+	(func)(info);
+	local_irq_enable();
+	return 0;
+}
+EXPORT_SYMBOL(smp_call_function_single);
-- 
cgit v1.2.3


From 649274d993212e7c23c0cb734572c2311c200872 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Sun, 11 Jan 2009 00:20:39 -0800
Subject: net_dma: acquire/release dma channels on ifup/ifdown

The recent dmaengine rework removed the capability to remove dma device
driver modules while net_dma is active.  Rather than notify
dmaengine-clients that channels are trying to be removed, we now rely on
clients to notify dmaengine when they no longer have a need for
channels.  Teach net_dma to release channels by taking dmaengine
references at netdevice open and dropping references at netdevice close.

Acked-by: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/dmaengine.h | 10 ++++++++++
 net/core/dev.c            | 13 ++++++++++---
 2 files changed, 20 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 64dea2ab326c..c73f1e2b59b7 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -270,8 +270,18 @@ struct dma_device {
 
 /* --- public DMA engine API --- */
 
+#ifdef CONFIG_DMA_ENGINE
 void dmaengine_get(void);
 void dmaengine_put(void);
+#else
+static inline void dmaengine_get(void)
+{
+}
+static inline void dmaengine_put(void)
+{
+}
+#endif
+
 dma_cookie_t dma_async_memcpy_buf_to_buf(struct dma_chan *chan,
 	void *dest, void *src, size_t len);
 dma_cookie_t dma_async_memcpy_buf_to_pg(struct dma_chan *chan,
diff --git a/net/core/dev.c b/net/core/dev.c
index 5f736f1ceeae..b715a55cccc4 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1087,6 +1087,11 @@ int dev_open(struct net_device *dev)
 		 */
 		dev->flags |= IFF_UP;
 
+		/*
+		 *	Enable NET_DMA
+		 */
+		dmaengine_get();
+
 		/*
 		 *	Initialize multicasting status
 		 */
@@ -1164,6 +1169,11 @@ int dev_close(struct net_device *dev)
 	 */
 	call_netdevice_notifiers(NETDEV_DOWN, dev);
 
+	/*
+	 *	Shutdown NET_DMA
+	 */
+	dmaengine_put();
+
 	return 0;
 }
 
@@ -5151,9 +5161,6 @@ static int __init net_dev_init(void)
 	hotcpu_notifier(dev_cpu_callback, 0);
 	dst_init();
 	dev_mcast_init();
-	#ifdef CONFIG_NET_DMA
-	dmaengine_get();
-	#endif
 	rc = 0;
 out:
 	return rc;
-- 
cgit v1.2.3


From 57de16e612d63138bd2c618449af9d8312466e25 Mon Sep 17 00:00:00 2001
From: Martin Bachem <m.bachem@gmx.de>
Date: Sun, 26 Oct 2008 13:30:09 +0100
Subject: BUGFIX: used NULL pointer at ioctl(sk,IMGETDEVINFO,&devinfo) when
 devinfo.id not registered

daxtar example # modprobe hfcsusb
daxtar example # modprobe mISDN_l1loop
daxtar example # ./misdnportinfo
Found 3 devices
        id:             0
        Dprotocols:     00000006
        Bprotocols:     0000000e
        protocol:       0
        nrbchan:        2
        name:           HFC-S_USB.1
        id:             1
        Dprotocols:     00000006
        Bprotocols:     0000000e
        protocol:       0
        nrbchan:        2
        name:           mISDN_l1loop.1
        id:             2
        Dprotocols:     00000006
        Bprotocols:     0000000e
        protocol:       0
        nrbchan:        2
        name:           mISDN_l1loop.2
daxtar example # rmmod hfcsusb
daxtar example # ./misdnportinfo
Found 2 devices
*Segmentation* *fault*

dmesg:

[ 9914.939718] BUG: unable to handle kernel NULL pointer dereference at 000000d4
[ 9914.939721] IP: [<f8f9f2dd>] :mISDN_core:get_mdevice+0x19/0x22
[ 9914.939729] *pde = 00000000
[ 9914.939732] Oops: 0000 [#14] PREEMPT SMP
[ 9914.939734] Modules linked in: mISDN_l1loop mISDN_core vmnet vmblock vmci vmmon coretemp w83627ehf hwmon_vid rfcomm l2cap blue
tooth usbhid snd_usb_audio snd_usb_lib snd_rawmidi snd_hwdep fuse nvidia(P) uhci_hcd i2c_i801 ehci_hcd snd_hda_intel atl1 usbcore i2c_core parport_seria
l [last unloaded: hfcsusb]
[ 9914.939751] Pid: 29618, comm: misdnportinfo Tainted: P      D   (2.6.27.3 #5)
[ 9914.939753] EIP: 0060:[<f8f9f2dd>] EFLAGS: 00210246 CPU: 0
[ 9914.939758] EIP is at get_mdevice+0x19/0x22 [mISDN_core]
[ 9914.939760] EAX: 00000000 EBX: f8fa791c ECX: f6afaa58 EDX: f7960cf4
[ 9914.939762] ESI: 80044944 EDI: bfc2e62c EBP: bfc2e62c ESP: f5adbef4
[ 9914.939763]  DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
[ 9914.939765] Process misdnportinfo (pid: 29618, ti=f5ada000 task=f6bec430 task.ti=f5ada000)
[ 9914.939767] Stack: f8f9f4e0 00000000 f8f9f867 bfc2e62c 0000000a c02461e8 00200246 c042dde8
[ 9914.939771]        00000003 c042dde4 00000000 00000001 00200082 c0114775 00000000 00000000
[ 9914.939775]        00000003 f7088010 00200282 f8fa791c 80044944 bfc2e62c bfc2e62c c02f6615
[ 9914.939780] Call Trace:
[ 9914.939782]  [<f8f9f4e0>] _get_mdevice+0x0/0x18 [mISDN_core]
[ 9914.939789]  [<f8f9f867>] base_sock_ioctl+0x7a/0x129 [mISDN_core]
[ 9914.939789]  [<c02461e8>] opost+0x171/0x182
[ 9914.939789]  [<c0114775>] __wake_up+0x29/0x39
[ 9914.939789]  [<c02f6615>] sock_ioctl+0x1b5/0x1d9
[ 9914.939789]  [<c02f6460>] sock_ioctl+0x0/0x1d9
[ 9914.939789]  [<c016794c>] vfs_ioctl+0x1c/0x5d
[ 9914.939789]  [<c0167bcb>] do_vfs_ioctl+0x23e/0x24e
[ 9914.939789]  [<c0167c07>] sys_ioctl+0x2c/0x45
[ 9914.939789]  [<c0102cbd>] sysenter_do_call+0x12/0x21
[ 9914.939789]  [<c0350000>] pci_fixup_i450gx+0x4e/0x56
[ 9914.939789]  =======================
[ 9914.939789] Code: 00 68 02 f0 f9 f8 e8 ae b4 2c c7 8b 44 24 04 5a 59 c3 83 ec 04 31 d2 89 04 24 89 e1 b8 ac df fa f8 68 e0 f4
f9 f8 e8 4a b5 2c c7 <8b> 80 d4 00 00 00 5a 59 c3 53 89 cb 8d 90 9c 00 00 00 89 c8 e8
[ 9914.939789] EIP: [<f8f9f2dd>] get_mdevice+0x19/0x22 [mISDN_core] SS:ESP 0068:f5adbef4
[ 9914.939858] ---[ end trace 50e18a715b019424 ]---

Signed-off-by: Martin Bachem <m.bachem@gmx.de>
Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 include/linux/mISDNif.h | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/mISDNif.h b/include/linux/mISDNif.h
index 557477ac3d5b..5da3d95b27f1 100644
--- a/include/linux/mISDNif.h
+++ b/include/linux/mISDNif.h
@@ -559,7 +559,10 @@ extern void	mISDN_unregister_clock(struct mISDNclock *);
 
 static inline struct mISDNdevice *dev_to_mISDN(struct device *dev)
 {
-	return dev_get_drvdata(dev);
+	if (dev)
+		return dev_get_drvdata(dev);
+	else
+		return NULL;
 }
 
 extern void	set_channel_address(struct mISDNchannel *, u_int, u_int);
-- 
cgit v1.2.3


From 6d612b0f943289856c6e8186c564cda922cd040e Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <peterz@infradead.org>
Date: Mon, 12 Jan 2009 12:52:23 +0100
Subject: locking, hpet: annotate false positive warning

Alexander Beregalov reported that this warning is caused by the HPET code:

> hpet0: at MMIO 0xfed00000, IRQs 2, 8, 0
> hpet0: 3 comparators, 64-bit 14.318180 MHz counter
> ODEBUG: object is on stack, but not annotated
> ------------[ cut here ]------------
> WARNING: at lib/debugobjects.c:251 __debug_object_init+0x2a4/0x352()

> Bisected down to 26afe5f2fbf06ea0765aaa316640c4dd472310c0
> (x86: HPET_MSI Initialise per-cpu HPET timers)

The commit is fine - but the on-stack workqueue entry needs annotation.

Reported-and-bisected-by: Alexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Tested-by: Alexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/hpet.c    | 2 +-
 include/linux/workqueue.h | 6 ++++++
 2 files changed, 7 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index cd759ad90690..bb2e0f0975ae 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -628,7 +628,7 @@ static int hpet_cpuhp_notify(struct notifier_block *n,
 
 	switch (action & 0xf) {
 	case CPU_ONLINE:
-		INIT_DELAYED_WORK(&work.work, hpet_work);
+		INIT_DELAYED_WORK_ON_STACK(&work.work, hpet_work);
 		init_completion(&work.complete);
 		/* FIXME: add schedule_work_on() */
 		schedule_delayed_work_on(cpu, &work.work, 0);
diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index b36291130f22..47151c8495aa 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -118,6 +118,12 @@ struct execute_work {
 		init_timer(&(_work)->timer);			\
 	} while (0)
 
+#define INIT_DELAYED_WORK_ON_STACK(_work, _func)		\
+	do {							\
+		INIT_WORK(&(_work)->work, (_func));		\
+		init_timer_on_stack(&(_work)->timer);		\
+	} while (0)
+
 #define INIT_DELAYED_WORK_DEFERRABLE(_work, _func)			\
 	do {							\
 		INIT_WORK(&(_work)->work, (_func));		\
-- 
cgit v1.2.3


From 2e4c77bea3d8b17d94f8ee382411f359b708560f Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert@linux-m68k.org>
Date: Tue, 30 Dec 2008 14:16:41 +0100
Subject: m68k: dio - Kill warn_unused_result warnings

warning: ignoring return value of 'device_register', declared with attribute
warn_unused_result
warning: ignoring return value of 'device_create_file', declared with
attribute warn_unused_result

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
---
 drivers/dio/dio-sysfs.c | 16 ++++++++++------
 drivers/dio/dio.c       | 18 +++++++++++++++---
 include/linux/dio.h     |  2 +-
 3 files changed, 26 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/dio/dio-sysfs.c b/drivers/dio/dio-sysfs.c
index f46463038847..91d5f4da8769 100644
--- a/drivers/dio/dio-sysfs.c
+++ b/drivers/dio/dio-sysfs.c
@@ -63,15 +63,19 @@ static ssize_t dio_show_resource(struct device *dev, struct device_attribute *at
 }
 static DEVICE_ATTR(resource, S_IRUGO, dio_show_resource, NULL);
 
-void dio_create_sysfs_dev_files(struct dio_dev *d)
+int dio_create_sysfs_dev_files(struct dio_dev *d)
 {
 	struct device *dev = &d->dev;
+	int error;
 
 	/* current configuration's attributes */
-	device_create_file(dev, &dev_attr_id);
-	device_create_file(dev, &dev_attr_ipl);
-	device_create_file(dev, &dev_attr_secid);
-	device_create_file(dev, &dev_attr_name);
-	device_create_file(dev, &dev_attr_resource);
+	if ((error = device_create_file(dev, &dev_attr_id)) ||
+	    (error = device_create_file(dev, &dev_attr_ipl)) ||
+	    (error = device_create_file(dev, &dev_attr_secid)) ||
+	    (error = device_create_file(dev, &dev_attr_name)) ||
+	    (error = device_create_file(dev, &dev_attr_resource)))
+		return error;
+
+	return 0;
 }
 
diff --git a/drivers/dio/dio.c b/drivers/dio/dio.c
index 07f274f853d9..10c3c498358c 100644
--- a/drivers/dio/dio.c
+++ b/drivers/dio/dio.c
@@ -173,6 +173,7 @@ static int __init dio_init(void)
 	mm_segment_t fs;
 	int i;
 	struct dio_dev *dev;
+	int error;
 
 	if (!MACH_IS_HP300)
 		return 0;
@@ -182,7 +183,11 @@ static int __init dio_init(void)
 	/* Initialize the DIO bus */ 
 	INIT_LIST_HEAD(&dio_bus.devices);
 	strcpy(dio_bus.dev.bus_id, "dio");
-	device_register(&dio_bus.dev);
+	error = device_register(&dio_bus.dev);
+	if (error) {
+		pr_err("DIO: Error registering dio_bus\n");
+		return error;
+	}
 
 	/* Request all resources */
 	dio_bus.num_resources = (hp300_model == HP_320 ? 1 : 2);
@@ -252,8 +257,15 @@ static int __init dio_init(void)
 
 		if (scode >= DIOII_SCBASE)
 			iounmap(va);
-		device_register(&dev->dev);
-		dio_create_sysfs_dev_files(dev);
+		error = device_register(&dev->dev);
+		if (error) {
+			pr_err("DIO: Error registering device %s\n",
+			       dev->name);
+			continue;
+		}
+		error = dio_create_sysfs_dev_files(dev);
+		if (error)
+			dev_err(&dev->dev, "Error creating sysfs files\n");
         }
 	return 0;
 }
diff --git a/include/linux/dio.h b/include/linux/dio.h
index 1e65ebc2a3db..b2dd31ca1710 100644
--- a/include/linux/dio.h
+++ b/include/linux/dio.h
@@ -241,7 +241,7 @@ struct dio_driver {
 
 extern int dio_find(int deviceid);
 extern unsigned long dio_scodetophysaddr(int scode);
-extern void dio_create_sysfs_dev_files(struct dio_dev *);
+extern int dio_create_sysfs_dev_files(struct dio_dev *);
 
 /* New-style probing */
 extern int dio_register_driver(struct dio_driver *);
-- 
cgit v1.2.3


From 985ebdb5ed54151eba734aa1b307460e8e4267ba Mon Sep 17 00:00:00 2001
From: Krzysztof Hałasa <khc@pm.waw.pl>
Date: Mon, 12 Jan 2009 16:32:13 -0800
Subject: net: Fix a comment in include/linux/netdevice.h.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix a comment in include/linux/netdevice.h.

Signed-off-by: Krzysztof Hałasa <khc@pm.waw.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index f24556813375..4647604c7ca9 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -467,7 +467,7 @@ struct netdev_queue {
  *     This function is called when network device transistions to the down
  *     state.
  *
- * int (*ndo_hard_start_xmit)(struct sk_buff *skb, struct net_device *dev);
+ * int (*ndo_start_xmit)(struct sk_buff *skb, struct net_device *dev);
  *	Called when a packet needs to be transmitted.
  *	Must return NETDEV_TX_OK , NETDEV_TX_BUSY, or NETDEV_TX_LOCKED,
  *	Required can not be NULL.
-- 
cgit v1.2.3


From daaf83d2b9277928739f3eb7ea64f49c1254fd62 Mon Sep 17 00:00:00 2001
From: Richard Kennedy <richard@rsk.demon.co.uk>
Date: Mon, 12 Jan 2009 00:06:11 +0000
Subject: netfilter 09/09: remove padding from struct xt_match on 64bit builds

reorder struct xt_match to remove 8 bytes of padding and make its size
128 bytes.

This saves a small amount of data space in each of the xt netfilter
modules and fits xt_match in one 128 byte cache line.

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/x_tables.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index e52ce475d19f..c7ee8744d26b 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -270,6 +270,7 @@ struct xt_match
 	struct list_head list;
 
 	const char name[XT_FUNCTION_MAXNAMELEN-1];
+	u_int8_t revision;
 
 	/* Return true or false: return FALSE and set *hotdrop = 1 to
            force immediate packet drop. */
@@ -302,7 +303,6 @@ struct xt_match
 	unsigned short proto;
 
 	unsigned short family;
-	u_int8_t revision;
 };
 
 /* Registration hooks for targets. */
-- 
cgit v1.2.3


From baf48f6577e581a9adb8fe849dc80e24b21d171d Mon Sep 17 00:00:00 2001
From: Mandeep Singh Baines <msb@google.com>
Date: Mon, 12 Jan 2009 21:15:17 -0800
Subject: softlock: fix false panic which can occur if softlockup_thresh is
 reduced

At run-time, if softlockup_thresh is changed to a much lower value,
touch_timestamp is likely to be much older than the new softlock_thresh.

This will cause a false softlockup to be detected. If softlockup_panic
is enabled, the system will panic.

The fix is to touch all watchdogs before changing softlockup_thresh.

Signed-off-by: Mandeep Singh Baines <msb@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h | 3 +++
 kernel/softlockup.c   | 9 +++++++++
 kernel/sysctl.c       | 2 +-
 3 files changed, 13 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4cae9b81a1f8..54cbabf3b871 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -293,6 +293,9 @@ extern void sched_show_task(struct task_struct *p);
 extern void softlockup_tick(void);
 extern void touch_softlockup_watchdog(void);
 extern void touch_all_softlockup_watchdogs(void);
+extern int proc_dosoftlockup_thresh(struct ctl_table *table, int write,
+				    struct file *filp, void __user *buffer,
+				    size_t *lenp, loff_t *ppos);
 extern unsigned int  softlockup_panic;
 extern unsigned long sysctl_hung_task_check_count;
 extern unsigned long sysctl_hung_task_timeout_secs;
diff --git a/kernel/softlockup.c b/kernel/softlockup.c
index d9188c66278a..85d5a2455103 100644
--- a/kernel/softlockup.c
+++ b/kernel/softlockup.c
@@ -16,6 +16,7 @@
 #include <linux/lockdep.h>
 #include <linux/notifier.h>
 #include <linux/module.h>
+#include <linux/sysctl.h>
 
 #include <asm/irq_regs.h>
 
@@ -88,6 +89,14 @@ void touch_all_softlockup_watchdogs(void)
 }
 EXPORT_SYMBOL(touch_all_softlockup_watchdogs);
 
+int proc_dosoftlockup_thresh(struct ctl_table *table, int write,
+			     struct file *filp, void __user *buffer,
+			     size_t *lenp, loff_t *ppos)
+{
+	touch_all_softlockup_watchdogs();
+	return proc_dointvec_minmax(table, write, filp, buffer, lenp, ppos);
+}
+
 /*
  * This callback runs from the timer interrupt, and checks
  * whether the watchdog thread has hung or not:
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 89d74436318c..596dc31a7116 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -800,7 +800,7 @@ static struct ctl_table kern_table[] = {
 		.data		= &softlockup_thresh,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_minmax,
+		.proc_handler	= &proc_dosoftlockup_thresh,
 		.strategy	= &sysctl_intvec,
 		.extra1		= &neg_one,
 		.extra2		= &sixty,
-- 
cgit v1.2.3


From 4c696ba7982501d43dea11dbbaabd2aa8a19cc42 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Wed, 14 Jan 2009 14:13:53 +0100
Subject: [CVE-2009-0029] Move compat system call declarations to compat header
 file

Move declarations to correct header file.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 include/linux/compat.h   | 13 +++++++++++++
 include/linux/syscalls.h | 12 ------------
 2 files changed, 13 insertions(+), 12 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/compat.h b/include/linux/compat.h
index e88f3ecf38b4..3fd2194ff573 100644
--- a/include/linux/compat.h
+++ b/include/linux/compat.h
@@ -280,5 +280,18 @@ asmlinkage long compat_sys_timerfd_settime(int ufd, int flags,
 asmlinkage long compat_sys_timerfd_gettime(int ufd,
 				   struct compat_itimerspec __user *otmr);
 
+asmlinkage long compat_sys_move_pages(pid_t pid, unsigned long nr_page,
+				      __u32 __user *pages,
+				      const int __user *nodes,
+				      int __user *status,
+				      int flags);
+asmlinkage long compat_sys_futimesat(unsigned int dfd, char __user *filename,
+				     struct compat_timeval __user *t);
+asmlinkage long compat_sys_newfstatat(unsigned int dfd, char __user * filename,
+				      struct compat_stat __user *statbuf,
+				      int flag);
+asmlinkage long compat_sys_openat(unsigned int dfd, const char __user *filename,
+				  int flags, int mode);
+
 #endif /* CONFIG_COMPAT */
 #endif /* _LINUX_COMPAT_H */
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 18d0a243a7b3..a7593f670ca6 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -530,11 +530,6 @@ asmlinkage long sys_move_pages(pid_t pid, unsigned long nr_pages,
 				const int __user *nodes,
 				int __user *status,
 				int flags);
-asmlinkage long compat_sys_move_pages(pid_t pid, unsigned long nr_page,
-				__u32 __user *pages,
-				const int __user *nodes,
-				int __user *status,
-				int flags);
 asmlinkage long sys_mbind(unsigned long start, unsigned long len,
 				unsigned long mode,
 				unsigned long __user *nmask,
@@ -583,13 +578,6 @@ asmlinkage long sys_readlinkat(int dfd, const char __user *path, char __user *bu
 			       int bufsiz);
 asmlinkage long sys_utimensat(int dfd, char __user *filename,
 				struct timespec __user *utimes, int flags);
-asmlinkage long compat_sys_futimesat(unsigned int dfd, char __user *filename,
-				     struct compat_timeval __user *t);
-asmlinkage long compat_sys_newfstatat(unsigned int dfd, char __user * filename,
-				      struct compat_stat __user *statbuf,
-				      int flag);
-asmlinkage long compat_sys_openat(unsigned int dfd, const char __user *filename,
-				   int flags, int mode);
 asmlinkage long sys_unshare(unsigned long unshare_flags);
 
 asmlinkage long sys_splice(int fd_in, loff_t __user *off_in,
-- 
cgit v1.2.3


From 2ed7c03ec17779afb4fcfa3b8c61df61bd4879ba Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Wed, 14 Jan 2009 14:13:54 +0100
Subject: [CVE-2009-0029] Convert all system calls to return a long

Convert all system calls to return a long. This should be a NOP since all
converted types should have the same size anyway.
With the exception of sys_exit_group which returned void. But that doesn't
matter since the system call doesn't return.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 fs/read_write.c          | 18 +++++------
 fs/xattr.c               | 12 ++++----
 include/linux/syscalls.h | 79 ++++++++++++++++++++++++------------------------
 ipc/mqueue.c             |  2 +-
 kernel/exit.c            |  4 ++-
 kernel/signal.c          |  2 +-
 kernel/timer.c           |  2 +-
 mm/filemap.c             |  2 +-
 mm/mmap.c                |  2 +-
 mm/mremap.c              |  2 +-
 mm/nommu.c               |  2 +-
 11 files changed, 64 insertions(+), 63 deletions(-)

(limited to 'include/linux')

diff --git a/fs/read_write.c b/fs/read_write.c
index 5cc6924eb158..940367f51f2a 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -147,7 +147,7 @@ loff_t vfs_llseek(struct file *file, loff_t offset, int origin)
 }
 EXPORT_SYMBOL(vfs_llseek);
 
-asmlinkage off_t sys_lseek(unsigned int fd, off_t offset, unsigned int origin)
+asmlinkage long sys_lseek(unsigned int fd, off_t offset, unsigned int origin)
 {
 	off_t retval;
 	struct file * file;
@@ -369,7 +369,7 @@ static inline void file_pos_write(struct file *file, loff_t pos)
 	file->f_pos = pos;
 }
 
-asmlinkage ssize_t sys_read(unsigned int fd, char __user * buf, size_t count)
+asmlinkage long sys_read(unsigned int fd, char __user * buf, size_t count)
 {
 	struct file *file;
 	ssize_t ret = -EBADF;
@@ -386,7 +386,7 @@ asmlinkage ssize_t sys_read(unsigned int fd, char __user * buf, size_t count)
 	return ret;
 }
 
-asmlinkage ssize_t sys_write(unsigned int fd, const char __user * buf, size_t count)
+asmlinkage long sys_write(unsigned int fd, const char __user * buf, size_t count)
 {
 	struct file *file;
 	ssize_t ret = -EBADF;
@@ -403,7 +403,7 @@ asmlinkage ssize_t sys_write(unsigned int fd, const char __user * buf, size_t co
 	return ret;
 }
 
-asmlinkage ssize_t sys_pread64(unsigned int fd, char __user *buf,
+asmlinkage long sys_pread64(unsigned int fd, char __user *buf,
 			     size_t count, loff_t pos)
 {
 	struct file *file;
@@ -424,7 +424,7 @@ asmlinkage ssize_t sys_pread64(unsigned int fd, char __user *buf,
 	return ret;
 }
 
-asmlinkage ssize_t sys_pwrite64(unsigned int fd, const char __user *buf,
+asmlinkage long sys_pwrite64(unsigned int fd, const char __user *buf,
 			      size_t count, loff_t pos)
 {
 	struct file *file;
@@ -672,7 +672,7 @@ ssize_t vfs_writev(struct file *file, const struct iovec __user *vec,
 
 EXPORT_SYMBOL(vfs_writev);
 
-asmlinkage ssize_t
+asmlinkage long
 sys_readv(unsigned long fd, const struct iovec __user *vec, unsigned long vlen)
 {
 	struct file *file;
@@ -693,7 +693,7 @@ sys_readv(unsigned long fd, const struct iovec __user *vec, unsigned long vlen)
 	return ret;
 }
 
-asmlinkage ssize_t
+asmlinkage long
 sys_writev(unsigned long fd, const struct iovec __user *vec, unsigned long vlen)
 {
 	struct file *file;
@@ -812,7 +812,7 @@ out:
 	return retval;
 }
 
-asmlinkage ssize_t sys_sendfile(int out_fd, int in_fd, off_t __user *offset, size_t count)
+asmlinkage long sys_sendfile(int out_fd, int in_fd, off_t __user *offset, size_t count)
 {
 	loff_t pos;
 	off_t off;
@@ -831,7 +831,7 @@ asmlinkage ssize_t sys_sendfile(int out_fd, int in_fd, off_t __user *offset, siz
 	return do_sendfile(out_fd, in_fd, NULL, count, 0);
 }
 
-asmlinkage ssize_t sys_sendfile64(int out_fd, int in_fd, loff_t __user *offset, size_t count)
+asmlinkage long sys_sendfile64(int out_fd, int in_fd, loff_t __user *offset, size_t count)
 {
 	loff_t pos;
 	ssize_t ret;
diff --git a/fs/xattr.c b/fs/xattr.c
index 237804cd6b56..d049ae27aae7 100644
--- a/fs/xattr.c
+++ b/fs/xattr.c
@@ -349,7 +349,7 @@ getxattr(struct dentry *d, const char __user *name, void __user *value,
 	return error;
 }
 
-asmlinkage ssize_t
+asmlinkage long
 sys_getxattr(const char __user *pathname, const char __user *name,
 	     void __user *value, size_t size)
 {
@@ -364,7 +364,7 @@ sys_getxattr(const char __user *pathname, const char __user *name,
 	return error;
 }
 
-asmlinkage ssize_t
+asmlinkage long
 sys_lgetxattr(const char __user *pathname, const char __user *name, void __user *value,
 	      size_t size)
 {
@@ -379,7 +379,7 @@ sys_lgetxattr(const char __user *pathname, const char __user *name, void __user
 	return error;
 }
 
-asmlinkage ssize_t
+asmlinkage long
 sys_fgetxattr(int fd, const char __user *name, void __user *value, size_t size)
 {
 	struct file *f;
@@ -424,7 +424,7 @@ listxattr(struct dentry *d, char __user *list, size_t size)
 	return error;
 }
 
-asmlinkage ssize_t
+asmlinkage long
 sys_listxattr(const char __user *pathname, char __user *list, size_t size)
 {
 	struct path path;
@@ -438,7 +438,7 @@ sys_listxattr(const char __user *pathname, char __user *list, size_t size)
 	return error;
 }
 
-asmlinkage ssize_t
+asmlinkage long
 sys_llistxattr(const char __user *pathname, char __user *list, size_t size)
 {
 	struct path path;
@@ -452,7 +452,7 @@ sys_llistxattr(const char __user *pathname, char __user *list, size_t size)
 	return error;
 }
 
-asmlinkage ssize_t
+asmlinkage long
 sys_flistxattr(int fd, char __user *list, size_t size)
 {
 	struct file *f;
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index a7593f670ca6..22290eeaf553 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -77,7 +77,7 @@ asmlinkage long sys_times(struct tms __user *tbuf);
 
 asmlinkage long sys_gettid(void);
 asmlinkage long sys_nanosleep(struct timespec __user *rqtp, struct timespec __user *rmtp);
-asmlinkage unsigned long sys_alarm(unsigned int seconds);
+asmlinkage long sys_alarm(unsigned int seconds);
 asmlinkage long sys_getpid(void);
 asmlinkage long sys_getppid(void);
 asmlinkage long sys_getuid(void);
@@ -166,7 +166,7 @@ asmlinkage long sys_kexec_load(unsigned long entry, unsigned long nr_segments,
 				unsigned long flags);
 
 asmlinkage long sys_exit(int error_code);
-asmlinkage void sys_exit_group(int error_code);
+asmlinkage long sys_exit_group(int error_code);
 asmlinkage long sys_wait4(pid_t pid, int __user *stat_addr,
 				int options, struct rusage __user *ru);
 asmlinkage long sys_waitid(int which, pid_t pid,
@@ -196,7 +196,7 @@ asmlinkage long sys_tkill(int pid, int sig);
 asmlinkage long sys_rt_sigqueueinfo(int pid, int sig, siginfo_t __user *uinfo);
 asmlinkage long sys_sgetmask(void);
 asmlinkage long sys_ssetmask(int newmask);
-asmlinkage unsigned long sys_signal(int sig, __sighandler_t handler);
+asmlinkage long sys_signal(int sig, __sighandler_t handler);
 asmlinkage long sys_pause(void);
 
 asmlinkage long sys_sync(void);
@@ -246,29 +246,29 @@ asmlinkage long sys_lsetxattr(const char __user *path, const char __user *name,
 			      const void __user *value, size_t size, int flags);
 asmlinkage long sys_fsetxattr(int fd, const char __user *name,
 			      const void __user *value, size_t size, int flags);
-asmlinkage ssize_t sys_getxattr(const char __user *path, const char __user *name,
-				void __user *value, size_t size);
-asmlinkage ssize_t sys_lgetxattr(const char __user *path, const char __user *name,
-				void __user *value, size_t size);
-asmlinkage ssize_t sys_fgetxattr(int fd, const char __user *name,
-				void __user *value, size_t size);
-asmlinkage ssize_t sys_listxattr(const char __user *path, char __user *list,
-				size_t size);
-asmlinkage ssize_t sys_llistxattr(const char __user *path, char __user *list,
-				size_t size);
-asmlinkage ssize_t sys_flistxattr(int fd, char __user *list, size_t size);
+asmlinkage long sys_getxattr(const char __user *path, const char __user *name,
+			     void __user *value, size_t size);
+asmlinkage long sys_lgetxattr(const char __user *path, const char __user *name,
+			      void __user *value, size_t size);
+asmlinkage long sys_fgetxattr(int fd, const char __user *name,
+			      void __user *value, size_t size);
+asmlinkage long sys_listxattr(const char __user *path, char __user *list,
+			      size_t size);
+asmlinkage long sys_llistxattr(const char __user *path, char __user *list,
+			       size_t size);
+asmlinkage long sys_flistxattr(int fd, char __user *list, size_t size);
 asmlinkage long sys_removexattr(const char __user *path,
 				const char __user *name);
 asmlinkage long sys_lremovexattr(const char __user *path,
 				 const char __user *name);
 asmlinkage long sys_fremovexattr(int fd, const char __user *name);
 
-asmlinkage unsigned long sys_brk(unsigned long brk);
+asmlinkage long sys_brk(unsigned long brk);
 asmlinkage long sys_mprotect(unsigned long start, size_t len,
 				unsigned long prot);
-asmlinkage unsigned long sys_mremap(unsigned long addr,
-				unsigned long old_len, unsigned long new_len,
-				unsigned long flags, unsigned long new_addr);
+asmlinkage long sys_mremap(unsigned long addr,
+			   unsigned long old_len, unsigned long new_len,
+			   unsigned long flags, unsigned long new_addr);
 asmlinkage long sys_remap_file_pages(unsigned long start, unsigned long size,
 			unsigned long prot, unsigned long pgoff,
 			unsigned long flags);
@@ -321,10 +321,10 @@ asmlinkage long sys_io_submit(aio_context_t, long,
 				struct iocb __user * __user *);
 asmlinkage long sys_io_cancel(aio_context_t ctx_id, struct iocb __user *iocb,
 			      struct io_event __user *result);
-asmlinkage ssize_t sys_sendfile(int out_fd, int in_fd,
-				off_t __user *offset, size_t count);
-asmlinkage ssize_t sys_sendfile64(int out_fd, int in_fd,
-				loff_t __user *offset, size_t count);
+asmlinkage long sys_sendfile(int out_fd, int in_fd,
+			     off_t __user *offset, size_t count);
+asmlinkage long sys_sendfile64(int out_fd, int in_fd,
+			       loff_t __user *offset, size_t count);
 asmlinkage long sys_readlink(const char __user *path,
 				char __user *buf, int bufsiz);
 asmlinkage long sys_creat(const char __user *pathname, int mode);
@@ -368,26 +368,25 @@ asmlinkage long sys_utime(char __user *filename,
 				struct utimbuf __user *times);
 asmlinkage long sys_utimes(char __user *filename,
 				struct timeval __user *utimes);
-asmlinkage off_t sys_lseek(unsigned int fd, off_t offset,
-				unsigned int origin);
+asmlinkage long sys_lseek(unsigned int fd, off_t offset,
+			  unsigned int origin);
 asmlinkage long sys_llseek(unsigned int fd, unsigned long offset_high,
 			unsigned long offset_low, loff_t __user *result,
 			unsigned int origin);
-asmlinkage ssize_t sys_read(unsigned int fd, char __user *buf,
-				size_t count);
-asmlinkage ssize_t sys_readahead(int fd, loff_t offset, size_t count);
-asmlinkage ssize_t sys_readv(unsigned long fd,
-				const struct iovec __user *vec,
-				unsigned long vlen);
-asmlinkage ssize_t sys_write(unsigned int fd, const char __user *buf,
-				size_t count);
-asmlinkage ssize_t sys_writev(unsigned long fd,
-				const struct iovec __user *vec,
-				unsigned long vlen);
-asmlinkage ssize_t sys_pread64(unsigned int fd, char __user *buf,
-				size_t count, loff_t pos);
-asmlinkage ssize_t sys_pwrite64(unsigned int fd, const char __user *buf,
-				size_t count, loff_t pos);
+asmlinkage long sys_read(unsigned int fd, char __user *buf, size_t count);
+asmlinkage long sys_readahead(int fd, loff_t offset, size_t count);
+asmlinkage long sys_readv(unsigned long fd,
+			  const struct iovec __user *vec,
+			  unsigned long vlen);
+asmlinkage long sys_write(unsigned int fd, const char __user *buf,
+			  size_t count);
+asmlinkage long sys_writev(unsigned long fd,
+			   const struct iovec __user *vec,
+			   unsigned long vlen);
+asmlinkage long sys_pread64(unsigned int fd, char __user *buf,
+			    size_t count, loff_t pos);
+asmlinkage long sys_pwrite64(unsigned int fd, const char __user *buf,
+			     size_t count, loff_t pos);
 asmlinkage long sys_getcwd(char __user *buf, unsigned long size);
 asmlinkage long sys_mkdir(const char __user *pathname, int mode);
 asmlinkage long sys_chdir(const char __user *filename);
@@ -476,7 +475,7 @@ asmlinkage long sys_shmctl(int shmid, int cmd, struct shmid_ds __user *buf);
 asmlinkage long sys_mq_open(const char __user *name, int oflag, mode_t mode, struct mq_attr __user *attr);
 asmlinkage long sys_mq_unlink(const char __user *name);
 asmlinkage long sys_mq_timedsend(mqd_t mqdes, const char __user *msg_ptr, size_t msg_len, unsigned int msg_prio, const struct timespec __user *abs_timeout);
-asmlinkage ssize_t sys_mq_timedreceive(mqd_t mqdes, char __user *msg_ptr, size_t msg_len, unsigned int __user *msg_prio, const struct timespec __user *abs_timeout);
+asmlinkage long sys_mq_timedreceive(mqd_t mqdes, char __user *msg_ptr, size_t msg_len, unsigned int __user *msg_prio, const struct timespec __user *abs_timeout);
 asmlinkage long sys_mq_notify(mqd_t mqdes, const struct sigevent __user *notification);
 asmlinkage long sys_mq_getsetattr(mqd_t mqdes, const struct mq_attr __user *mqstat, struct mq_attr __user *omqstat);
 
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 23fdb8492b8e..6df028b70543 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -907,7 +907,7 @@ out:
 	return ret;
 }
 
-asmlinkage ssize_t sys_mq_timedreceive(mqd_t mqdes, char __user *u_msg_ptr,
+asmlinkage long sys_mq_timedreceive(mqd_t mqdes, char __user *u_msg_ptr,
 	size_t msg_len, unsigned int __user *u_msg_prio,
 	const struct timespec __user *u_abs_timeout)
 {
diff --git a/kernel/exit.c b/kernel/exit.c
index c7740fa3252c..fac9b040af2c 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -1182,9 +1182,11 @@ do_group_exit(int exit_code)
  * wait4()-ing process will get the correct exit code - even if this
  * thread is not the thread group leader.
  */
-asmlinkage void sys_exit_group(int error_code)
+asmlinkage long sys_exit_group(int error_code)
 {
 	do_group_exit((error_code & 0xff) << 8);
+	/* NOTREACHED */
+	return 0;
 }
 
 static struct pid *task_pid_type(struct task_struct *task, enum pid_type type)
diff --git a/kernel/signal.c b/kernel/signal.c
index 3152ac3b62e2..856a5479d49d 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -2559,7 +2559,7 @@ sys_ssetmask(int newmask)
 /*
  * For backwards compatibility.  Functionality superseded by sigaction.
  */
-asmlinkage unsigned long
+asmlinkage long
 sys_signal(int sig, __sighandler_t handler)
 {
 	struct k_sigaction new_sa, old_sa;
diff --git a/kernel/timer.c b/kernel/timer.c
index dee3f641a7a7..7b8697d7f04d 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -1129,7 +1129,7 @@ void do_timer(unsigned long ticks)
  * For backwards compatibility?  This can be done in libc so Alpha
  * and all newer ports shouldn't need it.
  */
-asmlinkage unsigned long sys_alarm(unsigned int seconds)
+asmlinkage long sys_alarm(unsigned int seconds)
 {
 	return alarm_setitimer(seconds);
 }
diff --git a/mm/filemap.c b/mm/filemap.c
index ceba0bd03662..538b75ed6236 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1374,7 +1374,7 @@ do_readahead(struct address_space *mapping, struct file *filp,
 	return 0;
 }
 
-asmlinkage ssize_t sys_readahead(int fd, loff_t offset, size_t count)
+asmlinkage long sys_readahead(int fd, loff_t offset, size_t count)
 {
 	ssize_t ret;
 	struct file *file;
diff --git a/mm/mmap.c b/mm/mmap.c
index 749623196cb9..a970d890cb21 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -245,7 +245,7 @@ static struct vm_area_struct *remove_vma(struct vm_area_struct *vma)
 	return next;
 }
 
-asmlinkage unsigned long sys_brk(unsigned long brk)
+asmlinkage long sys_brk(unsigned long brk)
 {
 	unsigned long rlim, retval;
 	unsigned long newbrk, oldbrk;
diff --git a/mm/mremap.c b/mm/mremap.c
index 646de959aa58..5572e0825d80 100644
--- a/mm/mremap.c
+++ b/mm/mremap.c
@@ -420,7 +420,7 @@ out_nc:
 	return ret;
 }
 
-asmlinkage unsigned long sys_mremap(unsigned long addr,
+asmlinkage long sys_mremap(unsigned long addr,
 	unsigned long old_len, unsigned long new_len,
 	unsigned long flags, unsigned long new_addr)
 {
diff --git a/mm/nommu.c b/mm/nommu.c
index 60ed8375c986..ee3e78927739 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -416,7 +416,7 @@ EXPORT_SYMBOL(vm_insert_page);
  *  to a regular file.  in this case, the unmapping will need
  *  to invoke file system routines that need the global lock.
  */
-asmlinkage unsigned long sys_brk(unsigned long brk)
+asmlinkage long sys_brk(unsigned long brk)
 {
 	struct mm_struct *mm = current->mm;
 
-- 
cgit v1.2.3


From e55380edf68796d75bf41391a781c68ee678587d Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Wed, 14 Jan 2009 14:13:55 +0100
Subject: [CVE-2009-0029] Rename old_readdir to sys_old_readdir

This way it matches the generic system call name convention.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 arch/arm/kernel/calls.S              | 2 +-
 arch/cris/arch-v10/kernel/entry.S    | 2 +-
 arch/cris/arch-v32/kernel/entry.S    | 2 +-
 arch/h8300/kernel/syscalls.S         | 2 +-
 arch/m68k/kernel/entry.S             | 2 +-
 arch/m68knommu/kernel/syscalltable.S | 2 +-
 arch/mips/kernel/scall32-o32.S       | 2 +-
 arch/mn10300/kernel/entry.S          | 2 +-
 arch/powerpc/include/asm/systbl.h    | 2 +-
 arch/sh/kernel/syscalls_32.S         | 2 +-
 arch/sh/kernel/syscalls_64.S         | 2 +-
 arch/sparc/kernel/systbls_32.S       | 2 +-
 arch/x86/kernel/syscall_table_32.S   | 2 +-
 fs/readdir.c                         | 2 +-
 include/linux/syscalls.h             | 2 ++
 15 files changed, 16 insertions(+), 14 deletions(-)

(limited to 'include/linux')

diff --git a/arch/arm/kernel/calls.S b/arch/arm/kernel/calls.S
index 09a061cb7838..9ca8d13f05f7 100644
--- a/arch/arm/kernel/calls.S
+++ b/arch/arm/kernel/calls.S
@@ -98,7 +98,7 @@
 		CALL(sys_uselib)
 		CALL(sys_swapon)
 		CALL(sys_reboot)
-		CALL(OBSOLETE(old_readdir))	/* used by libc4 */
+		CALL(OBSOLETE(sys_old_readdir))	/* used by libc4 */
 /* 90 */	CALL(OBSOLETE(old_mmap))	/* used by libc4 */
 		CALL(sys_munmap)
 		CALL(sys_truncate)
diff --git a/arch/cris/arch-v10/kernel/entry.S b/arch/cris/arch-v10/kernel/entry.S
index ed171d389e65..72f5cd319b97 100644
--- a/arch/cris/arch-v10/kernel/entry.S
+++ b/arch/cris/arch-v10/kernel/entry.S
@@ -691,7 +691,7 @@ sys_call_table:
 	.long sys_uselib
 	.long sys_swapon
 	.long sys_reboot
-	.long old_readdir
+	.long sys_old_readdir
 	.long old_mmap		/* 90 */
 	.long sys_munmap
 	.long sys_truncate
diff --git a/arch/cris/arch-v32/kernel/entry.S b/arch/cris/arch-v32/kernel/entry.S
index 7f6f93e6b70e..5e674c8f7c51 100644
--- a/arch/cris/arch-v32/kernel/entry.S
+++ b/arch/cris/arch-v32/kernel/entry.S
@@ -614,7 +614,7 @@ sys_call_table:
 	.long sys_uselib
 	.long sys_swapon
 	.long sys_reboot
-	.long old_readdir
+	.long sys_old_readdir
 	.long old_mmap		/* 90 */
 	.long sys_munmap
 	.long sys_truncate
diff --git a/arch/h8300/kernel/syscalls.S b/arch/h8300/kernel/syscalls.S
index 54e21c3f2057..4eb67faac633 100644
--- a/arch/h8300/kernel/syscalls.S
+++ b/arch/h8300/kernel/syscalls.S
@@ -103,7 +103,7 @@ SYMBOL_NAME_LABEL(sys_call_table)
 	.long SYMBOL_NAME(sys_uselib)
 	.long SYMBOL_NAME(sys_swapon)
 	.long SYMBOL_NAME(sys_reboot)
-	.long SYMBOL_NAME(old_readdir)
+	.long SYMBOL_NAME(sys_old_readdir)
 	.long SYMBOL_NAME(old_mmap)		/* 90 */
 	.long SYMBOL_NAME(sys_munmap)
 	.long SYMBOL_NAME(sys_truncate)
diff --git a/arch/m68k/kernel/entry.S b/arch/m68k/kernel/entry.S
index 5b780826647c..5c332f2b9b83 100644
--- a/arch/m68k/kernel/entry.S
+++ b/arch/m68k/kernel/entry.S
@@ -513,7 +513,7 @@ sys_call_table:
 	.long sys_uselib
 	.long sys_swapon
 	.long sys_reboot
-	.long old_readdir
+	.long sys_old_readdir
 	.long old_mmap		/* 90 */
 	.long sys_munmap
 	.long sys_truncate
diff --git a/arch/m68knommu/kernel/syscalltable.S b/arch/m68knommu/kernel/syscalltable.S
index 812f8d8b7a85..5c3e3f62194a 100644
--- a/arch/m68knommu/kernel/syscalltable.S
+++ b/arch/m68knommu/kernel/syscalltable.S
@@ -107,7 +107,7 @@ ENTRY(sys_call_table)
 	.long sys_uselib
 	.long sys_ni_syscall	/* sys_swapon */
 	.long sys_reboot
-	.long old_readdir
+	.long sys_old_readdir
 	.long old_mmap		/* 90 */
 	.long sys_munmap
 	.long sys_truncate
diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S
index d0916a55cd77..51d1ba415b90 100644
--- a/arch/mips/kernel/scall32-o32.S
+++ b/arch/mips/kernel/scall32-o32.S
@@ -398,7 +398,7 @@ einval:	li	v0, -ENOSYS
 	sys	sys_uselib		1
 	sys	sys_swapon		2
 	sys	sys_reboot		3
-	sys	old_readdir		3
+	sys	sys_old_readdir		3
 	sys	old_mmap		6	/* 4090 */
 	sys	sys_munmap		2
 	sys	sys_truncate		2
diff --git a/arch/mn10300/kernel/entry.S b/arch/mn10300/kernel/entry.S
index 62fba8aa9b6e..ceeaaaa359e2 100644
--- a/arch/mn10300/kernel/entry.S
+++ b/arch/mn10300/kernel/entry.S
@@ -478,7 +478,7 @@ ENTRY(sys_call_table)
 	.long sys_uselib
 	.long sys_swapon
 	.long sys_reboot
-	.long old_readdir
+	.long sys_old_readdir
 	.long old_mmap		/* 90 */
 	.long sys_munmap
 	.long sys_truncate
diff --git a/arch/powerpc/include/asm/systbl.h b/arch/powerpc/include/asm/systbl.h
index 803def236654..72353f6070a4 100644
--- a/arch/powerpc/include/asm/systbl.h
+++ b/arch/powerpc/include/asm/systbl.h
@@ -92,7 +92,7 @@ COMPAT_SYS_SPU(readlink)
 SYSCALL(uselib)
 SYSCALL(swapon)
 SYSCALL(reboot)
-SYSX(sys_ni_syscall,compat_sys_old_readdir,old_readdir)
+SYSX(sys_ni_syscall,compat_sys_old_readdir,sys_old_readdir)
 SYSCALL_SPU(mmap)
 SYSCALL_SPU(munmap)
 SYSCALL_SPU(truncate)
diff --git a/arch/sh/kernel/syscalls_32.S b/arch/sh/kernel/syscalls_32.S
index 0af693e65764..a87ce076cfa2 100644
--- a/arch/sh/kernel/syscalls_32.S
+++ b/arch/sh/kernel/syscalls_32.S
@@ -105,7 +105,7 @@ ENTRY(sys_call_table)
 	.long sys_uselib
 	.long sys_swapon
 	.long sys_reboot
-	.long old_readdir
+	.long sys_old_readdir
 	.long old_mmap		/* 90 */
 	.long sys_munmap
 	.long sys_truncate
diff --git a/arch/sh/kernel/syscalls_64.S b/arch/sh/kernel/syscalls_64.S
index 0b436aa3cad7..557cb91f5caf 100644
--- a/arch/sh/kernel/syscalls_64.S
+++ b/arch/sh/kernel/syscalls_64.S
@@ -109,7 +109,7 @@ sys_call_table:
 	.long sys_uselib
 	.long sys_swapon
 	.long sys_reboot
-	.long old_readdir
+	.long sys_old_readdir
 	.long old_mmap			/* 90 */
 	.long sys_munmap
 	.long sys_truncate
diff --git a/arch/sparc/kernel/systbls_32.S b/arch/sparc/kernel/systbls_32.S
index 7d0807586442..8a434f51ba0f 100644
--- a/arch/sparc/kernel/systbls_32.S
+++ b/arch/sparc/kernel/systbls_32.S
@@ -56,7 +56,7 @@ sys_call_table:
 /*185*/	.long sys_setpgid, sys_fremovexattr, sys_tkill, sys_exit_group, sys_newuname
 /*190*/	.long sys_init_module, sys_personality, sparc_remap_file_pages, sys_epoll_create, sys_epoll_ctl
 /*195*/	.long sys_epoll_wait, sys_ioprio_set, sys_getppid, sparc_sigaction, sys_sgetmask
-/*200*/	.long sys_ssetmask, sys_sigsuspend, sys_newlstat, sys_uselib, old_readdir
+/*200*/	.long sys_ssetmask, sys_sigsuspend, sys_newlstat, sys_uselib, sys_old_readdir
 /*205*/	.long sys_readahead, sys_socketcall, sys_syslog, sys_lookup_dcookie, sys_fadvise64
 /*210*/	.long sys_fadvise64_64, sys_tgkill, sys_waitpid, sys_swapoff, sys_sysinfo
 /*215*/	.long sys_ipc, sys_sigreturn, sys_clone, sys_ioprio_get, sys_adjtimex
diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S
index d44395ff34c3..e2e86a08f31d 100644
--- a/arch/x86/kernel/syscall_table_32.S
+++ b/arch/x86/kernel/syscall_table_32.S
@@ -88,7 +88,7 @@ ENTRY(sys_call_table)
 	.long sys_uselib
 	.long sys_swapon
 	.long sys_reboot
-	.long old_readdir
+	.long sys_old_readdir
 	.long old_mmap		/* 90 */
 	.long sys_munmap
 	.long sys_truncate
diff --git a/fs/readdir.c b/fs/readdir.c
index b318d9b5af2e..8b4c2a0051a6 100644
--- a/fs/readdir.c
+++ b/fs/readdir.c
@@ -102,7 +102,7 @@ efault:
 	return -EFAULT;
 }
 
-asmlinkage long old_readdir(unsigned int fd, struct old_linux_dirent __user * dirent, unsigned int count)
+asmlinkage long sys_old_readdir(unsigned int fd, struct old_linux_dirent __user * dirent, unsigned int count)
 {
 	int error;
 	struct file * file;
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 22290eeaf553..ca079c3d09e3 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -54,6 +54,7 @@ struct compat_stat;
 struct compat_timeval;
 struct robust_list_head;
 struct getcpu_cache;
+struct old_linux_dirent;
 
 #include <linux/types.h>
 #include <linux/aio_abi.h>
@@ -608,6 +609,7 @@ asmlinkage long sys_timerfd_gettime(int ufd, struct itimerspec __user *otmr);
 asmlinkage long sys_eventfd(unsigned int count);
 asmlinkage long sys_eventfd2(unsigned int count, int flags);
 asmlinkage long sys_fallocate(int fd, int mode, loff_t offset, loff_t len);
+asmlinkage long sys_old_readdir(unsigned int, struct old_linux_dirent __user *, unsigned int);
 
 int kernel_execve(const char *filename, char *const argv[], char *const envp[]);
 
-- 
cgit v1.2.3


From 1a94bc34768e463a93cb3751819709ab0ea80a01 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Wed, 14 Jan 2009 14:13:59 +0100
Subject: [CVE-2009-0029] System call wrapper infrastructure

From: Martin Schwidefsky <schwidefsky@de.ibm.com>

By selecting HAVE_SYSCALL_WRAPPERS architectures can activate
system call wrappers in order to sign extend system call arguments.

All architectures where the ABI defines that the caller of a function
has to perform sign extension probably need this.

Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 arch/Kconfig             |  3 +++
 include/linux/syscalls.h | 62 ++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 65 insertions(+)

(limited to 'include/linux')

diff --git a/arch/Kconfig b/arch/Kconfig
index 2e13aa261929..550dab22daa1 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -62,6 +62,9 @@ config HAVE_EFFICIENT_UNALIGNED_ACCESS
 	  See Documentation/unaligned-memory-access.txt for more
 	  information on the topic of unaligned memory accesses.
 
+config HAVE_SYSCALL_WRAPPERS
+	bool
+
 config KRETPROBES
 	def_bool y
 	depends on KPROBES && HAVE_KRETPROBES
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index ca079c3d09e3..0bb537d7ba2e 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -66,6 +66,68 @@ struct old_linux_dirent;
 #include <linux/quota.h>
 #include <linux/key.h>
 
+#define __SC_DECL1(t1, a1)	t1 a1
+#define __SC_DECL2(t2, a2, ...) t2 a2, __SC_DECL1(__VA_ARGS__)
+#define __SC_DECL3(t3, a3, ...) t3 a3, __SC_DECL2(__VA_ARGS__)
+#define __SC_DECL4(t4, a4, ...) t4 a4, __SC_DECL3(__VA_ARGS__)
+#define __SC_DECL5(t5, a5, ...) t5 a5, __SC_DECL4(__VA_ARGS__)
+#define __SC_DECL6(t6, a6, ...) t6 a6, __SC_DECL5(__VA_ARGS__)
+
+#define __SC_LONG1(t1, a1) 	long a1
+#define __SC_LONG2(t2, a2, ...) long a2, __SC_LONG1(__VA_ARGS__)
+#define __SC_LONG3(t3, a3, ...) long a3, __SC_LONG2(__VA_ARGS__)
+#define __SC_LONG4(t4, a4, ...) long a4, __SC_LONG3(__VA_ARGS__)
+#define __SC_LONG5(t5, a5, ...) long a5, __SC_LONG4(__VA_ARGS__)
+#define __SC_LONG6(t6, a6, ...) long a6, __SC_LONG5(__VA_ARGS__)
+
+#define __SC_CAST1(t1, a1)	(t1) a1
+#define __SC_CAST2(t2, a2, ...) (t2) a2, __SC_CAST1(__VA_ARGS__)
+#define __SC_CAST3(t3, a3, ...) (t3) a3, __SC_CAST2(__VA_ARGS__)
+#define __SC_CAST4(t4, a4, ...) (t4) a4, __SC_CAST3(__VA_ARGS__)
+#define __SC_CAST5(t5, a5, ...) (t5) a5, __SC_CAST4(__VA_ARGS__)
+#define __SC_CAST6(t6, a6, ...) (t6) a6, __SC_CAST5(__VA_ARGS__)
+
+#define __SC_TEST(type)		BUILD_BUG_ON(sizeof(type) > sizeof(long))
+#define __SC_TEST1(t1, a1)	__SC_TEST(t1)
+#define __SC_TEST2(t2, a2, ...)	__SC_TEST(t2); __SC_TEST1(__VA_ARGS__)
+#define __SC_TEST3(t3, a3, ...)	__SC_TEST(t3); __SC_TEST2(__VA_ARGS__)
+#define __SC_TEST4(t4, a4, ...)	__SC_TEST(t4); __SC_TEST3(__VA_ARGS__)
+#define __SC_TEST5(t5, a5, ...)	__SC_TEST(t5); __SC_TEST4(__VA_ARGS__)
+#define __SC_TEST6(t6, a6, ...)	__SC_TEST(t6); __SC_TEST5(__VA_ARGS__)
+
+#define SYSCALL_DEFINE0(name)   asmlinkage long sys_##name(void)
+#define SYSCALL_DEFINE1(...)    SYSCALL_DEFINEx(1, __VA_ARGS__)
+#define SYSCALL_DEFINE2(...)    SYSCALL_DEFINEx(2, __VA_ARGS__)
+#define SYSCALL_DEFINE3(...)    SYSCALL_DEFINEx(3, __VA_ARGS__)
+#define SYSCALL_DEFINE4(...)    SYSCALL_DEFINEx(4, __VA_ARGS__)
+#define SYSCALL_DEFINE5(...)    SYSCALL_DEFINEx(5, __VA_ARGS__)
+#define SYSCALL_DEFINE6(...)    SYSCALL_DEFINEx(6, __VA_ARGS__)
+
+#define SYSCALL_ALIAS(alias, name)					\
+	asm ("\t.globl " #alias "\n\t.set " #alias ", " #name)
+
+#ifdef CONFIG_HAVE_SYSCALL_WRAPPERS
+
+#define SYSCALL_DEFINE(name) static inline long SYSC_##name
+#define SYSCALL_DEFINEx(x, name, ...)					\
+	asmlinkage long sys_##name(__SC_DECL##x(__VA_ARGS__));		\
+	static inline long SYSC_##name(__SC_DECL##x(__VA_ARGS__));	\
+	asmlinkage long SyS_##name(__SC_LONG##x(__VA_ARGS__))		\
+	{								\
+		__SC_TEST##x(__VA_ARGS__);				\
+		return (long) SYSC_##name(__SC_CAST##x(__VA_ARGS__));	\
+	}								\
+	SYSCALL_ALIAS(sys_##name, SyS_##name);				\
+	static inline long SYSC_##name(__SC_DECL##x(__VA_ARGS__))
+
+#else /* CONFIG_HAVE_SYSCALL_WRAPPERS */
+
+#define SYSCALL_DEFINE(name) asmlinkage long sys_##name
+#define SYSCALL_DEFINEx(x, name, ...)					\
+	asmlinkage long sys_##name(__SC_DECL##x(__VA_ARGS__))
+
+#endif /* CONFIG_HAVE_SYSCALL_WRAPPERS */
+
 asmlinkage long sys_time(time_t __user *tloc);
 asmlinkage long sys_stime(time_t __user *tptr);
 asmlinkage long sys_gettimeofday(struct timeval __user *tv,
-- 
cgit v1.2.3


From ee6a093222549ac0c72cfd296c69fa5e7d6daa34 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 14 Jan 2009 14:14:00 +0100
Subject: [CVE-2009-0029] powerpc: Enable syscall wrappers for 64-bit

This enables the use of syscall wrappers to do proper sign extension
for 64-bit programs.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 arch/powerpc/Kconfig     | 1 +
 include/linux/syscalls.h | 6 ++++++
 2 files changed, 7 insertions(+)

(limited to 'include/linux')

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 84b861316ce7..e39b73bc0ff8 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -123,6 +123,7 @@ config PPC
 	select HAVE_DMA_ATTRS if PPC64
 	select USE_GENERIC_SMP_HELPERS if SMP
 	select HAVE_OPROFILE
+	select HAVE_SYSCALL_WRAPPERS if PPC64
 
 config EARLY_PRINTK
 	bool
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 0bb537d7ba2e..90aa5eba87a2 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -103,8 +103,14 @@ struct old_linux_dirent;
 #define SYSCALL_DEFINE5(...)    SYSCALL_DEFINEx(5, __VA_ARGS__)
 #define SYSCALL_DEFINE6(...)    SYSCALL_DEFINEx(6, __VA_ARGS__)
 
+#ifdef CONFIG_PPC64
+#define SYSCALL_ALIAS(alias, name)					\
+	asm ("\t.globl " #alias "\n\t.set " #alias ", " #name "\n"	\
+	     "\t.globl ." #alias "\n\t.set ." #alias ", ." #name)
+#else
 #define SYSCALL_ALIAS(alias, name)					\
 	asm ("\t.globl " #alias "\n\t.set " #alias ", " #name)
+#endif
 
 #ifdef CONFIG_HAVE_SYSCALL_WRAPPERS
 
-- 
cgit v1.2.3


From d4e82042c4cfa87a7d51710b71f568fe80132551 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Wed, 14 Jan 2009 14:14:34 +0100
Subject: [CVE-2009-0029] System call wrappers part 32

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 fs/eventfd.c             |  5 ++---
 fs/pipe.c                |  2 +-
 fs/readdir.c             |  3 ++-
 fs/select.c              | 11 ++++++-----
 fs/timerfd.c             |  2 +-
 include/linux/syscalls.h |  7 +++++++
 kernel/signal.c          | 11 +++++------
 7 files changed, 24 insertions(+), 17 deletions(-)

(limited to 'include/linux')

diff --git a/fs/eventfd.c b/fs/eventfd.c
index 08bf558d0408..5de2c2db3aa2 100644
--- a/fs/eventfd.c
+++ b/fs/eventfd.c
@@ -198,7 +198,7 @@ struct file *eventfd_fget(int fd)
 	return file;
 }
 
-asmlinkage long sys_eventfd2(unsigned int count, int flags)
+SYSCALL_DEFINE2(eventfd2, unsigned int, count, int, flags)
 {
 	int fd;
 	struct eventfd_ctx *ctx;
@@ -228,8 +228,7 @@ asmlinkage long sys_eventfd2(unsigned int count, int flags)
 	return fd;
 }
 
-asmlinkage long sys_eventfd(unsigned int count)
+SYSCALL_DEFINE1(eventfd, unsigned int, count)
 {
 	return sys_eventfd2(count, 0);
 }
-
diff --git a/fs/pipe.c b/fs/pipe.c
index 0c64db86c919..b89c878588a9 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -1043,7 +1043,7 @@ int do_pipe(int *fd)
  * sys_pipe() is the normal C calling standard for creating
  * a pipe. It's not the way Unix traditionally does this, though.
  */
-asmlinkage long sys_pipe2(int __user *fildes, int flags)
+SYSCALL_DEFINE2(pipe2, int __user *, fildes, int, flags)
 {
 	int fd[2];
 	int error;
diff --git a/fs/readdir.c b/fs/readdir.c
index cf6a0e39819a..7723401f8d8b 100644
--- a/fs/readdir.c
+++ b/fs/readdir.c
@@ -102,7 +102,8 @@ efault:
 	return -EFAULT;
 }
 
-asmlinkage long sys_old_readdir(unsigned int fd, struct old_linux_dirent __user * dirent, unsigned int count)
+SYSCALL_DEFINE3(old_readdir, unsigned int, fd,
+		struct old_linux_dirent __user *, dirent, unsigned int, count)
 {
 	int error;
 	struct file * file;
diff --git a/fs/select.c b/fs/select.c
index 338f703403af..0fe0e1469df3 100644
--- a/fs/select.c
+++ b/fs/select.c
@@ -636,8 +636,9 @@ static long do_pselect(int n, fd_set __user *inp, fd_set __user *outp,
  * which has a pointer to the sigset_t itself followed by a size_t containing
  * the sigset size.
  */
-asmlinkage long sys_pselect6(int n, fd_set __user *inp, fd_set __user *outp,
-	fd_set __user *exp, struct timespec __user *tsp, void __user *sig)
+SYSCALL_DEFINE6(pselect6, int, n, fd_set __user *, inp, fd_set __user *, outp,
+		fd_set __user *, exp, struct timespec __user *, tsp,
+		void __user *, sig)
 {
 	size_t sigsetsize = 0;
 	sigset_t __user *up = NULL;
@@ -889,9 +890,9 @@ SYSCALL_DEFINE3(poll, struct pollfd __user *, ufds, unsigned int, nfds,
 }
 
 #ifdef HAVE_SET_RESTORE_SIGMASK
-asmlinkage long sys_ppoll(struct pollfd __user *ufds, unsigned int nfds,
-	struct timespec __user *tsp, const sigset_t __user *sigmask,
-	size_t sigsetsize)
+SYSCALL_DEFINE5(ppoll, struct pollfd __user *, ufds, unsigned int, nfds,
+		struct timespec __user *, tsp, const sigset_t __user *, sigmask,
+		size_t, sigsetsize)
 {
 	sigset_t ksigmask, sigsaved;
 	struct timespec ts, end_time, *to = NULL;
diff --git a/fs/timerfd.c b/fs/timerfd.c
index c8c14f58b96f..6a123b8ff3f5 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -265,7 +265,7 @@ SYSCALL_DEFINE4(timerfd_settime, int, ufd, int, flags,
 	return 0;
 }
 
-asmlinkage long sys_timerfd_gettime(int ufd, struct itimerspec __user *otmr)
+SYSCALL_DEFINE2(timerfd_gettime, int, ufd, struct itimerspec __user *, otmr)
 {
 	struct file *file;
 	struct timerfd_ctx *ctx;
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 90aa5eba87a2..56c400138b05 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -678,6 +678,13 @@ asmlinkage long sys_eventfd(unsigned int count);
 asmlinkage long sys_eventfd2(unsigned int count, int flags);
 asmlinkage long sys_fallocate(int fd, int mode, loff_t offset, loff_t len);
 asmlinkage long sys_old_readdir(unsigned int, struct old_linux_dirent __user *, unsigned int);
+asmlinkage long sys_pselect6(int, fd_set __user *, fd_set __user *,
+			     fd_set __user *, struct timespec __user *,
+			     void __user *);
+asmlinkage long sys_ppoll(struct pollfd __user *, unsigned int,
+			  struct timespec __user *, const sigset_t __user *,
+			  size_t);
+asmlinkage long sys_pipe2(int __user *, int);
 
 int kernel_execve(const char *filename, char *const argv[], char *const envp[]);
 
diff --git a/kernel/signal.c b/kernel/signal.c
index e2333929611a..e73759783dc8 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -2491,11 +2491,10 @@ out:
 #endif /* __ARCH_WANT_SYS_SIGPROCMASK */
 
 #ifdef __ARCH_WANT_SYS_RT_SIGACTION
-asmlinkage long
-sys_rt_sigaction(int sig,
-		 const struct sigaction __user *act,
-		 struct sigaction __user *oact,
-		 size_t sigsetsize)
+SYSCALL_DEFINE4(rt_sigaction, int, sig,
+		const struct sigaction __user *, act,
+		struct sigaction __user *, oact,
+		size_t, sigsetsize)
 {
 	struct k_sigaction new_sa, old_sa;
 	int ret = -EINVAL;
@@ -2578,7 +2577,7 @@ SYSCALL_DEFINE0(pause)
 #endif
 
 #ifdef __ARCH_WANT_SYS_RT_SIGSUSPEND
-asmlinkage long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize)
+SYSCALL_DEFINE2(rt_sigsuspend, sigset_t __user *, unewset, size_t, sigsetsize)
 {
 	sigset_t newset;
 
-- 
cgit v1.2.3


From 2b66421995d2e93c9d1a0111acf2581f8529c6e5 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Wed, 14 Jan 2009 14:14:35 +0100
Subject: [CVE-2009-0029] System call wrappers part 33

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 fs/pipe.c                | 2 +-
 include/linux/syscalls.h | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/fs/pipe.c b/fs/pipe.c
index b89c878588a9..3a48ba5179d5 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -1059,7 +1059,7 @@ SYSCALL_DEFINE2(pipe2, int __user *, fildes, int, flags)
 	return error;
 }
 
-asmlinkage long sys_pipe(int __user *fildes)
+SYSCALL_DEFINE1(pipe, int __user *, fildes)
 {
 	return sys_pipe2(fildes, 0);
 }
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 56c400138b05..16875f89e6a7 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -685,6 +685,7 @@ asmlinkage long sys_ppoll(struct pollfd __user *, unsigned int,
 			  struct timespec __user *, const sigset_t __user *,
 			  size_t);
 asmlinkage long sys_pipe2(int __user *, int);
+asmlinkage long sys_pipe(int __user *);
 
 int kernel_execve(const char *filename, char *const argv[], char *const envp[]);
 
-- 
cgit v1.2.3


From 18e6959c385f3edf3991fa6662a53dac4eb10d5b Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Wed, 14 Jan 2009 07:28:16 +0100
Subject: mm: fix assertion

This assertion is incorrect for lockless pagecache.  By definition if we
have an unpinned page that we are trying to take a speculative reference
to, it may become the tail of a compound page at any time (if it is
freed, then reallocated as a compound page).

It was still a valid assertion for the vmscan.c LRU isolation case, but
it doesn't seem incredibly helpful...  if somebody wants it, they can
put it back directly where it applies in the vmscan code.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/mm.h | 1 -
 1 file changed, 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/mm.h b/include/linux/mm.h
index b91a73fd1bcc..e8ddc98b8405 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -260,7 +260,6 @@ static inline int put_page_testzero(struct page *page)
  */
 static inline int get_page_unless_zero(struct page *page)
 {
-	VM_BUG_ON(PageTail(page));
 	return atomic_inc_not_zero(&page->_count);
 }
 
-- 
cgit v1.2.3


From b94b898f3107046b5c97c556e23529283ea5eadd Mon Sep 17 00:00:00 2001
From: Brandon Philips <brandon@ifup.org>
Date: Wed, 14 Jan 2009 19:19:02 +0100
Subject: it821x: Add ultra_mask quirk for Vortex86SX

On Vortex86SX with IDE controller revision 0x11 ultra DMA must be
disabled. This patch was tested by DMP and seems to work.

It is a cleaned up version of their older Kernel patch:
 http://www.dmp.com.tw/tech/vortex86sx/patch-2.6.24-DMP.gz

Tested-by: Shawn Lin <shawn@dmp.com.tw>
Signed-off-by: Brandon Philips <bphilips@suse.de>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/it821x.c    | 12 ++++++++++++
 include/linux/pci_ids.h |  1 +
 2 files changed, 13 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/ide/it821x.c b/drivers/ide/it821x.c
index 0be27ac1f077..983440a9a5f7 100644
--- a/drivers/ide/it821x.c
+++ b/drivers/ide/it821x.c
@@ -68,6 +68,8 @@
 
 #define DRV_NAME "it821x"
 
+#define QUIRK_VORTEX86 1
+
 struct it821x_dev
 {
 	unsigned int smart:1,		/* Are we in smart raid mode */
@@ -79,6 +81,7 @@ struct it821x_dev
 	u16	pio[2];			/* Cached PIO values */
 	u16	mwdma[2];		/* Cached MWDMA values */
 	u16	udma[2];		/* Cached UDMA values (per drive) */
+	u16	quirks;
 };
 
 #define ATA_66		0
@@ -577,6 +580,12 @@ static void __devinit init_hwif_it821x(ide_hwif_t *hwif)
 
 	hwif->ultra_mask = ATA_UDMA6;
 	hwif->mwdma_mask = ATA_MWDMA2;
+
+	/* Vortex86SX quirk: prevent Ultra-DMA mode to fix BadCRC issue */
+	if (idev->quirks & QUIRK_VORTEX86) {
+		if (dev->revision == 0x11)
+			hwif->ultra_mask = 0;
+	}
 }
 
 static void it8212_disable_raid(struct pci_dev *dev)
@@ -649,6 +658,8 @@ static int __devinit it821x_init_one(struct pci_dev *dev, const struct pci_devic
 		return -ENOMEM;
 	}
 
+	itdevs->quirks = id->driver_data;
+
 	rc = ide_pci_init_one(dev, &it821x_chipset, itdevs);
 	if (rc)
 		kfree(itdevs);
@@ -668,6 +679,7 @@ static void __devexit it821x_remove(struct pci_dev *dev)
 static const struct pci_device_id it821x_pci_tbl[] = {
 	{ PCI_VDEVICE(ITE, PCI_DEVICE_ID_ITE_8211), 0 },
 	{ PCI_VDEVICE(ITE, PCI_DEVICE_ID_ITE_8212), 0 },
+	{ PCI_VDEVICE(RDC, PCI_DEVICE_ID_RDC_D1010), QUIRK_VORTEX86 },
 	{ 0, },
 };
 
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index d543365518ab..d56ad9c21c09 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2174,6 +2174,7 @@
 #define PCI_DEVICE_ID_RDC_R6040		0x6040
 #define PCI_DEVICE_ID_RDC_R6060		0x6060
 #define PCI_DEVICE_ID_RDC_R6061		0x6061
+#define PCI_DEVICE_ID_RDC_D1010		0x1010
 
 #define PCI_VENDOR_ID_LENOVO		0x17aa
 
-- 
cgit v1.2.3


From e720b9e498b6bbb1b4f3b3d2f8e9a78578aafef7 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Wed, 14 Jan 2009 19:19:04 +0100
Subject: IDE: fix sparse signed-ness errors with host->host_busy

The host_busy field in struct ide_host defaults to a
signed-long, where most arch's test_and_set_bit_*
macros use an unsigned long.

Change to using an unsigned long, which on ARM removes
the following sparse errors:

drivers/ide/ide-io.c:681:8: warning: incorrect type in argument 2 (different signedness)
drivers/ide/ide-io.c:681:8:    expected unsigned long volatile *p
drivers/ide/ide-io.c:681:8:    got long volatile *<noident>
drivers/ide/ide-io.c:681:8: warning: incorrect type in argument 2 (different signedness)
drivers/ide/ide-io.c:681:8:    expected unsigned long volatile *p
drivers/ide/ide-io.c:681:8:    got long volatile *<noident>
drivers/ide/ide-io.c:695:3: warning: incorrect type in argument 2 (different signedness)
drivers/ide/ide-io.c:695:3:    expected unsigned long volatile *p
drivers/ide/ide-io.c:695:3:    got long volatile *<noident>
drivers/ide/ide-io.c:695:3: warning: incorrect type in argument 2 (different signedness)
drivers/ide/ide-io.c:695:3:    expected unsigned long volatile *p
drivers/ide/ide-io.c:695:3:    got long volatile *<noident>
drivers/ide/ide-io.c:695:3: warning: incorrect type in argument 2 (different signedness)
drivers/ide/ide-io.c:695:3:    expected unsigned long volatile *p
drivers/ide/ide-io.c:695:3:    got long volatile *<noident>
drivers/ide/ide-io.c:695:3: warning: incorrect type in argument 2 (different signedness)
drivers/ide/ide-io.c:695:3:    expected unsigned long volatile *p
drivers/ide/ide-io.c:695:3:    got long volatile *<noident>
drivers/ide/ide-io.c:695:3: warning: incorrect type in argument 2 (different signedness)
drivers/ide/ide-io.c:695:3:    expected unsigned long volatile *p
drivers/ide/ide-io.c:695:3:    got long volatile *<noident>
drivers/ide/ide-io.c:695:3: warning: incorrect type in argument 2 (different signedness)
drivers/ide/ide-io.c:695:3:    expected unsigned long volatile *p
drivers/ide/ide-io.c:695:3:    got long volatile *<noident>
drivers/ide/ide-io.c:695:3: warning: incorrect type in argument 2 (different signedness)
drivers/ide/ide-io.c:695:3:    expected unsigned long volatile *p
drivers/ide/ide-io.c:695:3:    got long volatile *<noident>
drivers/ide/ide-io.c:695:3: warning: incorrect type in argument 2 (different signedness)
drivers/ide/ide-io.c:695:3:    expected unsigned long volatile *p
drivers/ide/ide-io.c:695:3:    got long volatile *<noident>

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/linux/ide.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/ide.h b/include/linux/ide.h
index 3644f6323384..194da5a4b0d6 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -871,7 +871,7 @@ struct ide_host {
 	ide_hwif_t	*cur_port;	/* for hosts requiring serialization */
 
 	/* used for hosts requiring serialization */
-	volatile long	host_busy;
+	volatile unsigned long	host_busy;
 };
 
 #define IDE_HOST_BUSY 0
-- 
cgit v1.2.3


From 74d96f018673759d04d032c137d132f6447bfb1e Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Tue, 13 Jan 2009 19:27:09 -0800
Subject: byteorder: make swab.h include asm/swab.h like a regular header

Add swab.h to kbuild.asm and remove the individual entries from
each arch, mark as unifdef as some arches have some kernel-only
bits inside.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/alpha/include/asm/Kbuild          | 1 -
 arch/alpha/include/asm/byteorder.h     | 1 -
 arch/arm/include/asm/Kbuild            | 1 -
 arch/arm/include/asm/byteorder.h       | 2 --
 arch/avr32/include/asm/Kbuild          | 1 -
 arch/avr32/include/asm/byteorder.h     | 1 -
 arch/blackfin/include/asm/Kbuild       | 1 -
 arch/blackfin/include/asm/byteorder.h  | 1 -
 arch/cris/include/asm/Kbuild           | 1 -
 arch/cris/include/asm/byteorder.h      | 1 -
 arch/h8300/include/asm/Kbuild          | 1 -
 arch/h8300/include/asm/byteorder.h     | 1 -
 arch/ia64/include/asm/Kbuild           | 1 -
 arch/ia64/include/asm/byteorder.h      | 1 -
 arch/m68knommu/include/asm/Kbuild      | 2 --
 arch/m68knommu/include/asm/byteorder.h | 1 -
 arch/mips/include/asm/Kbuild           | 1 -
 arch/mips/include/asm/byteorder.h      | 2 --
 arch/parisc/include/asm/Kbuild         | 1 -
 arch/parisc/include/asm/byteorder.h    | 1 -
 arch/powerpc/include/asm/Kbuild        | 1 -
 arch/powerpc/include/asm/byteorder.h   | 2 --
 arch/s390/include/asm/Kbuild           | 1 -
 arch/s390/include/asm/byteorder.h      | 1 -
 arch/sh/include/asm/Kbuild             | 1 -
 arch/sh/include/asm/byteorder.h        | 2 --
 arch/sparc/include/asm/Kbuild          | 1 -
 arch/sparc/include/asm/byteorder.h     | 1 -
 arch/x86/include/asm/Kbuild            | 1 -
 arch/x86/include/asm/byteorder.h       | 1 -
 arch/xtensa/include/asm/Kbuild         | 2 --
 arch/xtensa/include/asm/byteorder.h    | 2 --
 include/asm-frv/Kbuild                 | 1 -
 include/asm-frv/byteorder.h            | 1 -
 include/asm-generic/Kbuild.asm         | 1 +
 include/asm-m32r/Kbuild                | 1 -
 include/asm-m32r/byteorder.h           | 2 --
 include/asm-m68k/Kbuild                | 1 -
 include/asm-m68k/byteorder.h           | 1 -
 include/asm-mn10300/Kbuild             | 1 -
 include/asm-mn10300/byteorder.h        | 1 -
 include/linux/swab.h                   | 2 +-
 42 files changed, 2 insertions(+), 49 deletions(-)

(limited to 'include/linux')

diff --git a/arch/alpha/include/asm/Kbuild b/arch/alpha/include/asm/Kbuild
index 4dad27360576..b7c8f188b313 100644
--- a/arch/alpha/include/asm/Kbuild
+++ b/arch/alpha/include/asm/Kbuild
@@ -9,4 +9,3 @@ unifdef-y += console.h
 unifdef-y += fpu.h
 unifdef-y += sysinfo.h
 unifdef-y += compiler.h
-unifdef-y += swab.h
diff --git a/arch/alpha/include/asm/byteorder.h b/arch/alpha/include/asm/byteorder.h
index 6772f3168701..73683093202d 100644
--- a/arch/alpha/include/asm/byteorder.h
+++ b/arch/alpha/include/asm/byteorder.h
@@ -1,7 +1,6 @@
 #ifndef _ALPHA_BYTEORDER_H
 #define _ALPHA_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/little_endian.h>
 
 #endif /* _ALPHA_BYTEORDER_H */
diff --git a/arch/arm/include/asm/Kbuild b/arch/arm/include/asm/Kbuild
index 43b0b2ba392f..73237bd130a2 100644
--- a/arch/arm/include/asm/Kbuild
+++ b/arch/arm/include/asm/Kbuild
@@ -1,4 +1,3 @@
 include include/asm-generic/Kbuild.asm
 
 unifdef-y += hwcap.h
-unifdef-y += swab.h
diff --git a/arch/arm/include/asm/byteorder.h b/arch/arm/include/asm/byteorder.h
index c02b6fc28e1a..77379748b171 100644
--- a/arch/arm/include/asm/byteorder.h
+++ b/arch/arm/include/asm/byteorder.h
@@ -15,8 +15,6 @@
 #ifndef __ASM_ARM_BYTEORDER_H
 #define __ASM_ARM_BYTEORDER_H
 
-#include <asm/swab.h>
-
 #ifdef __ARMEB__
 #include <linux/byteorder/big_endian.h>
 #else
diff --git a/arch/avr32/include/asm/Kbuild b/arch/avr32/include/asm/Kbuild
index 219822c8ad18..3136628ba8d2 100644
--- a/arch/avr32/include/asm/Kbuild
+++ b/arch/avr32/include/asm/Kbuild
@@ -1,4 +1,3 @@
 include include/asm-generic/Kbuild.asm
 
-header-y	+= swab.h
 header-y	+= cachectl.h
diff --git a/arch/avr32/include/asm/byteorder.h b/arch/avr32/include/asm/byteorder.h
index 2aba64b4e122..50abc21619a8 100644
--- a/arch/avr32/include/asm/byteorder.h
+++ b/arch/avr32/include/asm/byteorder.h
@@ -4,7 +4,6 @@
 #ifndef __ASM_AVR32_BYTEORDER_H
 #define __ASM_AVR32_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/big_endian.h>
 
 #endif /* __ASM_AVR32_BYTEORDER_H */
diff --git a/arch/blackfin/include/asm/Kbuild b/arch/blackfin/include/asm/Kbuild
index d0d1ac435544..606ecfdcc962 100644
--- a/arch/blackfin/include/asm/Kbuild
+++ b/arch/blackfin/include/asm/Kbuild
@@ -1,4 +1,3 @@
 include include/asm-generic/Kbuild.asm
 
 unifdef-y += fixed_code.h
-unifdef-y += swab.h
diff --git a/arch/blackfin/include/asm/byteorder.h b/arch/blackfin/include/asm/byteorder.h
index b9e797a497b4..3e69106a4d37 100644
--- a/arch/blackfin/include/asm/byteorder.h
+++ b/arch/blackfin/include/asm/byteorder.h
@@ -1,7 +1,6 @@
 #ifndef _BLACKFIN_BYTEORDER_H
 #define _BLACKFIN_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/little_endian.h>
 
 #endif				/* _BLACKFIN_BYTEORDER_H */
diff --git a/arch/cris/include/asm/Kbuild b/arch/cris/include/asm/Kbuild
index b79b7c6543a6..d5b631935ec8 100644
--- a/arch/cris/include/asm/Kbuild
+++ b/arch/cris/include/asm/Kbuild
@@ -9,4 +9,3 @@ header-y += sync_serial.h
 
 unifdef-y += etraxgpio.h
 unifdef-y += rs485.h
-unifdef-y += swab.h
diff --git a/arch/cris/include/asm/byteorder.h b/arch/cris/include/asm/byteorder.h
index 7678d86317ae..bcd189798e26 100644
--- a/arch/cris/include/asm/byteorder.h
+++ b/arch/cris/include/asm/byteorder.h
@@ -1,7 +1,6 @@
 #ifndef _CRIS_BYTEORDER_H
 #define _CRIS_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/little_endian.h>
 
 #endif
diff --git a/arch/h8300/include/asm/Kbuild b/arch/h8300/include/asm/Kbuild
index 27b108a86b39..c68e1680da01 100644
--- a/arch/h8300/include/asm/Kbuild
+++ b/arch/h8300/include/asm/Kbuild
@@ -1,2 +1 @@
 include include/asm-generic/Kbuild.asm
-unifdef-y += swab.h
diff --git a/arch/h8300/include/asm/byteorder.h b/arch/h8300/include/asm/byteorder.h
index c36b80a3dd84..13539da99efd 100644
--- a/arch/h8300/include/asm/byteorder.h
+++ b/arch/h8300/include/asm/byteorder.h
@@ -1,7 +1,6 @@
 #ifndef _H8300_BYTEORDER_H
 #define _H8300_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/big_endian.h>
 
 #endif /* _H8300_BYTEORDER_H */
diff --git a/arch/ia64/include/asm/Kbuild b/arch/ia64/include/asm/Kbuild
index 3b25bd9dca91..ccbe8ae47a61 100644
--- a/arch/ia64/include/asm/Kbuild
+++ b/arch/ia64/include/asm/Kbuild
@@ -14,4 +14,3 @@ unifdef-y += gcc_intrin.h
 unifdef-y += intrinsics.h
 unifdef-y += perfmon.h
 unifdef-y += ustack.h
-unifdef-y += swab.h
diff --git a/arch/ia64/include/asm/byteorder.h b/arch/ia64/include/asm/byteorder.h
index 0f84c5cb703d..a8dd73558150 100644
--- a/arch/ia64/include/asm/byteorder.h
+++ b/arch/ia64/include/asm/byteorder.h
@@ -1,7 +1,6 @@
 #ifndef _ASM_IA64_BYTEORDER_H
 #define _ASM_IA64_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/little_endian.h>
 
 #endif /* _ASM_IA64_BYTEORDER_H */
diff --git a/arch/m68knommu/include/asm/Kbuild b/arch/m68knommu/include/asm/Kbuild
index 58c02a454130..c68e1680da01 100644
--- a/arch/m68knommu/include/asm/Kbuild
+++ b/arch/m68knommu/include/asm/Kbuild
@@ -1,3 +1 @@
 include include/asm-generic/Kbuild.asm
-
-unifdef-y += swab.h
diff --git a/arch/m68knommu/include/asm/byteorder.h b/arch/m68knommu/include/asm/byteorder.h
index a6f0b8f7f622..9c6c76a15041 100644
--- a/arch/m68knommu/include/asm/byteorder.h
+++ b/arch/m68knommu/include/asm/byteorder.h
@@ -1,7 +1,6 @@
 #ifndef _M68KNOMMU_BYTEORDER_H
 #define _M68KNOMMU_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/big_endian.h>
 
 #endif /* _M68KNOMMU_BYTEORDER_H */
diff --git a/arch/mips/include/asm/Kbuild b/arch/mips/include/asm/Kbuild
index 023866c0c102..7897f05e3165 100644
--- a/arch/mips/include/asm/Kbuild
+++ b/arch/mips/include/asm/Kbuild
@@ -1,4 +1,3 @@
 include include/asm-generic/Kbuild.asm
 
 header-y += cachectl.h sgidefs.h sysmips.h
-header-y += swab.h
diff --git a/arch/mips/include/asm/byteorder.h b/arch/mips/include/asm/byteorder.h
index 607b71830707..9579051ff1c7 100644
--- a/arch/mips/include/asm/byteorder.h
+++ b/arch/mips/include/asm/byteorder.h
@@ -8,8 +8,6 @@
 #ifndef _ASM_BYTEORDER_H
 #define _ASM_BYTEORDER_H
 
-#include <asm/swab.h>
-
 #if defined(__MIPSEB__)
 #include <linux/byteorder/big_endian.h>
 #elif defined(__MIPSEL__)
diff --git a/arch/parisc/include/asm/Kbuild b/arch/parisc/include/asm/Kbuild
index 2121d99f8364..f88b252e419c 100644
--- a/arch/parisc/include/asm/Kbuild
+++ b/arch/parisc/include/asm/Kbuild
@@ -1,4 +1,3 @@
 include include/asm-generic/Kbuild.asm
 
 unifdef-y += pdc.h
-unifdef-y += swab.h
diff --git a/arch/parisc/include/asm/byteorder.h b/arch/parisc/include/asm/byteorder.h
index da66029c4cb2..58af2c5f5d61 100644
--- a/arch/parisc/include/asm/byteorder.h
+++ b/arch/parisc/include/asm/byteorder.h
@@ -1,7 +1,6 @@
 #ifndef _PARISC_BYTEORDER_H
 #define _PARISC_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/big_endian.h>
 
 #endif /* _PARISC_BYTEORDER_H */
diff --git a/arch/powerpc/include/asm/Kbuild b/arch/powerpc/include/asm/Kbuild
index 9268602de5d0..5ab7d7fe198c 100644
--- a/arch/powerpc/include/asm/Kbuild
+++ b/arch/powerpc/include/asm/Kbuild
@@ -35,4 +35,3 @@ unifdef-y += spu_info.h
 unifdef-y += termios.h
 unifdef-y += types.h
 unifdef-y += unistd.h
-unifdef-y += swab.h
diff --git a/arch/powerpc/include/asm/byteorder.h b/arch/powerpc/include/asm/byteorder.h
index 5cca27a41532..aa6cc4fac965 100644
--- a/arch/powerpc/include/asm/byteorder.h
+++ b/arch/powerpc/include/asm/byteorder.h
@@ -7,8 +7,6 @@
  * as published by the Free Software Foundation; either version
  * 2 of the License, or (at your option) any later version.
  */
-
-#include <asm/swab.h>
 #include <linux/byteorder/big_endian.h>
 
 #endif /* _ASM_POWERPC_BYTEORDER_H */
diff --git a/arch/s390/include/asm/Kbuild b/arch/s390/include/asm/Kbuild
index f2af4167bd5f..63a23415fba6 100644
--- a/arch/s390/include/asm/Kbuild
+++ b/arch/s390/include/asm/Kbuild
@@ -13,4 +13,3 @@ unifdef-y += cmb.h
 unifdef-y += debug.h
 unifdef-y += chpid.h
 unifdef-y += schid.h
-unifdef-y += swab.h
diff --git a/arch/s390/include/asm/byteorder.h b/arch/s390/include/asm/byteorder.h
index b95a2b2933fb..a332e59e26fc 100644
--- a/arch/s390/include/asm/byteorder.h
+++ b/arch/s390/include/asm/byteorder.h
@@ -1,7 +1,6 @@
 #ifndef _S390_BYTEORDER_H
 #define _S390_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/big_endian.h>
 
 #endif /* _S390_BYTEORDER_H */
diff --git a/arch/sh/include/asm/Kbuild b/arch/sh/include/asm/Kbuild
index f1a2a0d1c79c..43910cdf78a5 100644
--- a/arch/sh/include/asm/Kbuild
+++ b/arch/sh/include/asm/Kbuild
@@ -6,4 +6,3 @@ unifdef-y += unistd_32.h
 unifdef-y += unistd_64.h
 unifdef-y += posix_types_32.h
 unifdef-y += posix_types_64.h
-unifdef-y += swab.h
diff --git a/arch/sh/include/asm/byteorder.h b/arch/sh/include/asm/byteorder.h
index e95c41a5c8cc..db2f5d7cb17d 100644
--- a/arch/sh/include/asm/byteorder.h
+++ b/arch/sh/include/asm/byteorder.h
@@ -1,8 +1,6 @@
 #ifndef __ASM_SH_BYTEORDER_H
 #define __ASM_SH_BYTEORDER_H
 
-#include <asm/swab.h>
-
 #ifdef __LITTLE_ENDIAN__
 #include <linux/byteorder/little_endian.h>
 #else
diff --git a/arch/sparc/include/asm/Kbuild b/arch/sparc/include/asm/Kbuild
index 95e38a43dff0..deeb0fba8029 100644
--- a/arch/sparc/include/asm/Kbuild
+++ b/arch/sparc/include/asm/Kbuild
@@ -17,4 +17,3 @@ header-y += traps.h
 header-y += uctx.h
 header-y += utrap.h
 header-y += watchdog.h
-header-y += swab.h
diff --git a/arch/sparc/include/asm/byteorder.h b/arch/sparc/include/asm/byteorder.h
index 48a047cd6fa9..ccc1b6b7de6c 100644
--- a/arch/sparc/include/asm/byteorder.h
+++ b/arch/sparc/include/asm/byteorder.h
@@ -1,7 +1,6 @@
 #ifndef _SPARC_BYTEORDER_H
 #define _SPARC_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/big_endian.h>
 
 #endif /* _SPARC_BYTEORDER_H */
diff --git a/arch/x86/include/asm/Kbuild b/arch/x86/include/asm/Kbuild
index a9f8a814a1f7..4a8e80cdcfa5 100644
--- a/arch/x86/include/asm/Kbuild
+++ b/arch/x86/include/asm/Kbuild
@@ -22,4 +22,3 @@ unifdef-y += unistd_32.h
 unifdef-y += unistd_64.h
 unifdef-y += vm86.h
 unifdef-y += vsyscall.h
-unifdef-y += swab.h
diff --git a/arch/x86/include/asm/byteorder.h b/arch/x86/include/asm/byteorder.h
index 7c49917e3d9d..b13a7a88f3eb 100644
--- a/arch/x86/include/asm/byteorder.h
+++ b/arch/x86/include/asm/byteorder.h
@@ -1,7 +1,6 @@
 #ifndef _ASM_X86_BYTEORDER_H
 #define _ASM_X86_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/little_endian.h>
 
 #endif /* _ASM_X86_BYTEORDER_H */
diff --git a/arch/xtensa/include/asm/Kbuild b/arch/xtensa/include/asm/Kbuild
index 58c02a454130..c68e1680da01 100644
--- a/arch/xtensa/include/asm/Kbuild
+++ b/arch/xtensa/include/asm/Kbuild
@@ -1,3 +1 @@
 include include/asm-generic/Kbuild.asm
-
-unifdef-y += swab.h
diff --git a/arch/xtensa/include/asm/byteorder.h b/arch/xtensa/include/asm/byteorder.h
index 329b94591ca4..54eb6315349c 100644
--- a/arch/xtensa/include/asm/byteorder.h
+++ b/arch/xtensa/include/asm/byteorder.h
@@ -1,8 +1,6 @@
 #ifndef _XTENSA_BYTEORDER_H
 #define _XTENSA_BYTEORDER_H
 
-#include <asm/swab.h>
-
 #ifdef __XTENSA_EL__
 #include <linux/byteorder/little_endian.h>
 #elif defined(__XTENSA_EB__)
diff --git a/include/asm-frv/Kbuild b/include/asm-frv/Kbuild
index 1f44e7c76995..0f8956def738 100644
--- a/include/asm-frv/Kbuild
+++ b/include/asm-frv/Kbuild
@@ -3,4 +3,3 @@ include include/asm-generic/Kbuild.asm
 header-y += registers.h
 
 unifdef-y += termios.h
-unifdef-y += swab.h
diff --git a/include/asm-frv/byteorder.h b/include/asm-frv/byteorder.h
index 1187e51ecd13..f29b7593e088 100644
--- a/include/asm-frv/byteorder.h
+++ b/include/asm-frv/byteorder.h
@@ -1,7 +1,6 @@
 #ifndef _ASM_BYTEORDER_H
 #define _ASM_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/big_endian.h>
 
 #endif /* _ASM_BYTEORDER_H */
diff --git a/include/asm-generic/Kbuild.asm b/include/asm-generic/Kbuild.asm
index 1870d5e05f1c..70d185534b9d 100644
--- a/include/asm-generic/Kbuild.asm
+++ b/include/asm-generic/Kbuild.asm
@@ -31,6 +31,7 @@ unifdef-y += socket.h
 unifdef-y += sockios.h
 unifdef-y += stat.h
 unifdef-y += statfs.h
+unifdef-y += swab.h
 unifdef-y += termbits.h
 unifdef-y += termios.h
 unifdef-y += types.h
diff --git a/include/asm-m32r/Kbuild b/include/asm-m32r/Kbuild
index 27b108a86b39..c68e1680da01 100644
--- a/include/asm-m32r/Kbuild
+++ b/include/asm-m32r/Kbuild
@@ -1,2 +1 @@
 include include/asm-generic/Kbuild.asm
-unifdef-y += swab.h
diff --git a/include/asm-m32r/byteorder.h b/include/asm-m32r/byteorder.h
index 61ff9cfd8451..21855d8b028b 100644
--- a/include/asm-m32r/byteorder.h
+++ b/include/asm-m32r/byteorder.h
@@ -1,8 +1,6 @@
 #ifndef _ASM_M32R_BYTEORDER_H
 #define _ASM_M32R_BYTEORDER_H
 
-#include <asm/swab.h>
-
 #if defined(__LITTLE_ENDIAN__)
 #  include <linux/byteorder/little_endian.h>
 #else
diff --git a/include/asm-m68k/Kbuild b/include/asm-m68k/Kbuild
index 52fd96b4142a..1a922fad76f7 100644
--- a/include/asm-m68k/Kbuild
+++ b/include/asm-m68k/Kbuild
@@ -1,3 +1,2 @@
 include include/asm-generic/Kbuild.asm
 header-y += cachectl.h
-unifdef-y += swab.h
diff --git a/include/asm-m68k/byteorder.h b/include/asm-m68k/byteorder.h
index 300866523b86..31b260a88803 100644
--- a/include/asm-m68k/byteorder.h
+++ b/include/asm-m68k/byteorder.h
@@ -1,7 +1,6 @@
 #ifndef _M68K_BYTEORDER_H
 #define _M68K_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/big_endian.h>
 
 #endif /* _M68K_BYTEORDER_H */
diff --git a/include/asm-mn10300/Kbuild b/include/asm-mn10300/Kbuild
index 27b108a86b39..c68e1680da01 100644
--- a/include/asm-mn10300/Kbuild
+++ b/include/asm-mn10300/Kbuild
@@ -1,2 +1 @@
 include include/asm-generic/Kbuild.asm
-unifdef-y += swab.h
diff --git a/include/asm-mn10300/byteorder.h b/include/asm-mn10300/byteorder.h
index 45b18ded19e6..5dd0bdd9feee 100644
--- a/include/asm-mn10300/byteorder.h
+++ b/include/asm-mn10300/byteorder.h
@@ -1,7 +1,6 @@
 #ifndef _ASM_BYTEORDER_H
 #define _ASM_BYTEORDER_H
 
-#include <asm/swab.h>
 #include <linux/byteorder/little_endian.h>
 
 #endif /* _ASM_BYTEORDER_H */
diff --git a/include/linux/swab.h b/include/linux/swab.h
index be5284d4a053..ea0c02fd5163 100644
--- a/include/linux/swab.h
+++ b/include/linux/swab.h
@@ -3,7 +3,7 @@
 
 #include <linux/types.h>
 #include <linux/compiler.h>
-#include <asm/byteorder.h>
+#include <asm/swab.h>
 
 /*
  * casts are necessary for constants, because we never know how for sure
-- 
cgit v1.2.3


From 937f1ba56b4be37d9e2ad77412f95048662058d2 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 14 Jan 2009 21:05:05 -0800
Subject: net: Add init_dummy_netdev() and fix EMAC driver using it

This adds an init_dummy_netdev() function that gets a network device
structure (allocation and lifetime entirely under caller's control) and
initialize the minimum amount of fields so it can be used to schedule
NAPI polls without registering a full blown interface. This is to be
used by drivers that need to tie several hardware interfaces to a single
NAPI poll scheduler due to HW limitations.

It also updates the ibm_newemac driver to use that, this fixing the
oops on 2.6.29 due to passing NULL as "dev" to netif_napi_add()

Symbol is exported GPL only a I don't think we want binary drivers doing
that sort of acrobatics (if we want them at all).

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Tested-by: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ibm_newemac/mal.c |  4 +++-
 drivers/net/ibm_newemac/mal.h |  2 ++
 include/linux/netdevice.h     |  3 +++
 net/core/dev.c                | 39 +++++++++++++++++++++++++++++++++++++++
 4 files changed, 47 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/drivers/net/ibm_newemac/mal.c b/drivers/net/ibm_newemac/mal.c
index ecf9798987fa..2a2fc17b2878 100644
--- a/drivers/net/ibm_newemac/mal.c
+++ b/drivers/net/ibm_newemac/mal.c
@@ -613,7 +613,9 @@ static int __devinit mal_probe(struct of_device *ofdev,
 	INIT_LIST_HEAD(&mal->list);
 	spin_lock_init(&mal->lock);
 
-	netif_napi_add(NULL, &mal->napi, mal_poll,
+	init_dummy_netdev(&mal->dummy_dev);
+
+	netif_napi_add(&mal->dummy_dev, &mal->napi, mal_poll,
 		       CONFIG_IBM_NEW_EMAC_POLL_WEIGHT);
 
 	/* Load power-on reset defaults */
diff --git a/drivers/net/ibm_newemac/mal.h b/drivers/net/ibm_newemac/mal.h
index 2f0a87360844..9ededfbf0726 100644
--- a/drivers/net/ibm_newemac/mal.h
+++ b/drivers/net/ibm_newemac/mal.h
@@ -214,6 +214,8 @@ struct mal_instance {
 	int			index;
 	spinlock_t		lock;
 
+	struct net_device	dummy_dev;
+
 	unsigned int features;
 };
 
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 4647604c7ca9..ec54785d34f9 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -795,6 +795,7 @@ struct net_device
 	       NETREG_UNREGISTERING,	/* called unregister_netdevice */
 	       NETREG_UNREGISTERED,	/* completed unregister todo */
 	       NETREG_RELEASED,		/* called free_netdev */
+	       NETREG_DUMMY,		/* dummy device for NAPI poll */
 	} reg_state;
 
 	/* Called from unregister, can be used to call free_netdev */
@@ -1077,6 +1078,8 @@ extern void		free_netdev(struct net_device *dev);
 extern void		synchronize_net(void);
 extern int 		register_netdevice_notifier(struct notifier_block *nb);
 extern int		unregister_netdevice_notifier(struct notifier_block *nb);
+extern int		init_dummy_netdev(struct net_device *dev);
+
 extern int call_netdevice_notifiers(unsigned long val, struct net_device *dev);
 extern struct net_device	*dev_get_by_index(struct net *net, int ifindex);
 extern struct net_device	*__dev_get_by_index(struct net *net, int ifindex);
diff --git a/net/core/dev.c b/net/core/dev.c
index 60377b6c0a80..8d675975d85b 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4430,6 +4430,45 @@ err_uninit:
 	goto out;
 }
 
+/**
+ *	init_dummy_netdev	- init a dummy network device for NAPI
+ *	@dev: device to init
+ *
+ *	This takes a network device structure and initialize the minimum
+ *	amount of fields so it can be used to schedule NAPI polls without
+ *	registering a full blown interface. This is to be used by drivers
+ *	that need to tie several hardware interfaces to a single NAPI
+ *	poll scheduler due to HW limitations.
+ */
+int init_dummy_netdev(struct net_device *dev)
+{
+	/* Clear everything. Note we don't initialize spinlocks
+	 * are they aren't supposed to be taken by any of the
+	 * NAPI code and this dummy netdev is supposed to be
+	 * only ever used for NAPI polls
+	 */
+	memset(dev, 0, sizeof(struct net_device));
+
+	/* make sure we BUG if trying to hit standard
+	 * register/unregister code path
+	 */
+	dev->reg_state = NETREG_DUMMY;
+
+	/* initialize the ref count */
+	atomic_set(&dev->refcnt, 1);
+
+	/* NAPI wants this */
+	INIT_LIST_HEAD(&dev->napi_list);
+
+	/* a dummy interface is started by default */
+	set_bit(__LINK_STATE_PRESENT, &dev->state);
+	set_bit(__LINK_STATE_START, &dev->state);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(init_dummy_netdev);
+
+
 /**
  *	register_netdev	- register a network device
  *	@dev: device to register
-- 
cgit v1.2.3


From 45ce80fb6b6f9594d1396d44dd7e7c02d596fef8 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Thu, 15 Jan 2009 13:50:59 -0800
Subject: cgroups: consolidate cgroup documents

Move Documentation/cpusets.txt and Documentation/controllers/* to
Documentation/cgroups/

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/cgroups/cgroups.txt              |   5 +-
 Documentation/cgroups/cpuacct.txt              |  32 +
 Documentation/cgroups/cpusets.txt              | 808 +++++++++++++++++++++++++
 Documentation/cgroups/devices.txt              |  52 ++
 Documentation/cgroups/memcg_test.txt           | 342 +++++++++++
 Documentation/cgroups/memory.txt               | 399 ++++++++++++
 Documentation/cgroups/resource_counter.txt     | 181 ++++++
 Documentation/controllers/cpuacct.txt          |  32 -
 Documentation/controllers/devices.txt          |  52 --
 Documentation/controllers/memcg_test.txt       | 342 -----------
 Documentation/controllers/memory.txt           | 399 ------------
 Documentation/controllers/resource_counter.txt | 181 ------
 Documentation/cpusets.txt                      | 808 -------------------------
 Documentation/scheduler/sched-design-CFS.txt   |   2 +-
 include/linux/res_counter.h                    |   2 +-
 init/Kconfig                                   |   9 +-
 kernel/cpuset.c                                |   2 +-
 17 files changed, 1824 insertions(+), 1824 deletions(-)
 create mode 100644 Documentation/cgroups/cpuacct.txt
 create mode 100644 Documentation/cgroups/cpusets.txt
 create mode 100644 Documentation/cgroups/devices.txt
 create mode 100644 Documentation/cgroups/memcg_test.txt
 create mode 100644 Documentation/cgroups/memory.txt
 create mode 100644 Documentation/cgroups/resource_counter.txt
 delete mode 100644 Documentation/controllers/cpuacct.txt
 delete mode 100644 Documentation/controllers/devices.txt
 delete mode 100644 Documentation/controllers/memcg_test.txt
 delete mode 100644 Documentation/controllers/memory.txt
 delete mode 100644 Documentation/controllers/resource_counter.txt
 delete mode 100644 Documentation/cpusets.txt

(limited to 'include/linux')

diff --git a/Documentation/cgroups/cgroups.txt b/Documentation/cgroups/cgroups.txt
index e33ee74eee77..d9e5d6f41b92 100644
--- a/Documentation/cgroups/cgroups.txt
+++ b/Documentation/cgroups/cgroups.txt
@@ -1,7 +1,8 @@
 				CGROUPS
 				-------
 
-Written by Paul Menage <menage@google.com> based on Documentation/cpusets.txt
+Written by Paul Menage <menage@google.com> based on
+Documentation/cgroups/cpusets.txt
 
 Original copyright statements from cpusets.txt:
 Portions Copyright (C) 2004 BULL SA.
@@ -68,7 +69,7 @@ On their own, the only use for cgroups is for simple job
 tracking. The intention is that other subsystems hook into the generic
 cgroup support to provide new attributes for cgroups, such as
 accounting/limiting the resources which processes in a cgroup can
-access. For example, cpusets (see Documentation/cpusets.txt) allows
+access. For example, cpusets (see Documentation/cgroups/cpusets.txt) allows
 you to associate a set of CPUs and a set of memory nodes with the
 tasks in each cgroup.
 
diff --git a/Documentation/cgroups/cpuacct.txt b/Documentation/cgroups/cpuacct.txt
new file mode 100644
index 000000000000..bb775fbe43d7
--- /dev/null
+++ b/Documentation/cgroups/cpuacct.txt
@@ -0,0 +1,32 @@
+CPU Accounting Controller
+-------------------------
+
+The CPU accounting controller is used to group tasks using cgroups and
+account the CPU usage of these groups of tasks.
+
+The CPU accounting controller supports multi-hierarchy groups. An accounting
+group accumulates the CPU usage of all of its child groups and the tasks
+directly present in its group.
+
+Accounting groups can be created by first mounting the cgroup filesystem.
+
+# mkdir /cgroups
+# mount -t cgroup -ocpuacct none /cgroups
+
+With the above step, the initial or the parent accounting group
+becomes visible at /cgroups. At bootup, this group includes all the
+tasks in the system. /cgroups/tasks lists the tasks in this cgroup.
+/cgroups/cpuacct.usage gives the CPU time (in nanoseconds) obtained by
+this group which is essentially the CPU time obtained by all the tasks
+in the system.
+
+New accounting groups can be created under the parent group /cgroups.
+
+# cd /cgroups
+# mkdir g1
+# echo $$ > g1
+
+The above steps create a new group g1 and move the current shell
+process (bash) into it. CPU time consumed by this bash and its children
+can be obtained from g1/cpuacct.usage and the same is accumulated in
+/cgroups/cpuacct.usage also.
diff --git a/Documentation/cgroups/cpusets.txt b/Documentation/cgroups/cpusets.txt
new file mode 100644
index 000000000000..5c86c258c791
--- /dev/null
+++ b/Documentation/cgroups/cpusets.txt
@@ -0,0 +1,808 @@
+				CPUSETS
+				-------
+
+Copyright (C) 2004 BULL SA.
+Written by Simon.Derr@bull.net
+
+Portions Copyright (c) 2004-2006 Silicon Graphics, Inc.
+Modified by Paul Jackson <pj@sgi.com>
+Modified by Christoph Lameter <clameter@sgi.com>
+Modified by Paul Menage <menage@google.com>
+Modified by Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
+
+CONTENTS:
+=========
+
+1. Cpusets
+  1.1 What are cpusets ?
+  1.2 Why are cpusets needed ?
+  1.3 How are cpusets implemented ?
+  1.4 What are exclusive cpusets ?
+  1.5 What is memory_pressure ?
+  1.6 What is memory spread ?
+  1.7 What is sched_load_balance ?
+  1.8 What is sched_relax_domain_level ?
+  1.9 How do I use cpusets ?
+2. Usage Examples and Syntax
+  2.1 Basic Usage
+  2.2 Adding/removing cpus
+  2.3 Setting flags
+  2.4 Attaching processes
+3. Questions
+4. Contact
+
+1. Cpusets
+==========
+
+1.1 What are cpusets ?
+----------------------
+
+Cpusets provide a mechanism for assigning a set of CPUs and Memory
+Nodes to a set of tasks.   In this document "Memory Node" refers to
+an on-line node that contains memory.
+
+Cpusets constrain the CPU and Memory placement of tasks to only
+the resources within a tasks current cpuset.  They form a nested
+hierarchy visible in a virtual file system.  These are the essential
+hooks, beyond what is already present, required to manage dynamic
+job placement on large systems.
+
+Cpusets use the generic cgroup subsystem described in
+Documentation/cgroups/cgroups.txt.
+
+Requests by a task, using the sched_setaffinity(2) system call to
+include CPUs in its CPU affinity mask, and using the mbind(2) and
+set_mempolicy(2) system calls to include Memory Nodes in its memory
+policy, are both filtered through that tasks cpuset, filtering out any
+CPUs or Memory Nodes not in that cpuset.  The scheduler will not
+schedule a task on a CPU that is not allowed in its cpus_allowed
+vector, and the kernel page allocator will not allocate a page on a
+node that is not allowed in the requesting tasks mems_allowed vector.
+
+User level code may create and destroy cpusets by name in the cgroup
+virtual file system, manage the attributes and permissions of these
+cpusets and which CPUs and Memory Nodes are assigned to each cpuset,
+specify and query to which cpuset a task is assigned, and list the
+task pids assigned to a cpuset.
+
+
+1.2 Why are cpusets needed ?
+----------------------------
+
+The management of large computer systems, with many processors (CPUs),
+complex memory cache hierarchies and multiple Memory Nodes having
+non-uniform access times (NUMA) presents additional challenges for
+the efficient scheduling and memory placement of processes.
+
+Frequently more modest sized systems can be operated with adequate
+efficiency just by letting the operating system automatically share
+the available CPU and Memory resources amongst the requesting tasks.
+
+But larger systems, which benefit more from careful processor and
+memory placement to reduce memory access times and contention,
+and which typically represent a larger investment for the customer,
+can benefit from explicitly placing jobs on properly sized subsets of
+the system.
+
+This can be especially valuable on:
+
+    * Web Servers running multiple instances of the same web application,
+    * Servers running different applications (for instance, a web server
+      and a database), or
+    * NUMA systems running large HPC applications with demanding
+      performance characteristics.
+
+These subsets, or "soft partitions" must be able to be dynamically
+adjusted, as the job mix changes, without impacting other concurrently
+executing jobs. The location of the running jobs pages may also be moved
+when the memory locations are changed.
+
+The kernel cpuset patch provides the minimum essential kernel
+mechanisms required to efficiently implement such subsets.  It
+leverages existing CPU and Memory Placement facilities in the Linux
+kernel to avoid any additional impact on the critical scheduler or
+memory allocator code.
+
+
+1.3 How are cpusets implemented ?
+---------------------------------
+
+Cpusets provide a Linux kernel mechanism to constrain which CPUs and
+Memory Nodes are used by a process or set of processes.
+
+The Linux kernel already has a pair of mechanisms to specify on which
+CPUs a task may be scheduled (sched_setaffinity) and on which Memory
+Nodes it may obtain memory (mbind, set_mempolicy).
+
+Cpusets extends these two mechanisms as follows:
+
+ - Cpusets are sets of allowed CPUs and Memory Nodes, known to the
+   kernel.
+ - Each task in the system is attached to a cpuset, via a pointer
+   in the task structure to a reference counted cgroup structure.
+ - Calls to sched_setaffinity are filtered to just those CPUs
+   allowed in that tasks cpuset.
+ - Calls to mbind and set_mempolicy are filtered to just
+   those Memory Nodes allowed in that tasks cpuset.
+ - The root cpuset contains all the systems CPUs and Memory
+   Nodes.
+ - For any cpuset, one can define child cpusets containing a subset
+   of the parents CPU and Memory Node resources.
+ - The hierarchy of cpusets can be mounted at /dev/cpuset, for
+   browsing and manipulation from user space.
+ - A cpuset may be marked exclusive, which ensures that no other
+   cpuset (except direct ancestors and descendents) may contain
+   any overlapping CPUs or Memory Nodes.
+ - You can list all the tasks (by pid) attached to any cpuset.
+
+The implementation of cpusets requires a few, simple hooks
+into the rest of the kernel, none in performance critical paths:
+
+ - in init/main.c, to initialize the root cpuset at system boot.
+ - in fork and exit, to attach and detach a task from its cpuset.
+ - in sched_setaffinity, to mask the requested CPUs by what's
+   allowed in that tasks cpuset.
+ - in sched.c migrate_all_tasks(), to keep migrating tasks within
+   the CPUs allowed by their cpuset, if possible.
+ - in the mbind and set_mempolicy system calls, to mask the requested
+   Memory Nodes by what's allowed in that tasks cpuset.
+ - in page_alloc.c, to restrict memory to allowed nodes.
+ - in vmscan.c, to restrict page recovery to the current cpuset.
+
+You should mount the "cgroup" filesystem type in order to enable
+browsing and modifying the cpusets presently known to the kernel.  No
+new system calls are added for cpusets - all support for querying and
+modifying cpusets is via this cpuset file system.
+
+The /proc/<pid>/status file for each task has four added lines,
+displaying the tasks cpus_allowed (on which CPUs it may be scheduled)
+and mems_allowed (on which Memory Nodes it may obtain memory),
+in the two formats seen in the following example:
+
+  Cpus_allowed:   ffffffff,ffffffff,ffffffff,ffffffff
+  Cpus_allowed_list:      0-127
+  Mems_allowed:   ffffffff,ffffffff
+  Mems_allowed_list:      0-63
+
+Each cpuset is represented by a directory in the cgroup file system
+containing (on top of the standard cgroup files) the following
+files describing that cpuset:
+
+ - cpus: list of CPUs in that cpuset
+ - mems: list of Memory Nodes in that cpuset
+ - memory_migrate flag: if set, move pages to cpusets nodes
+ - cpu_exclusive flag: is cpu placement exclusive?
+ - mem_exclusive flag: is memory placement exclusive?
+ - mem_hardwall flag:  is memory allocation hardwalled
+ - memory_pressure: measure of how much paging pressure in cpuset
+
+In addition, the root cpuset only has the following file:
+ - memory_pressure_enabled flag: compute memory_pressure?
+
+New cpusets are created using the mkdir system call or shell
+command.  The properties of a cpuset, such as its flags, allowed
+CPUs and Memory Nodes, and attached tasks, are modified by writing
+to the appropriate file in that cpusets directory, as listed above.
+
+The named hierarchical structure of nested cpusets allows partitioning
+a large system into nested, dynamically changeable, "soft-partitions".
+
+The attachment of each task, automatically inherited at fork by any
+children of that task, to a cpuset allows organizing the work load
+on a system into related sets of tasks such that each set is constrained
+to using the CPUs and Memory Nodes of a particular cpuset.  A task
+may be re-attached to any other cpuset, if allowed by the permissions
+on the necessary cpuset file system directories.
+
+Such management of a system "in the large" integrates smoothly with
+the detailed placement done on individual tasks and memory regions
+using the sched_setaffinity, mbind and set_mempolicy system calls.
+
+The following rules apply to each cpuset:
+
+ - Its CPUs and Memory Nodes must be a subset of its parents.
+ - It can't be marked exclusive unless its parent is.
+ - If its cpu or memory is exclusive, they may not overlap any sibling.
+
+These rules, and the natural hierarchy of cpusets, enable efficient
+enforcement of the exclusive guarantee, without having to scan all
+cpusets every time any of them change to ensure nothing overlaps a
+exclusive cpuset.  Also, the use of a Linux virtual file system (vfs)
+to represent the cpuset hierarchy provides for a familiar permission
+and name space for cpusets, with a minimum of additional kernel code.
+
+The cpus and mems files in the root (top_cpuset) cpuset are
+read-only.  The cpus file automatically tracks the value of
+cpu_online_map using a CPU hotplug notifier, and the mems file
+automatically tracks the value of node_states[N_HIGH_MEMORY]--i.e.,
+nodes with memory--using the cpuset_track_online_nodes() hook.
+
+
+1.4 What are exclusive cpusets ?
+--------------------------------
+
+If a cpuset is cpu or mem exclusive, no other cpuset, other than
+a direct ancestor or descendent, may share any of the same CPUs or
+Memory Nodes.
+
+A cpuset that is mem_exclusive *or* mem_hardwall is "hardwalled",
+i.e. it restricts kernel allocations for page, buffer and other data
+commonly shared by the kernel across multiple users.  All cpusets,
+whether hardwalled or not, restrict allocations of memory for user
+space.  This enables configuring a system so that several independent
+jobs can share common kernel data, such as file system pages, while
+isolating each job's user allocation in its own cpuset.  To do this,
+construct a large mem_exclusive cpuset to hold all the jobs, and
+construct child, non-mem_exclusive cpusets for each individual job.
+Only a small amount of typical kernel memory, such as requests from
+interrupt handlers, is allowed to be taken outside even a
+mem_exclusive cpuset.
+
+
+1.5 What is memory_pressure ?
+-----------------------------
+The memory_pressure of a cpuset provides a simple per-cpuset metric
+of the rate that the tasks in a cpuset are attempting to free up in
+use memory on the nodes of the cpuset to satisfy additional memory
+requests.
+
+This enables batch managers monitoring jobs running in dedicated
+cpusets to efficiently detect what level of memory pressure that job
+is causing.
+
+This is useful both on tightly managed systems running a wide mix of
+submitted jobs, which may choose to terminate or re-prioritize jobs that
+are trying to use more memory than allowed on the nodes assigned them,
+and with tightly coupled, long running, massively parallel scientific
+computing jobs that will dramatically fail to meet required performance
+goals if they start to use more memory than allowed to them.
+
+This mechanism provides a very economical way for the batch manager
+to monitor a cpuset for signs of memory pressure.  It's up to the
+batch manager or other user code to decide what to do about it and
+take action.
+
+==> Unless this feature is enabled by writing "1" to the special file
+    /dev/cpuset/memory_pressure_enabled, the hook in the rebalance
+    code of __alloc_pages() for this metric reduces to simply noticing
+    that the cpuset_memory_pressure_enabled flag is zero.  So only
+    systems that enable this feature will compute the metric.
+
+Why a per-cpuset, running average:
+
+    Because this meter is per-cpuset, rather than per-task or mm,
+    the system load imposed by a batch scheduler monitoring this
+    metric is sharply reduced on large systems, because a scan of
+    the tasklist can be avoided on each set of queries.
+
+    Because this meter is a running average, instead of an accumulating
+    counter, a batch scheduler can detect memory pressure with a
+    single read, instead of having to read and accumulate results
+    for a period of time.
+
+    Because this meter is per-cpuset rather than per-task or mm,
+    the batch scheduler can obtain the key information, memory
+    pressure in a cpuset, with a single read, rather than having to
+    query and accumulate results over all the (dynamically changing)
+    set of tasks in the cpuset.
+
+A per-cpuset simple digital filter (requires a spinlock and 3 words
+of data per-cpuset) is kept, and updated by any task attached to that
+cpuset, if it enters the synchronous (direct) page reclaim code.
+
+A per-cpuset file provides an integer number representing the recent
+(half-life of 10 seconds) rate of direct page reclaims caused by
+the tasks in the cpuset, in units of reclaims attempted per second,
+times 1000.
+
+
+1.6 What is memory spread ?
+---------------------------
+There are two boolean flag files per cpuset that control where the
+kernel allocates pages for the file system buffers and related in
+kernel data structures.  They are called 'memory_spread_page' and
+'memory_spread_slab'.
+
+If the per-cpuset boolean flag file 'memory_spread_page' is set, then
+the kernel will spread the file system buffers (page cache) evenly
+over all the nodes that the faulting task is allowed to use, instead
+of preferring to put those pages on the node where the task is running.
+
+If the per-cpuset boolean flag file 'memory_spread_slab' is set,
+then the kernel will spread some file system related slab caches,
+such as for inodes and dentries evenly over all the nodes that the
+faulting task is allowed to use, instead of preferring to put those
+pages on the node where the task is running.
+
+The setting of these flags does not affect anonymous data segment or
+stack segment pages of a task.
+
+By default, both kinds of memory spreading are off, and memory
+pages are allocated on the node local to where the task is running,
+except perhaps as modified by the tasks NUMA mempolicy or cpuset
+configuration, so long as sufficient free memory pages are available.
+
+When new cpusets are created, they inherit the memory spread settings
+of their parent.
+
+Setting memory spreading causes allocations for the affected page
+or slab caches to ignore the tasks NUMA mempolicy and be spread
+instead.    Tasks using mbind() or set_mempolicy() calls to set NUMA
+mempolicies will not notice any change in these calls as a result of
+their containing tasks memory spread settings.  If memory spreading
+is turned off, then the currently specified NUMA mempolicy once again
+applies to memory page allocations.
+
+Both 'memory_spread_page' and 'memory_spread_slab' are boolean flag
+files.  By default they contain "0", meaning that the feature is off
+for that cpuset.  If a "1" is written to that file, then that turns
+the named feature on.
+
+The implementation is simple.
+
+Setting the flag 'memory_spread_page' turns on a per-process flag
+PF_SPREAD_PAGE for each task that is in that cpuset or subsequently
+joins that cpuset.  The page allocation calls for the page cache
+is modified to perform an inline check for this PF_SPREAD_PAGE task
+flag, and if set, a call to a new routine cpuset_mem_spread_node()
+returns the node to prefer for the allocation.
+
+Similarly, setting 'memory_spread_slab' turns on the flag
+PF_SPREAD_SLAB, and appropriately marked slab caches will allocate
+pages from the node returned by cpuset_mem_spread_node().
+
+The cpuset_mem_spread_node() routine is also simple.  It uses the
+value of a per-task rotor cpuset_mem_spread_rotor to select the next
+node in the current tasks mems_allowed to prefer for the allocation.
+
+This memory placement policy is also known (in other contexts) as
+round-robin or interleave.
+
+This policy can provide substantial improvements for jobs that need
+to place thread local data on the corresponding node, but that need
+to access large file system data sets that need to be spread across
+the several nodes in the jobs cpuset in order to fit.  Without this
+policy, especially for jobs that might have one thread reading in the
+data set, the memory allocation across the nodes in the jobs cpuset
+can become very uneven.
+
+1.7 What is sched_load_balance ?
+--------------------------------
+
+The kernel scheduler (kernel/sched.c) automatically load balances
+tasks.  If one CPU is underutilized, kernel code running on that
+CPU will look for tasks on other more overloaded CPUs and move those
+tasks to itself, within the constraints of such placement mechanisms
+as cpusets and sched_setaffinity.
+
+The algorithmic cost of load balancing and its impact on key shared
+kernel data structures such as the task list increases more than
+linearly with the number of CPUs being balanced.  So the scheduler
+has support to  partition the systems CPUs into a number of sched
+domains such that it only load balances within each sched domain.
+Each sched domain covers some subset of the CPUs in the system;
+no two sched domains overlap; some CPUs might not be in any sched
+domain and hence won't be load balanced.
+
+Put simply, it costs less to balance between two smaller sched domains
+than one big one, but doing so means that overloads in one of the
+two domains won't be load balanced to the other one.
+
+By default, there is one sched domain covering all CPUs, except those
+marked isolated using the kernel boot time "isolcpus=" argument.
+
+This default load balancing across all CPUs is not well suited for
+the following two situations:
+ 1) On large systems, load balancing across many CPUs is expensive.
+    If the system is managed using cpusets to place independent jobs
+    on separate sets of CPUs, full load balancing is unnecessary.
+ 2) Systems supporting realtime on some CPUs need to minimize
+    system overhead on those CPUs, including avoiding task load
+    balancing if that is not needed.
+
+When the per-cpuset flag "sched_load_balance" is enabled (the default
+setting), it requests that all the CPUs in that cpusets allowed 'cpus'
+be contained in a single sched domain, ensuring that load balancing
+can move a task (not otherwised pinned, as by sched_setaffinity)
+from any CPU in that cpuset to any other.
+
+When the per-cpuset flag "sched_load_balance" is disabled, then the
+scheduler will avoid load balancing across the CPUs in that cpuset,
+--except-- in so far as is necessary because some overlapping cpuset
+has "sched_load_balance" enabled.
+
+So, for example, if the top cpuset has the flag "sched_load_balance"
+enabled, then the scheduler will have one sched domain covering all
+CPUs, and the setting of the "sched_load_balance" flag in any other
+cpusets won't matter, as we're already fully load balancing.
+
+Therefore in the above two situations, the top cpuset flag
+"sched_load_balance" should be disabled, and only some of the smaller,
+child cpusets have this flag enabled.
+
+When doing this, you don't usually want to leave any unpinned tasks in
+the top cpuset that might use non-trivial amounts of CPU, as such tasks
+may be artificially constrained to some subset of CPUs, depending on
+the particulars of this flag setting in descendent cpusets.  Even if
+such a task could use spare CPU cycles in some other CPUs, the kernel
+scheduler might not consider the possibility of load balancing that
+task to that underused CPU.
+
+Of course, tasks pinned to a particular CPU can be left in a cpuset
+that disables "sched_load_balance" as those tasks aren't going anywhere
+else anyway.
+
+There is an impedance mismatch here, between cpusets and sched domains.
+Cpusets are hierarchical and nest.  Sched domains are flat; they don't
+overlap and each CPU is in at most one sched domain.
+
+It is necessary for sched domains to be flat because load balancing
+across partially overlapping sets of CPUs would risk unstable dynamics
+that would be beyond our understanding.  So if each of two partially
+overlapping cpusets enables the flag 'sched_load_balance', then we
+form a single sched domain that is a superset of both.  We won't move
+a task to a CPU outside it cpuset, but the scheduler load balancing
+code might waste some compute cycles considering that possibility.
+
+This mismatch is why there is not a simple one-to-one relation
+between which cpusets have the flag "sched_load_balance" enabled,
+and the sched domain configuration.  If a cpuset enables the flag, it
+will get balancing across all its CPUs, but if it disables the flag,
+it will only be assured of no load balancing if no other overlapping
+cpuset enables the flag.
+
+If two cpusets have partially overlapping 'cpus' allowed, and only
+one of them has this flag enabled, then the other may find its
+tasks only partially load balanced, just on the overlapping CPUs.
+This is just the general case of the top_cpuset example given a few
+paragraphs above.  In the general case, as in the top cpuset case,
+don't leave tasks that might use non-trivial amounts of CPU in
+such partially load balanced cpusets, as they may be artificially
+constrained to some subset of the CPUs allowed to them, for lack of
+load balancing to the other CPUs.
+
+1.7.1 sched_load_balance implementation details.
+------------------------------------------------
+
+The per-cpuset flag 'sched_load_balance' defaults to enabled (contrary
+to most cpuset flags.)  When enabled for a cpuset, the kernel will
+ensure that it can load balance across all the CPUs in that cpuset
+(makes sure that all the CPUs in the cpus_allowed of that cpuset are
+in the same sched domain.)
+
+If two overlapping cpusets both have 'sched_load_balance' enabled,
+then they will be (must be) both in the same sched domain.
+
+If, as is the default, the top cpuset has 'sched_load_balance' enabled,
+then by the above that means there is a single sched domain covering
+the whole system, regardless of any other cpuset settings.
+
+The kernel commits to user space that it will avoid load balancing
+where it can.  It will pick as fine a granularity partition of sched
+domains as it can while still providing load balancing for any set
+of CPUs allowed to a cpuset having 'sched_load_balance' enabled.
+
+The internal kernel cpuset to scheduler interface passes from the
+cpuset code to the scheduler code a partition of the load balanced
+CPUs in the system. This partition is a set of subsets (represented
+as an array of cpumask_t) of CPUs, pairwise disjoint, that cover all
+the CPUs that must be load balanced.
+
+Whenever the 'sched_load_balance' flag changes, or CPUs come or go
+from a cpuset with this flag enabled, or a cpuset with this flag
+enabled is removed, the cpuset code builds a new such partition and
+passes it to the scheduler sched domain setup code, to have the sched
+domains rebuilt as necessary.
+
+This partition exactly defines what sched domains the scheduler should
+setup - one sched domain for each element (cpumask_t) in the partition.
+
+The scheduler remembers the currently active sched domain partitions.
+When the scheduler routine partition_sched_domains() is invoked from
+the cpuset code to update these sched domains, it compares the new
+partition requested with the current, and updates its sched domains,
+removing the old and adding the new, for each change.
+
+
+1.8 What is sched_relax_domain_level ?
+--------------------------------------
+
+In sched domain, the scheduler migrates tasks in 2 ways; periodic load
+balance on tick, and at time of some schedule events.
+
+When a task is woken up, scheduler try to move the task on idle CPU.
+For example, if a task A running on CPU X activates another task B
+on the same CPU X, and if CPU Y is X's sibling and performing idle,
+then scheduler migrate task B to CPU Y so that task B can start on
+CPU Y without waiting task A on CPU X.
+
+And if a CPU run out of tasks in its runqueue, the CPU try to pull
+extra tasks from other busy CPUs to help them before it is going to
+be idle.
+
+Of course it takes some searching cost to find movable tasks and/or
+idle CPUs, the scheduler might not search all CPUs in the domain
+everytime.  In fact, in some architectures, the searching ranges on
+events are limited in the same socket or node where the CPU locates,
+while the load balance on tick searchs all.
+
+For example, assume CPU Z is relatively far from CPU X.  Even if CPU Z
+is idle while CPU X and the siblings are busy, scheduler can't migrate
+woken task B from X to Z since it is out of its searching range.
+As the result, task B on CPU X need to wait task A or wait load balance
+on the next tick.  For some applications in special situation, waiting
+1 tick may be too long.
+
+The 'sched_relax_domain_level' file allows you to request changing
+this searching range as you like.  This file takes int value which
+indicates size of searching range in levels ideally as follows,
+otherwise initial value -1 that indicates the cpuset has no request.
+
+  -1  : no request. use system default or follow request of others.
+   0  : no search.
+   1  : search siblings (hyperthreads in a core).
+   2  : search cores in a package.
+   3  : search cpus in a node [= system wide on non-NUMA system]
+ ( 4  : search nodes in a chunk of node [on NUMA system] )
+ ( 5  : search system wide [on NUMA system] )
+
+The system default is architecture dependent.  The system default
+can be changed using the relax_domain_level= boot parameter.
+
+This file is per-cpuset and affect the sched domain where the cpuset
+belongs to.  Therefore if the flag 'sched_load_balance' of a cpuset
+is disabled, then 'sched_relax_domain_level' have no effect since
+there is no sched domain belonging the cpuset.
+
+If multiple cpusets are overlapping and hence they form a single sched
+domain, the largest value among those is used.  Be careful, if one
+requests 0 and others are -1 then 0 is used.
+
+Note that modifying this file will have both good and bad effects,
+and whether it is acceptable or not will be depend on your situation.
+Don't modify this file if you are not sure.
+
+If your situation is:
+ - The migration costs between each cpu can be assumed considerably
+   small(for you) due to your special application's behavior or
+   special hardware support for CPU cache etc.
+ - The searching cost doesn't have impact(for you) or you can make
+   the searching cost enough small by managing cpuset to compact etc.
+ - The latency is required even it sacrifices cache hit rate etc.
+then increasing 'sched_relax_domain_level' would benefit you.
+
+
+1.9 How do I use cpusets ?
+--------------------------
+
+In order to minimize the impact of cpusets on critical kernel
+code, such as the scheduler, and due to the fact that the kernel
+does not support one task updating the memory placement of another
+task directly, the impact on a task of changing its cpuset CPU
+or Memory Node placement, or of changing to which cpuset a task
+is attached, is subtle.
+
+If a cpuset has its Memory Nodes modified, then for each task attached
+to that cpuset, the next time that the kernel attempts to allocate
+a page of memory for that task, the kernel will notice the change
+in the tasks cpuset, and update its per-task memory placement to
+remain within the new cpusets memory placement.  If the task was using
+mempolicy MPOL_BIND, and the nodes to which it was bound overlap with
+its new cpuset, then the task will continue to use whatever subset
+of MPOL_BIND nodes are still allowed in the new cpuset.  If the task
+was using MPOL_BIND and now none of its MPOL_BIND nodes are allowed
+in the new cpuset, then the task will be essentially treated as if it
+was MPOL_BIND bound to the new cpuset (even though its numa placement,
+as queried by get_mempolicy(), doesn't change).  If a task is moved
+from one cpuset to another, then the kernel will adjust the tasks
+memory placement, as above, the next time that the kernel attempts
+to allocate a page of memory for that task.
+
+If a cpuset has its 'cpus' modified, then each task in that cpuset
+will have its allowed CPU placement changed immediately.  Similarly,
+if a tasks pid is written to a cpusets 'tasks' file, in either its
+current cpuset or another cpuset, then its allowed CPU placement is
+changed immediately.  If such a task had been bound to some subset
+of its cpuset using the sched_setaffinity() call, the task will be
+allowed to run on any CPU allowed in its new cpuset, negating the
+affect of the prior sched_setaffinity() call.
+
+In summary, the memory placement of a task whose cpuset is changed is
+updated by the kernel, on the next allocation of a page for that task,
+but the processor placement is not updated, until that tasks pid is
+rewritten to the 'tasks' file of its cpuset.  This is done to avoid
+impacting the scheduler code in the kernel with a check for changes
+in a tasks processor placement.
+
+Normally, once a page is allocated (given a physical page
+of main memory) then that page stays on whatever node it
+was allocated, so long as it remains allocated, even if the
+cpusets memory placement policy 'mems' subsequently changes.
+If the cpuset flag file 'memory_migrate' is set true, then when
+tasks are attached to that cpuset, any pages that task had
+allocated to it on nodes in its previous cpuset are migrated
+to the tasks new cpuset. The relative placement of the page within
+the cpuset is preserved during these migration operations if possible.
+For example if the page was on the second valid node of the prior cpuset
+then the page will be placed on the second valid node of the new cpuset.
+
+Also if 'memory_migrate' is set true, then if that cpusets
+'mems' file is modified, pages allocated to tasks in that
+cpuset, that were on nodes in the previous setting of 'mems',
+will be moved to nodes in the new setting of 'mems.'
+Pages that were not in the tasks prior cpuset, or in the cpusets
+prior 'mems' setting, will not be moved.
+
+There is an exception to the above.  If hotplug functionality is used
+to remove all the CPUs that are currently assigned to a cpuset,
+then all the tasks in that cpuset will be moved to the nearest ancestor
+with non-empty cpus.  But the moving of some (or all) tasks might fail if
+cpuset is bound with another cgroup subsystem which has some restrictions
+on task attaching.  In this failing case, those tasks will stay
+in the original cpuset, and the kernel will automatically update
+their cpus_allowed to allow all online CPUs.  When memory hotplug
+functionality for removing Memory Nodes is available, a similar exception
+is expected to apply there as well.  In general, the kernel prefers to
+violate cpuset placement, over starving a task that has had all
+its allowed CPUs or Memory Nodes taken offline.
+
+There is a second exception to the above.  GFP_ATOMIC requests are
+kernel internal allocations that must be satisfied, immediately.
+The kernel may drop some request, in rare cases even panic, if a
+GFP_ATOMIC alloc fails.  If the request cannot be satisfied within
+the current tasks cpuset, then we relax the cpuset, and look for
+memory anywhere we can find it.  It's better to violate the cpuset
+than stress the kernel.
+
+To start a new job that is to be contained within a cpuset, the steps are:
+
+ 1) mkdir /dev/cpuset
+ 2) mount -t cgroup -ocpuset cpuset /dev/cpuset
+ 3) Create the new cpuset by doing mkdir's and write's (or echo's) in
+    the /dev/cpuset virtual file system.
+ 4) Start a task that will be the "founding father" of the new job.
+ 5) Attach that task to the new cpuset by writing its pid to the
+    /dev/cpuset tasks file for that cpuset.
+ 6) fork, exec or clone the job tasks from this founding father task.
+
+For example, the following sequence of commands will setup a cpuset
+named "Charlie", containing just CPUs 2 and 3, and Memory Node 1,
+and then start a subshell 'sh' in that cpuset:
+
+  mount -t cgroup -ocpuset cpuset /dev/cpuset
+  cd /dev/cpuset
+  mkdir Charlie
+  cd Charlie
+  /bin/echo 2-3 > cpus
+  /bin/echo 1 > mems
+  /bin/echo $$ > tasks
+  sh
+  # The subshell 'sh' is now running in cpuset Charlie
+  # The next line should display '/Charlie'
+  cat /proc/self/cpuset
+
+In the future, a C library interface to cpusets will likely be
+available.  For now, the only way to query or modify cpusets is
+via the cpuset file system, using the various cd, mkdir, echo, cat,
+rmdir commands from the shell, or their equivalent from C.
+
+The sched_setaffinity calls can also be done at the shell prompt using
+SGI's runon or Robert Love's taskset.  The mbind and set_mempolicy
+calls can be done at the shell prompt using the numactl command
+(part of Andi Kleen's numa package).
+
+2. Usage Examples and Syntax
+============================
+
+2.1 Basic Usage
+---------------
+
+Creating, modifying, using the cpusets can be done through the cpuset
+virtual filesystem.
+
+To mount it, type:
+# mount -t cgroup -o cpuset cpuset /dev/cpuset
+
+Then under /dev/cpuset you can find a tree that corresponds to the
+tree of the cpusets in the system. For instance, /dev/cpuset
+is the cpuset that holds the whole system.
+
+If you want to create a new cpuset under /dev/cpuset:
+# cd /dev/cpuset
+# mkdir my_cpuset
+
+Now you want to do something with this cpuset.
+# cd my_cpuset
+
+In this directory you can find several files:
+# ls
+cpu_exclusive  memory_migrate      mems                      tasks
+cpus           memory_pressure     notify_on_release
+mem_exclusive  memory_spread_page  sched_load_balance
+mem_hardwall   memory_spread_slab  sched_relax_domain_level
+
+Reading them will give you information about the state of this cpuset:
+the CPUs and Memory Nodes it can use, the processes that are using
+it, its properties.  By writing to these files you can manipulate
+the cpuset.
+
+Set some flags:
+# /bin/echo 1 > cpu_exclusive
+
+Add some cpus:
+# /bin/echo 0-7 > cpus
+
+Add some mems:
+# /bin/echo 0-7 > mems
+
+Now attach your shell to this cpuset:
+# /bin/echo $$ > tasks
+
+You can also create cpusets inside your cpuset by using mkdir in this
+directory.
+# mkdir my_sub_cs
+
+To remove a cpuset, just use rmdir:
+# rmdir my_sub_cs
+This will fail if the cpuset is in use (has cpusets inside, or has
+processes attached).
+
+Note that for legacy reasons, the "cpuset" filesystem exists as a
+wrapper around the cgroup filesystem.
+
+The command
+
+mount -t cpuset X /dev/cpuset
+
+is equivalent to
+
+mount -t cgroup -ocpuset X /dev/cpuset
+echo "/sbin/cpuset_release_agent" > /dev/cpuset/release_agent
+
+2.2 Adding/removing cpus
+------------------------
+
+This is the syntax to use when writing in the cpus or mems files
+in cpuset directories:
+
+# /bin/echo 1-4 > cpus		-> set cpus list to cpus 1,2,3,4
+# /bin/echo 1,2,3,4 > cpus	-> set cpus list to cpus 1,2,3,4
+
+2.3 Setting flags
+-----------------
+
+The syntax is very simple:
+
+# /bin/echo 1 > cpu_exclusive 	-> set flag 'cpu_exclusive'
+# /bin/echo 0 > cpu_exclusive 	-> unset flag 'cpu_exclusive'
+
+2.4 Attaching processes
+-----------------------
+
+# /bin/echo PID > tasks
+
+Note that it is PID, not PIDs. You can only attach ONE task at a time.
+If you have several tasks to attach, you have to do it one after another:
+
+# /bin/echo PID1 > tasks
+# /bin/echo PID2 > tasks
+	...
+# /bin/echo PIDn > tasks
+
+
+3. Questions
+============
+
+Q: what's up with this '/bin/echo' ?
+A: bash's builtin 'echo' command does not check calls to write() against
+   errors. If you use it in the cpuset file system, you won't be
+   able to tell whether a command succeeded or failed.
+
+Q: When I attach processes, only the first of the line gets really attached !
+A: We can only return one error code per call to write(). So you should also
+   put only ONE pid.
+
+4. Contact
+==========
+
+Web: http://www.bullopensource.org/cpuset
diff --git a/Documentation/cgroups/devices.txt b/Documentation/cgroups/devices.txt
new file mode 100644
index 000000000000..7cc6e6a60672
--- /dev/null
+++ b/Documentation/cgroups/devices.txt
@@ -0,0 +1,52 @@
+Device Whitelist Controller
+
+1. Description:
+
+Implement a cgroup to track and enforce open and mknod restrictions
+on device files.  A device cgroup associates a device access
+whitelist with each cgroup.  A whitelist entry has 4 fields.
+'type' is a (all), c (char), or b (block).  'all' means it applies
+to all types and all major and minor numbers.  Major and minor are
+either an integer or * for all.  Access is a composition of r
+(read), w (write), and m (mknod).
+
+The root device cgroup starts with rwm to 'all'.  A child device
+cgroup gets a copy of the parent.  Administrators can then remove
+devices from the whitelist or add new entries.  A child cgroup can
+never receive a device access which is denied by its parent.  However
+when a device access is removed from a parent it will not also be
+removed from the child(ren).
+
+2. User Interface
+
+An entry is added using devices.allow, and removed using
+devices.deny.  For instance
+
+	echo 'c 1:3 mr' > /cgroups/1/devices.allow
+
+allows cgroup 1 to read and mknod the device usually known as
+/dev/null.  Doing
+
+	echo a > /cgroups/1/devices.deny
+
+will remove the default 'a *:* rwm' entry. Doing
+
+	echo a > /cgroups/1/devices.allow
+
+will add the 'a *:* rwm' entry to the whitelist.
+
+3. Security
+
+Any task can move itself between cgroups.  This clearly won't
+suffice, but we can decide the best way to adequately restrict
+movement as people get some experience with this.  We may just want
+to require CAP_SYS_ADMIN, which at least is a separate bit from
+CAP_MKNOD.  We may want to just refuse moving to a cgroup which
+isn't a descendent of the current one.  Or we may want to use
+CAP_MAC_ADMIN, since we really are trying to lock down root.
+
+CAP_SYS_ADMIN is needed to modify the whitelist or move another
+task to a new cgroup.  (Again we'll probably want to change that).
+
+A cgroup may not be granted more permissions than the cgroup's
+parent has.
diff --git a/Documentation/cgroups/memcg_test.txt b/Documentation/cgroups/memcg_test.txt
new file mode 100644
index 000000000000..19533f93b7a2
--- /dev/null
+++ b/Documentation/cgroups/memcg_test.txt
@@ -0,0 +1,342 @@
+Memory Resource Controller(Memcg)  Implementation Memo.
+Last Updated: 2008/12/15
+Base Kernel Version: based on 2.6.28-rc8-mm.
+
+Because VM is getting complex (one of reasons is memcg...), memcg's behavior
+is complex. This is a document for memcg's internal behavior.
+Please note that implementation details can be changed.
+
+(*) Topics on API should be in Documentation/cgroups/memory.txt)
+
+0. How to record usage ?
+   2 objects are used.
+
+   page_cgroup ....an object per page.
+	Allocated at boot or memory hotplug. Freed at memory hot removal.
+
+   swap_cgroup ... an entry per swp_entry.
+	Allocated at swapon(). Freed at swapoff().
+
+   The page_cgroup has USED bit and double count against a page_cgroup never
+   occurs. swap_cgroup is used only when a charged page is swapped-out.
+
+1. Charge
+
+   a page/swp_entry may be charged (usage += PAGE_SIZE) at
+
+	mem_cgroup_newpage_charge()
+	  Called at new page fault and Copy-On-Write.
+
+	mem_cgroup_try_charge_swapin()
+	  Called at do_swap_page() (page fault on swap entry) and swapoff.
+	  Followed by charge-commit-cancel protocol. (With swap accounting)
+	  At commit, a charge recorded in swap_cgroup is removed.
+
+	mem_cgroup_cache_charge()
+	  Called at add_to_page_cache()
+
+	mem_cgroup_cache_charge_swapin()
+	  Called at shmem's swapin.
+
+	mem_cgroup_prepare_migration()
+	  Called before migration. "extra" charge is done and followed by
+	  charge-commit-cancel protocol.
+	  At commit, charge against oldpage or newpage will be committed.
+
+2. Uncharge
+  a page/swp_entry may be uncharged (usage -= PAGE_SIZE) by
+
+	mem_cgroup_uncharge_page()
+	  Called when an anonymous page is fully unmapped. I.e., mapcount goes
+	  to 0. If the page is SwapCache, uncharge is delayed until
+	  mem_cgroup_uncharge_swapcache().
+
+	mem_cgroup_uncharge_cache_page()
+	  Called when a page-cache is deleted from radix-tree. If the page is
+	  SwapCache, uncharge is delayed until mem_cgroup_uncharge_swapcache().
+
+	mem_cgroup_uncharge_swapcache()
+	  Called when SwapCache is removed from radix-tree. The charge itself
+	  is moved to swap_cgroup. (If mem+swap controller is disabled, no
+	  charge to swap occurs.)
+
+	mem_cgroup_uncharge_swap()
+	  Called when swp_entry's refcnt goes down to 0. A charge against swap
+	  disappears.
+
+	mem_cgroup_end_migration(old, new)
+	At success of migration old is uncharged (if necessary), a charge
+	to new page is committed. At failure, charge to old page is committed.
+
+3. charge-commit-cancel
+	In some case, we can't know this "charge" is valid or not at charging
+	(because of races).
+	To handle such case, there are charge-commit-cancel functions.
+		mem_cgroup_try_charge_XXX
+		mem_cgroup_commit_charge_XXX
+		mem_cgroup_cancel_charge_XXX
+	these are used in swap-in and migration.
+
+	At try_charge(), there are no flags to say "this page is charged".
+	at this point, usage += PAGE_SIZE.
+
+	At commit(), the function checks the page should be charged or not
+	and set flags or avoid charging.(usage -= PAGE_SIZE)
+
+	At cancel(), simply usage -= PAGE_SIZE.
+
+Under below explanation, we assume CONFIG_MEM_RES_CTRL_SWAP=y.
+
+4. Anonymous
+	Anonymous page is newly allocated at
+		  - page fault into MAP_ANONYMOUS mapping.
+		  - Copy-On-Write.
+ 	It is charged right after it's allocated before doing any page table
+	related operations. Of course, it's uncharged when another page is used
+	for the fault address.
+
+	At freeing anonymous page (by exit() or munmap()), zap_pte() is called
+	and pages for ptes are freed one by one.(see mm/memory.c). Uncharges
+	are done at page_remove_rmap() when page_mapcount() goes down to 0.
+
+	Another page freeing is by page-reclaim (vmscan.c) and anonymous
+	pages are swapped out. In this case, the page is marked as
+	PageSwapCache(). uncharge() routine doesn't uncharge the page marked
+	as SwapCache(). It's delayed until __delete_from_swap_cache().
+
+	4.1 Swap-in.
+	At swap-in, the page is taken from swap-cache. There are 2 cases.
+
+	(a) If the SwapCache is newly allocated and read, it has no charges.
+	(b) If the SwapCache has been mapped by processes, it has been
+	    charged already.
+
+	This swap-in is one of the most complicated work. In do_swap_page(),
+	following events occur when pte is unchanged.
+
+	(1) the page (SwapCache) is looked up.
+	(2) lock_page()
+	(3) try_charge_swapin()
+	(4) reuse_swap_page() (may call delete_swap_cache())
+	(5) commit_charge_swapin()
+	(6) swap_free().
+
+	Considering following situation for example.
+
+	(A) The page has not been charged before (2) and reuse_swap_page()
+	    doesn't call delete_from_swap_cache().
+	(B) The page has not been charged before (2) and reuse_swap_page()
+	    calls delete_from_swap_cache().
+	(C) The page has been charged before (2) and reuse_swap_page() doesn't
+	    call delete_from_swap_cache().
+	(D) The page has been charged before (2) and reuse_swap_page() calls
+	    delete_from_swap_cache().
+
+	    memory.usage/memsw.usage changes to this page/swp_entry will be
+	 Case          (A)      (B)       (C)     (D)
+         Event
+       Before (2)     0/ 1     0/ 1      1/ 1    1/ 1
+          ===========================================
+          (3)        +1/+1    +1/+1     +1/+1   +1/+1
+          (4)          -       0/ 0       -     -1/ 0
+          (5)         0/-1     0/ 0     -1/-1    0/ 0
+          (6)          -       0/-1       -      0/-1
+          ===========================================
+       Result         1/ 1     1/ 1      1/ 1    1/ 1
+
+       In any cases, charges to this page should be 1/ 1.
+
+	4.2 Swap-out.
+	At swap-out, typical state transition is below.
+
+	(a) add to swap cache. (marked as SwapCache)
+	    swp_entry's refcnt += 1.
+	(b) fully unmapped.
+	    swp_entry's refcnt += # of ptes.
+	(c) write back to swap.
+	(d) delete from swap cache. (remove from SwapCache)
+	    swp_entry's refcnt -= 1.
+
+
+	At (b), the page is marked as SwapCache and not uncharged.
+	At (d), the page is removed from SwapCache and a charge in page_cgroup
+	is moved to swap_cgroup.
+
+	Finally, at task exit,
+	(e) zap_pte() is called and swp_entry's refcnt -=1 -> 0.
+	Here, a charge in swap_cgroup disappears.
+
+5. Page Cache
+   	Page Cache is charged at
+	- add_to_page_cache_locked().
+
+	uncharged at
+	- __remove_from_page_cache().
+
+	The logic is very clear. (About migration, see below)
+	Note: __remove_from_page_cache() is called by remove_from_page_cache()
+	and __remove_mapping().
+
+6. Shmem(tmpfs) Page Cache
+	Memcg's charge/uncharge have special handlers of shmem. The best way
+	to understand shmem's page state transition is to read mm/shmem.c.
+	But brief explanation of the behavior of memcg around shmem will be
+	helpful to understand the logic.
+
+	Shmem's page (just leaf page, not direct/indirect block) can be on
+		- radix-tree of shmem's inode.
+		- SwapCache.
+		- Both on radix-tree and SwapCache. This happens at swap-in
+		  and swap-out,
+
+	It's charged when...
+	- A new page is added to shmem's radix-tree.
+	- A swp page is read. (move a charge from swap_cgroup to page_cgroup)
+	It's uncharged when
+	- A page is removed from radix-tree and not SwapCache.
+	- When SwapCache is removed, a charge is moved to swap_cgroup.
+	- When swp_entry's refcnt goes down to 0, a charge in swap_cgroup
+	  disappears.
+
+7. Page Migration
+   	One of the most complicated functions is page-migration-handler.
+	Memcg has 2 routines. Assume that we are migrating a page's contents
+	from OLDPAGE to NEWPAGE.
+
+	Usual migration logic is..
+	(a) remove the page from LRU.
+	(b) allocate NEWPAGE (migration target)
+	(c) lock by lock_page().
+	(d) unmap all mappings.
+	(e-1) If necessary, replace entry in radix-tree.
+	(e-2) move contents of a page.
+	(f) map all mappings again.
+	(g) pushback the page to LRU.
+	(-) OLDPAGE will be freed.
+
+	Before (g), memcg should complete all necessary charge/uncharge to
+	NEWPAGE/OLDPAGE.
+
+	The point is....
+	- If OLDPAGE is anonymous, all charges will be dropped at (d) because
+          try_to_unmap() drops all mapcount and the page will not be
+	  SwapCache.
+
+	- If OLDPAGE is SwapCache, charges will be kept at (g) because
+	  __delete_from_swap_cache() isn't called at (e-1)
+
+	- If OLDPAGE is page-cache, charges will be kept at (g) because
+	  remove_from_swap_cache() isn't called at (e-1)
+
+	memcg provides following hooks.
+
+	- mem_cgroup_prepare_migration(OLDPAGE)
+	  Called after (b) to account a charge (usage += PAGE_SIZE) against
+	  memcg which OLDPAGE belongs to.
+
+        - mem_cgroup_end_migration(OLDPAGE, NEWPAGE)
+	  Called after (f) before (g).
+	  If OLDPAGE is used, commit OLDPAGE again. If OLDPAGE is already
+	  charged, a charge by prepare_migration() is automatically canceled.
+	  If NEWPAGE is used, commit NEWPAGE and uncharge OLDPAGE.
+
+	  But zap_pte() (by exit or munmap) can be called while migration,
+	  we have to check if OLDPAGE/NEWPAGE is a valid page after commit().
+
+8. LRU
+        Each memcg has its own private LRU. Now, it's handling is under global
+	VM's control (means that it's handled under global zone->lru_lock).
+	Almost all routines around memcg's LRU is called by global LRU's
+	list management functions under zone->lru_lock().
+
+	A special function is mem_cgroup_isolate_pages(). This scans
+	memcg's private LRU and call __isolate_lru_page() to extract a page
+	from LRU.
+	(By __isolate_lru_page(), the page is removed from both of global and
+	 private LRU.)
+
+
+9. Typical Tests.
+
+ Tests for racy cases.
+
+ 9.1 Small limit to memcg.
+	When you do test to do racy case, it's good test to set memcg's limit
+	to be very small rather than GB. Many races found in the test under
+	xKB or xxMB limits.
+	(Memory behavior under GB and Memory behavior under MB shows very
+	 different situation.)
+
+ 9.2 Shmem
+	Historically, memcg's shmem handling was poor and we saw some amount
+	of troubles here. This is because shmem is page-cache but can be
+	SwapCache. Test with shmem/tmpfs is always good test.
+
+ 9.3 Migration
+	For NUMA, migration is an another special case. To do easy test, cpuset
+	is useful. Following is a sample script to do migration.
+
+	mount -t cgroup -o cpuset none /opt/cpuset
+
+	mkdir /opt/cpuset/01
+	echo 1 > /opt/cpuset/01/cpuset.cpus
+	echo 0 > /opt/cpuset/01/cpuset.mems
+	echo 1 > /opt/cpuset/01/cpuset.memory_migrate
+	mkdir /opt/cpuset/02
+	echo 1 > /opt/cpuset/02/cpuset.cpus
+	echo 1 > /opt/cpuset/02/cpuset.mems
+	echo 1 > /opt/cpuset/02/cpuset.memory_migrate
+
+	In above set, when you moves a task from 01 to 02, page migration to
+	node 0 to node 1 will occur. Following is a script to migrate all
+	under cpuset.
+	--
+	move_task()
+	{
+	for pid in $1
+        do
+                /bin/echo $pid >$2/tasks 2>/dev/null
+		echo -n $pid
+		echo -n " "
+        done
+	echo END
+	}
+
+	G1_TASK=`cat ${G1}/tasks`
+	G2_TASK=`cat ${G2}/tasks`
+	move_task "${G1_TASK}" ${G2} &
+	--
+ 9.4 Memory hotplug.
+	memory hotplug test is one of good test.
+	to offline memory, do following.
+	# echo offline > /sys/devices/system/memory/memoryXXX/state
+	(XXX is the place of memory)
+	This is an easy way to test page migration, too.
+
+ 9.5 mkdir/rmdir
+	When using hierarchy, mkdir/rmdir test should be done.
+	Use tests like the following.
+
+	echo 1 >/opt/cgroup/01/memory/use_hierarchy
+	mkdir /opt/cgroup/01/child_a
+	mkdir /opt/cgroup/01/child_b
+
+	set limit to 01.
+	add limit to 01/child_b
+	run jobs under child_a and child_b
+
+	create/delete following groups at random while jobs are running.
+	/opt/cgroup/01/child_a/child_aa
+	/opt/cgroup/01/child_b/child_bb
+	/opt/cgroup/01/child_c
+
+	running new jobs in new group is also good.
+
+ 9.6 Mount with other subsystems.
+	Mounting with other subsystems is a good test because there is a
+	race and lock dependency with other cgroup subsystems.
+
+	example)
+	# mount -t cgroup none /cgroup -t cpuset,memory,cpu,devices
+
+	and do task move, mkdir, rmdir etc...under this.
diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt
new file mode 100644
index 000000000000..e1501964df1e
--- /dev/null
+++ b/Documentation/cgroups/memory.txt
@@ -0,0 +1,399 @@
+Memory Resource Controller
+
+NOTE: The Memory Resource Controller has been generically been referred
+to as the memory controller in this document. Do not confuse memory controller
+used here with the memory controller that is used in hardware.
+
+Salient features
+
+a. Enable control of both RSS (mapped) and Page Cache (unmapped) pages
+b. The infrastructure allows easy addition of other types of memory to control
+c. Provides *zero overhead* for non memory controller users
+d. Provides a double LRU: global memory pressure causes reclaim from the
+   global LRU; a cgroup on hitting a limit, reclaims from the per
+   cgroup LRU
+
+NOTE: Swap Cache (unmapped) is not accounted now.
+
+Benefits and Purpose of the memory controller
+
+The memory controller isolates the memory behaviour of a group of tasks
+from the rest of the system. The article on LWN [12] mentions some probable
+uses of the memory controller. The memory controller can be used to
+
+a. Isolate an application or a group of applications
+   Memory hungry applications can be isolated and limited to a smaller
+   amount of memory.
+b. Create a cgroup with limited amount of memory, this can be used
+   as a good alternative to booting with mem=XXXX.
+c. Virtualization solutions can control the amount of memory they want
+   to assign to a virtual machine instance.
+d. A CD/DVD burner could control the amount of memory used by the
+   rest of the system to ensure that burning does not fail due to lack
+   of available memory.
+e. There are several other use cases, find one or use the controller just
+   for fun (to learn and hack on the VM subsystem).
+
+1. History
+
+The memory controller has a long history. A request for comments for the memory
+controller was posted by Balbir Singh [1]. At the time the RFC was posted
+there were several implementations for memory control. The goal of the
+RFC was to build consensus and agreement for the minimal features required
+for memory control. The first RSS controller was posted by Balbir Singh[2]
+in Feb 2007. Pavel Emelianov [3][4][5] has since posted three versions of the
+RSS controller. At OLS, at the resource management BoF, everyone suggested
+that we handle both page cache and RSS together. Another request was raised
+to allow user space handling of OOM. The current memory controller is
+at version 6; it combines both mapped (RSS) and unmapped Page
+Cache Control [11].
+
+2. Memory Control
+
+Memory is a unique resource in the sense that it is present in a limited
+amount. If a task requires a lot of CPU processing, the task can spread
+its processing over a period of hours, days, months or years, but with
+memory, the same physical memory needs to be reused to accomplish the task.
+
+The memory controller implementation has been divided into phases. These
+are:
+
+1. Memory controller
+2. mlock(2) controller
+3. Kernel user memory accounting and slab control
+4. user mappings length controller
+
+The memory controller is the first controller developed.
+
+2.1. Design
+
+The core of the design is a counter called the res_counter. The res_counter
+tracks the current memory usage and limit of the group of processes associated
+with the controller. Each cgroup has a memory controller specific data
+structure (mem_cgroup) associated with it.
+
+2.2. Accounting
+
+		+--------------------+
+		|  mem_cgroup     |
+		|  (res_counter)     |
+		+--------------------+
+		 /            ^      \
+		/             |       \
+           +---------------+  |        +---------------+
+           | mm_struct     |  |....    | mm_struct     |
+           |               |  |        |               |
+           +---------------+  |        +---------------+
+                              |
+                              + --------------+
+                                              |
+           +---------------+           +------+--------+
+           | page          +---------->  page_cgroup|
+           |               |           |               |
+           +---------------+           +---------------+
+
+             (Figure 1: Hierarchy of Accounting)
+
+
+Figure 1 shows the important aspects of the controller
+
+1. Accounting happens per cgroup
+2. Each mm_struct knows about which cgroup it belongs to
+3. Each page has a pointer to the page_cgroup, which in turn knows the
+   cgroup it belongs to
+
+The accounting is done as follows: mem_cgroup_charge() is invoked to setup
+the necessary data structures and check if the cgroup that is being charged
+is over its limit. If it is then reclaim is invoked on the cgroup.
+More details can be found in the reclaim section of this document.
+If everything goes well, a page meta-data-structure called page_cgroup is
+allocated and associated with the page.  This routine also adds the page to
+the per cgroup LRU.
+
+2.2.1 Accounting details
+
+All mapped anon pages (RSS) and cache pages (Page Cache) are accounted.
+(some pages which never be reclaimable and will not be on global LRU
+ are not accounted. we just accounts pages under usual vm management.)
+
+RSS pages are accounted at page_fault unless they've already been accounted
+for earlier. A file page will be accounted for as Page Cache when it's
+inserted into inode (radix-tree). While it's mapped into the page tables of
+processes, duplicate accounting is carefully avoided.
+
+A RSS page is unaccounted when it's fully unmapped. A PageCache page is
+unaccounted when it's removed from radix-tree.
+
+At page migration, accounting information is kept.
+
+Note: we just account pages-on-lru because our purpose is to control amount
+of used pages. not-on-lru pages are tend to be out-of-control from vm view.
+
+2.3 Shared Page Accounting
+
+Shared pages are accounted on the basis of the first touch approach. The
+cgroup that first touches a page is accounted for the page. The principle
+behind this approach is that a cgroup that aggressively uses a shared
+page will eventually get charged for it (once it is uncharged from
+the cgroup that brought it in -- this will happen on memory pressure).
+
+Exception: If CONFIG_CGROUP_CGROUP_MEM_RES_CTLR_SWAP is not used..
+When you do swapoff and make swapped-out pages of shmem(tmpfs) to
+be backed into memory in force, charges for pages are accounted against the
+caller of swapoff rather than the users of shmem.
+
+
+2.4 Swap Extension (CONFIG_CGROUP_MEM_RES_CTLR_SWAP)
+Swap Extension allows you to record charge for swap. A swapped-in page is
+charged back to original page allocator if possible.
+
+When swap is accounted, following files are added.
+ - memory.memsw.usage_in_bytes.
+ - memory.memsw.limit_in_bytes.
+
+usage of mem+swap is limited by memsw.limit_in_bytes.
+
+Note: why 'mem+swap' rather than swap.
+The global LRU(kswapd) can swap out arbitrary pages. Swap-out means
+to move account from memory to swap...there is no change in usage of
+mem+swap.
+
+In other words, when we want to limit the usage of swap without affecting
+global LRU, mem+swap limit is better than just limiting swap from OS point
+of view.
+
+2.5 Reclaim
+
+Each cgroup maintains a per cgroup LRU that consists of an active
+and inactive list. When a cgroup goes over its limit, we first try
+to reclaim memory from the cgroup so as to make space for the new
+pages that the cgroup has touched. If the reclaim is unsuccessful,
+an OOM routine is invoked to select and kill the bulkiest task in the
+cgroup.
+
+The reclaim algorithm has not been modified for cgroups, except that
+pages that are selected for reclaiming come from the per cgroup LRU
+list.
+
+2. Locking
+
+The memory controller uses the following hierarchy
+
+1. zone->lru_lock is used for selecting pages to be isolated
+2. mem->per_zone->lru_lock protects the per cgroup LRU (per zone)
+3. lock_page_cgroup() is used to protect page->page_cgroup
+
+3. User Interface
+
+0. Configuration
+
+a. Enable CONFIG_CGROUPS
+b. Enable CONFIG_RESOURCE_COUNTERS
+c. Enable CONFIG_CGROUP_MEM_RES_CTLR
+
+1. Prepare the cgroups
+# mkdir -p /cgroups
+# mount -t cgroup none /cgroups -o memory
+
+2. Make the new group and move bash into it
+# mkdir /cgroups/0
+# echo $$ >  /cgroups/0/tasks
+
+Since now we're in the 0 cgroup,
+We can alter the memory limit:
+# echo 4M > /cgroups/0/memory.limit_in_bytes
+
+NOTE: We can use a suffix (k, K, m, M, g or G) to indicate values in kilo,
+mega or gigabytes.
+
+# cat /cgroups/0/memory.limit_in_bytes
+4194304
+
+NOTE: The interface has now changed to display the usage in bytes
+instead of pages
+
+We can check the usage:
+# cat /cgroups/0/memory.usage_in_bytes
+1216512
+
+A successful write to this file does not guarantee a successful set of
+this limit to the value written into the file.  This can be due to a
+number of factors, such as rounding up to page boundaries or the total
+availability of memory on the system.  The user is required to re-read
+this file after a write to guarantee the value committed by the kernel.
+
+# echo 1 > memory.limit_in_bytes
+# cat memory.limit_in_bytes
+4096
+
+The memory.failcnt field gives the number of times that the cgroup limit was
+exceeded.
+
+The memory.stat file gives accounting information. Now, the number of
+caches, RSS and Active pages/Inactive pages are shown.
+
+4. Testing
+
+Balbir posted lmbench, AIM9, LTP and vmmstress results [10] and [11].
+Apart from that v6 has been tested with several applications and regular
+daily use. The controller has also been tested on the PPC64, x86_64 and
+UML platforms.
+
+4.1 Troubleshooting
+
+Sometimes a user might find that the application under a cgroup is
+terminated. There are several causes for this:
+
+1. The cgroup limit is too low (just too low to do anything useful)
+2. The user is using anonymous memory and swap is turned off or too low
+
+A sync followed by echo 1 > /proc/sys/vm/drop_caches will help get rid of
+some of the pages cached in the cgroup (page cache pages).
+
+4.2 Task migration
+
+When a task migrates from one cgroup to another, it's charge is not
+carried forward. The pages allocated from the original cgroup still
+remain charged to it, the charge is dropped when the page is freed or
+reclaimed.
+
+4.3 Removing a cgroup
+
+A cgroup can be removed by rmdir, but as discussed in sections 4.1 and 4.2, a
+cgroup might have some charge associated with it, even though all
+tasks have migrated away from it.
+Such charges are freed(at default) or moved to its parent. When moved,
+both of RSS and CACHES are moved to parent.
+If both of them are busy, rmdir() returns -EBUSY. See 5.1 Also.
+
+Charges recorded in swap information is not updated at removal of cgroup.
+Recorded information is discarded and a cgroup which uses swap (swapcache)
+will be charged as a new owner of it.
+
+
+5. Misc. interfaces.
+
+5.1 force_empty
+  memory.force_empty interface is provided to make cgroup's memory usage empty.
+  You can use this interface only when the cgroup has no tasks.
+  When writing anything to this
+
+  # echo 0 > memory.force_empty
+
+  Almost all pages tracked by this memcg will be unmapped and freed. Some of
+  pages cannot be freed because it's locked or in-use. Such pages are moved
+  to parent and this cgroup will be empty. But this may return -EBUSY in
+  some too busy case.
+
+  Typical use case of this interface is that calling this before rmdir().
+  Because rmdir() moves all pages to parent, some out-of-use page caches can be
+  moved to the parent. If you want to avoid that, force_empty will be useful.
+
+5.2 stat file
+  memory.stat file includes following statistics (now)
+	cache			- # of pages from page-cache and shmem.
+	rss			- # of pages from anonymous memory.
+	pgpgin			- # of event of charging
+	pgpgout			- # of event of uncharging
+	active_anon		- # of pages on active lru of anon, shmem.
+	inactive_anon 		- # of pages on active lru of anon, shmem
+	active_file		- # of pages on active lru of file-cache
+	inactive_file		- # of pages on inactive lru of file cache
+	unevictable		- # of pages cannot be reclaimed.(mlocked etc)
+
+	Below is depend on CONFIG_DEBUG_VM.
+	inactive_ratio		- VM inernal parameter. (see mm/page_alloc.c)
+	recent_rotated_anon	- VM internal parameter. (see mm/vmscan.c)
+	recent_rotated_file	- VM internal parameter. (see mm/vmscan.c)
+	recent_scanned_anon 	- VM internal parameter. (see mm/vmscan.c)
+	recent_scanned_file 	- VM internal parameter. (see mm/vmscan.c)
+
+  Memo:
+	recent_rotated means recent frequency of lru rotation.
+	recent_scanned means recent # of scans to lru.
+	showing for better debug please see the code for meanings.
+
+
+5.3 swappiness
+  Similar to /proc/sys/vm/swappiness, but affecting a hierarchy of groups only.
+
+  Following cgroup's swapiness can't be changed.
+  - root cgroup (uses /proc/sys/vm/swappiness).
+  - a cgroup which uses hierarchy and it has child cgroup.
+  - a cgroup which uses hierarchy and not the root of hierarchy.
+
+
+6. Hierarchy support
+
+The memory controller supports a deep hierarchy and hierarchical accounting.
+The hierarchy is created by creating the appropriate cgroups in the
+cgroup filesystem. Consider for example, the following cgroup filesystem
+hierarchy
+
+		root
+	     /  |   \
+           /	|    \
+	  a	b	c
+			| \
+			|  \
+			d   e
+
+In the diagram above, with hierarchical accounting enabled, all memory
+usage of e, is accounted to its ancestors up until the root (i.e, c and root),
+that has memory.use_hierarchy enabled.  If one of the ancestors goes over its
+limit, the reclaim algorithm reclaims from the tasks in the ancestor and the
+children of the ancestor.
+
+6.1 Enabling hierarchical accounting and reclaim
+
+The memory controller by default disables the hierarchy feature. Support
+can be enabled by writing 1 to memory.use_hierarchy file of the root cgroup
+
+# echo 1 > memory.use_hierarchy
+
+The feature can be disabled by
+
+# echo 0 > memory.use_hierarchy
+
+NOTE1: Enabling/disabling will fail if the cgroup already has other
+cgroups created below it.
+
+NOTE2: This feature can be enabled/disabled per subtree.
+
+7. TODO
+
+1. Add support for accounting huge pages (as a separate controller)
+2. Make per-cgroup scanner reclaim not-shared pages first
+3. Teach controller to account for shared-pages
+4. Start reclamation in the background when the limit is
+   not yet hit but the usage is getting closer
+
+Summary
+
+Overall, the memory controller has been a stable controller and has been
+commented and discussed quite extensively in the community.
+
+References
+
+1. Singh, Balbir. RFC: Memory Controller, http://lwn.net/Articles/206697/
+2. Singh, Balbir. Memory Controller (RSS Control),
+   http://lwn.net/Articles/222762/
+3. Emelianov, Pavel. Resource controllers based on process cgroups
+   http://lkml.org/lkml/2007/3/6/198
+4. Emelianov, Pavel. RSS controller based on process cgroups (v2)
+   http://lkml.org/lkml/2007/4/9/78
+5. Emelianov, Pavel. RSS controller based on process cgroups (v3)
+   http://lkml.org/lkml/2007/5/30/244
+6. Menage, Paul. Control Groups v10, http://lwn.net/Articles/236032/
+7. Vaidyanathan, Srinivasan, Control Groups: Pagecache accounting and control
+   subsystem (v3), http://lwn.net/Articles/235534/
+8. Singh, Balbir. RSS controller v2 test results (lmbench),
+   http://lkml.org/lkml/2007/5/17/232
+9. Singh, Balbir. RSS controller v2 AIM9 results
+   http://lkml.org/lkml/2007/5/18/1
+10. Singh, Balbir. Memory controller v6 test results,
+    http://lkml.org/lkml/2007/8/19/36
+11. Singh, Balbir. Memory controller introduction (v6),
+    http://lkml.org/lkml/2007/8/17/69
+12. Corbet, Jonathan, Controlling memory use in cgroups,
+    http://lwn.net/Articles/243795/
diff --git a/Documentation/cgroups/resource_counter.txt b/Documentation/cgroups/resource_counter.txt
new file mode 100644
index 000000000000..f196ac1d7d25
--- /dev/null
+++ b/Documentation/cgroups/resource_counter.txt
@@ -0,0 +1,181 @@
+
+		The Resource Counter
+
+The resource counter, declared at include/linux/res_counter.h,
+is supposed to facilitate the resource management by controllers
+by providing common stuff for accounting.
+
+This "stuff" includes the res_counter structure and routines
+to work with it.
+
+
+
+1. Crucial parts of the res_counter structure
+
+ a. unsigned long long usage
+
+ 	The usage value shows the amount of a resource that is consumed
+	by a group at a given time. The units of measurement should be
+	determined by the controller that uses this counter. E.g. it can
+	be bytes, items or any other unit the controller operates on.
+
+ b. unsigned long long max_usage
+
+ 	The maximal value of the usage over time.
+
+ 	This value is useful when gathering statistical information about
+	the particular group, as it shows the actual resource requirements
+	for a particular group, not just some usage snapshot.
+
+ c. unsigned long long limit
+
+ 	The maximal allowed amount of resource to consume by the group. In
+	case the group requests for more resources, so that the usage value
+	would exceed the limit, the resource allocation is rejected (see
+	the next section).
+
+ d. unsigned long long failcnt
+
+ 	The failcnt stands for "failures counter". This is the number of
+	resource allocation attempts that failed.
+
+ c. spinlock_t lock
+
+ 	Protects changes of the above values.
+
+
+
+2. Basic accounting routines
+
+ a. void res_counter_init(struct res_counter *rc)
+
+ 	Initializes the resource counter. As usual, should be the first
+	routine called for a new counter.
+
+ b. int res_counter_charge[_locked]
+			(struct res_counter *rc, unsigned long val)
+
+	When a resource is about to be allocated it has to be accounted
+	with the appropriate resource counter (controller should determine
+	which one to use on its own). This operation is called "charging".
+
+	This is not very important which operation - resource allocation
+	or charging - is performed first, but
+	  * if the allocation is performed first, this may create a
+	    temporary resource over-usage by the time resource counter is
+	    charged;
+	  * if the charging is performed first, then it should be uncharged
+	    on error path (if the one is called).
+
+ c. void res_counter_uncharge[_locked]
+			(struct res_counter *rc, unsigned long val)
+
+	When a resource is released (freed) it should be de-accounted
+	from the resource counter it was accounted to.  This is called
+	"uncharging".
+
+    The _locked routines imply that the res_counter->lock is taken.
+
+
+ 2.1 Other accounting routines
+
+    There are more routines that may help you with common needs, like
+    checking whether the limit is reached or resetting the max_usage
+    value. They are all declared in include/linux/res_counter.h.
+
+
+
+3. Analyzing the resource counter registrations
+
+ a. If the failcnt value constantly grows, this means that the counter's
+    limit is too tight. Either the group is misbehaving and consumes too
+    many resources, or the configuration is not suitable for the group
+    and the limit should be increased.
+
+ b. The max_usage value can be used to quickly tune the group. One may
+    set the limits to maximal values and either load the container with
+    a common pattern or leave one for a while. After this the max_usage
+    value shows the amount of memory the container would require during
+    its common activity.
+
+    Setting the limit a bit above this value gives a pretty good
+    configuration that works in most of the cases.
+
+ c. If the max_usage is much less than the limit, but the failcnt value
+    is growing, then the group tries to allocate a big chunk of resource
+    at once.
+
+ d. If the max_usage is much less than the limit, but the failcnt value
+    is 0, then this group is given too high limit, that it does not
+    require. It is better to lower the limit a bit leaving more resource
+    for other groups.
+
+
+
+4. Communication with the control groups subsystem (cgroups)
+
+All the resource controllers that are using cgroups and resource counters
+should provide files (in the cgroup filesystem) to work with the resource
+counter fields. They are recommended to adhere to the following rules:
+
+ a. File names
+
+ 	Field name	File name
+	---------------------------------------------------
+	usage		usage_in_<unit_of_measurement>
+	max_usage	max_usage_in_<unit_of_measurement>
+	limit		limit_in_<unit_of_measurement>
+	failcnt		failcnt
+	lock		no file :)
+
+ b. Reading from file should show the corresponding field value in the
+    appropriate format.
+
+ c. Writing to file
+
+ 	Field		Expected behavior
+	----------------------------------
+	usage		prohibited
+	max_usage	reset to usage
+	limit		set the limit
+	failcnt		reset to zero
+
+
+
+5. Usage example
+
+ a. Declare a task group (take a look at cgroups subsystem for this) and
+    fold a res_counter into it
+
+	struct my_group {
+		struct res_counter res;
+
+		<other fields>
+	}
+
+ b. Put hooks in resource allocation/release paths
+
+ 	int alloc_something(...)
+	{
+		if (res_counter_charge(res_counter_ptr, amount) < 0)
+			return -ENOMEM;
+
+		<allocate the resource and return to the caller>
+	}
+
+	void release_something(...)
+	{
+		res_counter_uncharge(res_counter_ptr, amount);
+
+		<release the resource>
+	}
+
+    In order to keep the usage value self-consistent, both the
+    "res_counter_ptr" and the "amount" in release_something() should be
+    the same as they were in the alloc_something() when the releasing
+    resource was allocated.
+
+ c. Provide the way to read res_counter values and set them (the cgroups
+    still can help with it).
+
+ c. Compile and run :)
diff --git a/Documentation/controllers/cpuacct.txt b/Documentation/controllers/cpuacct.txt
deleted file mode 100644
index bb775fbe43d7..000000000000
--- a/Documentation/controllers/cpuacct.txt
+++ /dev/null
@@ -1,32 +0,0 @@
-CPU Accounting Controller
--------------------------
-
-The CPU accounting controller is used to group tasks using cgroups and
-account the CPU usage of these groups of tasks.
-
-The CPU accounting controller supports multi-hierarchy groups. An accounting
-group accumulates the CPU usage of all of its child groups and the tasks
-directly present in its group.
-
-Accounting groups can be created by first mounting the cgroup filesystem.
-
-# mkdir /cgroups
-# mount -t cgroup -ocpuacct none /cgroups
-
-With the above step, the initial or the parent accounting group
-becomes visible at /cgroups. At bootup, this group includes all the
-tasks in the system. /cgroups/tasks lists the tasks in this cgroup.
-/cgroups/cpuacct.usage gives the CPU time (in nanoseconds) obtained by
-this group which is essentially the CPU time obtained by all the tasks
-in the system.
-
-New accounting groups can be created under the parent group /cgroups.
-
-# cd /cgroups
-# mkdir g1
-# echo $$ > g1
-
-The above steps create a new group g1 and move the current shell
-process (bash) into it. CPU time consumed by this bash and its children
-can be obtained from g1/cpuacct.usage and the same is accumulated in
-/cgroups/cpuacct.usage also.
diff --git a/Documentation/controllers/devices.txt b/Documentation/controllers/devices.txt
deleted file mode 100644
index 7cc6e6a60672..000000000000
--- a/Documentation/controllers/devices.txt
+++ /dev/null
@@ -1,52 +0,0 @@
-Device Whitelist Controller
-
-1. Description:
-
-Implement a cgroup to track and enforce open and mknod restrictions
-on device files.  A device cgroup associates a device access
-whitelist with each cgroup.  A whitelist entry has 4 fields.
-'type' is a (all), c (char), or b (block).  'all' means it applies
-to all types and all major and minor numbers.  Major and minor are
-either an integer or * for all.  Access is a composition of r
-(read), w (write), and m (mknod).
-
-The root device cgroup starts with rwm to 'all'.  A child device
-cgroup gets a copy of the parent.  Administrators can then remove
-devices from the whitelist or add new entries.  A child cgroup can
-never receive a device access which is denied by its parent.  However
-when a device access is removed from a parent it will not also be
-removed from the child(ren).
-
-2. User Interface
-
-An entry is added using devices.allow, and removed using
-devices.deny.  For instance
-
-	echo 'c 1:3 mr' > /cgroups/1/devices.allow
-
-allows cgroup 1 to read and mknod the device usually known as
-/dev/null.  Doing
-
-	echo a > /cgroups/1/devices.deny
-
-will remove the default 'a *:* rwm' entry. Doing
-
-	echo a > /cgroups/1/devices.allow
-
-will add the 'a *:* rwm' entry to the whitelist.
-
-3. Security
-
-Any task can move itself between cgroups.  This clearly won't
-suffice, but we can decide the best way to adequately restrict
-movement as people get some experience with this.  We may just want
-to require CAP_SYS_ADMIN, which at least is a separate bit from
-CAP_MKNOD.  We may want to just refuse moving to a cgroup which
-isn't a descendent of the current one.  Or we may want to use
-CAP_MAC_ADMIN, since we really are trying to lock down root.
-
-CAP_SYS_ADMIN is needed to modify the whitelist or move another
-task to a new cgroup.  (Again we'll probably want to change that).
-
-A cgroup may not be granted more permissions than the cgroup's
-parent has.
diff --git a/Documentation/controllers/memcg_test.txt b/Documentation/controllers/memcg_test.txt
deleted file mode 100644
index 08d4d3ea0d79..000000000000
--- a/Documentation/controllers/memcg_test.txt
+++ /dev/null
@@ -1,342 +0,0 @@
-Memory Resource Controller(Memcg)  Implementation Memo.
-Last Updated: 2008/12/15
-Base Kernel Version: based on 2.6.28-rc8-mm.
-
-Because VM is getting complex (one of reasons is memcg...), memcg's behavior
-is complex. This is a document for memcg's internal behavior.
-Please note that implementation details can be changed.
-
-(*) Topics on API should be in Documentation/controllers/memory.txt)
-
-0. How to record usage ?
-   2 objects are used.
-
-   page_cgroup ....an object per page.
-	Allocated at boot or memory hotplug. Freed at memory hot removal.
-
-   swap_cgroup ... an entry per swp_entry.
-	Allocated at swapon(). Freed at swapoff().
-
-   The page_cgroup has USED bit and double count against a page_cgroup never
-   occurs. swap_cgroup is used only when a charged page is swapped-out.
-
-1. Charge
-
-   a page/swp_entry may be charged (usage += PAGE_SIZE) at
-
-	mem_cgroup_newpage_charge()
-	  Called at new page fault and Copy-On-Write.
-
-	mem_cgroup_try_charge_swapin()
-	  Called at do_swap_page() (page fault on swap entry) and swapoff.
-	  Followed by charge-commit-cancel protocol. (With swap accounting)
-	  At commit, a charge recorded in swap_cgroup is removed.
-
-	mem_cgroup_cache_charge()
-	  Called at add_to_page_cache()
-
-	mem_cgroup_cache_charge_swapin()
-	  Called at shmem's swapin.
-
-	mem_cgroup_prepare_migration()
-	  Called before migration. "extra" charge is done and followed by
-	  charge-commit-cancel protocol.
-	  At commit, charge against oldpage or newpage will be committed.
-
-2. Uncharge
-  a page/swp_entry may be uncharged (usage -= PAGE_SIZE) by
-
-	mem_cgroup_uncharge_page()
-	  Called when an anonymous page is fully unmapped. I.e., mapcount goes
-	  to 0. If the page is SwapCache, uncharge is delayed until
-	  mem_cgroup_uncharge_swapcache().
-
-	mem_cgroup_uncharge_cache_page()
-	  Called when a page-cache is deleted from radix-tree. If the page is
-	  SwapCache, uncharge is delayed until mem_cgroup_uncharge_swapcache().
-
-	mem_cgroup_uncharge_swapcache()
-	  Called when SwapCache is removed from radix-tree. The charge itself
-	  is moved to swap_cgroup. (If mem+swap controller is disabled, no
-	  charge to swap occurs.)
-
-	mem_cgroup_uncharge_swap()
-	  Called when swp_entry's refcnt goes down to 0. A charge against swap
-	  disappears.
-
-	mem_cgroup_end_migration(old, new)
-	At success of migration old is uncharged (if necessary), a charge
-	to new page is committed. At failure, charge to old page is committed.
-
-3. charge-commit-cancel
-	In some case, we can't know this "charge" is valid or not at charging
-	(because of races).
-	To handle such case, there are charge-commit-cancel functions.
-		mem_cgroup_try_charge_XXX
-		mem_cgroup_commit_charge_XXX
-		mem_cgroup_cancel_charge_XXX
-	these are used in swap-in and migration.
-
-	At try_charge(), there are no flags to say "this page is charged".
-	at this point, usage += PAGE_SIZE.
-
-	At commit(), the function checks the page should be charged or not
-	and set flags or avoid charging.(usage -= PAGE_SIZE)
-
-	At cancel(), simply usage -= PAGE_SIZE.
-
-Under below explanation, we assume CONFIG_MEM_RES_CTRL_SWAP=y.
-
-4. Anonymous
-	Anonymous page is newly allocated at
-		  - page fault into MAP_ANONYMOUS mapping.
-		  - Copy-On-Write.
- 	It is charged right after it's allocated before doing any page table
-	related operations. Of course, it's uncharged when another page is used
-	for the fault address.
-
-	At freeing anonymous page (by exit() or munmap()), zap_pte() is called
-	and pages for ptes are freed one by one.(see mm/memory.c). Uncharges
-	are done at page_remove_rmap() when page_mapcount() goes down to 0.
-
-	Another page freeing is by page-reclaim (vmscan.c) and anonymous
-	pages are swapped out. In this case, the page is marked as
-	PageSwapCache(). uncharge() routine doesn't uncharge the page marked
-	as SwapCache(). It's delayed until __delete_from_swap_cache().
-
-	4.1 Swap-in.
-	At swap-in, the page is taken from swap-cache. There are 2 cases.
-
-	(a) If the SwapCache is newly allocated and read, it has no charges.
-	(b) If the SwapCache has been mapped by processes, it has been
-	    charged already.
-
-	This swap-in is one of the most complicated work. In do_swap_page(),
-	following events occur when pte is unchanged.
-
-	(1) the page (SwapCache) is looked up.
-	(2) lock_page()
-	(3) try_charge_swapin()
-	(4) reuse_swap_page() (may call delete_swap_cache())
-	(5) commit_charge_swapin()
-	(6) swap_free().
-
-	Considering following situation for example.
-
-	(A) The page has not been charged before (2) and reuse_swap_page()
-	    doesn't call delete_from_swap_cache().
-	(B) The page has not been charged before (2) and reuse_swap_page()
-	    calls delete_from_swap_cache().
-	(C) The page has been charged before (2) and reuse_swap_page() doesn't
-	    call delete_from_swap_cache().
-	(D) The page has been charged before (2) and reuse_swap_page() calls
-	    delete_from_swap_cache().
-
-	    memory.usage/memsw.usage changes to this page/swp_entry will be
-	 Case          (A)      (B)       (C)     (D)
-         Event
-       Before (2)     0/ 1     0/ 1      1/ 1    1/ 1
-          ===========================================
-          (3)        +1/+1    +1/+1     +1/+1   +1/+1
-          (4)          -       0/ 0       -     -1/ 0
-          (5)         0/-1     0/ 0     -1/-1    0/ 0
-          (6)          -       0/-1       -      0/-1
-          ===========================================
-       Result         1/ 1     1/ 1      1/ 1    1/ 1
-
-       In any cases, charges to this page should be 1/ 1.
-
-	4.2 Swap-out.
-	At swap-out, typical state transition is below.
-
-	(a) add to swap cache. (marked as SwapCache)
-	    swp_entry's refcnt += 1.
-	(b) fully unmapped.
-	    swp_entry's refcnt += # of ptes.
-	(c) write back to swap.
-	(d) delete from swap cache. (remove from SwapCache)
-	    swp_entry's refcnt -= 1.
-
-
-	At (b), the page is marked as SwapCache and not uncharged.
-	At (d), the page is removed from SwapCache and a charge in page_cgroup
-	is moved to swap_cgroup.
-
-	Finally, at task exit,
-	(e) zap_pte() is called and swp_entry's refcnt -=1 -> 0.
-	Here, a charge in swap_cgroup disappears.
-
-5. Page Cache
-   	Page Cache is charged at
-	- add_to_page_cache_locked().
-
-	uncharged at
-	- __remove_from_page_cache().
-
-	The logic is very clear. (About migration, see below)
-	Note: __remove_from_page_cache() is called by remove_from_page_cache()
-	and __remove_mapping().
-
-6. Shmem(tmpfs) Page Cache
-	Memcg's charge/uncharge have special handlers of shmem. The best way
-	to understand shmem's page state transition is to read mm/shmem.c.
-	But brief explanation of the behavior of memcg around shmem will be
-	helpful to understand the logic.
-
-	Shmem's page (just leaf page, not direct/indirect block) can be on
-		- radix-tree of shmem's inode.
-		- SwapCache.
-		- Both on radix-tree and SwapCache. This happens at swap-in
-		  and swap-out,
-
-	It's charged when...
-	- A new page is added to shmem's radix-tree.
-	- A swp page is read. (move a charge from swap_cgroup to page_cgroup)
-	It's uncharged when
-	- A page is removed from radix-tree and not SwapCache.
-	- When SwapCache is removed, a charge is moved to swap_cgroup.
-	- When swp_entry's refcnt goes down to 0, a charge in swap_cgroup
-	  disappears.
-
-7. Page Migration
-   	One of the most complicated functions is page-migration-handler.
-	Memcg has 2 routines. Assume that we are migrating a page's contents
-	from OLDPAGE to NEWPAGE.
-
-	Usual migration logic is..
-	(a) remove the page from LRU.
-	(b) allocate NEWPAGE (migration target)
-	(c) lock by lock_page().
-	(d) unmap all mappings.
-	(e-1) If necessary, replace entry in radix-tree.
-	(e-2) move contents of a page.
-	(f) map all mappings again.
-	(g) pushback the page to LRU.
-	(-) OLDPAGE will be freed.
-
-	Before (g), memcg should complete all necessary charge/uncharge to
-	NEWPAGE/OLDPAGE.
-
-	The point is....
-	- If OLDPAGE is anonymous, all charges will be dropped at (d) because
-          try_to_unmap() drops all mapcount and the page will not be
-	  SwapCache.
-
-	- If OLDPAGE is SwapCache, charges will be kept at (g) because
-	  __delete_from_swap_cache() isn't called at (e-1)
-
-	- If OLDPAGE is page-cache, charges will be kept at (g) because
-	  remove_from_swap_cache() isn't called at (e-1)
-
-	memcg provides following hooks.
-
-	- mem_cgroup_prepare_migration(OLDPAGE)
-	  Called after (b) to account a charge (usage += PAGE_SIZE) against
-	  memcg which OLDPAGE belongs to.
-
-        - mem_cgroup_end_migration(OLDPAGE, NEWPAGE)
-	  Called after (f) before (g).
-	  If OLDPAGE is used, commit OLDPAGE again. If OLDPAGE is already
-	  charged, a charge by prepare_migration() is automatically canceled.
-	  If NEWPAGE is used, commit NEWPAGE and uncharge OLDPAGE.
-
-	  But zap_pte() (by exit or munmap) can be called while migration,
-	  we have to check if OLDPAGE/NEWPAGE is a valid page after commit().
-
-8. LRU
-        Each memcg has its own private LRU. Now, it's handling is under global
-	VM's control (means that it's handled under global zone->lru_lock).
-	Almost all routines around memcg's LRU is called by global LRU's
-	list management functions under zone->lru_lock().
-
-	A special function is mem_cgroup_isolate_pages(). This scans
-	memcg's private LRU and call __isolate_lru_page() to extract a page
-	from LRU.
-	(By __isolate_lru_page(), the page is removed from both of global and
-	 private LRU.)
-
-
-9. Typical Tests.
-
- Tests for racy cases.
-
- 9.1 Small limit to memcg.
-	When you do test to do racy case, it's good test to set memcg's limit
-	to be very small rather than GB. Many races found in the test under
-	xKB or xxMB limits.
-	(Memory behavior under GB and Memory behavior under MB shows very
-	 different situation.)
-
- 9.2 Shmem
-	Historically, memcg's shmem handling was poor and we saw some amount
-	of troubles here. This is because shmem is page-cache but can be
-	SwapCache. Test with shmem/tmpfs is always good test.
-
- 9.3 Migration
-	For NUMA, migration is an another special case. To do easy test, cpuset
-	is useful. Following is a sample script to do migration.
-
-	mount -t cgroup -o cpuset none /opt/cpuset
-
-	mkdir /opt/cpuset/01
-	echo 1 > /opt/cpuset/01/cpuset.cpus
-	echo 0 > /opt/cpuset/01/cpuset.mems
-	echo 1 > /opt/cpuset/01/cpuset.memory_migrate
-	mkdir /opt/cpuset/02
-	echo 1 > /opt/cpuset/02/cpuset.cpus
-	echo 1 > /opt/cpuset/02/cpuset.mems
-	echo 1 > /opt/cpuset/02/cpuset.memory_migrate
-
-	In above set, when you moves a task from 01 to 02, page migration to
-	node 0 to node 1 will occur. Following is a script to migrate all
-	under cpuset.
-	--
-	move_task()
-	{
-	for pid in $1
-        do
-                /bin/echo $pid >$2/tasks 2>/dev/null
-		echo -n $pid
-		echo -n " "
-        done
-	echo END
-	}
-
-	G1_TASK=`cat ${G1}/tasks`
-	G2_TASK=`cat ${G2}/tasks`
-	move_task "${G1_TASK}" ${G2} &
-	--
- 9.4 Memory hotplug.
-	memory hotplug test is one of good test.
-	to offline memory, do following.
-	# echo offline > /sys/devices/system/memory/memoryXXX/state
-	(XXX is the place of memory)
-	This is an easy way to test page migration, too.
-
- 9.5 mkdir/rmdir
-	When using hierarchy, mkdir/rmdir test should be done.
-	Use tests like the following.
-
-	echo 1 >/opt/cgroup/01/memory/use_hierarchy
-	mkdir /opt/cgroup/01/child_a
-	mkdir /opt/cgroup/01/child_b
-
-	set limit to 01.
-	add limit to 01/child_b
-	run jobs under child_a and child_b
-
-	create/delete following groups at random while jobs are running.
-	/opt/cgroup/01/child_a/child_aa
-	/opt/cgroup/01/child_b/child_bb
-	/opt/cgroup/01/child_c
-
-	running new jobs in new group is also good.
-
- 9.6 Mount with other subsystems.
-	Mounting with other subsystems is a good test because there is a
-	race and lock dependency with other cgroup subsystems.
-
-	example)
-	# mount -t cgroup none /cgroup -t cpuset,memory,cpu,devices
-
-	and do task move, mkdir, rmdir etc...under this.
diff --git a/Documentation/controllers/memory.txt b/Documentation/controllers/memory.txt
deleted file mode 100644
index e1501964df1e..000000000000
--- a/Documentation/controllers/memory.txt
+++ /dev/null
@@ -1,399 +0,0 @@
-Memory Resource Controller
-
-NOTE: The Memory Resource Controller has been generically been referred
-to as the memory controller in this document. Do not confuse memory controller
-used here with the memory controller that is used in hardware.
-
-Salient features
-
-a. Enable control of both RSS (mapped) and Page Cache (unmapped) pages
-b. The infrastructure allows easy addition of other types of memory to control
-c. Provides *zero overhead* for non memory controller users
-d. Provides a double LRU: global memory pressure causes reclaim from the
-   global LRU; a cgroup on hitting a limit, reclaims from the per
-   cgroup LRU
-
-NOTE: Swap Cache (unmapped) is not accounted now.
-
-Benefits and Purpose of the memory controller
-
-The memory controller isolates the memory behaviour of a group of tasks
-from the rest of the system. The article on LWN [12] mentions some probable
-uses of the memory controller. The memory controller can be used to
-
-a. Isolate an application or a group of applications
-   Memory hungry applications can be isolated and limited to a smaller
-   amount of memory.
-b. Create a cgroup with limited amount of memory, this can be used
-   as a good alternative to booting with mem=XXXX.
-c. Virtualization solutions can control the amount of memory they want
-   to assign to a virtual machine instance.
-d. A CD/DVD burner could control the amount of memory used by the
-   rest of the system to ensure that burning does not fail due to lack
-   of available memory.
-e. There are several other use cases, find one or use the controller just
-   for fun (to learn and hack on the VM subsystem).
-
-1. History
-
-The memory controller has a long history. A request for comments for the memory
-controller was posted by Balbir Singh [1]. At the time the RFC was posted
-there were several implementations for memory control. The goal of the
-RFC was to build consensus and agreement for the minimal features required
-for memory control. The first RSS controller was posted by Balbir Singh[2]
-in Feb 2007. Pavel Emelianov [3][4][5] has since posted three versions of the
-RSS controller. At OLS, at the resource management BoF, everyone suggested
-that we handle both page cache and RSS together. Another request was raised
-to allow user space handling of OOM. The current memory controller is
-at version 6; it combines both mapped (RSS) and unmapped Page
-Cache Control [11].
-
-2. Memory Control
-
-Memory is a unique resource in the sense that it is present in a limited
-amount. If a task requires a lot of CPU processing, the task can spread
-its processing over a period of hours, days, months or years, but with
-memory, the same physical memory needs to be reused to accomplish the task.
-
-The memory controller implementation has been divided into phases. These
-are:
-
-1. Memory controller
-2. mlock(2) controller
-3. Kernel user memory accounting and slab control
-4. user mappings length controller
-
-The memory controller is the first controller developed.
-
-2.1. Design
-
-The core of the design is a counter called the res_counter. The res_counter
-tracks the current memory usage and limit of the group of processes associated
-with the controller. Each cgroup has a memory controller specific data
-structure (mem_cgroup) associated with it.
-
-2.2. Accounting
-
-		+--------------------+
-		|  mem_cgroup     |
-		|  (res_counter)     |
-		+--------------------+
-		 /            ^      \
-		/             |       \
-           +---------------+  |        +---------------+
-           | mm_struct     |  |....    | mm_struct     |
-           |               |  |        |               |
-           +---------------+  |        +---------------+
-                              |
-                              + --------------+
-                                              |
-           +---------------+           +------+--------+
-           | page          +---------->  page_cgroup|
-           |               |           |               |
-           +---------------+           +---------------+
-
-             (Figure 1: Hierarchy of Accounting)
-
-
-Figure 1 shows the important aspects of the controller
-
-1. Accounting happens per cgroup
-2. Each mm_struct knows about which cgroup it belongs to
-3. Each page has a pointer to the page_cgroup, which in turn knows the
-   cgroup it belongs to
-
-The accounting is done as follows: mem_cgroup_charge() is invoked to setup
-the necessary data structures and check if the cgroup that is being charged
-is over its limit. If it is then reclaim is invoked on the cgroup.
-More details can be found in the reclaim section of this document.
-If everything goes well, a page meta-data-structure called page_cgroup is
-allocated and associated with the page.  This routine also adds the page to
-the per cgroup LRU.
-
-2.2.1 Accounting details
-
-All mapped anon pages (RSS) and cache pages (Page Cache) are accounted.
-(some pages which never be reclaimable and will not be on global LRU
- are not accounted. we just accounts pages under usual vm management.)
-
-RSS pages are accounted at page_fault unless they've already been accounted
-for earlier. A file page will be accounted for as Page Cache when it's
-inserted into inode (radix-tree). While it's mapped into the page tables of
-processes, duplicate accounting is carefully avoided.
-
-A RSS page is unaccounted when it's fully unmapped. A PageCache page is
-unaccounted when it's removed from radix-tree.
-
-At page migration, accounting information is kept.
-
-Note: we just account pages-on-lru because our purpose is to control amount
-of used pages. not-on-lru pages are tend to be out-of-control from vm view.
-
-2.3 Shared Page Accounting
-
-Shared pages are accounted on the basis of the first touch approach. The
-cgroup that first touches a page is accounted for the page. The principle
-behind this approach is that a cgroup that aggressively uses a shared
-page will eventually get charged for it (once it is uncharged from
-the cgroup that brought it in -- this will happen on memory pressure).
-
-Exception: If CONFIG_CGROUP_CGROUP_MEM_RES_CTLR_SWAP is not used..
-When you do swapoff and make swapped-out pages of shmem(tmpfs) to
-be backed into memory in force, charges for pages are accounted against the
-caller of swapoff rather than the users of shmem.
-
-
-2.4 Swap Extension (CONFIG_CGROUP_MEM_RES_CTLR_SWAP)
-Swap Extension allows you to record charge for swap. A swapped-in page is
-charged back to original page allocator if possible.
-
-When swap is accounted, following files are added.
- - memory.memsw.usage_in_bytes.
- - memory.memsw.limit_in_bytes.
-
-usage of mem+swap is limited by memsw.limit_in_bytes.
-
-Note: why 'mem+swap' rather than swap.
-The global LRU(kswapd) can swap out arbitrary pages. Swap-out means
-to move account from memory to swap...there is no change in usage of
-mem+swap.
-
-In other words, when we want to limit the usage of swap without affecting
-global LRU, mem+swap limit is better than just limiting swap from OS point
-of view.
-
-2.5 Reclaim
-
-Each cgroup maintains a per cgroup LRU that consists of an active
-and inactive list. When a cgroup goes over its limit, we first try
-to reclaim memory from the cgroup so as to make space for the new
-pages that the cgroup has touched. If the reclaim is unsuccessful,
-an OOM routine is invoked to select and kill the bulkiest task in the
-cgroup.
-
-The reclaim algorithm has not been modified for cgroups, except that
-pages that are selected for reclaiming come from the per cgroup LRU
-list.
-
-2. Locking
-
-The memory controller uses the following hierarchy
-
-1. zone->lru_lock is used for selecting pages to be isolated
-2. mem->per_zone->lru_lock protects the per cgroup LRU (per zone)
-3. lock_page_cgroup() is used to protect page->page_cgroup
-
-3. User Interface
-
-0. Configuration
-
-a. Enable CONFIG_CGROUPS
-b. Enable CONFIG_RESOURCE_COUNTERS
-c. Enable CONFIG_CGROUP_MEM_RES_CTLR
-
-1. Prepare the cgroups
-# mkdir -p /cgroups
-# mount -t cgroup none /cgroups -o memory
-
-2. Make the new group and move bash into it
-# mkdir /cgroups/0
-# echo $$ >  /cgroups/0/tasks
-
-Since now we're in the 0 cgroup,
-We can alter the memory limit:
-# echo 4M > /cgroups/0/memory.limit_in_bytes
-
-NOTE: We can use a suffix (k, K, m, M, g or G) to indicate values in kilo,
-mega or gigabytes.
-
-# cat /cgroups/0/memory.limit_in_bytes
-4194304
-
-NOTE: The interface has now changed to display the usage in bytes
-instead of pages
-
-We can check the usage:
-# cat /cgroups/0/memory.usage_in_bytes
-1216512
-
-A successful write to this file does not guarantee a successful set of
-this limit to the value written into the file.  This can be due to a
-number of factors, such as rounding up to page boundaries or the total
-availability of memory on the system.  The user is required to re-read
-this file after a write to guarantee the value committed by the kernel.
-
-# echo 1 > memory.limit_in_bytes
-# cat memory.limit_in_bytes
-4096
-
-The memory.failcnt field gives the number of times that the cgroup limit was
-exceeded.
-
-The memory.stat file gives accounting information. Now, the number of
-caches, RSS and Active pages/Inactive pages are shown.
-
-4. Testing
-
-Balbir posted lmbench, AIM9, LTP and vmmstress results [10] and [11].
-Apart from that v6 has been tested with several applications and regular
-daily use. The controller has also been tested on the PPC64, x86_64 and
-UML platforms.
-
-4.1 Troubleshooting
-
-Sometimes a user might find that the application under a cgroup is
-terminated. There are several causes for this:
-
-1. The cgroup limit is too low (just too low to do anything useful)
-2. The user is using anonymous memory and swap is turned off or too low
-
-A sync followed by echo 1 > /proc/sys/vm/drop_caches will help get rid of
-some of the pages cached in the cgroup (page cache pages).
-
-4.2 Task migration
-
-When a task migrates from one cgroup to another, it's charge is not
-carried forward. The pages allocated from the original cgroup still
-remain charged to it, the charge is dropped when the page is freed or
-reclaimed.
-
-4.3 Removing a cgroup
-
-A cgroup can be removed by rmdir, but as discussed in sections 4.1 and 4.2, a
-cgroup might have some charge associated with it, even though all
-tasks have migrated away from it.
-Such charges are freed(at default) or moved to its parent. When moved,
-both of RSS and CACHES are moved to parent.
-If both of them are busy, rmdir() returns -EBUSY. See 5.1 Also.
-
-Charges recorded in swap information is not updated at removal of cgroup.
-Recorded information is discarded and a cgroup which uses swap (swapcache)
-will be charged as a new owner of it.
-
-
-5. Misc. interfaces.
-
-5.1 force_empty
-  memory.force_empty interface is provided to make cgroup's memory usage empty.
-  You can use this interface only when the cgroup has no tasks.
-  When writing anything to this
-
-  # echo 0 > memory.force_empty
-
-  Almost all pages tracked by this memcg will be unmapped and freed. Some of
-  pages cannot be freed because it's locked or in-use. Such pages are moved
-  to parent and this cgroup will be empty. But this may return -EBUSY in
-  some too busy case.
-
-  Typical use case of this interface is that calling this before rmdir().
-  Because rmdir() moves all pages to parent, some out-of-use page caches can be
-  moved to the parent. If you want to avoid that, force_empty will be useful.
-
-5.2 stat file
-  memory.stat file includes following statistics (now)
-	cache			- # of pages from page-cache and shmem.
-	rss			- # of pages from anonymous memory.
-	pgpgin			- # of event of charging
-	pgpgout			- # of event of uncharging
-	active_anon		- # of pages on active lru of anon, shmem.
-	inactive_anon 		- # of pages on active lru of anon, shmem
-	active_file		- # of pages on active lru of file-cache
-	inactive_file		- # of pages on inactive lru of file cache
-	unevictable		- # of pages cannot be reclaimed.(mlocked etc)
-
-	Below is depend on CONFIG_DEBUG_VM.
-	inactive_ratio		- VM inernal parameter. (see mm/page_alloc.c)
-	recent_rotated_anon	- VM internal parameter. (see mm/vmscan.c)
-	recent_rotated_file	- VM internal parameter. (see mm/vmscan.c)
-	recent_scanned_anon 	- VM internal parameter. (see mm/vmscan.c)
-	recent_scanned_file 	- VM internal parameter. (see mm/vmscan.c)
-
-  Memo:
-	recent_rotated means recent frequency of lru rotation.
-	recent_scanned means recent # of scans to lru.
-	showing for better debug please see the code for meanings.
-
-
-5.3 swappiness
-  Similar to /proc/sys/vm/swappiness, but affecting a hierarchy of groups only.
-
-  Following cgroup's swapiness can't be changed.
-  - root cgroup (uses /proc/sys/vm/swappiness).
-  - a cgroup which uses hierarchy and it has child cgroup.
-  - a cgroup which uses hierarchy and not the root of hierarchy.
-
-
-6. Hierarchy support
-
-The memory controller supports a deep hierarchy and hierarchical accounting.
-The hierarchy is created by creating the appropriate cgroups in the
-cgroup filesystem. Consider for example, the following cgroup filesystem
-hierarchy
-
-		root
-	     /  |   \
-           /	|    \
-	  a	b	c
-			| \
-			|  \
-			d   e
-
-In the diagram above, with hierarchical accounting enabled, all memory
-usage of e, is accounted to its ancestors up until the root (i.e, c and root),
-that has memory.use_hierarchy enabled.  If one of the ancestors goes over its
-limit, the reclaim algorithm reclaims from the tasks in the ancestor and the
-children of the ancestor.
-
-6.1 Enabling hierarchical accounting and reclaim
-
-The memory controller by default disables the hierarchy feature. Support
-can be enabled by writing 1 to memory.use_hierarchy file of the root cgroup
-
-# echo 1 > memory.use_hierarchy
-
-The feature can be disabled by
-
-# echo 0 > memory.use_hierarchy
-
-NOTE1: Enabling/disabling will fail if the cgroup already has other
-cgroups created below it.
-
-NOTE2: This feature can be enabled/disabled per subtree.
-
-7. TODO
-
-1. Add support for accounting huge pages (as a separate controller)
-2. Make per-cgroup scanner reclaim not-shared pages first
-3. Teach controller to account for shared-pages
-4. Start reclamation in the background when the limit is
-   not yet hit but the usage is getting closer
-
-Summary
-
-Overall, the memory controller has been a stable controller and has been
-commented and discussed quite extensively in the community.
-
-References
-
-1. Singh, Balbir. RFC: Memory Controller, http://lwn.net/Articles/206697/
-2. Singh, Balbir. Memory Controller (RSS Control),
-   http://lwn.net/Articles/222762/
-3. Emelianov, Pavel. Resource controllers based on process cgroups
-   http://lkml.org/lkml/2007/3/6/198
-4. Emelianov, Pavel. RSS controller based on process cgroups (v2)
-   http://lkml.org/lkml/2007/4/9/78
-5. Emelianov, Pavel. RSS controller based on process cgroups (v3)
-   http://lkml.org/lkml/2007/5/30/244
-6. Menage, Paul. Control Groups v10, http://lwn.net/Articles/236032/
-7. Vaidyanathan, Srinivasan, Control Groups: Pagecache accounting and control
-   subsystem (v3), http://lwn.net/Articles/235534/
-8. Singh, Balbir. RSS controller v2 test results (lmbench),
-   http://lkml.org/lkml/2007/5/17/232
-9. Singh, Balbir. RSS controller v2 AIM9 results
-   http://lkml.org/lkml/2007/5/18/1
-10. Singh, Balbir. Memory controller v6 test results,
-    http://lkml.org/lkml/2007/8/19/36
-11. Singh, Balbir. Memory controller introduction (v6),
-    http://lkml.org/lkml/2007/8/17/69
-12. Corbet, Jonathan, Controlling memory use in cgroups,
-    http://lwn.net/Articles/243795/
diff --git a/Documentation/controllers/resource_counter.txt b/Documentation/controllers/resource_counter.txt
deleted file mode 100644
index f196ac1d7d25..000000000000
--- a/Documentation/controllers/resource_counter.txt
+++ /dev/null
@@ -1,181 +0,0 @@
-
-		The Resource Counter
-
-The resource counter, declared at include/linux/res_counter.h,
-is supposed to facilitate the resource management by controllers
-by providing common stuff for accounting.
-
-This "stuff" includes the res_counter structure and routines
-to work with it.
-
-
-
-1. Crucial parts of the res_counter structure
-
- a. unsigned long long usage
-
- 	The usage value shows the amount of a resource that is consumed
-	by a group at a given time. The units of measurement should be
-	determined by the controller that uses this counter. E.g. it can
-	be bytes, items or any other unit the controller operates on.
-
- b. unsigned long long max_usage
-
- 	The maximal value of the usage over time.
-
- 	This value is useful when gathering statistical information about
-	the particular group, as it shows the actual resource requirements
-	for a particular group, not just some usage snapshot.
-
- c. unsigned long long limit
-
- 	The maximal allowed amount of resource to consume by the group. In
-	case the group requests for more resources, so that the usage value
-	would exceed the limit, the resource allocation is rejected (see
-	the next section).
-
- d. unsigned long long failcnt
-
- 	The failcnt stands for "failures counter". This is the number of
-	resource allocation attempts that failed.
-
- c. spinlock_t lock
-
- 	Protects changes of the above values.
-
-
-
-2. Basic accounting routines
-
- a. void res_counter_init(struct res_counter *rc)
-
- 	Initializes the resource counter. As usual, should be the first
-	routine called for a new counter.
-
- b. int res_counter_charge[_locked]
-			(struct res_counter *rc, unsigned long val)
-
-	When a resource is about to be allocated it has to be accounted
-	with the appropriate resource counter (controller should determine
-	which one to use on its own). This operation is called "charging".
-
-	This is not very important which operation - resource allocation
-	or charging - is performed first, but
-	  * if the allocation is performed first, this may create a
-	    temporary resource over-usage by the time resource counter is
-	    charged;
-	  * if the charging is performed first, then it should be uncharged
-	    on error path (if the one is called).
-
- c. void res_counter_uncharge[_locked]
-			(struct res_counter *rc, unsigned long val)
-
-	When a resource is released (freed) it should be de-accounted
-	from the resource counter it was accounted to.  This is called
-	"uncharging".
-
-    The _locked routines imply that the res_counter->lock is taken.
-
-
- 2.1 Other accounting routines
-
-    There are more routines that may help you with common needs, like
-    checking whether the limit is reached or resetting the max_usage
-    value. They are all declared in include/linux/res_counter.h.
-
-
-
-3. Analyzing the resource counter registrations
-
- a. If the failcnt value constantly grows, this means that the counter's
-    limit is too tight. Either the group is misbehaving and consumes too
-    many resources, or the configuration is not suitable for the group
-    and the limit should be increased.
-
- b. The max_usage value can be used to quickly tune the group. One may
-    set the limits to maximal values and either load the container with
-    a common pattern or leave one for a while. After this the max_usage
-    value shows the amount of memory the container would require during
-    its common activity.
-
-    Setting the limit a bit above this value gives a pretty good
-    configuration that works in most of the cases.
-
- c. If the max_usage is much less than the limit, but the failcnt value
-    is growing, then the group tries to allocate a big chunk of resource
-    at once.
-
- d. If the max_usage is much less than the limit, but the failcnt value
-    is 0, then this group is given too high limit, that it does not
-    require. It is better to lower the limit a bit leaving more resource
-    for other groups.
-
-
-
-4. Communication with the control groups subsystem (cgroups)
-
-All the resource controllers that are using cgroups and resource counters
-should provide files (in the cgroup filesystem) to work with the resource
-counter fields. They are recommended to adhere to the following rules:
-
- a. File names
-
- 	Field name	File name
-	---------------------------------------------------
-	usage		usage_in_<unit_of_measurement>
-	max_usage	max_usage_in_<unit_of_measurement>
-	limit		limit_in_<unit_of_measurement>
-	failcnt		failcnt
-	lock		no file :)
-
- b. Reading from file should show the corresponding field value in the
-    appropriate format.
-
- c. Writing to file
-
- 	Field		Expected behavior
-	----------------------------------
-	usage		prohibited
-	max_usage	reset to usage
-	limit		set the limit
-	failcnt		reset to zero
-
-
-
-5. Usage example
-
- a. Declare a task group (take a look at cgroups subsystem for this) and
-    fold a res_counter into it
-
-	struct my_group {
-		struct res_counter res;
-
-		<other fields>
-	}
-
- b. Put hooks in resource allocation/release paths
-
- 	int alloc_something(...)
-	{
-		if (res_counter_charge(res_counter_ptr, amount) < 0)
-			return -ENOMEM;
-
-		<allocate the resource and return to the caller>
-	}
-
-	void release_something(...)
-	{
-		res_counter_uncharge(res_counter_ptr, amount);
-
-		<release the resource>
-	}
-
-    In order to keep the usage value self-consistent, both the
-    "res_counter_ptr" and the "amount" in release_something() should be
-    the same as they were in the alloc_something() when the releasing
-    resource was allocated.
-
- c. Provide the way to read res_counter values and set them (the cgroups
-    still can help with it).
-
- c. Compile and run :)
diff --git a/Documentation/cpusets.txt b/Documentation/cpusets.txt
deleted file mode 100644
index 5c86c258c791..000000000000
--- a/Documentation/cpusets.txt
+++ /dev/null
@@ -1,808 +0,0 @@
-				CPUSETS
-				-------
-
-Copyright (C) 2004 BULL SA.
-Written by Simon.Derr@bull.net
-
-Portions Copyright (c) 2004-2006 Silicon Graphics, Inc.
-Modified by Paul Jackson <pj@sgi.com>
-Modified by Christoph Lameter <clameter@sgi.com>
-Modified by Paul Menage <menage@google.com>
-Modified by Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
-
-CONTENTS:
-=========
-
-1. Cpusets
-  1.1 What are cpusets ?
-  1.2 Why are cpusets needed ?
-  1.3 How are cpusets implemented ?
-  1.4 What are exclusive cpusets ?
-  1.5 What is memory_pressure ?
-  1.6 What is memory spread ?
-  1.7 What is sched_load_balance ?
-  1.8 What is sched_relax_domain_level ?
-  1.9 How do I use cpusets ?
-2. Usage Examples and Syntax
-  2.1 Basic Usage
-  2.2 Adding/removing cpus
-  2.3 Setting flags
-  2.4 Attaching processes
-3. Questions
-4. Contact
-
-1. Cpusets
-==========
-
-1.1 What are cpusets ?
-----------------------
-
-Cpusets provide a mechanism for assigning a set of CPUs and Memory
-Nodes to a set of tasks.   In this document "Memory Node" refers to
-an on-line node that contains memory.
-
-Cpusets constrain the CPU and Memory placement of tasks to only
-the resources within a tasks current cpuset.  They form a nested
-hierarchy visible in a virtual file system.  These are the essential
-hooks, beyond what is already present, required to manage dynamic
-job placement on large systems.
-
-Cpusets use the generic cgroup subsystem described in
-Documentation/cgroups/cgroups.txt.
-
-Requests by a task, using the sched_setaffinity(2) system call to
-include CPUs in its CPU affinity mask, and using the mbind(2) and
-set_mempolicy(2) system calls to include Memory Nodes in its memory
-policy, are both filtered through that tasks cpuset, filtering out any
-CPUs or Memory Nodes not in that cpuset.  The scheduler will not
-schedule a task on a CPU that is not allowed in its cpus_allowed
-vector, and the kernel page allocator will not allocate a page on a
-node that is not allowed in the requesting tasks mems_allowed vector.
-
-User level code may create and destroy cpusets by name in the cgroup
-virtual file system, manage the attributes and permissions of these
-cpusets and which CPUs and Memory Nodes are assigned to each cpuset,
-specify and query to which cpuset a task is assigned, and list the
-task pids assigned to a cpuset.
-
-
-1.2 Why are cpusets needed ?
-----------------------------
-
-The management of large computer systems, with many processors (CPUs),
-complex memory cache hierarchies and multiple Memory Nodes having
-non-uniform access times (NUMA) presents additional challenges for
-the efficient scheduling and memory placement of processes.
-
-Frequently more modest sized systems can be operated with adequate
-efficiency just by letting the operating system automatically share
-the available CPU and Memory resources amongst the requesting tasks.
-
-But larger systems, which benefit more from careful processor and
-memory placement to reduce memory access times and contention,
-and which typically represent a larger investment for the customer,
-can benefit from explicitly placing jobs on properly sized subsets of
-the system.
-
-This can be especially valuable on:
-
-    * Web Servers running multiple instances of the same web application,
-    * Servers running different applications (for instance, a web server
-      and a database), or
-    * NUMA systems running large HPC applications with demanding
-      performance characteristics.
-
-These subsets, or "soft partitions" must be able to be dynamically
-adjusted, as the job mix changes, without impacting other concurrently
-executing jobs. The location of the running jobs pages may also be moved
-when the memory locations are changed.
-
-The kernel cpuset patch provides the minimum essential kernel
-mechanisms required to efficiently implement such subsets.  It
-leverages existing CPU and Memory Placement facilities in the Linux
-kernel to avoid any additional impact on the critical scheduler or
-memory allocator code.
-
-
-1.3 How are cpusets implemented ?
----------------------------------
-
-Cpusets provide a Linux kernel mechanism to constrain which CPUs and
-Memory Nodes are used by a process or set of processes.
-
-The Linux kernel already has a pair of mechanisms to specify on which
-CPUs a task may be scheduled (sched_setaffinity) and on which Memory
-Nodes it may obtain memory (mbind, set_mempolicy).
-
-Cpusets extends these two mechanisms as follows:
-
- - Cpusets are sets of allowed CPUs and Memory Nodes, known to the
-   kernel.
- - Each task in the system is attached to a cpuset, via a pointer
-   in the task structure to a reference counted cgroup structure.
- - Calls to sched_setaffinity are filtered to just those CPUs
-   allowed in that tasks cpuset.
- - Calls to mbind and set_mempolicy are filtered to just
-   those Memory Nodes allowed in that tasks cpuset.
- - The root cpuset contains all the systems CPUs and Memory
-   Nodes.
- - For any cpuset, one can define child cpusets containing a subset
-   of the parents CPU and Memory Node resources.
- - The hierarchy of cpusets can be mounted at /dev/cpuset, for
-   browsing and manipulation from user space.
- - A cpuset may be marked exclusive, which ensures that no other
-   cpuset (except direct ancestors and descendents) may contain
-   any overlapping CPUs or Memory Nodes.
- - You can list all the tasks (by pid) attached to any cpuset.
-
-The implementation of cpusets requires a few, simple hooks
-into the rest of the kernel, none in performance critical paths:
-
- - in init/main.c, to initialize the root cpuset at system boot.
- - in fork and exit, to attach and detach a task from its cpuset.
- - in sched_setaffinity, to mask the requested CPUs by what's
-   allowed in that tasks cpuset.
- - in sched.c migrate_all_tasks(), to keep migrating tasks within
-   the CPUs allowed by their cpuset, if possible.
- - in the mbind and set_mempolicy system calls, to mask the requested
-   Memory Nodes by what's allowed in that tasks cpuset.
- - in page_alloc.c, to restrict memory to allowed nodes.
- - in vmscan.c, to restrict page recovery to the current cpuset.
-
-You should mount the "cgroup" filesystem type in order to enable
-browsing and modifying the cpusets presently known to the kernel.  No
-new system calls are added for cpusets - all support for querying and
-modifying cpusets is via this cpuset file system.
-
-The /proc/<pid>/status file for each task has four added lines,
-displaying the tasks cpus_allowed (on which CPUs it may be scheduled)
-and mems_allowed (on which Memory Nodes it may obtain memory),
-in the two formats seen in the following example:
-
-  Cpus_allowed:   ffffffff,ffffffff,ffffffff,ffffffff
-  Cpus_allowed_list:      0-127
-  Mems_allowed:   ffffffff,ffffffff
-  Mems_allowed_list:      0-63
-
-Each cpuset is represented by a directory in the cgroup file system
-containing (on top of the standard cgroup files) the following
-files describing that cpuset:
-
- - cpus: list of CPUs in that cpuset
- - mems: list of Memory Nodes in that cpuset
- - memory_migrate flag: if set, move pages to cpusets nodes
- - cpu_exclusive flag: is cpu placement exclusive?
- - mem_exclusive flag: is memory placement exclusive?
- - mem_hardwall flag:  is memory allocation hardwalled
- - memory_pressure: measure of how much paging pressure in cpuset
-
-In addition, the root cpuset only has the following file:
- - memory_pressure_enabled flag: compute memory_pressure?
-
-New cpusets are created using the mkdir system call or shell
-command.  The properties of a cpuset, such as its flags, allowed
-CPUs and Memory Nodes, and attached tasks, are modified by writing
-to the appropriate file in that cpusets directory, as listed above.
-
-The named hierarchical structure of nested cpusets allows partitioning
-a large system into nested, dynamically changeable, "soft-partitions".
-
-The attachment of each task, automatically inherited at fork by any
-children of that task, to a cpuset allows organizing the work load
-on a system into related sets of tasks such that each set is constrained
-to using the CPUs and Memory Nodes of a particular cpuset.  A task
-may be re-attached to any other cpuset, if allowed by the permissions
-on the necessary cpuset file system directories.
-
-Such management of a system "in the large" integrates smoothly with
-the detailed placement done on individual tasks and memory regions
-using the sched_setaffinity, mbind and set_mempolicy system calls.
-
-The following rules apply to each cpuset:
-
- - Its CPUs and Memory Nodes must be a subset of its parents.
- - It can't be marked exclusive unless its parent is.
- - If its cpu or memory is exclusive, they may not overlap any sibling.
-
-These rules, and the natural hierarchy of cpusets, enable efficient
-enforcement of the exclusive guarantee, without having to scan all
-cpusets every time any of them change to ensure nothing overlaps a
-exclusive cpuset.  Also, the use of a Linux virtual file system (vfs)
-to represent the cpuset hierarchy provides for a familiar permission
-and name space for cpusets, with a minimum of additional kernel code.
-
-The cpus and mems files in the root (top_cpuset) cpuset are
-read-only.  The cpus file automatically tracks the value of
-cpu_online_map using a CPU hotplug notifier, and the mems file
-automatically tracks the value of node_states[N_HIGH_MEMORY]--i.e.,
-nodes with memory--using the cpuset_track_online_nodes() hook.
-
-
-1.4 What are exclusive cpusets ?
---------------------------------
-
-If a cpuset is cpu or mem exclusive, no other cpuset, other than
-a direct ancestor or descendent, may share any of the same CPUs or
-Memory Nodes.
-
-A cpuset that is mem_exclusive *or* mem_hardwall is "hardwalled",
-i.e. it restricts kernel allocations for page, buffer and other data
-commonly shared by the kernel across multiple users.  All cpusets,
-whether hardwalled or not, restrict allocations of memory for user
-space.  This enables configuring a system so that several independent
-jobs can share common kernel data, such as file system pages, while
-isolating each job's user allocation in its own cpuset.  To do this,
-construct a large mem_exclusive cpuset to hold all the jobs, and
-construct child, non-mem_exclusive cpusets for each individual job.
-Only a small amount of typical kernel memory, such as requests from
-interrupt handlers, is allowed to be taken outside even a
-mem_exclusive cpuset.
-
-
-1.5 What is memory_pressure ?
------------------------------
-The memory_pressure of a cpuset provides a simple per-cpuset metric
-of the rate that the tasks in a cpuset are attempting to free up in
-use memory on the nodes of the cpuset to satisfy additional memory
-requests.
-
-This enables batch managers monitoring jobs running in dedicated
-cpusets to efficiently detect what level of memory pressure that job
-is causing.
-
-This is useful both on tightly managed systems running a wide mix of
-submitted jobs, which may choose to terminate or re-prioritize jobs that
-are trying to use more memory than allowed on the nodes assigned them,
-and with tightly coupled, long running, massively parallel scientific
-computing jobs that will dramatically fail to meet required performance
-goals if they start to use more memory than allowed to them.
-
-This mechanism provides a very economical way for the batch manager
-to monitor a cpuset for signs of memory pressure.  It's up to the
-batch manager or other user code to decide what to do about it and
-take action.
-
-==> Unless this feature is enabled by writing "1" to the special file
-    /dev/cpuset/memory_pressure_enabled, the hook in the rebalance
-    code of __alloc_pages() for this metric reduces to simply noticing
-    that the cpuset_memory_pressure_enabled flag is zero.  So only
-    systems that enable this feature will compute the metric.
-
-Why a per-cpuset, running average:
-
-    Because this meter is per-cpuset, rather than per-task or mm,
-    the system load imposed by a batch scheduler monitoring this
-    metric is sharply reduced on large systems, because a scan of
-    the tasklist can be avoided on each set of queries.
-
-    Because this meter is a running average, instead of an accumulating
-    counter, a batch scheduler can detect memory pressure with a
-    single read, instead of having to read and accumulate results
-    for a period of time.
-
-    Because this meter is per-cpuset rather than per-task or mm,
-    the batch scheduler can obtain the key information, memory
-    pressure in a cpuset, with a single read, rather than having to
-    query and accumulate results over all the (dynamically changing)
-    set of tasks in the cpuset.
-
-A per-cpuset simple digital filter (requires a spinlock and 3 words
-of data per-cpuset) is kept, and updated by any task attached to that
-cpuset, if it enters the synchronous (direct) page reclaim code.
-
-A per-cpuset file provides an integer number representing the recent
-(half-life of 10 seconds) rate of direct page reclaims caused by
-the tasks in the cpuset, in units of reclaims attempted per second,
-times 1000.
-
-
-1.6 What is memory spread ?
----------------------------
-There are two boolean flag files per cpuset that control where the
-kernel allocates pages for the file system buffers and related in
-kernel data structures.  They are called 'memory_spread_page' and
-'memory_spread_slab'.
-
-If the per-cpuset boolean flag file 'memory_spread_page' is set, then
-the kernel will spread the file system buffers (page cache) evenly
-over all the nodes that the faulting task is allowed to use, instead
-of preferring to put those pages on the node where the task is running.
-
-If the per-cpuset boolean flag file 'memory_spread_slab' is set,
-then the kernel will spread some file system related slab caches,
-such as for inodes and dentries evenly over all the nodes that the
-faulting task is allowed to use, instead of preferring to put those
-pages on the node where the task is running.
-
-The setting of these flags does not affect anonymous data segment or
-stack segment pages of a task.
-
-By default, both kinds of memory spreading are off, and memory
-pages are allocated on the node local to where the task is running,
-except perhaps as modified by the tasks NUMA mempolicy or cpuset
-configuration, so long as sufficient free memory pages are available.
-
-When new cpusets are created, they inherit the memory spread settings
-of their parent.
-
-Setting memory spreading causes allocations for the affected page
-or slab caches to ignore the tasks NUMA mempolicy and be spread
-instead.    Tasks using mbind() or set_mempolicy() calls to set NUMA
-mempolicies will not notice any change in these calls as a result of
-their containing tasks memory spread settings.  If memory spreading
-is turned off, then the currently specified NUMA mempolicy once again
-applies to memory page allocations.
-
-Both 'memory_spread_page' and 'memory_spread_slab' are boolean flag
-files.  By default they contain "0", meaning that the feature is off
-for that cpuset.  If a "1" is written to that file, then that turns
-the named feature on.
-
-The implementation is simple.
-
-Setting the flag 'memory_spread_page' turns on a per-process flag
-PF_SPREAD_PAGE for each task that is in that cpuset or subsequently
-joins that cpuset.  The page allocation calls for the page cache
-is modified to perform an inline check for this PF_SPREAD_PAGE task
-flag, and if set, a call to a new routine cpuset_mem_spread_node()
-returns the node to prefer for the allocation.
-
-Similarly, setting 'memory_spread_slab' turns on the flag
-PF_SPREAD_SLAB, and appropriately marked slab caches will allocate
-pages from the node returned by cpuset_mem_spread_node().
-
-The cpuset_mem_spread_node() routine is also simple.  It uses the
-value of a per-task rotor cpuset_mem_spread_rotor to select the next
-node in the current tasks mems_allowed to prefer for the allocation.
-
-This memory placement policy is also known (in other contexts) as
-round-robin or interleave.
-
-This policy can provide substantial improvements for jobs that need
-to place thread local data on the corresponding node, but that need
-to access large file system data sets that need to be spread across
-the several nodes in the jobs cpuset in order to fit.  Without this
-policy, especially for jobs that might have one thread reading in the
-data set, the memory allocation across the nodes in the jobs cpuset
-can become very uneven.
-
-1.7 What is sched_load_balance ?
---------------------------------
-
-The kernel scheduler (kernel/sched.c) automatically load balances
-tasks.  If one CPU is underutilized, kernel code running on that
-CPU will look for tasks on other more overloaded CPUs and move those
-tasks to itself, within the constraints of such placement mechanisms
-as cpusets and sched_setaffinity.
-
-The algorithmic cost of load balancing and its impact on key shared
-kernel data structures such as the task list increases more than
-linearly with the number of CPUs being balanced.  So the scheduler
-has support to  partition the systems CPUs into a number of sched
-domains such that it only load balances within each sched domain.
-Each sched domain covers some subset of the CPUs in the system;
-no two sched domains overlap; some CPUs might not be in any sched
-domain and hence won't be load balanced.
-
-Put simply, it costs less to balance between two smaller sched domains
-than one big one, but doing so means that overloads in one of the
-two domains won't be load balanced to the other one.
-
-By default, there is one sched domain covering all CPUs, except those
-marked isolated using the kernel boot time "isolcpus=" argument.
-
-This default load balancing across all CPUs is not well suited for
-the following two situations:
- 1) On large systems, load balancing across many CPUs is expensive.
-    If the system is managed using cpusets to place independent jobs
-    on separate sets of CPUs, full load balancing is unnecessary.
- 2) Systems supporting realtime on some CPUs need to minimize
-    system overhead on those CPUs, including avoiding task load
-    balancing if that is not needed.
-
-When the per-cpuset flag "sched_load_balance" is enabled (the default
-setting), it requests that all the CPUs in that cpusets allowed 'cpus'
-be contained in a single sched domain, ensuring that load balancing
-can move a task (not otherwised pinned, as by sched_setaffinity)
-from any CPU in that cpuset to any other.
-
-When the per-cpuset flag "sched_load_balance" is disabled, then the
-scheduler will avoid load balancing across the CPUs in that cpuset,
---except-- in so far as is necessary because some overlapping cpuset
-has "sched_load_balance" enabled.
-
-So, for example, if the top cpuset has the flag "sched_load_balance"
-enabled, then the scheduler will have one sched domain covering all
-CPUs, and the setting of the "sched_load_balance" flag in any other
-cpusets won't matter, as we're already fully load balancing.
-
-Therefore in the above two situations, the top cpuset flag
-"sched_load_balance" should be disabled, and only some of the smaller,
-child cpusets have this flag enabled.
-
-When doing this, you don't usually want to leave any unpinned tasks in
-the top cpuset that might use non-trivial amounts of CPU, as such tasks
-may be artificially constrained to some subset of CPUs, depending on
-the particulars of this flag setting in descendent cpusets.  Even if
-such a task could use spare CPU cycles in some other CPUs, the kernel
-scheduler might not consider the possibility of load balancing that
-task to that underused CPU.
-
-Of course, tasks pinned to a particular CPU can be left in a cpuset
-that disables "sched_load_balance" as those tasks aren't going anywhere
-else anyway.
-
-There is an impedance mismatch here, between cpusets and sched domains.
-Cpusets are hierarchical and nest.  Sched domains are flat; they don't
-overlap and each CPU is in at most one sched domain.
-
-It is necessary for sched domains to be flat because load balancing
-across partially overlapping sets of CPUs would risk unstable dynamics
-that would be beyond our understanding.  So if each of two partially
-overlapping cpusets enables the flag 'sched_load_balance', then we
-form a single sched domain that is a superset of both.  We won't move
-a task to a CPU outside it cpuset, but the scheduler load balancing
-code might waste some compute cycles considering that possibility.
-
-This mismatch is why there is not a simple one-to-one relation
-between which cpusets have the flag "sched_load_balance" enabled,
-and the sched domain configuration.  If a cpuset enables the flag, it
-will get balancing across all its CPUs, but if it disables the flag,
-it will only be assured of no load balancing if no other overlapping
-cpuset enables the flag.
-
-If two cpusets have partially overlapping 'cpus' allowed, and only
-one of them has this flag enabled, then the other may find its
-tasks only partially load balanced, just on the overlapping CPUs.
-This is just the general case of the top_cpuset example given a few
-paragraphs above.  In the general case, as in the top cpuset case,
-don't leave tasks that might use non-trivial amounts of CPU in
-such partially load balanced cpusets, as they may be artificially
-constrained to some subset of the CPUs allowed to them, for lack of
-load balancing to the other CPUs.
-
-1.7.1 sched_load_balance implementation details.
-------------------------------------------------
-
-The per-cpuset flag 'sched_load_balance' defaults to enabled (contrary
-to most cpuset flags.)  When enabled for a cpuset, the kernel will
-ensure that it can load balance across all the CPUs in that cpuset
-(makes sure that all the CPUs in the cpus_allowed of that cpuset are
-in the same sched domain.)
-
-If two overlapping cpusets both have 'sched_load_balance' enabled,
-then they will be (must be) both in the same sched domain.
-
-If, as is the default, the top cpuset has 'sched_load_balance' enabled,
-then by the above that means there is a single sched domain covering
-the whole system, regardless of any other cpuset settings.
-
-The kernel commits to user space that it will avoid load balancing
-where it can.  It will pick as fine a granularity partition of sched
-domains as it can while still providing load balancing for any set
-of CPUs allowed to a cpuset having 'sched_load_balance' enabled.
-
-The internal kernel cpuset to scheduler interface passes from the
-cpuset code to the scheduler code a partition of the load balanced
-CPUs in the system. This partition is a set of subsets (represented
-as an array of cpumask_t) of CPUs, pairwise disjoint, that cover all
-the CPUs that must be load balanced.
-
-Whenever the 'sched_load_balance' flag changes, or CPUs come or go
-from a cpuset with this flag enabled, or a cpuset with this flag
-enabled is removed, the cpuset code builds a new such partition and
-passes it to the scheduler sched domain setup code, to have the sched
-domains rebuilt as necessary.
-
-This partition exactly defines what sched domains the scheduler should
-setup - one sched domain for each element (cpumask_t) in the partition.
-
-The scheduler remembers the currently active sched domain partitions.
-When the scheduler routine partition_sched_domains() is invoked from
-the cpuset code to update these sched domains, it compares the new
-partition requested with the current, and updates its sched domains,
-removing the old and adding the new, for each change.
-
-
-1.8 What is sched_relax_domain_level ?
---------------------------------------
-
-In sched domain, the scheduler migrates tasks in 2 ways; periodic load
-balance on tick, and at time of some schedule events.
-
-When a task is woken up, scheduler try to move the task on idle CPU.
-For example, if a task A running on CPU X activates another task B
-on the same CPU X, and if CPU Y is X's sibling and performing idle,
-then scheduler migrate task B to CPU Y so that task B can start on
-CPU Y without waiting task A on CPU X.
-
-And if a CPU run out of tasks in its runqueue, the CPU try to pull
-extra tasks from other busy CPUs to help them before it is going to
-be idle.
-
-Of course it takes some searching cost to find movable tasks and/or
-idle CPUs, the scheduler might not search all CPUs in the domain
-everytime.  In fact, in some architectures, the searching ranges on
-events are limited in the same socket or node where the CPU locates,
-while the load balance on tick searchs all.
-
-For example, assume CPU Z is relatively far from CPU X.  Even if CPU Z
-is idle while CPU X and the siblings are busy, scheduler can't migrate
-woken task B from X to Z since it is out of its searching range.
-As the result, task B on CPU X need to wait task A or wait load balance
-on the next tick.  For some applications in special situation, waiting
-1 tick may be too long.
-
-The 'sched_relax_domain_level' file allows you to request changing
-this searching range as you like.  This file takes int value which
-indicates size of searching range in levels ideally as follows,
-otherwise initial value -1 that indicates the cpuset has no request.
-
-  -1  : no request. use system default or follow request of others.
-   0  : no search.
-   1  : search siblings (hyperthreads in a core).
-   2  : search cores in a package.
-   3  : search cpus in a node [= system wide on non-NUMA system]
- ( 4  : search nodes in a chunk of node [on NUMA system] )
- ( 5  : search system wide [on NUMA system] )
-
-The system default is architecture dependent.  The system default
-can be changed using the relax_domain_level= boot parameter.
-
-This file is per-cpuset and affect the sched domain where the cpuset
-belongs to.  Therefore if the flag 'sched_load_balance' of a cpuset
-is disabled, then 'sched_relax_domain_level' have no effect since
-there is no sched domain belonging the cpuset.
-
-If multiple cpusets are overlapping and hence they form a single sched
-domain, the largest value among those is used.  Be careful, if one
-requests 0 and others are -1 then 0 is used.
-
-Note that modifying this file will have both good and bad effects,
-and whether it is acceptable or not will be depend on your situation.
-Don't modify this file if you are not sure.
-
-If your situation is:
- - The migration costs between each cpu can be assumed considerably
-   small(for you) due to your special application's behavior or
-   special hardware support for CPU cache etc.
- - The searching cost doesn't have impact(for you) or you can make
-   the searching cost enough small by managing cpuset to compact etc.
- - The latency is required even it sacrifices cache hit rate etc.
-then increasing 'sched_relax_domain_level' would benefit you.
-
-
-1.9 How do I use cpusets ?
---------------------------
-
-In order to minimize the impact of cpusets on critical kernel
-code, such as the scheduler, and due to the fact that the kernel
-does not support one task updating the memory placement of another
-task directly, the impact on a task of changing its cpuset CPU
-or Memory Node placement, or of changing to which cpuset a task
-is attached, is subtle.
-
-If a cpuset has its Memory Nodes modified, then for each task attached
-to that cpuset, the next time that the kernel attempts to allocate
-a page of memory for that task, the kernel will notice the change
-in the tasks cpuset, and update its per-task memory placement to
-remain within the new cpusets memory placement.  If the task was using
-mempolicy MPOL_BIND, and the nodes to which it was bound overlap with
-its new cpuset, then the task will continue to use whatever subset
-of MPOL_BIND nodes are still allowed in the new cpuset.  If the task
-was using MPOL_BIND and now none of its MPOL_BIND nodes are allowed
-in the new cpuset, then the task will be essentially treated as if it
-was MPOL_BIND bound to the new cpuset (even though its numa placement,
-as queried by get_mempolicy(), doesn't change).  If a task is moved
-from one cpuset to another, then the kernel will adjust the tasks
-memory placement, as above, the next time that the kernel attempts
-to allocate a page of memory for that task.
-
-If a cpuset has its 'cpus' modified, then each task in that cpuset
-will have its allowed CPU placement changed immediately.  Similarly,
-if a tasks pid is written to a cpusets 'tasks' file, in either its
-current cpuset or another cpuset, then its allowed CPU placement is
-changed immediately.  If such a task had been bound to some subset
-of its cpuset using the sched_setaffinity() call, the task will be
-allowed to run on any CPU allowed in its new cpuset, negating the
-affect of the prior sched_setaffinity() call.
-
-In summary, the memory placement of a task whose cpuset is changed is
-updated by the kernel, on the next allocation of a page for that task,
-but the processor placement is not updated, until that tasks pid is
-rewritten to the 'tasks' file of its cpuset.  This is done to avoid
-impacting the scheduler code in the kernel with a check for changes
-in a tasks processor placement.
-
-Normally, once a page is allocated (given a physical page
-of main memory) then that page stays on whatever node it
-was allocated, so long as it remains allocated, even if the
-cpusets memory placement policy 'mems' subsequently changes.
-If the cpuset flag file 'memory_migrate' is set true, then when
-tasks are attached to that cpuset, any pages that task had
-allocated to it on nodes in its previous cpuset are migrated
-to the tasks new cpuset. The relative placement of the page within
-the cpuset is preserved during these migration operations if possible.
-For example if the page was on the second valid node of the prior cpuset
-then the page will be placed on the second valid node of the new cpuset.
-
-Also if 'memory_migrate' is set true, then if that cpusets
-'mems' file is modified, pages allocated to tasks in that
-cpuset, that were on nodes in the previous setting of 'mems',
-will be moved to nodes in the new setting of 'mems.'
-Pages that were not in the tasks prior cpuset, or in the cpusets
-prior 'mems' setting, will not be moved.
-
-There is an exception to the above.  If hotplug functionality is used
-to remove all the CPUs that are currently assigned to a cpuset,
-then all the tasks in that cpuset will be moved to the nearest ancestor
-with non-empty cpus.  But the moving of some (or all) tasks might fail if
-cpuset is bound with another cgroup subsystem which has some restrictions
-on task attaching.  In this failing case, those tasks will stay
-in the original cpuset, and the kernel will automatically update
-their cpus_allowed to allow all online CPUs.  When memory hotplug
-functionality for removing Memory Nodes is available, a similar exception
-is expected to apply there as well.  In general, the kernel prefers to
-violate cpuset placement, over starving a task that has had all
-its allowed CPUs or Memory Nodes taken offline.
-
-There is a second exception to the above.  GFP_ATOMIC requests are
-kernel internal allocations that must be satisfied, immediately.
-The kernel may drop some request, in rare cases even panic, if a
-GFP_ATOMIC alloc fails.  If the request cannot be satisfied within
-the current tasks cpuset, then we relax the cpuset, and look for
-memory anywhere we can find it.  It's better to violate the cpuset
-than stress the kernel.
-
-To start a new job that is to be contained within a cpuset, the steps are:
-
- 1) mkdir /dev/cpuset
- 2) mount -t cgroup -ocpuset cpuset /dev/cpuset
- 3) Create the new cpuset by doing mkdir's and write's (or echo's) in
-    the /dev/cpuset virtual file system.
- 4) Start a task that will be the "founding father" of the new job.
- 5) Attach that task to the new cpuset by writing its pid to the
-    /dev/cpuset tasks file for that cpuset.
- 6) fork, exec or clone the job tasks from this founding father task.
-
-For example, the following sequence of commands will setup a cpuset
-named "Charlie", containing just CPUs 2 and 3, and Memory Node 1,
-and then start a subshell 'sh' in that cpuset:
-
-  mount -t cgroup -ocpuset cpuset /dev/cpuset
-  cd /dev/cpuset
-  mkdir Charlie
-  cd Charlie
-  /bin/echo 2-3 > cpus
-  /bin/echo 1 > mems
-  /bin/echo $$ > tasks
-  sh
-  # The subshell 'sh' is now running in cpuset Charlie
-  # The next line should display '/Charlie'
-  cat /proc/self/cpuset
-
-In the future, a C library interface to cpusets will likely be
-available.  For now, the only way to query or modify cpusets is
-via the cpuset file system, using the various cd, mkdir, echo, cat,
-rmdir commands from the shell, or their equivalent from C.
-
-The sched_setaffinity calls can also be done at the shell prompt using
-SGI's runon or Robert Love's taskset.  The mbind and set_mempolicy
-calls can be done at the shell prompt using the numactl command
-(part of Andi Kleen's numa package).
-
-2. Usage Examples and Syntax
-============================
-
-2.1 Basic Usage
----------------
-
-Creating, modifying, using the cpusets can be done through the cpuset
-virtual filesystem.
-
-To mount it, type:
-# mount -t cgroup -o cpuset cpuset /dev/cpuset
-
-Then under /dev/cpuset you can find a tree that corresponds to the
-tree of the cpusets in the system. For instance, /dev/cpuset
-is the cpuset that holds the whole system.
-
-If you want to create a new cpuset under /dev/cpuset:
-# cd /dev/cpuset
-# mkdir my_cpuset
-
-Now you want to do something with this cpuset.
-# cd my_cpuset
-
-In this directory you can find several files:
-# ls
-cpu_exclusive  memory_migrate      mems                      tasks
-cpus           memory_pressure     notify_on_release
-mem_exclusive  memory_spread_page  sched_load_balance
-mem_hardwall   memory_spread_slab  sched_relax_domain_level
-
-Reading them will give you information about the state of this cpuset:
-the CPUs and Memory Nodes it can use, the processes that are using
-it, its properties.  By writing to these files you can manipulate
-the cpuset.
-
-Set some flags:
-# /bin/echo 1 > cpu_exclusive
-
-Add some cpus:
-# /bin/echo 0-7 > cpus
-
-Add some mems:
-# /bin/echo 0-7 > mems
-
-Now attach your shell to this cpuset:
-# /bin/echo $$ > tasks
-
-You can also create cpusets inside your cpuset by using mkdir in this
-directory.
-# mkdir my_sub_cs
-
-To remove a cpuset, just use rmdir:
-# rmdir my_sub_cs
-This will fail if the cpuset is in use (has cpusets inside, or has
-processes attached).
-
-Note that for legacy reasons, the "cpuset" filesystem exists as a
-wrapper around the cgroup filesystem.
-
-The command
-
-mount -t cpuset X /dev/cpuset
-
-is equivalent to
-
-mount -t cgroup -ocpuset X /dev/cpuset
-echo "/sbin/cpuset_release_agent" > /dev/cpuset/release_agent
-
-2.2 Adding/removing cpus
-------------------------
-
-This is the syntax to use when writing in the cpus or mems files
-in cpuset directories:
-
-# /bin/echo 1-4 > cpus		-> set cpus list to cpus 1,2,3,4
-# /bin/echo 1,2,3,4 > cpus	-> set cpus list to cpus 1,2,3,4
-
-2.3 Setting flags
------------------
-
-The syntax is very simple:
-
-# /bin/echo 1 > cpu_exclusive 	-> set flag 'cpu_exclusive'
-# /bin/echo 0 > cpu_exclusive 	-> unset flag 'cpu_exclusive'
-
-2.4 Attaching processes
------------------------
-
-# /bin/echo PID > tasks
-
-Note that it is PID, not PIDs. You can only attach ONE task at a time.
-If you have several tasks to attach, you have to do it one after another:
-
-# /bin/echo PID1 > tasks
-# /bin/echo PID2 > tasks
-	...
-# /bin/echo PIDn > tasks
-
-
-3. Questions
-============
-
-Q: what's up with this '/bin/echo' ?
-A: bash's builtin 'echo' command does not check calls to write() against
-   errors. If you use it in the cpuset file system, you won't be
-   able to tell whether a command succeeded or failed.
-
-Q: When I attach processes, only the first of the line gets really attached !
-A: We can only return one error code per call to write(). So you should also
-   put only ONE pid.
-
-4. Contact
-==========
-
-Web: http://www.bullopensource.org/cpuset
diff --git a/Documentation/scheduler/sched-design-CFS.txt b/Documentation/scheduler/sched-design-CFS.txt
index 8398ca4ff4ed..6f33593e59e2 100644
--- a/Documentation/scheduler/sched-design-CFS.txt
+++ b/Documentation/scheduler/sched-design-CFS.txt
@@ -231,7 +231,7 @@ CPU bandwidth control purposes:
 
    This options needs CONFIG_CGROUPS to be defined, and lets the administrator
    create arbitrary groups of tasks, using the "cgroup" pseudo filesystem.  See
-   Documentation/cgroups.txt for more information about this filesystem.
+   Documentation/cgroups/cgroups.txt for more information about this filesystem.
 
 Only one of these options to group tasks can be chosen and not both.
 
diff --git a/include/linux/res_counter.h b/include/linux/res_counter.h
index dede0a2cfc45..4c5bcf6ca7e8 100644
--- a/include/linux/res_counter.h
+++ b/include/linux/res_counter.h
@@ -9,7 +9,7 @@
  *
  * Author: Pavel Emelianov <xemul@openvz.org>
  *
- * See Documentation/controllers/resource_counter.txt for more
+ * See Documentation/cgroups/resource_counter.txt for more
  * info about what this counter is.
  */
 
diff --git a/init/Kconfig b/init/Kconfig
index 56fd93c63c77..2af83825634e 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -323,8 +323,8 @@ config CGROUP_SCHED
 	  This option allows you to create arbitrary task groups
 	  using the "cgroup" pseudo filesystem and control
 	  the cpu bandwidth allocated to each such task group.
-	  Refer to Documentation/cgroups.txt for more information
-	  on "cgroup" pseudo filesystem.
+	  Refer to Documentation/cgroups/cgroups.txt for more
+	  information on "cgroup" pseudo filesystem.
 
 endchoice
 
@@ -335,10 +335,9 @@ menuconfig CGROUPS
 	  use with process control subsystems such as Cpusets, CFS, memory
 	  controls or device isolation.
 	  See
-		- Documentation/cpusets.txt	(Cpusets)
 		- Documentation/scheduler/sched-design-CFS.txt	(CFS)
-		- Documentation/cgroups/ (features for grouping, isolation)
-		- Documentation/controllers/ (features for resource control)
+		- Documentation/cgroups/ (features for grouping, isolation
+					  and resource control)
 
 	  Say N if unsure.
 
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 647c77a88fcb..a85678865c5e 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -568,7 +568,7 @@ update_domain_attr_tree(struct sched_domain_attr *dattr, struct cpuset *c)
  * load balancing domains (sched domains) as specified by that partial
  * partition.
  *
- * See "What is sched_load_balance" in Documentation/cpusets.txt
+ * See "What is sched_load_balance" in Documentation/cgroups/cpusets.txt
  * for a background explanation of this.
  *
  * Does not return errors, on the theory that the callers of this
-- 
cgit v1.2.3


From 3eabdb76a03bbe8f556162738c264dbfb24cff6a Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Thu, 15 Jan 2009 13:51:01 -0800
Subject: jbd: fix missing kernel-doc

Fix jbd header file kernel-doc notation:

Warning(linux-2.6.28-git13//include/linux/jbd.h:823): No description found for parameter 'j_average_commit_time'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/jbd.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/jbd.h b/include/linux/jbd.h
index 6384b19efe64..64246dce5663 100644
--- a/include/linux/jbd.h
+++ b/include/linux/jbd.h
@@ -614,6 +614,8 @@ struct transaction_s
  * @j_wbufsize: maximum number of buffer_heads allowed in j_wbuf, the
  *	number that will fit in j_blocksize
  * @j_last_sync_writer: most recent pid which did a synchronous write
+ * @j_average_commit_time: the average amount of time in nanoseconds it
+ *	takes to commit a transaction to the disk.
  * @j_private: An opaque pointer to fs-private information.
  */
 
-- 
cgit v1.2.3


From 6ae301e85c9c58d2f430a8a7057ce488b7ff76df Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Thu, 15 Jan 2009 13:51:01 -0800
Subject: resources: fix parameter name and kernel-doc

Fix __request_region() parameter kernel-doc notation and parameter name:

Warning(linux-2.6.28-git10//kernel/resource.c:627): No description found for parameter 'flags'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/ioport.h | 3 ++-
 kernel/resource.c      | 1 +
 2 files changed, 3 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/ioport.h b/include/linux/ioport.h
index f6bb2ca8e3ba..32e4b2f72294 100644
--- a/include/linux/ioport.h
+++ b/include/linux/ioport.h
@@ -143,7 +143,8 @@ static inline unsigned long resource_type(struct resource *res)
 
 extern struct resource * __request_region(struct resource *,
 					resource_size_t start,
-					resource_size_t n, const char *name, int relaxed);
+					resource_size_t n,
+					const char *name, int flags);
 
 /* Compatibility cruft */
 #define release_region(start,n)	__release_region(&ioport_resource, (start), (n))
diff --git a/kernel/resource.c b/kernel/resource.c
index ca6a1536b205..fd5d7d574bb9 100644
--- a/kernel/resource.c
+++ b/kernel/resource.c
@@ -620,6 +620,7 @@ resource_size_t resource_alignment(struct resource *res)
  * @start: resource start address
  * @n: resource region size
  * @name: reserving caller's ID string
+ * @flags: IO resource flags
  */
 struct resource * __request_region(struct resource *parent,
 				   resource_size_t start, resource_size_t n,
-- 
cgit v1.2.3


From 1bcbf31337391a2f54ef6c1e8871c2de5944a7dc Mon Sep 17 00:00:00 2001
From: Qinghuang Feng <qhfeng.kernel@gmail.com>
Date: Thu, 15 Jan 2009 13:51:03 -0800
Subject: btrfs & squashfs: Move btrfs and squashfsto's magic number to
 <linux/magic.h>

Use the standard magic.h for btrfs and squashfs.

Signed-off-by: Qinghuang Feng <qhfeng.kernel@gmail.com>
Cc: Phillip Lougher <phillip@lougher.demon.co.uk>
Cc: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/btrfs/super.c          | 2 +-
 fs/squashfs/squashfs_fs.h | 1 -
 fs/squashfs/super.c       | 1 +
 include/linux/magic.h     | 2 ++
 4 files changed, 4 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 0a14b495532f..7256cf242eb0 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -38,6 +38,7 @@
 #include <linux/namei.h>
 #include <linux/miscdevice.h>
 #include <linux/version.h>
+#include <linux/magic.h>
 #include "compat.h"
 #include "ctree.h"
 #include "disk-io.h"
@@ -51,7 +52,6 @@
 #include "export.h"
 #include "compression.h"
 
-#define BTRFS_SUPER_MAGIC 0x9123683E
 
 static struct super_operations btrfs_super_ops;
 
diff --git a/fs/squashfs/squashfs_fs.h b/fs/squashfs/squashfs_fs.h
index 6840da1bf21e..283daafc568e 100644
--- a/fs/squashfs/squashfs_fs.h
+++ b/fs/squashfs/squashfs_fs.h
@@ -26,7 +26,6 @@
 #define SQUASHFS_CACHED_FRAGMENTS	CONFIG_SQUASHFS_FRAGMENT_CACHE_SIZE
 #define SQUASHFS_MAJOR			4
 #define SQUASHFS_MINOR			0
-#define SQUASHFS_MAGIC			0x73717368
 #define SQUASHFS_START			0
 
 /* size of metadata (inode and directory) blocks */
diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c
index a0466d7467b2..071df5b5b491 100644
--- a/fs/squashfs/super.c
+++ b/fs/squashfs/super.c
@@ -35,6 +35,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/zlib.h>
+#include <linux/magic.h>
 
 #include "squashfs_fs.h"
 #include "squashfs_fs_sb.h"
diff --git a/include/linux/magic.h b/include/linux/magic.h
index 439f6f3cb0c4..0b4df7eba852 100644
--- a/include/linux/magic.h
+++ b/include/linux/magic.h
@@ -10,11 +10,13 @@
 #define SYSFS_MAGIC		0x62656572
 #define SECURITYFS_MAGIC	0x73636673
 #define TMPFS_MAGIC		0x01021994
+#define SQUASHFS_MAGIC		0x73717368
 #define EFS_SUPER_MAGIC		0x414A53
 #define EXT2_SUPER_MAGIC	0xEF53
 #define EXT3_SUPER_MAGIC	0xEF53
 #define XENFS_SUPER_MAGIC	0xabba1974
 #define EXT4_SUPER_MAGIC	0xEF53
+#define BTRFS_SUPER_MAGIC	0x9123683E
 #define HPFS_SUPER_MAGIC	0xf995e849
 #define ISOFS_SUPER_MAGIC	0x9660
 #define JFFS2_SUPER_MAGIC	0x72b6
-- 
cgit v1.2.3


From 00bfddaf7f68a6551319b536f052040c370756b0 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@infradead.org>
Date: Thu, 15 Jan 2009 13:51:26 -0800
Subject: include of <linux/types.h> is preferred over <asm/types.h>

Impact: fix 15 make headers_check warnings:

include of <linux/types.h> is preferred over <asm/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/agpgart.h      | 1 -
 include/linux/atm_idt77105.h | 2 +-
 include/linux/capi.h         | 2 +-
 include/linux/connector.h    | 2 +-
 include/linux/cyclades.h     | 2 --
 include/linux/fb.h           | 2 +-
 include/linux/if_pppol2tp.h  | 2 +-
 include/linux/if_pppox.h     | 2 +-
 include/linux/input.h        | 2 +-
 include/linux/joystick.h     | 2 +-
 include/linux/kvm.h          | 2 +-
 include/linux/loop.h         | 2 +-
 include/linux/matroxfb.h     | 2 +-
 include/linux/phantom.h      | 2 +-
 include/linux/radeonfb.h     | 2 +-
 15 files changed, 13 insertions(+), 16 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/agpgart.h b/include/linux/agpgart.h
index c8fdb6e658e1..110c600c885f 100644
--- a/include/linux/agpgart.h
+++ b/include/linux/agpgart.h
@@ -52,7 +52,6 @@
 
 #ifndef __KERNEL__
 #include <linux/types.h>
-#include <asm/types.h>
 
 struct agp_version {
 	__u16 major;
diff --git a/include/linux/atm_idt77105.h b/include/linux/atm_idt77105.h
index 05621cf20709..8b724000aa50 100644
--- a/include/linux/atm_idt77105.h
+++ b/include/linux/atm_idt77105.h
@@ -7,7 +7,7 @@
 #ifndef LINUX_ATM_IDT77105_H
 #define LINUX_ATM_IDT77105_H
 
-#include <asm/types.h>
+#include <linux/types.h>
 #include <linux/atmioc.h>
 #include <linux/atmdev.h>
 
diff --git a/include/linux/capi.h b/include/linux/capi.h
index fdebaaa9f66e..65100d6cb89b 100644
--- a/include/linux/capi.h
+++ b/include/linux/capi.h
@@ -12,7 +12,7 @@
 #ifndef __LINUX_CAPI_H__
 #define __LINUX_CAPI_H__
 
-#include <asm/types.h>
+#include <linux/types.h>
 #include <linux/ioctl.h>
 #ifndef __KERNEL__
 #include <linux/kernelcapi.h>
diff --git a/include/linux/connector.h b/include/linux/connector.h
index 5c7f9468f753..34f2789d9b9b 100644
--- a/include/linux/connector.h
+++ b/include/linux/connector.h
@@ -22,7 +22,7 @@
 #ifndef __CONNECTOR_H
 #define __CONNECTOR_H
 
-#include <asm/types.h>
+#include <linux/types.h>
 
 #define CN_IDX_CONNECTOR		0xffffffff
 #define CN_VAL_CONNECTOR		0xffffffff
diff --git a/include/linux/cyclades.h b/include/linux/cyclades.h
index 2d3d1e04ba92..d06fbf286346 100644
--- a/include/linux/cyclades.h
+++ b/include/linux/cyclades.h
@@ -150,8 +150,6 @@ struct CYZ_BOOT_CTRL {
  *	architectures and compilers.
  */
 
-#include <asm/types.h>
-
 typedef __u64  ucdouble;		/* 64 bits, unsigned */
 typedef __u32  uclong;			/* 32 bits, unsigned */
 typedef __u16  ucshort;		/* 16 bits, unsigned */
diff --git a/include/linux/fb.h b/include/linux/fb.h
index 1ee63df5be92..818fe21257e8 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -1,7 +1,7 @@
 #ifndef _LINUX_FB_H
 #define _LINUX_FB_H
 
-#include <asm/types.h>
+#include <linux/types.h>
 #include <linux/i2c.h>
 
 struct dentry;
diff --git a/include/linux/if_pppol2tp.h b/include/linux/if_pppol2tp.h
index a7d6a2234b31..c7a66882b6d0 100644
--- a/include/linux/if_pppol2tp.h
+++ b/include/linux/if_pppol2tp.h
@@ -15,7 +15,7 @@
 #ifndef __LINUX_IF_PPPOL2TP_H
 #define __LINUX_IF_PPPOL2TP_H
 
-#include <asm/types.h>
+#include <linux/types.h>
 
 #ifdef __KERNEL__
 #include <linux/in.h>
diff --git a/include/linux/if_pppox.h b/include/linux/if_pppox.h
index 6fb7f1788570..30c88b2245ff 100644
--- a/include/linux/if_pppox.h
+++ b/include/linux/if_pppox.h
@@ -17,7 +17,7 @@
 #define __LINUX_IF_PPPOX_H
 
 
-#include <asm/types.h>
+#include <linux/types.h>
 #include <asm/byteorder.h>
 
 #ifdef  __KERNEL__
diff --git a/include/linux/input.h b/include/linux/input.h
index 9a6355f74db2..1249a0c20a38 100644
--- a/include/linux/input.h
+++ b/include/linux/input.h
@@ -16,7 +16,7 @@
 #include <sys/time.h>
 #include <sys/ioctl.h>
 #include <sys/types.h>
-#include <asm/types.h>
+#include <linux/types.h>
 #endif
 
 /*
diff --git a/include/linux/joystick.h b/include/linux/joystick.h
index b5e051295a67..9e20c29c1e14 100644
--- a/include/linux/joystick.h
+++ b/include/linux/joystick.h
@@ -27,7 +27,7 @@
  * Vojtech Pavlik, Ucitelska 1576, Prague 8, 182 00 Czech Republic
  */
 
-#include <asm/types.h>
+#include <linux/types.h>
 #include <linux/input.h>
 
 /*
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 35525ac63337..5715f1907601 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -7,7 +7,7 @@
  * Note: you must update KVM_API_VERSION if you change this interface.
  */
 
-#include <asm/types.h>
+#include <linux/types.h>
 #include <linux/compiler.h>
 #include <linux/ioctl.h>
 #include <asm/kvm.h>
diff --git a/include/linux/loop.h b/include/linux/loop.h
index 46169a7b559b..6ffd6db5bb0d 100644
--- a/include/linux/loop.h
+++ b/include/linux/loop.h
@@ -80,7 +80,7 @@ enum {
 };
 
 #include <asm/posix_types.h>	/* for __kernel_old_dev_t */
-#include <asm/types.h>		/* for __u64 */
+#include <linux/types.h>	/* for __u64 */
 
 /* Backwards compatibility version */
 struct loop_info {
diff --git a/include/linux/matroxfb.h b/include/linux/matroxfb.h
index ae5b09493062..404f678e734b 100644
--- a/include/linux/matroxfb.h
+++ b/include/linux/matroxfb.h
@@ -2,7 +2,7 @@
 #define __LINUX_MATROXFB_H__
 
 #include <asm/ioctl.h>
-#include <asm/types.h>
+#include <linux/types.h>
 #include <linux/videodev2.h>
 
 struct matroxioc_output_mode {
diff --git a/include/linux/phantom.h b/include/linux/phantom.h
index 02268c54c250..94dd6645c60a 100644
--- a/include/linux/phantom.h
+++ b/include/linux/phantom.h
@@ -10,7 +10,7 @@
 #ifndef __PHANTOM_H
 #define __PHANTOM_H
 
-#include <asm/types.h>
+#include <linux/types.h>
 
 /* PHN_(G/S)ET_REG param */
 struct phm_reg {
diff --git a/include/linux/radeonfb.h b/include/linux/radeonfb.h
index 5bd8975ed78e..8c4bbdecc44f 100644
--- a/include/linux/radeonfb.h
+++ b/include/linux/radeonfb.h
@@ -2,7 +2,7 @@
 #define __LINUX_RADEONFB_H__
 
 #include <asm/ioctl.h>
-#include <asm/types.h>
+#include <linux/types.h>
 
 #define ATY_RADEON_LCD_ON	0x00000001
 #define ATY_RADEON_CRT_ON	0x00000002
-- 
cgit v1.2.3


From 94be9a58d7e683ac3c1df1858a17f09ebade8da0 Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Fri, 16 Jan 2009 10:17:09 -0500
Subject: [libata] get-identity ioctl: Fix use of invalid memory pointer for
 SAS drivers.

Caught by Ke Wei (and team?) at Marvell.

Also, move the ata_scsi_ioctl export to libata-scsi.c, as that seems to be the
general trend.

Acked-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c           |  1 -
 drivers/ata/libata-scsi.c           | 17 +++++++++++++----
 drivers/scsi/ipr.c                  |  2 +-
 drivers/scsi/libsas/sas_scsi_host.c |  2 +-
 include/linux/libata.h              |  2 ++
 5 files changed, 17 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 71218d76d75e..552ecae13434 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -6638,7 +6638,6 @@ EXPORT_SYMBOL_GPL(ata_dev_pair);
 EXPORT_SYMBOL_GPL(ata_port_disable);
 EXPORT_SYMBOL_GPL(ata_ratelimit);
 EXPORT_SYMBOL_GPL(ata_wait_register);
-EXPORT_SYMBOL_GPL(ata_scsi_ioctl);
 EXPORT_SYMBOL_GPL(ata_scsi_queuecmd);
 EXPORT_SYMBOL_GPL(ata_scsi_slave_config);
 EXPORT_SYMBOL_GPL(ata_scsi_slave_destroy);
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 9e92107691f2..a1a6e6298c33 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -423,9 +423,9 @@ int ata_std_bios_param(struct scsi_device *sdev, struct block_device *bdev,
  *	RETURNS:
  *	Zero on success, negative errno on error.
  */
-static int ata_get_identity(struct scsi_device *sdev, void __user *arg)
+static int ata_get_identity(struct ata_port *ap, struct scsi_device *sdev,
+			    void __user *arg)
 {
-	struct ata_port *ap = ata_shost_to_port(sdev->host);
 	struct ata_device *dev = ata_scsi_find_dev(ap, sdev);
 	u16 __user *dst = arg;
 	char buf[40];
@@ -645,7 +645,8 @@ int ata_task_ioctl(struct scsi_device *scsidev, void __user *arg)
 	return rc;
 }
 
-int ata_scsi_ioctl(struct scsi_device *scsidev, int cmd, void __user *arg)
+int ata_sas_scsi_ioctl(struct ata_port *ap, struct scsi_device *scsidev,
+		     int cmd, void __user *arg)
 {
 	int val = -EINVAL, rc = -EINVAL;
 
@@ -663,7 +664,7 @@ int ata_scsi_ioctl(struct scsi_device *scsidev, int cmd, void __user *arg)
 		return 0;
 
 	case HDIO_GET_IDENTITY:
-		return ata_get_identity(scsidev, arg);
+		return ata_get_identity(ap, scsidev, arg);
 
 	case HDIO_DRIVE_CMD:
 		if (!capable(CAP_SYS_ADMIN) || !capable(CAP_SYS_RAWIO))
@@ -682,6 +683,14 @@ int ata_scsi_ioctl(struct scsi_device *scsidev, int cmd, void __user *arg)
 
 	return rc;
 }
+EXPORT_SYMBOL_GPL(ata_sas_scsi_ioctl);
+
+int ata_scsi_ioctl(struct scsi_device *scsidev, int cmd, void __user *arg)
+{
+	return ata_sas_scsi_ioctl(ata_shost_to_port(scsidev->host),
+				scsidev, cmd, arg);
+}
+EXPORT_SYMBOL_GPL(ata_scsi_ioctl);
 
 /**
  *	ata_scsi_qc_new - acquire new ata_queued_cmd reference
diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c
index 841f460edbc4..07829009a8be 100644
--- a/drivers/scsi/ipr.c
+++ b/drivers/scsi/ipr.c
@@ -4912,7 +4912,7 @@ static int ipr_ioctl(struct scsi_device *sdev, int cmd, void __user *arg)
 	if (res && ipr_is_gata(res)) {
 		if (cmd == HDIO_GET_IDENTITY)
 			return -ENOTTY;
-		return ata_scsi_ioctl(sdev, cmd, arg);
+		return ata_sas_scsi_ioctl(res->sata_port->ap, sdev, cmd, arg);
 	}
 
 	return -EINVAL;
diff --git a/drivers/scsi/libsas/sas_scsi_host.c b/drivers/scsi/libsas/sas_scsi_host.c
index 744838780ada..1c558d3bce18 100644
--- a/drivers/scsi/libsas/sas_scsi_host.c
+++ b/drivers/scsi/libsas/sas_scsi_host.c
@@ -717,7 +717,7 @@ int sas_ioctl(struct scsi_device *sdev, int cmd, void __user *arg)
 	struct domain_device *dev = sdev_to_domain_dev(sdev);
 
 	if (dev_is_sata(dev))
-		return ata_scsi_ioctl(sdev, cmd, arg);
+		return ata_sas_scsi_ioctl(dev->sata_dev.ap, sdev, cmd, arg);
 
 	return -EINVAL;
 }
diff --git a/include/linux/libata.h b/include/linux/libata.h
index b6b8a7f3ec66..73b69c7071c5 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -927,6 +927,8 @@ extern void ata_host_init(struct ata_host *, struct device *,
 extern int ata_scsi_detect(struct scsi_host_template *sht);
 extern int ata_scsi_ioctl(struct scsi_device *dev, int cmd, void __user *arg);
 extern int ata_scsi_queuecmd(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *));
+extern int ata_sas_scsi_ioctl(struct ata_port *ap, struct scsi_device *dev,
+			    int cmd, void __user *arg);
 extern void ata_sas_port_destroy(struct ata_port *);
 extern struct ata_port *ata_sas_port_alloc(struct ata_host *,
 					   struct ata_port_info *, struct Scsi_Host *);
-- 
cgit v1.2.3


From 3ada9c126499dd4700dcdbd5b9fe8110ad17f578 Mon Sep 17 00:00:00 2001
From: David Daney <ddaney@caviumnetworks.com>
Date: Thu, 15 Jan 2009 17:45:31 -0800
Subject: libata: Add another column to the ata_timing table.

The forthcoming OCTEON SOC Compact Flash driver needs an additional
timing value that was not available in the ata_timing table.  I add a
new column for dmack_hold time.  The values were obtained from the
Compact Flash specification Rev 4.1.

Signed-off-by: David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c | 72 ++++++++++++++++++++++++-----------------------
 include/linux/libata.h    |  9 ++++--
 2 files changed, 43 insertions(+), 38 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 552ecae13434..88c242856dae 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -3029,33 +3029,33 @@ int sata_set_spd(struct ata_link *link)
  */
 
 static const struct ata_timing ata_timing[] = {
-/*	{ XFER_PIO_SLOW, 120, 290, 240, 960, 290, 240, 960,   0 }, */
-	{ XFER_PIO_0,     70, 290, 240, 600, 165, 150, 600,   0 },
-	{ XFER_PIO_1,     50, 290,  93, 383, 125, 100, 383,   0 },
-	{ XFER_PIO_2,     30, 290,  40, 330, 100,  90, 240,   0 },
-	{ XFER_PIO_3,     30,  80,  70, 180,  80,  70, 180,   0 },
-	{ XFER_PIO_4,     25,  70,  25, 120,  70,  25, 120,   0 },
-	{ XFER_PIO_5,     15,  65,  25, 100,  65,  25, 100,   0 },
-	{ XFER_PIO_6,     10,  55,  20,  80,  55,  20,  80,   0 },
-
-	{ XFER_SW_DMA_0, 120,   0,   0,   0, 480, 480, 960,   0 },
-	{ XFER_SW_DMA_1,  90,   0,   0,   0, 240, 240, 480,   0 },
-	{ XFER_SW_DMA_2,  60,   0,   0,   0, 120, 120, 240,   0 },
-
-	{ XFER_MW_DMA_0,  60,   0,   0,   0, 215, 215, 480,   0 },
-	{ XFER_MW_DMA_1,  45,   0,   0,   0,  80,  50, 150,   0 },
-	{ XFER_MW_DMA_2,  25,   0,   0,   0,  70,  25, 120,   0 },
-	{ XFER_MW_DMA_3,  25,   0,   0,   0,  65,  25, 100,   0 },
-	{ XFER_MW_DMA_4,  25,   0,   0,   0,  55,  20,  80,   0 },
-
-/*	{ XFER_UDMA_SLOW,  0,   0,   0,   0,   0,   0,   0, 150 }, */
-	{ XFER_UDMA_0,     0,   0,   0,   0,   0,   0,   0, 120 },
-	{ XFER_UDMA_1,     0,   0,   0,   0,   0,   0,   0,  80 },
-	{ XFER_UDMA_2,     0,   0,   0,   0,   0,   0,   0,  60 },
-	{ XFER_UDMA_3,     0,   0,   0,   0,   0,   0,   0,  45 },
-	{ XFER_UDMA_4,     0,   0,   0,   0,   0,   0,   0,  30 },
-	{ XFER_UDMA_5,     0,   0,   0,   0,   0,   0,   0,  20 },
-	{ XFER_UDMA_6,     0,   0,   0,   0,   0,   0,   0,  15 },
+/*	{ XFER_PIO_SLOW, 120, 290, 240, 960, 290, 240, 0,  960,   0 }, */
+	{ XFER_PIO_0,     70, 290, 240, 600, 165, 150, 0,  600,   0 },
+	{ XFER_PIO_1,     50, 290,  93, 383, 125, 100, 0,  383,   0 },
+	{ XFER_PIO_2,     30, 290,  40, 330, 100,  90, 0,  240,   0 },
+	{ XFER_PIO_3,     30,  80,  70, 180,  80,  70, 0,  180,   0 },
+	{ XFER_PIO_4,     25,  70,  25, 120,  70,  25, 0,  120,   0 },
+	{ XFER_PIO_5,     15,  65,  25, 100,  65,  25, 0,  100,   0 },
+	{ XFER_PIO_6,     10,  55,  20,  80,  55,  20, 0,   80,   0 },
+
+	{ XFER_SW_DMA_0, 120,   0,   0,   0, 480, 480, 50, 960,   0 },
+	{ XFER_SW_DMA_1,  90,   0,   0,   0, 240, 240, 30, 480,   0 },
+	{ XFER_SW_DMA_2,  60,   0,   0,   0, 120, 120, 20, 240,   0 },
+
+	{ XFER_MW_DMA_0,  60,   0,   0,   0, 215, 215, 20, 480,   0 },
+	{ XFER_MW_DMA_1,  45,   0,   0,   0,  80,  50, 5,  150,   0 },
+	{ XFER_MW_DMA_2,  25,   0,   0,   0,  70,  25, 5,  120,   0 },
+	{ XFER_MW_DMA_3,  25,   0,   0,   0,  65,  25, 5,  100,   0 },
+	{ XFER_MW_DMA_4,  25,   0,   0,   0,  55,  20, 5,   80,   0 },
+
+/*	{ XFER_UDMA_SLOW,  0,   0,   0,   0,   0,   0, 0,    0, 150 }, */
+	{ XFER_UDMA_0,     0,   0,   0,   0,   0,   0, 0,    0, 120 },
+	{ XFER_UDMA_1,     0,   0,   0,   0,   0,   0, 0,    0,  80 },
+	{ XFER_UDMA_2,     0,   0,   0,   0,   0,   0, 0,    0,  60 },
+	{ XFER_UDMA_3,     0,   0,   0,   0,   0,   0, 0,    0,  45 },
+	{ XFER_UDMA_4,     0,   0,   0,   0,   0,   0, 0,    0,  30 },
+	{ XFER_UDMA_5,     0,   0,   0,   0,   0,   0, 0,    0,  20 },
+	{ XFER_UDMA_6,     0,   0,   0,   0,   0,   0, 0,    0,  15 },
 
 	{ 0xFF }
 };
@@ -3065,14 +3065,15 @@ static const struct ata_timing ata_timing[] = {
 
 static void ata_timing_quantize(const struct ata_timing *t, struct ata_timing *q, int T, int UT)
 {
-	q->setup   = EZ(t->setup   * 1000,  T);
-	q->act8b   = EZ(t->act8b   * 1000,  T);
-	q->rec8b   = EZ(t->rec8b   * 1000,  T);
-	q->cyc8b   = EZ(t->cyc8b   * 1000,  T);
-	q->active  = EZ(t->active  * 1000,  T);
-	q->recover = EZ(t->recover * 1000,  T);
-	q->cycle   = EZ(t->cycle   * 1000,  T);
-	q->udma    = EZ(t->udma    * 1000, UT);
+	q->setup	= EZ(t->setup      * 1000,  T);
+	q->act8b	= EZ(t->act8b      * 1000,  T);
+	q->rec8b	= EZ(t->rec8b      * 1000,  T);
+	q->cyc8b	= EZ(t->cyc8b      * 1000,  T);
+	q->active	= EZ(t->active     * 1000,  T);
+	q->recover	= EZ(t->recover    * 1000,  T);
+	q->dmack_hold	= EZ(t->dmack_hold * 1000,  T);
+	q->cycle	= EZ(t->cycle      * 1000,  T);
+	q->udma		= EZ(t->udma       * 1000, UT);
 }
 
 void ata_timing_merge(const struct ata_timing *a, const struct ata_timing *b,
@@ -3084,6 +3085,7 @@ void ata_timing_merge(const struct ata_timing *a, const struct ata_timing *b,
 	if (what & ATA_TIMING_CYC8B  ) m->cyc8b   = max(a->cyc8b,   b->cyc8b);
 	if (what & ATA_TIMING_ACTIVE ) m->active  = max(a->active,  b->active);
 	if (what & ATA_TIMING_RECOVER) m->recover = max(a->recover, b->recover);
+	if (what & ATA_TIMING_DMACK_HOLD) m->dmack_hold = max(a->dmack_hold, b->dmack_hold);
 	if (what & ATA_TIMING_CYCLE  ) m->cycle   = max(a->cycle,   b->cycle);
 	if (what & ATA_TIMING_UDMA   ) m->udma    = max(a->udma,    b->udma);
 }
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 73b69c7071c5..2c6bd66209ff 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -401,12 +401,14 @@ enum {
 				  ATA_TIMING_CYC8B,
 	ATA_TIMING_ACTIVE	= (1 << 4),
 	ATA_TIMING_RECOVER	= (1 << 5),
-	ATA_TIMING_CYCLE	= (1 << 6),
-	ATA_TIMING_UDMA		= (1 << 7),
+	ATA_TIMING_DMACK_HOLD	= (1 << 6),
+	ATA_TIMING_CYCLE	= (1 << 7),
+	ATA_TIMING_UDMA		= (1 << 8),
 	ATA_TIMING_ALL		= ATA_TIMING_SETUP | ATA_TIMING_ACT8B |
 				  ATA_TIMING_REC8B | ATA_TIMING_CYC8B |
 				  ATA_TIMING_ACTIVE | ATA_TIMING_RECOVER |
-				  ATA_TIMING_CYCLE | ATA_TIMING_UDMA,
+				  ATA_TIMING_DMACK_HOLD | ATA_TIMING_CYCLE |
+				  ATA_TIMING_UDMA,
 };
 
 enum ata_xfer_mask {
@@ -866,6 +868,7 @@ struct ata_timing {
 	unsigned short cyc8b;		/* t0 for 8-bit I/O */
 	unsigned short active;		/* t2 or tD */
 	unsigned short recover;		/* t2i or tK */
+	unsigned short dmack_hold;	/* tj */
 	unsigned short cycle;		/* t0 */
 	unsigned short udma;		/* t2CYCTYP/2 */
 };
-- 
cgit v1.2.3


From 08ec8c3878cea0bf91f2ba3c0badf44b383752d0 Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Fri, 16 Jan 2009 11:57:00 -0500
Subject: jbd2: On a __journal_expect() assertion failure printk "JBD2", not
 "EXT3-fs"

Otherwise it can be very confusing to find a "EXT3-fs: " failure in
the middle of EXT4-fs failures, and it makes it harder to track the
source of the failure.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 include/linux/jbd2.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index b45109c61fba..b28b37eb11c6 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -308,7 +308,8 @@ void buffer_assertion_failure(struct buffer_head *bh);
 		int val = (expr);					     \
 		if (!val) {						     \
 			printk(KERN_ERR					     \
-				"EXT3-fs unexpected failure: %s;\n",# expr); \
+			       "JBD2 unexpected failure: %s: %s;\n",	     \
+			       __func__, #expr);			     \
 			printk(KERN_ERR why "\n");			     \
 		}							     \
 		val;							     \
-- 
cgit v1.2.3


From cc33412fb1f11613e20f9dfc2919a77ecd63fbc4 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Mon, 12 Jan 2009 17:23:05 +0100
Subject: quota: Improve locking

We implement dqget() and dqput() that need neither dqonoff_mutex nor dqptr_sem.
Then move dqget() and dqput() calls so that they are not called from under
dqptr_sem. This is important because filesystem callbacks aren't called from
under dqptr_sem which used to cause *lots* of problems with lock ranking
(and with OCFS2 they became close to unsolvable).

The patch also removes two functions which were introduced solely because OCFS2
needed them to cope with the old locking scheme. As time showed, they were not
enough for OCFS2 anyway and it would be unnecessary work to adapt them to the
new locking scheme in which they aren't needed.  As a result OCFS2 needs the
following patch to compile properly with quotas.  Sorry to any bisecters which
hit this in advance.

Signed-off-by: Jan Kara <jack@suse.cz>
---
 fs/dquot.c               | 218 ++++++++++++++++++++++++++---------------------
 include/linux/quotaops.h |   2 -
 2 files changed, 122 insertions(+), 98 deletions(-)

(limited to 'include/linux')

diff --git a/fs/dquot.c b/fs/dquot.c
index 48c0571f831d..bca3cac4bee7 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -87,14 +87,17 @@
 #define __DQUOT_PARANOIA
 
 /*
- * There are two quota SMP locks. dq_list_lock protects all lists with quotas
- * and quota formats and also dqstats structure containing statistics about the
- * lists. dq_data_lock protects data from dq_dqb and also mem_dqinfo structures
- * and also guards consistency of dquot->dq_dqb with inode->i_blocks, i_bytes.
+ * There are three quota SMP locks. dq_list_lock protects all lists with quotas
+ * and quota formats, dqstats structure containing statistics about the lists
+ * dq_data_lock protects data from dq_dqb and also mem_dqinfo structures and
+ * also guards consistency of dquot->dq_dqb with inode->i_blocks, i_bytes.
  * i_blocks and i_bytes updates itself are guarded by i_lock acquired directly
- * in inode_add_bytes() and inode_sub_bytes().
+ * in inode_add_bytes() and inode_sub_bytes(). dq_state_lock protects
+ * modifications of quota state (on quotaon and quotaoff) and readers who care
+ * about latest values take it as well.
  *
- * The spinlock ordering is hence: dq_data_lock > dq_list_lock > i_lock
+ * The spinlock ordering is hence: dq_data_lock > dq_list_lock > i_lock,
+ *   dq_list_lock > dq_state_lock
  *
  * Note that some things (eg. sb pointer, type, id) doesn't change during
  * the life of the dquot structure and so needn't to be protected by a lock
@@ -103,12 +106,7 @@
  * operation is just reading pointers from inode (or not using them at all) the
  * read lock is enough. If pointers are altered function must hold write lock
  * (these locking rules also apply for S_NOQUOTA flag in the inode - note that
- * for altering the flag i_mutex is also needed).  If operation is holding
- * reference to dquot in other way (e.g. quotactl ops) it must be guarded by
- * dqonoff_mutex.
- * This locking assures that:
- *   a) update/access to dquot pointers in inode is serialized
- *   b) everyone is guarded against invalidate_dquots()
+ * for altering the flag i_mutex is also needed).
  *
  * Each dquot has its dq_lock mutex. Locked dquots might not be referenced
  * from inodes (dquot_alloc_space() and such don't check the dq_lock).
@@ -122,10 +120,17 @@
  * Lock ordering (including related VFS locks) is the following:
  *   i_mutex > dqonoff_sem > journal_lock > dqptr_sem > dquot->dq_lock >
  *   dqio_mutex
+ * The lock ordering of dqptr_sem imposed by quota code is only dqonoff_sem >
+ * dqptr_sem. But filesystem has to count with the fact that functions such as
+ * dquot_alloc_space() acquire dqptr_sem and they usually have to be called
+ * from inside a transaction to keep filesystem consistency after a crash. Also
+ * filesystems usually want to do some IO on dquot from ->mark_dirty which is
+ * called with dqptr_sem held.
  * i_mutex on quota files is special (it's below dqio_mutex)
  */
 
 static DEFINE_SPINLOCK(dq_list_lock);
+static DEFINE_SPINLOCK(dq_state_lock);
 DEFINE_SPINLOCK(dq_data_lock);
 
 static char *quotatypes[] = INITQFNAMES;
@@ -428,7 +433,7 @@ static inline void do_destroy_dquot(struct dquot *dquot)
  * quota is disabled and pointers from inodes removed so there cannot be new
  * quota users. There can still be some users of quotas due to inodes being
  * just deleted or pruned by prune_icache() (those are not attached to any
- * list). We have to wait for such users.
+ * list) or parallel quotactl call. We have to wait for such users.
  */
 static void invalidate_dquots(struct super_block *sb, int type)
 {
@@ -600,7 +605,6 @@ static struct shrinker dqcache_shrinker = {
 /*
  * Put reference to dquot
  * NOTE: If you change this function please check whether dqput_blocks() works right...
- * MUST be called with either dqptr_sem or dqonoff_mutex held
  */
 void dqput(struct dquot *dquot)
 {
@@ -696,37 +700,31 @@ static struct dquot *get_empty_dquot(struct super_block *sb, int type)
 	return dquot;
 }
 
-/*
- * Check whether dquot is in memory.
- * MUST be called with either dqptr_sem or dqonoff_mutex held
- */
-int dquot_is_cached(struct super_block *sb, unsigned int id, int type)
-{
-	unsigned int hashent = hashfn(sb, id, type);
-	int ret = 0;
-
-        if (!sb_has_quota_active(sb, type))
-		return 0;
-	spin_lock(&dq_list_lock);
-	if (find_dquot(hashent, sb, id, type) != NODQUOT)
-		ret = 1;
-	spin_unlock(&dq_list_lock);
-	return ret;
-}
-
 /*
  * Get reference to dquot
- * MUST be called with either dqptr_sem or dqonoff_mutex held
+ *
+ * Locking is slightly tricky here. We are guarded from parallel quotaoff()
+ * destroying our dquot by:
+ *   a) checking for quota flags under dq_list_lock and
+ *   b) getting a reference to dquot before we release dq_list_lock
  */
 struct dquot *dqget(struct super_block *sb, unsigned int id, int type)
 {
 	unsigned int hashent = hashfn(sb, id, type);
-	struct dquot *dquot, *empty = NODQUOT;
+	struct dquot *dquot = NODQUOT, *empty = NODQUOT;
 
         if (!sb_has_quota_active(sb, type))
 		return NODQUOT;
 we_slept:
 	spin_lock(&dq_list_lock);
+	spin_lock(&dq_state_lock);
+	if (!sb_has_quota_active(sb, type)) {
+		spin_unlock(&dq_state_lock);
+		spin_unlock(&dq_list_lock);
+		goto out;
+	}
+	spin_unlock(&dq_state_lock);
+
 	if ((dquot = find_dquot(hashent, sb, id, type)) == NODQUOT) {
 		if (empty == NODQUOT) {
 			spin_unlock(&dq_list_lock);
@@ -735,6 +733,7 @@ we_slept:
 			goto we_slept;
 		}
 		dquot = empty;
+		empty = NODQUOT;
 		dquot->dq_id = id;
 		/* all dquots go on the inuse_list */
 		put_inuse(dquot);
@@ -749,8 +748,6 @@ we_slept:
 		dqstats.cache_hits++;
 		dqstats.lookups++;
 		spin_unlock(&dq_list_lock);
-		if (empty)
-			do_destroy_dquot(empty);
 	}
 	/* Wait for dq_lock - after this we know that either dquot_release() is already
 	 * finished or it will be canceled due to dq_count > 1 test */
@@ -758,11 +755,15 @@ we_slept:
 	/* Read the dquot and instantiate it (everything done only if needed) */
 	if (!test_bit(DQ_ACTIVE_B, &dquot->dq_flags) && sb->dq_op->acquire_dquot(dquot) < 0) {
 		dqput(dquot);
-		return NODQUOT;
+		dquot = NODQUOT;
+		goto out;
 	}
 #ifdef __DQUOT_PARANOIA
 	BUG_ON(!dquot->dq_sb);	/* Has somebody invalidated entry under us? */
 #endif
+out:
+	if (empty)
+		do_destroy_dquot(empty);
 
 	return dquot;
 }
@@ -1198,63 +1199,76 @@ static int info_bdq_free(struct dquot *dquot, qsize_t space)
 }
 /*
  *	Initialize quota pointers in inode
- *	Transaction must be started at entry
+ *	We do things in a bit complicated way but by that we avoid calling
+ *	dqget() and thus filesystem callbacks under dqptr_sem.
  */
 int dquot_initialize(struct inode *inode, int type)
 {
 	unsigned int id = 0;
 	int cnt, ret = 0;
+	struct dquot *got[MAXQUOTAS] = { NODQUOT, NODQUOT };
+	struct super_block *sb = inode->i_sb;
 
 	/* First test before acquiring mutex - solves deadlocks when we
          * re-enter the quota code and are already holding the mutex */
 	if (IS_NOQUOTA(inode))
 		return 0;
-	down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
+
+	/* First get references to structures we might need. */
+	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
+		if (type != -1 && cnt != type)
+			continue;
+		switch (cnt) {
+		case USRQUOTA:
+			id = inode->i_uid;
+			break;
+		case GRPQUOTA:
+			id = inode->i_gid;
+			break;
+		}
+		got[cnt] = dqget(sb, id, cnt);
+	}
+
+	down_write(&sb_dqopt(sb)->dqptr_sem);
 	/* Having dqptr_sem we know NOQUOTA flags can't be altered... */
 	if (IS_NOQUOTA(inode))
 		goto out_err;
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		if (type != -1 && cnt != type)
 			continue;
+		/* Avoid races with quotaoff() */
+		if (!sb_has_quota_active(sb, cnt))
+			continue;
 		if (inode->i_dquot[cnt] == NODQUOT) {
-			switch (cnt) {
-				case USRQUOTA:
-					id = inode->i_uid;
-					break;
-				case GRPQUOTA:
-					id = inode->i_gid;
-					break;
-			}
-			inode->i_dquot[cnt] = dqget(inode->i_sb, id, cnt);
+			inode->i_dquot[cnt] = got[cnt];
+			got[cnt] = NODQUOT;
 		}
 	}
 out_err:
-	up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	up_write(&sb_dqopt(sb)->dqptr_sem);
+	/* Drop unused references */
+	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
+		dqput(got[cnt]);
 	return ret;
 }
 
 /*
  * 	Release all quotas referenced by inode
- *	Transaction must be started at an entry
  */
-int dquot_drop_locked(struct inode *inode)
+int dquot_drop(struct inode *inode)
 {
 	int cnt;
+	struct dquot *put[MAXQUOTAS];
 
+	down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
-		if (inode->i_dquot[cnt] != NODQUOT) {
-			dqput(inode->i_dquot[cnt]);
-			inode->i_dquot[cnt] = NODQUOT;
-		}
+		put[cnt] = inode->i_dquot[cnt];
+		inode->i_dquot[cnt] = NODQUOT;
 	}
-	return 0;
-}
-
-int dquot_drop(struct inode *inode)
-{
-	down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
-	dquot_drop_locked(inode);
 	up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
+
+	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
+		dqput(put[cnt]);
 	return 0;
 }
 
@@ -1470,8 +1484,9 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr)
 	qsize_t space;
 	struct dquot *transfer_from[MAXQUOTAS];
 	struct dquot *transfer_to[MAXQUOTAS];
-	int cnt, ret = NO_QUOTA, chuid = (iattr->ia_valid & ATTR_UID) && inode->i_uid != iattr->ia_uid,
-	    chgid = (iattr->ia_valid & ATTR_GID) && inode->i_gid != iattr->ia_gid;
+	int cnt, ret = QUOTA_OK;
+	int chuid = iattr->ia_valid & ATTR_UID && inode->i_uid != iattr->ia_uid,
+	    chgid = iattr->ia_valid & ATTR_GID && inode->i_gid != iattr->ia_gid;
 	char warntype_to[MAXQUOTAS];
 	char warntype_from_inodes[MAXQUOTAS], warntype_from_space[MAXQUOTAS];
 
@@ -1479,21 +1494,11 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr)
          * re-enter the quota code and are already holding the mutex */
 	if (IS_NOQUOTA(inode))
 		return QUOTA_OK;
-	/* Clear the arrays */
+	/* Initialize the arrays */
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
-		transfer_to[cnt] = transfer_from[cnt] = NODQUOT;
+		transfer_from[cnt] = NODQUOT;
+		transfer_to[cnt] = NODQUOT;
 		warntype_to[cnt] = QUOTA_NL_NOWARN;
-	}
-	down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
-	/* Now recheck reliably when holding dqptr_sem */
-	if (IS_NOQUOTA(inode)) {	/* File without quota accounting? */
-		up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
-		return QUOTA_OK;
-	}
-	/* First build the transfer_to list - here we can block on
-	 * reading/instantiating of dquots.  We know that the transaction for
-	 * us was already started so we don't violate lock ranking here */
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		switch (cnt) {
 			case USRQUOTA:
 				if (!chuid)
@@ -1507,6 +1512,13 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr)
 				break;
 		}
 	}
+
+	down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	/* Now recheck reliably when holding dqptr_sem */
+	if (IS_NOQUOTA(inode)) {	/* File without quota accounting? */
+		up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
+		goto put_all;
+	}
 	spin_lock(&dq_data_lock);
 	space = inode_get_bytes(inode);
 	/* Build the transfer_from list and check the limits */
@@ -1517,7 +1529,7 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr)
 		if (check_idq(transfer_to[cnt], 1, warntype_to + cnt) ==
 		    NO_QUOTA || check_bdq(transfer_to[cnt], space, 0,
 		    warntype_to + cnt) == NO_QUOTA)
-			goto warn_put_all;
+			goto over_quota;
 	}
 
 	/*
@@ -1545,28 +1557,37 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr)
 
 		inode->i_dquot[cnt] = transfer_to[cnt];
 	}
-	ret = QUOTA_OK;
-warn_put_all:
 	spin_unlock(&dq_data_lock);
+	up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
+
 	/* Dirtify all the dquots - this can block when journalling */
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		if (transfer_from[cnt])
 			mark_dquot_dirty(transfer_from[cnt]);
-		if (transfer_to[cnt])
+		if (transfer_to[cnt]) {
 			mark_dquot_dirty(transfer_to[cnt]);
+			/* The reference we got is transferred to the inode */
+			transfer_to[cnt] = NODQUOT;
+		}
 	}
+warn_put_all:
 	flush_warnings(transfer_to, warntype_to);
 	flush_warnings(transfer_from, warntype_from_inodes);
 	flush_warnings(transfer_from, warntype_from_space);
-	
+put_all:
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
-		if (ret == QUOTA_OK && transfer_from[cnt] != NODQUOT)
-			dqput(transfer_from[cnt]);
-		if (ret == NO_QUOTA && transfer_to[cnt] != NODQUOT)
-			dqput(transfer_to[cnt]);
+		dqput(transfer_from[cnt]);
+		dqput(transfer_to[cnt]);
 	}
-	up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	return ret;
+over_quota:
+	spin_unlock(&dq_data_lock);
+	up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	/* Clear dquot pointers we don't want to dqput() */
+	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
+		transfer_from[cnt] = NODQUOT;
+	ret = NO_QUOTA;
+	goto warn_put_all;
 }
 
 /* Wrapper for transferring ownership of an inode */
@@ -1651,19 +1672,24 @@ int vfs_quota_disable(struct super_block *sb, int type, unsigned int flags)
 			continue;
 
 		if (flags & DQUOT_SUSPENDED) {
+			spin_lock(&dq_state_lock);
 			dqopt->flags |=
 				dquot_state_flag(DQUOT_SUSPENDED, cnt);
+			spin_unlock(&dq_state_lock);
 		} else {
+			spin_lock(&dq_state_lock);
 			dqopt->flags &= ~dquot_state_flag(flags, cnt);
 			/* Turning off suspended quotas? */
 			if (!sb_has_quota_loaded(sb, cnt) &&
 			    sb_has_quota_suspended(sb, cnt)) {
 				dqopt->flags &=	~dquot_state_flag(
 							DQUOT_SUSPENDED, cnt);
+				spin_unlock(&dq_state_lock);
 				iput(dqopt->files[cnt]);
 				dqopt->files[cnt] = NULL;
 				continue;
 			}
+			spin_unlock(&dq_state_lock);
 		}
 
 		/* We still have to keep quota loaded? */
@@ -1830,7 +1856,9 @@ static int vfs_load_quota_inode(struct inode *inode, int type, int format_id,
 	}
 	mutex_unlock(&dqopt->dqio_mutex);
 	mutex_unlock(&inode->i_mutex);
+	spin_lock(&dq_state_lock);
 	dqopt->flags |= dquot_state_flag(flags, type);
+	spin_unlock(&dq_state_lock);
 
 	add_dquot_ref(sb, type);
 	mutex_unlock(&dqopt->dqonoff_mutex);
@@ -1872,9 +1900,11 @@ static int vfs_quota_on_remount(struct super_block *sb, int type)
 	}
 	inode = dqopt->files[type];
 	dqopt->files[type] = NULL;
+	spin_lock(&dq_state_lock);
 	flags = dqopt->flags & dquot_state_flag(DQUOT_USAGE_ENABLED |
 						DQUOT_LIMITS_ENABLED, type);
 	dqopt->flags &= ~dquot_state_flag(DQUOT_STATE_FLAGS, type);
+	spin_unlock(&dq_state_lock);
 	mutex_unlock(&dqopt->dqonoff_mutex);
 
 	flags = dquot_generic_flag(flags, type);
@@ -1952,7 +1982,9 @@ int vfs_quota_enable(struct inode *inode, int type, int format_id,
 			ret = -EBUSY;
 			goto out_lock;
 		}
+		spin_lock(&dq_state_lock);
 		sb_dqopt(sb)->flags |= dquot_state_flag(flags, type);
+		spin_unlock(&dq_state_lock);
 out_lock:
 		mutex_unlock(&dqopt->dqonoff_mutex);
 		return ret;
@@ -2039,14 +2071,12 @@ int vfs_get_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *d
 {
 	struct dquot *dquot;
 
-	mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
-	if (!(dquot = dqget(sb, id, type))) {
-		mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
+	dquot = dqget(sb, id, type);
+	if (dquot == NODQUOT)
 		return -ESRCH;
-	}
 	do_get_dqblk(dquot, di);
 	dqput(dquot);
-	mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
+
 	return 0;
 }
 
@@ -2130,7 +2160,6 @@ int vfs_set_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *d
 	struct dquot *dquot;
 	int rc;
 
-	mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
 	dquot = dqget(sb, id, type);
 	if (!dquot) {
 		rc = -ESRCH;
@@ -2139,7 +2168,6 @@ int vfs_set_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *d
 	rc = do_set_dqblk(dquot, di);
 	dqput(dquot);
 out:
-	mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
 	return rc;
 }
 
@@ -2370,11 +2398,9 @@ EXPORT_SYMBOL(dquot_release);
 EXPORT_SYMBOL(dquot_mark_dquot_dirty);
 EXPORT_SYMBOL(dquot_initialize);
 EXPORT_SYMBOL(dquot_drop);
-EXPORT_SYMBOL(dquot_drop_locked);
 EXPORT_SYMBOL(vfs_dq_drop);
 EXPORT_SYMBOL(dqget);
 EXPORT_SYMBOL(dqput);
-EXPORT_SYMBOL(dquot_is_cached);
 EXPORT_SYMBOL(dquot_alloc_space);
 EXPORT_SYMBOL(dquot_alloc_inode);
 EXPORT_SYMBOL(dquot_free_space);
diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h
index 21b781a3350f..0b35b3a1be05 100644
--- a/include/linux/quotaops.h
+++ b/include/linux/quotaops.h
@@ -24,10 +24,8 @@ void sync_dquots(struct super_block *sb, int type);
 
 int dquot_initialize(struct inode *inode, int type);
 int dquot_drop(struct inode *inode);
-int dquot_drop_locked(struct inode *inode);
 struct dquot *dqget(struct super_block *sb, unsigned int id, int type);
 void dqput(struct dquot *dquot);
-int dquot_is_cached(struct super_block *sb, unsigned int id, int type);
 int dquot_scan_active(struct super_block *sb,
 		      int (*fn)(struct dquot *dquot, unsigned long priv),
 		      unsigned long priv);
-- 
cgit v1.2.3


From aa8c6c93747f7b55fa11e1624fec8ca33763a805 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Fri, 16 Jan 2009 21:54:43 +0100
Subject: PCI PM: Restore standard config registers of all devices early

There is a problem in our handling of suspend-resume of PCI devices that
many of them have their standard config registers restored with
interrupts enabled and they are put into the full power state with
interrupts enabled as well.  This may lead to the following scenario:
  * an interrupt vector is shared between two or more devices
  * one device is resumed earlier and generates an interrupt
  * the interrupt handler of another device tries to handle it and
    attempts to access the device the config space of which hasn't been
    restored yet and/or which still is in a low power state
  * the system crashes as a result

To prevent this from happening we should restore the standard
configuration registers of all devices with interrupts disabled and we
should put them into the D0 power state right after that.
Unfortunately, this cannot be done using the existing
pci_set_power_state(), because it can sleep.  Also, to do it we have to
make sure that the config spaces of all devices were actually saved
during suspend.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci-driver.c | 91 ++++++++++++++----------------------------------
 drivers/pci/pci.c        | 63 +++++++++++++++++++++++++++++----
 drivers/pci/pci.h        |  6 ++++
 include/linux/pci.h      |  5 +++
 4 files changed, 95 insertions(+), 70 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index c697f2680856..9de07b75b993 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -355,17 +355,27 @@ static int pci_legacy_suspend(struct device *dev, pm_message_t state)
 	int i = 0;
 
 	if (drv && drv->suspend) {
+		pci_dev->state_saved = false;
+
 		i = drv->suspend(pci_dev, state);
 		suspend_report_result(drv->suspend, i);
-	} else {
-		pci_save_state(pci_dev);
-		/*
-		 * This is for compatibility with existing code with legacy PM
-		 * support.
-		 */
-		pci_pm_set_unknown_state(pci_dev);
+		if (i)
+			return i;
+
+		if (pci_dev->state_saved)
+			goto Fixup;
+
+		if (WARN_ON_ONCE(pci_dev->current_state != PCI_D0))
+			goto Fixup;
 	}
 
+	pci_save_state(pci_dev);
+	/*
+	 * This is for compatibility with existing code with legacy PM support.
+	 */
+	pci_pm_set_unknown_state(pci_dev);
+
+ Fixup:
 	pci_fixup_device(pci_fixup_suspend, pci_dev);
 
 	return i;
@@ -386,81 +396,34 @@ static int pci_legacy_suspend_late(struct device *dev, pm_message_t state)
 
 static int pci_legacy_resume_early(struct device *dev)
 {
-	int error = 0;
 	struct pci_dev * pci_dev = to_pci_dev(dev);
 	struct pci_driver * drv = pci_dev->driver;
 
-	pci_fixup_device(pci_fixup_resume_early, pci_dev);
-
-	if (drv && drv->resume_early)
-		error = drv->resume_early(pci_dev);
-	return error;
+	return drv && drv->resume_early ?
+			drv->resume_early(pci_dev) : 0;
 }
 
 static int pci_legacy_resume(struct device *dev)
 {
-	int error;
 	struct pci_dev * pci_dev = to_pci_dev(dev);
 	struct pci_driver * drv = pci_dev->driver;
 
 	pci_fixup_device(pci_fixup_resume, pci_dev);
 
-	if (drv && drv->resume) {
-		error = drv->resume(pci_dev);
-	} else {
-		/* restore the PCI config space */
-		pci_restore_state(pci_dev);
-		error = pci_pm_reenable_device(pci_dev);
-	}
-	return error;
+	return drv && drv->resume ?
+			drv->resume(pci_dev) : pci_pm_reenable_device(pci_dev);
 }
 
 /* Auxiliary functions used by the new power management framework */
 
-static int pci_restore_standard_config(struct pci_dev *pci_dev)
-{
-	struct pci_dev *parent = pci_dev->bus->self;
-	int error = 0;
-
-	/* Check if the device's bus is operational */
-	if (!parent || parent->current_state == PCI_D0) {
-		pci_restore_state(pci_dev);
-		pci_update_current_state(pci_dev, PCI_D0);
-	} else {
-		dev_warn(&pci_dev->dev, "unable to restore config, "
-			"bridge %s in low power state D%d\n", pci_name(parent),
-			parent->current_state);
-		pci_dev->current_state = PCI_UNKNOWN;
-		error = -EAGAIN;
-	}
-
-	return error;
-}
-
-static bool pci_is_bridge(struct pci_dev *pci_dev)
-{
-	return !!(pci_dev->subordinate);
-}
-
 static void pci_pm_default_resume_noirq(struct pci_dev *pci_dev)
 {
-	if (pci_restore_standard_config(pci_dev))
-		pci_fixup_device(pci_fixup_resume_early, pci_dev);
+	pci_restore_standard_config(pci_dev);
+	pci_fixup_device(pci_fixup_resume_early, pci_dev);
 }
 
 static int pci_pm_default_resume(struct pci_dev *pci_dev)
 {
-	/*
-	 * pci_restore_standard_config() should have been called once already,
-	 * but it would have failed if the device's parent bridge had not been
-	 * in power state D0 at that time.  Check it and try again if necessary.
-	 */
-	if (pci_dev->current_state == PCI_UNKNOWN) {
-		int error = pci_restore_standard_config(pci_dev);
-		if (error)
-			return error;
-	}
-
 	pci_fixup_device(pci_fixup_resume, pci_dev);
 
 	if (!pci_is_bridge(pci_dev))
@@ -575,11 +538,11 @@ static int pci_pm_resume_noirq(struct device *dev)
 	struct device_driver *drv = dev->driver;
 	int error = 0;
 
+	pci_pm_default_resume_noirq(pci_dev);
+
 	if (pci_has_legacy_pm_support(pci_dev))
 		return pci_legacy_resume_early(dev);
 
-	pci_pm_default_resume_noirq(pci_dev);
-
 	if (drv && drv->pm && drv->pm->resume_noirq)
 		error = drv->pm->resume_noirq(dev);
 
@@ -730,11 +693,11 @@ static int pci_pm_restore_noirq(struct device *dev)
 	struct device_driver *drv = dev->driver;
 	int error = 0;
 
+	pci_pm_default_resume_noirq(pci_dev);
+
 	if (pci_has_legacy_pm_support(pci_dev))
 		return pci_legacy_resume_early(dev);
 
-	pci_pm_default_resume_noirq(pci_dev);
-
 	if (drv && drv->pm && drv->pm->restore_noirq)
 		error = drv->pm->restore_noirq(dev);
 
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index e491fdedf705..17bd9325a245 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -22,7 +22,7 @@
 #include <asm/dma.h>	/* isa_dma_bridge_buggy */
 #include "pci.h"
 
-unsigned int pci_pm_d3_delay = 10;
+unsigned int pci_pm_d3_delay = PCI_PM_D3_WAIT;
 
 #ifdef CONFIG_PCI_DOMAINS
 int pci_domains_supported = 1;
@@ -426,6 +426,7 @@ static inline int platform_pci_sleep_wake(struct pci_dev *dev, bool enable)
  *                           given PCI device
  * @dev: PCI device to handle.
  * @state: PCI power state (D0, D1, D2, D3hot) to put the device into.
+ * @wait: If 'true', wait for the device to change its power state
  *
  * RETURN VALUE:
  * -EINVAL if the requested state is invalid.
@@ -435,7 +436,7 @@ static inline int platform_pci_sleep_wake(struct pci_dev *dev, bool enable)
  * 0 if device's power state has been successfully changed.
  */
 static int
-pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state)
+pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state, bool wait)
 {
 	u16 pmcsr;
 	bool need_restore = false;
@@ -480,8 +481,10 @@ pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state)
 		break;
 	case PCI_UNKNOWN: /* Boot-up */
 		if ((pmcsr & PCI_PM_CTRL_STATE_MASK) == PCI_D3hot
-		 && !(pmcsr & PCI_PM_CTRL_NO_SOFT_RESET))
+		 && !(pmcsr & PCI_PM_CTRL_NO_SOFT_RESET)) {
 			need_restore = true;
+			wait = true;
+		}
 		/* Fall-through: force to D0 */
 	default:
 		pmcsr = 0;
@@ -491,12 +494,15 @@ pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state)
 	/* enter specified state */
 	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, pmcsr);
 
+	if (!wait)
+		return 0;
+
 	/* Mandatory power management transition delays */
 	/* see PCI PM 1.1 5.6.1 table 18 */
 	if (state == PCI_D3hot || dev->current_state == PCI_D3hot)
 		msleep(pci_pm_d3_delay);
 	else if (state == PCI_D2 || dev->current_state == PCI_D2)
-		udelay(200);
+		udelay(PCI_PM_D2_DELAY);
 
 	dev->current_state = state;
 
@@ -515,7 +521,7 @@ pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state)
 	if (need_restore)
 		pci_restore_bars(dev);
 
-	if (dev->bus->self)
+	if (wait && dev->bus->self)
 		pcie_aspm_pm_state_change(dev->bus->self);
 
 	return 0;
@@ -585,7 +591,7 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state)
 	if (state == PCI_D3hot && (dev->dev_flags & PCI_DEV_FLAGS_NO_D3))
 		return 0;
 
-	error = pci_raw_set_power_state(dev, state);
+	error = pci_raw_set_power_state(dev, state, true);
 
 	if (state > PCI_D0 && platform_pci_power_manageable(dev)) {
 		/* Allow the platform to finalize the transition */
@@ -730,6 +736,7 @@ pci_save_state(struct pci_dev *dev)
 	/* XXX: 100% dword access ok here? */
 	for (i = 0; i < 16; i++)
 		pci_read_config_dword(dev, i * 4,&dev->saved_config_space[i]);
+	dev->state_saved = true;
 	if ((i = pci_save_pcie_state(dev)) != 0)
 		return i;
 	if ((i = pci_save_pcix_state(dev)) != 0)
@@ -1373,6 +1380,50 @@ void pci_allocate_cap_save_buffers(struct pci_dev *dev)
 			"unable to preallocate PCI-X save buffer\n");
 }
 
+/**
+ * pci_restore_standard_config - restore standard config registers of PCI device
+ * @dev: PCI device to handle
+ *
+ * This function assumes that the device's configuration space is accessible.
+ * If the device needs to be powered up, the function will wait for it to
+ * change the state.
+ */
+int pci_restore_standard_config(struct pci_dev *dev)
+{
+	pci_power_t prev_state;
+	int error;
+
+	pci_restore_state(dev);
+	pci_update_current_state(dev, PCI_D0);
+
+	prev_state = dev->current_state;
+	if (prev_state == PCI_D0)
+		return 0;
+
+	error = pci_raw_set_power_state(dev, PCI_D0, false);
+	if (error)
+		return error;
+
+	if (pci_is_bridge(dev)) {
+		if (prev_state > PCI_D1)
+			mdelay(PCI_PM_BUS_WAIT);
+	} else {
+		switch(prev_state) {
+		case PCI_D3cold:
+		case PCI_D3hot:
+			mdelay(pci_pm_d3_delay);
+			break;
+		case PCI_D2:
+			udelay(PCI_PM_D2_DELAY);
+			break;
+		}
+	}
+
+	dev->current_state = PCI_D0;
+
+	return 0;
+}
+
 /**
  * pci_enable_ari - enable ARI forwarding if hardware support it
  * @dev: the PCI device
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index 1351bb4addde..26ddf78ac300 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -49,6 +49,12 @@ extern void pci_disable_enabled_device(struct pci_dev *dev);
 extern void pci_pm_init(struct pci_dev *dev);
 extern void platform_pci_wakeup_init(struct pci_dev *dev);
 extern void pci_allocate_cap_save_buffers(struct pci_dev *dev);
+extern int pci_restore_standard_config(struct pci_dev *dev);
+
+static inline bool pci_is_bridge(struct pci_dev *pci_dev)
+{
+	return !!(pci_dev->subordinate);
+}
 
 extern int pci_user_read_config_byte(struct pci_dev *dev, int where, u8 *val);
 extern int pci_user_read_config_word(struct pci_dev *dev, int where, u16 *val);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 80f8b8b65fde..48890cf3f96e 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -117,6 +117,10 @@ typedef int __bitwise pci_power_t;
 #define PCI_UNKNOWN	((pci_power_t __force) 5)
 #define PCI_POWER_ERROR	((pci_power_t __force) -1)
 
+#define PCI_PM_D2_DELAY	200
+#define PCI_PM_D3_WAIT	10
+#define PCI_PM_BUS_WAIT	50
+
 /** The pci_channel state describes connectivity between the CPU and
  *  the pci device.  If some PCI bus between here and the pci device
  *  has crashed or locked up, this info is reflected here.
@@ -252,6 +256,7 @@ struct pci_dev {
 	unsigned int	ari_enabled:1;	/* ARI forwarding */
 	unsigned int	is_managed:1;
 	unsigned int	is_pcie:1;
+	unsigned int	state_saved:1;
 	pci_dev_flags_t dev_flags;
 	atomic_t	enable_cnt;	/* pci_enable_device has been called */
 
-- 
cgit v1.2.3


From c50331e8be32eaba5e1949f98c70d50b891262db Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Mon, 19 Jan 2009 15:33:14 -0700
Subject: dmaengine: dma_issue_pending_all == nop when CONFIG_DMA_ENGINE=n

The device list will always be empty in this configuration, so no need
to walk the list.

Reported-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 include/linux/dmaengine.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 64dea2ab326c..c4a560e72ab7 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -390,11 +390,16 @@ static inline enum dma_status dma_async_is_complete(dma_cookie_t cookie,
 enum dma_status dma_sync_wait(struct dma_chan *chan, dma_cookie_t cookie);
 #ifdef CONFIG_DMA_ENGINE
 enum dma_status dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx);
+void dma_issue_pending_all(void);
 #else
 static inline enum dma_status dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx)
 {
 	return DMA_SUCCESS;
 }
+static inline void dma_issue_pending_all(void)
+{
+	do { } while (0);
+}
 #endif
 
 /* --- DMA device --- */
@@ -403,7 +408,6 @@ int dma_async_device_register(struct dma_device *device);
 void dma_async_device_unregister(struct dma_device *device);
 void dma_run_dependencies(struct dma_async_tx_descriptor *tx);
 struct dma_chan *dma_find_channel(enum dma_transaction_type tx_type);
-void dma_issue_pending_all(void);
 #define dma_request_channel(mask, x, y) __dma_request_channel(&(mask), x, y)
 struct dma_chan *__dma_request_channel(dma_cap_mask_t *mask, dma_filter_fn fn, void *fn_param);
 void dma_release_channel(struct dma_chan *chan);
-- 
cgit v1.2.3


From ef560682a97491f62ef538931a4861b57d66c52c Mon Sep 17 00:00:00 2001
From: Guennadi Liakhovetski <lg@denx.de>
Date: Mon, 19 Jan 2009 15:36:21 -0700
Subject: dmaengine: add async_tx_clear_ack() macro

To complete the DMA_CTRL_ACK handling API add a async_tx_clear_ack() macro.

Signed-off-by: Guennadi Liakhovetski <lg@denx.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 include/linux/dmaengine.h | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index c4a560e72ab7..34f124d7fb94 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -287,6 +287,11 @@ static inline void async_tx_ack(struct dma_async_tx_descriptor *tx)
 	tx->flags |= DMA_CTRL_ACK;
 }
 
+static inline void async_tx_clear_ack(struct dma_async_tx_descriptor *tx)
+{
+	tx->flags &= ~DMA_CTRL_ACK;
+}
+
 static inline bool async_tx_test_ack(struct dma_async_tx_descriptor *tx)
 {
 	return (tx->flags & DMA_CTRL_ACK) == DMA_CTRL_ACK;
-- 
cgit v1.2.3


From 8adb711f3668b034e7b956fac951ed08b53e0d55 Mon Sep 17 00:00:00 2001
From: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Date: Tue, 20 Jan 2009 12:17:28 -0800
Subject: debugfs: introduce stub for debugfs_create_size_t() when DEBUG_FS=n
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Toralf Förster <toralf.foerster@gmx.de> reported a build failure in
the WiMAX stack when CONFIG_DEBUG_FS=n

http://linuxwimax.org/pipermail/wimax/2009-January/000449.html

This is due to debugfs_create_size_t() missing an stub that returns
-ENODEV when the DEBUGFS subsystem is not configured in (like the rest
of the debugfs API).

This patch adds said stub.

Reported-by: Toralf Förster <toralf.foerster@gmx.de>
Signed-off-by: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/debugfs.h | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/debugfs.h b/include/linux/debugfs.h
index 23936b16426b..0f5c33b0bd3e 100644
--- a/include/linux/debugfs.h
+++ b/include/linux/debugfs.h
@@ -162,6 +162,13 @@ static inline struct dentry *debugfs_create_x32(const char *name, mode_t mode,
 	return ERR_PTR(-ENODEV);
 }
 
+struct dentry *debugfs_create_size_t(const char *name, mode_t mode,
+				     struct dentry *parent,
+				     size_t *value)
+{
+	return ERR_PTR(-ENODEV);
+}
+
 static inline struct dentry *debugfs_create_bool(const char *name, mode_t mode,
 						 struct dentry *parent,
 						 u32 *value)
-- 
cgit v1.2.3


From c0e69a5bbc6fc74184aa043aadb9a53bc58f953b Mon Sep 17 00:00:00 2001
From: Jesper Nilsson <Jesper.Nilsson@axis.com>
Date: Wed, 14 Jan 2009 11:19:08 +0100
Subject: klist.c: bit 0 in pointer can't be used as flag

The commit a1ed5b0cffe4b16a93a6a3390e8cee0fbef94f86
(klist: don't iterate over deleted entries) introduces use of the
low bit in a pointer to indicate if the knode is dead or not,
assuming that this bit is always free.

This is not true for all architectures, CRIS for example may align data
on byte borders.

The result is a bunch of warnings on bootup, devices not being
added correctly etc, reported by Hinko Kocevar <hinko.kocevar@cetrtapot.si>:

------------[ cut here ]------------
WARNING: at lib/klist.c:62 ()
Modules linked in:

Stack from c1fe1cf0:
       c01cc7f4 c1fe1d11 c000eb4e c000e4de 00000000 00000000 c1f4f78f c1f50c2d
       c01d008c c1fdd1a0 c1fdd1a0 c1fe1d38 c0192954 c1fe0000 00000000 c1fe1dc0
       00000002 7fffffff c1fe1da8 c0192d50 c1fe1dc0 00000002 7fffffff c1ff9fcc
Call Trace: [<c000eb4e>] [<c000e4de>] [<c0192954>] [<c0192d50>] [<c001d49e>] [<c000b688>] [<c0192a3c>]
       [<c000b63e>] [<c000b63e>] [<c001a542>] [<c00b55b0>] [<c00411c0>] [<c00b559c>] [<c01918e6>] [<c0191988>]
       [<c01919d0>] [<c00cd9c8>] [<c00cdd6a>] [<c0034178>] [<c000409a>] [<c0015576>] [<c0029130>] [<c0029078>]
       [<c0029170>] [<c0012336>] [<c00b4076>] [<c00b4770>] [<c006d6e4>] [<c006d974>] [<c006dca0>] [<c0028d6c>]
       [<c0028e12>] [<c0006424>] <4>---[ end trace 4eaa2a86a8e2da22 ]---
------------[ cut here ]------------
Repeat ad nauseam.

Wed, Jan 14, 2009 at 12:11:32AM +0100, Bastien ROUCARIES wrote:
> Perhaps using a pointerhackalign trick on this structure where
> #define pointerhackalign(x) __attribute__ ((aligned (x)))
> and declare
> struct klist_node {
> ...
> }  pointerhackalign(2);
>
> Because  __attribute__ ((aligned (x))) could only increase alignment
> it will safe to do that and serve as documentation purpose :)

That works, but we need to do it not for the struct klist_node,
but for the struct we insert into the void * in klist_node,
which is struct klist.

Reported-by: Hinko Kocevar <hinko.kocevar@cetrtapot.si
Cc: Bastien ROUCARIES <roucaries.bastien@gmail.com>
Signed-off-by: Jesper Nilsson <jesper.nilsson@axis.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/klist.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/klist.h b/include/linux/klist.h
index d5a27af9dba5..e91a4e59b771 100644
--- a/include/linux/klist.h
+++ b/include/linux/klist.h
@@ -22,7 +22,7 @@ struct klist {
 	struct list_head	k_list;
 	void			(*get)(struct klist_node *);
 	void			(*put)(struct klist_node *);
-};
+} __attribute__ ((aligned (4)));
 
 #define KLIST_INIT(_name, _get, _put)					\
 	{ .k_lock	= __SPIN_LOCK_UNLOCKED(_name.k_lock),		\
-- 
cgit v1.2.3


From 0b491eee46012772cbf029450d123e933c2e7940 Mon Sep 17 00:00:00 2001
From: Steve Glendinning <steve.glendinning@smsc.com>
Date: Wed, 21 Jan 2009 12:35:43 -0800
Subject: usbnet: allow type check of devdbg arguments in non-debug build

Improve usbnet's devdbg to always type-check diagnostic arguments,
like dev_dbg (device.h).  This makes no change to the resulting size of
usbnet modules.

This patch also removes an #ifdef DEBUG directive from rndis_wlan so
it's devdbg statements are always type-checked at compile time.

Signed-off-by: Steve Glendinning <steve.glendinning@smsc.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/wireless/rndis_wlan.c | 2 --
 include/linux/usb/usbnet.h        | 4 +++-
 2 files changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c
index 607ce9f61b54..ed93ac41297f 100644
--- a/drivers/net/wireless/rndis_wlan.c
+++ b/drivers/net/wireless/rndis_wlan.c
@@ -1649,9 +1649,7 @@ static char *rndis_translate_scan(struct net_device *dev,
 				  char *end_buf,
 				  struct ndis_80211_bssid_ex *bssid)
 {
-#ifdef DEBUG
 	struct usbnet *usbdev = netdev_priv(dev);
-#endif
 	u8 *ie;
 	char *current_val;
 	int bssid_len, ie_len, i;
diff --git a/include/linux/usb/usbnet.h b/include/linux/usb/usbnet.h
index ba09fe88adda..7d3822243074 100644
--- a/include/linux/usb/usbnet.h
+++ b/include/linux/usb/usbnet.h
@@ -197,7 +197,9 @@ extern int usbnet_nway_reset(struct net_device *net);
 #define devdbg(usbnet, fmt, arg...) \
 	printk(KERN_DEBUG "%s: " fmt "\n" , (usbnet)->net->name , ## arg)
 #else
-#define devdbg(usbnet, fmt, arg...) do {} while(0)
+#define devdbg(usbnet, fmt, arg...) \
+	({ if (0) printk(KERN_DEBUG "%s: " fmt "\n" , (usbnet)->net->name , \
+		## arg); 0; })
 #endif
 
 #define deverr(usbnet, fmt, arg...) \
-- 
cgit v1.2.3


From 336f6c322d87806ef93afad6308ac65083a865e5 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 22 Jan 2009 09:50:44 +0100
Subject: debugobjects: add and use INIT_WORK_ON_STACK

Impact: Fix debugobjects warning

debugobject enabled kernels spit out a warning in hpet code due to a
workqueue which is initialized on stack.

Add INIT_WORK_ON_STACK() which calls init_timer_on_stack() and use it
in hpet.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/hpet.c    | 3 ++-
 include/linux/workqueue.h | 6 ++++++
 2 files changed, 8 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index cd759ad90690..64d5ad0b8add 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -628,11 +628,12 @@ static int hpet_cpuhp_notify(struct notifier_block *n,
 
 	switch (action & 0xf) {
 	case CPU_ONLINE:
-		INIT_DELAYED_WORK(&work.work, hpet_work);
+		INIT_DELAYED_WORK_ON_STACK(&work.work, hpet_work);
 		init_completion(&work.complete);
 		/* FIXME: add schedule_work_on() */
 		schedule_delayed_work_on(cpu, &work.work, 0);
 		wait_for_completion(&work.complete);
+		destroy_timer_on_stack(&work.work.timer);
 		break;
 	case CPU_DEAD:
 		if (hdev) {
diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index b36291130f22..20b59eb1facd 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -124,6 +124,12 @@ struct execute_work {
 		init_timer_deferrable(&(_work)->timer);		\
 	} while (0)
 
+#define INIT_DELAYED_WORK_ON_STACK(_work, _func)		\
+	do {							\
+		INIT_WORK(&(_work)->work, (_func));		\
+		init_timer_on_stack(&(_work)->timer);		\
+	} while (0)
+
 /**
  * work_pending - Find out whether a work item is currently pending
  * @work: The work item in question
-- 
cgit v1.2.3


From e4d866cdea24543ee16ce6d07d80c513e86ba983 Mon Sep 17 00:00:00 2001
From: "JosephChan@via.com.tw" <JosephChan@via.com.tw>
Date: Fri, 23 Jan 2009 15:37:39 +0800
Subject: [libata] pata_via: support VX855, future chips whose IDE controller
 use 0x0571

It supports VX855 and future chips whose IDE controller uses PCI ID 0x0571.

Signed-off-by: Joseph Chan <josephchan@via.com.tw>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/pata_via.c  | 22 +++++++++++++++++-----
 include/linux/pci_ids.h |  4 ++++
 2 files changed, 21 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c
index 681169c9c640..79a6c9a0b721 100644
--- a/drivers/ata/pata_via.c
+++ b/drivers/ata/pata_via.c
@@ -86,6 +86,10 @@ enum {
 	VIA_SATA_PATA	= 0x800, /* SATA/PATA combined configuration */
 };
 
+enum {
+	VIA_IDFLAG_SINGLE = (1 << 0), /* single channel controller) */
+};
+
 /*
  * VIA SouthBridge chips.
  */
@@ -97,8 +101,12 @@ static const struct via_isa_bridge {
 	u8 rev_max;
 	u16 flags;
 } via_isa_bridges[] = {
+	{ "vx855",	PCI_DEVICE_ID_VIA_VX855,    0x00, 0x2f,
+	  VIA_UDMA_133 | VIA_BAD_AST | VIA_SATA_PATA },
 	{ "vx800",	PCI_DEVICE_ID_VIA_VX800,    0x00, 0x2f, VIA_UDMA_133 |
 	VIA_BAD_AST | VIA_SATA_PATA },
+	{ "vt8261",	PCI_DEVICE_ID_VIA_8261,     0x00, 0x2f,
+	  VIA_UDMA_133 | VIA_BAD_AST },
 	{ "vt8237s",	PCI_DEVICE_ID_VIA_8237S,    0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST },
 	{ "vt8251",	PCI_DEVICE_ID_VIA_8251,     0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST },
 	{ "cx700",	PCI_DEVICE_ID_VIA_CX700,    0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST | VIA_SATA_PATA },
@@ -122,6 +130,8 @@ static const struct via_isa_bridge {
 	{ "vt82c586",	PCI_DEVICE_ID_VIA_82C586_0, 0x00, 0x0f, VIA_UDMA_NONE | VIA_SET_FIFO },
 	{ "vt82c576",	PCI_DEVICE_ID_VIA_82C576,   0x00, 0x2f, VIA_UDMA_NONE | VIA_SET_FIFO | VIA_NO_UNMASK },
 	{ "vt82c576",	PCI_DEVICE_ID_VIA_82C576,   0x00, 0x2f, VIA_UDMA_NONE | VIA_SET_FIFO | VIA_NO_UNMASK | VIA_BAD_ID },
+	{ "vtxxxx",	PCI_DEVICE_ID_VIA_ANON,    0x00, 0x2f,
+	  VIA_UDMA_133 | VIA_BAD_AST },
 	{ NULL }
 };
 
@@ -460,6 +470,7 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	static int printed_version;
 	u8 enable;
 	u32 timing;
+	unsigned long flags = id->driver_data;
 	int rc;
 
 	if (!printed_version++)
@@ -469,9 +480,13 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (rc)
 		return rc;
 
+	if (flags & VIA_IDFLAG_SINGLE)
+		ppi[1] = &ata_dummy_port_info;
+
 	/* To find out how the IDE will behave and what features we
 	   actually have to look at the bridge not the IDE controller */
-	for (config = via_isa_bridges; config->id; config++)
+	for (config = via_isa_bridges; config->id != PCI_DEVICE_ID_VIA_ANON;
+	     config++)
 		if ((isa = pci_get_device(PCI_VENDOR_ID_VIA +
 			!!(config->flags & VIA_BAD_ID),
 			config->id, NULL))) {
@@ -482,10 +497,6 @@ static int via_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 			pci_dev_put(isa);
 		}
 
-	if (!config->id) {
-		printk(KERN_WARNING "via: Unknown VIA SouthBridge, disabling.\n");
-		return -ENODEV;
-	}
 	pci_dev_put(isa);
 
 	if (!(config->flags & VIA_NO_ENABLES)) {
@@ -587,6 +598,7 @@ static const struct pci_device_id via[] = {
 	{ PCI_VDEVICE(VIA, 0x1571), },
 	{ PCI_VDEVICE(VIA, 0x3164), },
 	{ PCI_VDEVICE(VIA, 0x5324), },
+	{ PCI_VDEVICE(VIA, 0xC409), VIA_IDFLAG_SINGLE },
 
 	{ },
 };
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index d56ad9c21c09..febc10ed3858 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1357,6 +1357,7 @@
 #define PCI_DEVICE_ID_VIA_8783_0	0x3208
 #define PCI_DEVICE_ID_VIA_8237		0x3227
 #define PCI_DEVICE_ID_VIA_8251		0x3287
+#define PCI_DEVICE_ID_VIA_8261		0x3402
 #define PCI_DEVICE_ID_VIA_8237A		0x3337
 #define PCI_DEVICE_ID_VIA_8237S		0x3372
 #define PCI_DEVICE_ID_VIA_SATA_EIDE	0x5324
@@ -1366,10 +1367,13 @@
 #define PCI_DEVICE_ID_VIA_CX700		0x8324
 #define PCI_DEVICE_ID_VIA_CX700_IDE	0x0581
 #define PCI_DEVICE_ID_VIA_VX800		0x8353
+#define PCI_DEVICE_ID_VIA_VX855		0x8409
 #define PCI_DEVICE_ID_VIA_8371_1	0x8391
 #define PCI_DEVICE_ID_VIA_82C598_1	0x8598
 #define PCI_DEVICE_ID_VIA_838X_1	0xB188
 #define PCI_DEVICE_ID_VIA_83_87XX_1	0xB198
+#define PCI_DEVICE_ID_VIA_C409_IDE	0XC409
+#define PCI_DEVICE_ID_VIA_ANON		0xFFFF
 
 #define PCI_VENDOR_ID_SIEMENS           0x110A
 #define PCI_DEVICE_ID_SIEMENS_DSCC4     0x2102
-- 
cgit v1.2.3


From e2ba42ed0f8ba174302ebfabfa063fb456d5d6f5 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Mon, 26 Jan 2009 21:19:52 +0100
Subject: i2c: Delete 10 unused driver IDs

Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 include/linux/i2c-id.h | 10 ----------
 1 file changed, 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h
index 01d67ba9e985..82b7eee22636 100644
--- a/include/linux/i2c-id.h
+++ b/include/linux/i2c-id.h
@@ -40,9 +40,7 @@
 #define I2C_DRIVERID_SAA7185B	13	/* video encoder		*/
 #define I2C_DRIVERID_SAA7110	22	/* video decoder		*/
 #define I2C_DRIVERID_SAA5249	24	/* SAA5249 and compatibles	*/
-#define I2C_DRIVERID_PCF8583	25	/* real time clock		*/
 #define I2C_DRIVERID_TDA7432	27	/* Stereo sound processor	*/
-#define I2C_DRIVERID_TVMIXER    28      /* Mixer driver for tv cards    */
 #define I2C_DRIVERID_TVAUDIO    29      /* Generic TV sound driver      */
 #define I2C_DRIVERID_TDA9875    32      /* TV sound decoder chip        */
 #define I2C_DRIVERID_BT819	40     /* video decoder			*/
@@ -54,7 +52,6 @@
 #define I2C_DRIVERID_SAA7191	57	/* video decoder		*/
 #define I2C_DRIVERID_INDYCAM	58	/* SGI IndyCam			*/
 #define I2C_DRIVERID_OVCAMCHIP	61	/* OmniVision CMOS image sens.	*/
-#define I2C_DRIVERID_MAX6900	63	/* MAX6900 real-time clock	*/
 #define I2C_DRIVERID_SAA6752HS	67	/* MPEG2 encoder		*/
 #define I2C_DRIVERID_TVEEPROM	68	/* TV EEPROM			*/
 #define I2C_DRIVERID_WM8775	69	/* wm8775 audio processor	*/
@@ -62,23 +59,16 @@
 #define I2C_DRIVERID_CX25840	71	/* cx2584x video encoder	*/
 #define I2C_DRIVERID_SAA7127	72	/* saa7127 video encoder	*/
 #define I2C_DRIVERID_SAA711X	73	/* saa711x video encoders	*/
-#define I2C_DRIVERID_AKITAIOEXP	74	/* IO Expander on Sharp SL-C1000 */
 #define I2C_DRIVERID_INFRARED	75	/* I2C InfraRed on Video boards */
 #define I2C_DRIVERID_TVP5150	76	/* TVP5150 video decoder        */
 #define I2C_DRIVERID_WM8739	77	/* wm8739 audio processor	*/
 #define I2C_DRIVERID_UPD64083	78	/* upd64083 video processor	*/
 #define I2C_DRIVERID_UPD64031A	79	/* upd64031a video processor	*/
 #define I2C_DRIVERID_SAA717X	80	/* saa717x video encoder	*/
-#define I2C_DRIVERID_DS1672	81	/* Dallas/Maxim DS1672 RTC	*/
 #define I2C_DRIVERID_BT866	85	/* Conexant bt866 video encoder */
 #define I2C_DRIVERID_KS0127	86	/* Samsung ks0127 video decoder */
 #define I2C_DRIVERID_TLV320AIC23B 87	/* TI TLV320AIC23B audio codec  */
-#define I2C_DRIVERID_WM8731	89	/* Wolfson WM8731 audio codec */
-#define I2C_DRIVERID_WM8750	90	/* Wolfson WM8750 audio codec */
-#define I2C_DRIVERID_WM8753	91	/* Wolfson WM8753 audio codec */
-#define I2C_DRIVERID_LM4857 	92 	/* LM4857 Audio Amplifier */
 #define I2C_DRIVERID_VP27SMPX	93	/* Panasonic VP27s tuner internal MPX */
-#define I2C_DRIVERID_CS4270	94	/* Cirrus Logic 4270 audio codec */
 #define I2C_DRIVERID_M52790 	95      /* Mitsubishi M52790SP/FP AV switch */
 #define I2C_DRIVERID_CS5345	96	/* cs5345 audio processor	*/
 
-- 
cgit v1.2.3


From 1745522ccbabd990bfc7511861aa9fa98287cba0 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Mon, 26 Jan 2009 21:19:52 +0100
Subject: i2c: Delete many unused adapter IDs

Signed-off-by: Jean Delvare <khali@linux-fr.org>
---
 drivers/gpu/drm/i915/intel_i2c.c       |  4 ---
 drivers/i2c/busses/i2c-acorn.c         |  1 -
 drivers/i2c/busses/i2c-ali1535.c       |  1 -
 drivers/i2c/busses/i2c-ali1563.c       |  1 -
 drivers/i2c/busses/i2c-ali15x3.c       |  1 -
 drivers/i2c/busses/i2c-amd756.c        |  1 -
 drivers/i2c/busses/i2c-amd8111.c       |  1 -
 drivers/i2c/busses/i2c-au1550.c        |  1 -
 drivers/i2c/busses/i2c-bfin-twi.c      |  1 -
 drivers/i2c/busses/i2c-elektor.c       |  1 -
 drivers/i2c/busses/i2c-hydra.c         |  1 -
 drivers/i2c/busses/i2c-i801.c          |  1 -
 drivers/i2c/busses/i2c-ibm_iic.c       |  1 -
 drivers/i2c/busses/i2c-iop3xx.c        |  1 -
 drivers/i2c/busses/i2c-ixp2000.c       |  1 -
 drivers/i2c/busses/i2c-mpc.c           |  1 -
 drivers/i2c/busses/i2c-mv64xxx.c       |  1 -
 drivers/i2c/busses/i2c-nforce2.c       |  1 -
 drivers/i2c/busses/i2c-parport-light.c |  1 -
 drivers/i2c/busses/i2c-parport.c       |  1 -
 drivers/i2c/busses/i2c-pca-isa.c       |  1 -
 drivers/i2c/busses/i2c-piix4.c         |  1 -
 drivers/i2c/busses/i2c-sibyte.c        |  2 --
 drivers/i2c/busses/i2c-sis5595.c       |  1 -
 drivers/i2c/busses/i2c-sis630.c        |  1 -
 drivers/i2c/busses/i2c-sis96x.c        |  1 -
 drivers/i2c/busses/i2c-via.c           |  1 -
 drivers/i2c/busses/i2c-viapro.c        |  1 -
 drivers/i2c/busses/i2c-voodoo3.c       |  2 --
 drivers/i2c/busses/scx200_acb.c        |  1 -
 drivers/i2c/busses/scx200_i2c.c        |  1 -
 drivers/ieee1394/pcilynx.c             |  1 -
 drivers/video/aty/radeon_i2c.c         |  1 -
 drivers/video/i810/i810-i2c.c          |  1 -
 drivers/video/intelfb/intelfb_i2c.c    |  1 -
 drivers/video/nvidia/nv_i2c.c          |  1 -
 drivers/video/savage/savagefb-i2c.c    |  1 -
 include/linux/i2c-id.h                 | 51 ----------------------------------
 38 files changed, 93 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c
index a5a2f5339e9e..5ee9d4c25753 100644
--- a/drivers/gpu/drm/i915/intel_i2c.c
+++ b/drivers/gpu/drm/i915/intel_i2c.c
@@ -137,10 +137,6 @@ struct intel_i2c_chan *intel_i2c_create(struct drm_device *dev, const u32 reg,
 	chan->reg = reg;
 	snprintf(chan->adapter.name, I2C_NAME_SIZE, "intel drm %s", name);
 	chan->adapter.owner = THIS_MODULE;
-#ifndef I2C_HW_B_INTELFB
-#define I2C_HW_B_INTELFB I2C_HW_B_I810
-#endif
-	chan->adapter.id = I2C_HW_B_INTELFB;
 	chan->adapter.algo_data	= &chan->algo;
 	chan->adapter.dev.parent = &dev->pdev->dev;
 	chan->algo.setsda = set_data;
diff --git a/drivers/i2c/busses/i2c-acorn.c b/drivers/i2c/busses/i2c-acorn.c
index 75089febbc13..9fee3ca17344 100644
--- a/drivers/i2c/busses/i2c-acorn.c
+++ b/drivers/i2c/busses/i2c-acorn.c
@@ -83,7 +83,6 @@ static struct i2c_algo_bit_data ioc_data = {
 };
 
 static struct i2c_adapter ioc_ops = {
-	.id			= I2C_HW_B_IOC,
 	.algo_data		= &ioc_data,
 };
 
diff --git a/drivers/i2c/busses/i2c-ali1535.c b/drivers/i2c/busses/i2c-ali1535.c
index 9cead9b9458e..981e080b32ae 100644
--- a/drivers/i2c/busses/i2c-ali1535.c
+++ b/drivers/i2c/busses/i2c-ali1535.c
@@ -476,7 +476,6 @@ static const struct i2c_algorithm smbus_algorithm = {
 
 static struct i2c_adapter ali1535_adapter = {
 	.owner		= THIS_MODULE,
-	.id		= I2C_HW_SMBUS_ALI1535,
 	.class          = I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
diff --git a/drivers/i2c/busses/i2c-ali1563.c b/drivers/i2c/busses/i2c-ali1563.c
index dd9e796fad69..f70f46582c6c 100644
--- a/drivers/i2c/busses/i2c-ali1563.c
+++ b/drivers/i2c/busses/i2c-ali1563.c
@@ -386,7 +386,6 @@ static const struct i2c_algorithm ali1563_algorithm = {
 
 static struct i2c_adapter ali1563_adapter = {
 	.owner	= THIS_MODULE,
-	.id	= I2C_HW_SMBUS_ALI1563,
 	.class	= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo	= &ali1563_algorithm,
 };
diff --git a/drivers/i2c/busses/i2c-ali15x3.c b/drivers/i2c/busses/i2c-ali15x3.c
index 234fdde7d40e..39066dee46e3 100644
--- a/drivers/i2c/busses/i2c-ali15x3.c
+++ b/drivers/i2c/busses/i2c-ali15x3.c
@@ -473,7 +473,6 @@ static const struct i2c_algorithm smbus_algorithm = {
 
 static struct i2c_adapter ali15x3_adapter = {
 	.owner		= THIS_MODULE,
-	.id		= I2C_HW_SMBUS_ALI15X3,
 	.class          = I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
diff --git a/drivers/i2c/busses/i2c-amd756.c b/drivers/i2c/busses/i2c-amd756.c
index 36bee5b9c952..220f4a1eee1d 100644
--- a/drivers/i2c/busses/i2c-amd756.c
+++ b/drivers/i2c/busses/i2c-amd756.c
@@ -298,7 +298,6 @@ static const struct i2c_algorithm smbus_algorithm = {
 
 struct i2c_adapter amd756_smbus = {
 	.owner		= THIS_MODULE,
-	.id		= I2C_HW_SMBUS_AMD756,
 	.class          = I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
diff --git a/drivers/i2c/busses/i2c-amd8111.c b/drivers/i2c/busses/i2c-amd8111.c
index 3972208876b3..edab51973bf5 100644
--- a/drivers/i2c/busses/i2c-amd8111.c
+++ b/drivers/i2c/busses/i2c-amd8111.c
@@ -387,7 +387,6 @@ static int __devinit amd8111_probe(struct pci_dev *dev,
 	smbus->adapter.owner = THIS_MODULE;
 	snprintf(smbus->adapter.name, sizeof(smbus->adapter.name),
 		"SMBus2 AMD8111 adapter at %04x", smbus->base);
-	smbus->adapter.id = I2C_HW_SMBUS_AMD8111;
 	smbus->adapter.class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	smbus->adapter.algo = &smbus_algorithm;
 	smbus->adapter.algo_data = smbus;
diff --git a/drivers/i2c/busses/i2c-au1550.c b/drivers/i2c/busses/i2c-au1550.c
index 66a04c2c660f..f78ce523e3db 100644
--- a/drivers/i2c/busses/i2c-au1550.c
+++ b/drivers/i2c/busses/i2c-au1550.c
@@ -400,7 +400,6 @@ i2c_au1550_probe(struct platform_device *pdev)
 	priv->xfer_timeout = 200;
 	priv->ack_timeout = 200;
 
-	priv->adap.id = I2C_HW_AU1550_PSC;
 	priv->adap.nr = pdev->id;
 	priv->adap.algo = &au1550_algo;
 	priv->adap.algo_data = priv;
diff --git a/drivers/i2c/busses/i2c-bfin-twi.c b/drivers/i2c/busses/i2c-bfin-twi.c
index 3fd2c417c1e0..fc548b3d002e 100644
--- a/drivers/i2c/busses/i2c-bfin-twi.c
+++ b/drivers/i2c/busses/i2c-bfin-twi.c
@@ -651,7 +651,6 @@ static int i2c_bfin_twi_probe(struct platform_device *pdev)
 	iface->timeout_timer.data = (unsigned long)iface;
 
 	p_adap = &iface->adap;
-	p_adap->id = I2C_HW_BLACKFIN;
 	p_adap->nr = pdev->id;
 	strlcpy(p_adap->name, pdev->name, sizeof(p_adap->name));
 	p_adap->algo = &bfin_twi_algorithm;
diff --git a/drivers/i2c/busses/i2c-elektor.c b/drivers/i2c/busses/i2c-elektor.c
index 0ed3ccb81b63..448b4bf35eb7 100644
--- a/drivers/i2c/busses/i2c-elektor.c
+++ b/drivers/i2c/busses/i2c-elektor.c
@@ -202,7 +202,6 @@ static struct i2c_algo_pcf_data pcf_isa_data = {
 static struct i2c_adapter pcf_isa_ops = {
 	.owner		= THIS_MODULE,
 	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
-	.id		= I2C_HW_P_ELEK,
 	.algo_data	= &pcf_isa_data,
 	.name		= "i2c-elektor",
 };
diff --git a/drivers/i2c/busses/i2c-hydra.c b/drivers/i2c/busses/i2c-hydra.c
index 648aa7baff83..bec9b845dd16 100644
--- a/drivers/i2c/busses/i2c-hydra.c
+++ b/drivers/i2c/busses/i2c-hydra.c
@@ -102,7 +102,6 @@ static struct i2c_algo_bit_data hydra_bit_data = {
 static struct i2c_adapter hydra_adap = {
 	.owner		= THIS_MODULE,
 	.name		= "Hydra i2c",
-	.id		= I2C_HW_B_HYDRA,
 	.algo_data	= &hydra_bit_data,
 };
 
diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c
index 526625eaa84b..230238df56c4 100644
--- a/drivers/i2c/busses/i2c-i801.c
+++ b/drivers/i2c/busses/i2c-i801.c
@@ -556,7 +556,6 @@ static const struct i2c_algorithm smbus_algorithm = {
 
 static struct i2c_adapter i801_adapter = {
 	.owner		= THIS_MODULE,
-	.id		= I2C_HW_SMBUS_I801,
 	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
diff --git a/drivers/i2c/busses/i2c-ibm_iic.c b/drivers/i2c/busses/i2c-ibm_iic.c
index 651f2f1ae5b7..88f0db73b364 100644
--- a/drivers/i2c/busses/i2c-ibm_iic.c
+++ b/drivers/i2c/busses/i2c-ibm_iic.c
@@ -746,7 +746,6 @@ static int __devinit iic_probe(struct of_device *ofdev,
 	adap->dev.parent = &ofdev->dev;
 	strlcpy(adap->name, "IBM IIC", sizeof(adap->name));
 	i2c_set_adapdata(adap, dev);
-	adap->id = I2C_HW_OCP;
 	adap->class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	adap->algo = &iic_algo;
 	adap->timeout = 1;
diff --git a/drivers/i2c/busses/i2c-iop3xx.c b/drivers/i2c/busses/i2c-iop3xx.c
index fc2714ac0c0f..3190690c26ce 100644
--- a/drivers/i2c/busses/i2c-iop3xx.c
+++ b/drivers/i2c/busses/i2c-iop3xx.c
@@ -480,7 +480,6 @@ iop3xx_i2c_probe(struct platform_device *pdev)
 	}
 
 	memcpy(new_adapter->name, pdev->name, strlen(pdev->name));
-	new_adapter->id = I2C_HW_IOP3XX;
 	new_adapter->owner = THIS_MODULE;
 	new_adapter->class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	new_adapter->dev.parent = &pdev->dev;
diff --git a/drivers/i2c/busses/i2c-ixp2000.c b/drivers/i2c/busses/i2c-ixp2000.c
index 05d72e981353..8e8467970481 100644
--- a/drivers/i2c/busses/i2c-ixp2000.c
+++ b/drivers/i2c/busses/i2c-ixp2000.c
@@ -116,7 +116,6 @@ static int ixp2000_i2c_probe(struct platform_device *plat_dev)
 	drv_data->algo_data.udelay = 6;
 	drv_data->algo_data.timeout = 100;
 
-	drv_data->adapter.id = I2C_HW_B_IXP2000,
 	strlcpy(drv_data->adapter.name, plat_dev->dev.driver->name,
 		sizeof(drv_data->adapter.name));
 	drv_data->adapter.algo_data = &drv_data->algo_data,
diff --git a/drivers/i2c/busses/i2c-mpc.c b/drivers/i2c/busses/i2c-mpc.c
index a9a45fcc8544..aedbbe6618db 100644
--- a/drivers/i2c/busses/i2c-mpc.c
+++ b/drivers/i2c/busses/i2c-mpc.c
@@ -310,7 +310,6 @@ static const struct i2c_algorithm mpc_algo = {
 static struct i2c_adapter mpc_ops = {
 	.owner = THIS_MODULE,
 	.name = "MPC adapter",
-	.id = I2C_HW_MPC107,
 	.algo = &mpc_algo,
 	.timeout = 1,
 };
diff --git a/drivers/i2c/busses/i2c-mv64xxx.c b/drivers/i2c/busses/i2c-mv64xxx.c
index 9e8118d2fe64..eeda276f8f16 100644
--- a/drivers/i2c/busses/i2c-mv64xxx.c
+++ b/drivers/i2c/busses/i2c-mv64xxx.c
@@ -527,7 +527,6 @@ mv64xxx_i2c_probe(struct platform_device *pd)
 		goto exit_unmap_regs;
 	}
 	drv_data->adapter.dev.parent = &pd->dev;
-	drv_data->adapter.id = I2C_HW_MV64XXX;
 	drv_data->adapter.algo = &mv64xxx_i2c_algo;
 	drv_data->adapter.owner = THIS_MODULE;
 	drv_data->adapter.class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
diff --git a/drivers/i2c/busses/i2c-nforce2.c b/drivers/i2c/busses/i2c-nforce2.c
index 3b19bc41a60b..05af6cd7f270 100644
--- a/drivers/i2c/busses/i2c-nforce2.c
+++ b/drivers/i2c/busses/i2c-nforce2.c
@@ -355,7 +355,6 @@ static int __devinit nforce2_probe_smb (struct pci_dev *dev, int bar,
 		return -EBUSY;
 	}
 	smbus->adapter.owner = THIS_MODULE;
-	smbus->adapter.id = I2C_HW_SMBUS_NFORCE2;
 	smbus->adapter.class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	smbus->adapter.algo = &smbus_algorithm;
 	smbus->adapter.algo_data = smbus;
diff --git a/drivers/i2c/busses/i2c-parport-light.c b/drivers/i2c/busses/i2c-parport-light.c
index b2b8380f6602..322c5691e38e 100644
--- a/drivers/i2c/busses/i2c-parport-light.c
+++ b/drivers/i2c/busses/i2c-parport-light.c
@@ -115,7 +115,6 @@ static struct i2c_algo_bit_data parport_algo_data = {
 static struct i2c_adapter parport_adapter = {
 	.owner		= THIS_MODULE,
 	.class		= I2C_CLASS_HWMON,
-	.id		= I2C_HW_B_LP,
 	.algo_data	= &parport_algo_data,
 	.name		= "Parallel port adapter (light)",
 };
diff --git a/drivers/i2c/busses/i2c-parport.c b/drivers/i2c/busses/i2c-parport.c
index a257cd5cd134..0d8998610c74 100644
--- a/drivers/i2c/busses/i2c-parport.c
+++ b/drivers/i2c/busses/i2c-parport.c
@@ -164,7 +164,6 @@ static void i2c_parport_attach (struct parport *port)
 	/* Fill the rest of the structure */
 	adapter->adapter.owner = THIS_MODULE;
 	adapter->adapter.class = I2C_CLASS_HWMON;
-	adapter->adapter.id = I2C_HW_B_LP;
 	strlcpy(adapter->adapter.name, "Parallel port adapter",
 		sizeof(adapter->adapter.name));
 	adapter->algo_data = parport_algo_data;
diff --git a/drivers/i2c/busses/i2c-pca-isa.c b/drivers/i2c/busses/i2c-pca-isa.c
index 9eb76268ec78..4aa8138cb0a9 100644
--- a/drivers/i2c/busses/i2c-pca-isa.c
+++ b/drivers/i2c/busses/i2c-pca-isa.c
@@ -101,7 +101,6 @@ static struct i2c_algo_pca_data pca_isa_data = {
 
 static struct i2c_adapter pca_isa_ops = {
 	.owner          = THIS_MODULE,
-	.id		= I2C_HW_A_ISA,
 	.algo_data	= &pca_isa_data,
 	.name		= "PCA9564 ISA Adapter",
 	.timeout	= 100,
diff --git a/drivers/i2c/busses/i2c-piix4.c b/drivers/i2c/busses/i2c-piix4.c
index eaa9b387543e..761f9dd53620 100644
--- a/drivers/i2c/busses/i2c-piix4.c
+++ b/drivers/i2c/busses/i2c-piix4.c
@@ -403,7 +403,6 @@ static const struct i2c_algorithm smbus_algorithm = {
 
 static struct i2c_adapter piix4_adapter = {
 	.owner		= THIS_MODULE,
-	.id		= I2C_HW_SMBUS_PIIX4,
 	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
diff --git a/drivers/i2c/busses/i2c-sibyte.c b/drivers/i2c/busses/i2c-sibyte.c
index 4ddefbf238e9..98b1ec489159 100644
--- a/drivers/i2c/busses/i2c-sibyte.c
+++ b/drivers/i2c/busses/i2c-sibyte.c
@@ -155,7 +155,6 @@ static struct i2c_algo_sibyte_data sibyte_board_data[2] = {
 static struct i2c_adapter sibyte_board_adapter[2] = {
 	{
 		.owner		= THIS_MODULE,
-		.id		= I2C_HW_SIBYTE,
 		.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 		.algo		= NULL,
 		.algo_data	= &sibyte_board_data[0],
@@ -164,7 +163,6 @@ static struct i2c_adapter sibyte_board_adapter[2] = {
 	},
 	{
 		.owner		= THIS_MODULE,
-		.id		= I2C_HW_SIBYTE,
 		.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 		.algo		= NULL,
 		.algo_data	= &sibyte_board_data[1],
diff --git a/drivers/i2c/busses/i2c-sis5595.c b/drivers/i2c/busses/i2c-sis5595.c
index 8ce2daff985c..f320ab27da46 100644
--- a/drivers/i2c/busses/i2c-sis5595.c
+++ b/drivers/i2c/busses/i2c-sis5595.c
@@ -365,7 +365,6 @@ static const struct i2c_algorithm smbus_algorithm = {
 
 static struct i2c_adapter sis5595_adapter = {
 	.owner		= THIS_MODULE,
-	.id		= I2C_HW_SMBUS_SIS5595,
 	.class          = I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
diff --git a/drivers/i2c/busses/i2c-sis630.c b/drivers/i2c/busses/i2c-sis630.c
index 9c9c016ff2b5..50c3610e6028 100644
--- a/drivers/i2c/busses/i2c-sis630.c
+++ b/drivers/i2c/busses/i2c-sis630.c
@@ -464,7 +464,6 @@ static const struct i2c_algorithm smbus_algorithm = {
 
 static struct i2c_adapter sis630_adapter = {
 	.owner		= THIS_MODULE,
-	.id		= I2C_HW_SMBUS_SIS630,
 	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
diff --git a/drivers/i2c/busses/i2c-sis96x.c b/drivers/i2c/busses/i2c-sis96x.c
index f1bba6396641..7e1594b40579 100644
--- a/drivers/i2c/busses/i2c-sis96x.c
+++ b/drivers/i2c/busses/i2c-sis96x.c
@@ -241,7 +241,6 @@ static const struct i2c_algorithm smbus_algorithm = {
 
 static struct i2c_adapter sis96x_adapter = {
 	.owner		= THIS_MODULE,
-	.id		= I2C_HW_SMBUS_SIS96X,
 	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
diff --git a/drivers/i2c/busses/i2c-via.c b/drivers/i2c/busses/i2c-via.c
index 29cef0433f34..8b24f192103a 100644
--- a/drivers/i2c/busses/i2c-via.c
+++ b/drivers/i2c/busses/i2c-via.c
@@ -83,7 +83,6 @@ static struct i2c_algo_bit_data bit_data = {
 
 static struct i2c_adapter vt586b_adapter = {
 	.owner		= THIS_MODULE,
-	.id		= I2C_HW_B_VIA,
 	.class          = I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.name		= "VIA i2c",
 	.algo_data	= &bit_data,
diff --git a/drivers/i2c/busses/i2c-viapro.c b/drivers/i2c/busses/i2c-viapro.c
index 9f194d9efd91..02e6f724b05f 100644
--- a/drivers/i2c/busses/i2c-viapro.c
+++ b/drivers/i2c/busses/i2c-viapro.c
@@ -321,7 +321,6 @@ static const struct i2c_algorithm smbus_algorithm = {
 
 static struct i2c_adapter vt596_adapter = {
 	.owner		= THIS_MODULE,
-	.id		= I2C_HW_SMBUS_VIA2,
 	.class		= I2C_CLASS_HWMON | I2C_CLASS_SPD,
 	.algo		= &smbus_algorithm,
 };
diff --git a/drivers/i2c/busses/i2c-voodoo3.c b/drivers/i2c/busses/i2c-voodoo3.c
index 1d4ae26ba73d..1a474acc0ddd 100644
--- a/drivers/i2c/busses/i2c-voodoo3.c
+++ b/drivers/i2c/busses/i2c-voodoo3.c
@@ -163,7 +163,6 @@ static struct i2c_algo_bit_data voo_i2c_bit_data = {
 
 static struct i2c_adapter voodoo3_i2c_adapter = {
 	.owner		= THIS_MODULE,
-	.id		= I2C_HW_B_VOO,
 	.class		= I2C_CLASS_TV_ANALOG, 
 	.name		= "I2C Voodoo3/Banshee adapter",
 	.algo_data	= &voo_i2c_bit_data,
@@ -180,7 +179,6 @@ static struct i2c_algo_bit_data voo_ddc_bit_data = {
 
 static struct i2c_adapter voodoo3_ddc_adapter = {
 	.owner		= THIS_MODULE,
-	.id		= I2C_HW_B_VOO,
 	.class		= I2C_CLASS_DDC, 
 	.name		= "DDC Voodoo3/Banshee adapter",
 	.algo_data	= &voo_ddc_bit_data,
diff --git a/drivers/i2c/busses/scx200_acb.c b/drivers/i2c/busses/scx200_acb.c
index ed794b145a11..648ecc6f60e6 100644
--- a/drivers/i2c/busses/scx200_acb.c
+++ b/drivers/i2c/busses/scx200_acb.c
@@ -440,7 +440,6 @@ static __init struct scx200_acb_iface *scx200_create_iface(const char *text,
 	i2c_set_adapdata(adapter, iface);
 	snprintf(adapter->name, sizeof(adapter->name), "%s ACB%d", text, index);
 	adapter->owner = THIS_MODULE;
-	adapter->id = I2C_HW_SMBUS_SCX200;
 	adapter->algo = &scx200_acb_algorithm;
 	adapter->class = I2C_CLASS_HWMON | I2C_CLASS_SPD;
 	adapter->dev.parent = dev;
diff --git a/drivers/i2c/busses/scx200_i2c.c b/drivers/i2c/busses/scx200_i2c.c
index e4c98539c517..162b74a04886 100644
--- a/drivers/i2c/busses/scx200_i2c.c
+++ b/drivers/i2c/busses/scx200_i2c.c
@@ -82,7 +82,6 @@ static struct i2c_algo_bit_data scx200_i2c_data = {
 static struct i2c_adapter scx200_i2c_ops = {
 	.owner		   = THIS_MODULE,
 	.class             = I2C_CLASS_HWMON | I2C_CLASS_SPD,
-	.id		   = I2C_HW_B_SCX200,
 	.algo_data	   = &scx200_i2c_data,
 	.name	= "NatSemi SCx200 I2C",
 };
diff --git a/drivers/ieee1394/pcilynx.c b/drivers/ieee1394/pcilynx.c
index dc15cadb06ef..38f712036201 100644
--- a/drivers/ieee1394/pcilynx.c
+++ b/drivers/ieee1394/pcilynx.c
@@ -1419,7 +1419,6 @@ static int __devinit add_card(struct pci_dev *dev,
 		i2c_ad = kzalloc(sizeof(*i2c_ad), GFP_KERNEL);
         	if (!i2c_ad) FAIL("failed to allocate I2C adapter memory");
 
-		i2c_ad->id = I2C_HW_B_PCILYNX;
 		strlcpy(i2c_ad->name, "PCILynx I2C", sizeof(i2c_ad->name));
                 i2c_adapter_data = bit_data;
                 i2c_ad->algo_data = &i2c_adapter_data;
diff --git a/drivers/video/aty/radeon_i2c.c b/drivers/video/aty/radeon_i2c.c
index 2c5567175dca..359fc64e761a 100644
--- a/drivers/video/aty/radeon_i2c.c
+++ b/drivers/video/aty/radeon_i2c.c
@@ -72,7 +72,6 @@ static int radeon_setup_i2c_bus(struct radeon_i2c_chan *chan, const char *name)
 	snprintf(chan->adapter.name, sizeof(chan->adapter.name),
 		 "radeonfb %s", name);
 	chan->adapter.owner		= THIS_MODULE;
-	chan->adapter.id		= I2C_HW_B_RADEON;
 	chan->adapter.algo_data		= &chan->algo;
 	chan->adapter.dev.parent	= &chan->rinfo->pdev->dev;
 	chan->algo.setsda		= radeon_gpio_setsda;
diff --git a/drivers/video/i810/i810-i2c.c b/drivers/video/i810/i810-i2c.c
index 7787c3322ffb..9dd55e5324a1 100644
--- a/drivers/video/i810/i810-i2c.c
+++ b/drivers/video/i810/i810-i2c.c
@@ -90,7 +90,6 @@ static int i810_setup_i2c_bus(struct i810fb_i2c_chan *chan, const char *name)
         chan->adapter.owner             = THIS_MODULE;
         chan->adapter.algo_data         = &chan->algo;
         chan->adapter.dev.parent        = &chan->par->dev->dev;
-	chan->adapter.id                = I2C_HW_B_I810;
 	chan->algo.setsda               = i810i2c_setsda;
 	chan->algo.setscl               = i810i2c_setscl;
 	chan->algo.getsda               = i810i2c_getsda;
diff --git a/drivers/video/intelfb/intelfb_i2c.c b/drivers/video/intelfb/intelfb_i2c.c
index 5d896b81f4e0..b3065492bb20 100644
--- a/drivers/video/intelfb/intelfb_i2c.c
+++ b/drivers/video/intelfb/intelfb_i2c.c
@@ -111,7 +111,6 @@ static int intelfb_setup_i2c_bus(struct intelfb_info *dinfo,
 		 "intelfb %s", name);
 	chan->adapter.class		= class;
 	chan->adapter.owner		= THIS_MODULE;
-	chan->adapter.id		= I2C_HW_B_INTELFB;
 	chan->adapter.algo_data		= &chan->algo;
 	chan->adapter.dev.parent	= &chan->dinfo->pdev->dev;
 	chan->algo.setsda		= intelfb_gpio_setsda;
diff --git a/drivers/video/nvidia/nv_i2c.c b/drivers/video/nvidia/nv_i2c.c
index 6fd7cb8f9b8e..6aaddb4f6788 100644
--- a/drivers/video/nvidia/nv_i2c.c
+++ b/drivers/video/nvidia/nv_i2c.c
@@ -87,7 +87,6 @@ static int nvidia_setup_i2c_bus(struct nvidia_i2c_chan *chan, const char *name,
 
 	strcpy(chan->adapter.name, name);
 	chan->adapter.owner = THIS_MODULE;
-	chan->adapter.id = I2C_HW_B_NVIDIA;
 	chan->adapter.class = i2c_class;
 	chan->adapter.algo_data = &chan->algo;
 	chan->adapter.dev.parent = &chan->par->pci_dev->dev;
diff --git a/drivers/video/savage/savagefb-i2c.c b/drivers/video/savage/savagefb-i2c.c
index 783d4adffb93..574b29e9f8f2 100644
--- a/drivers/video/savage/savagefb-i2c.c
+++ b/drivers/video/savage/savagefb-i2c.c
@@ -137,7 +137,6 @@ static int savage_setup_i2c_bus(struct savagefb_i2c_chan *chan,
 	if (chan->par) {
 		strcpy(chan->adapter.name, name);
 		chan->adapter.owner		= THIS_MODULE;
-		chan->adapter.id		= I2C_HW_B_SAVAGE;
 		chan->adapter.algo_data		= &chan->algo;
 		chan->adapter.dev.parent	= &chan->par->pcidev->dev;
 		chan->algo.udelay		= 10;
diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h
index 82b7eee22636..1ffc23bc5d1e 100644
--- a/include/linux/i2c-id.h
+++ b/include/linux/i2c-id.h
@@ -79,74 +79,23 @@
  */
 
 /* --- Bit algorithm adapters						*/
-#define I2C_HW_B_LP		0x010000 /* Parallel port Philips style */
 #define I2C_HW_B_BT848		0x010005 /* BT848 video boards */
-#define I2C_HW_B_VIA		0x010007 /* Via vt82c586b */
-#define I2C_HW_B_HYDRA		0x010008 /* Apple Hydra Mac I/O */
-#define I2C_HW_B_I810		0x01000a /* Intel I810 */
-#define I2C_HW_B_VOO		0x01000b /* 3dfx Voodoo 3 / Banshee */
-#define I2C_HW_B_SCX200		0x01000e /* Nat'l Semi SCx200 I2C */
 #define I2C_HW_B_RIVA		0x010010 /* Riva based graphics cards */
-#define I2C_HW_B_IOC		0x010011 /* IOC bit-wiggling */
-#define I2C_HW_B_IXP2000	0x010016 /* GPIO on IXP2000 systems */
 #define I2C_HW_B_ZR36067	0x010019 /* Zoran-36057/36067 based boards */
-#define I2C_HW_B_PCILYNX	0x01001a /* TI PCILynx I2C adapter */
 #define I2C_HW_B_CX2388x	0x01001b /* connexant 2388x based tv cards */
-#define I2C_HW_B_NVIDIA		0x01001c /* nvidia framebuffer driver */
-#define I2C_HW_B_SAVAGE		0x01001d /* savage framebuffer driver */
-#define I2C_HW_B_RADEON		0x01001e /* radeon framebuffer driver */
 #define I2C_HW_B_EM28XX		0x01001f /* em28xx video capture cards */
 #define I2C_HW_B_CX2341X	0x010020 /* Conexant CX2341X MPEG encoder cards */
-#define I2C_HW_B_INTELFB	0x010021 /* intel framebuffer driver */
 #define I2C_HW_B_CX23885	0x010022 /* conexant 23885 based tv cards (bus1) */
 #define I2C_HW_B_AU0828		0x010023 /* auvitek au0828 usb bridge */
 
-/* --- PCF 8584 based algorithms					*/
-#define I2C_HW_P_ELEK		0x020002 /* Elektor ISA Bus inteface card */
-
-/* --- PCA 9564 based algorithms */
-#define I2C_HW_A_ISA		0x1a0000 /* generic ISA Bus interface card */
-
-/* --- PowerPC on-chip adapters						*/
-#define I2C_HW_OCP		0x120000 /* IBM on-chip I2C adapter */
-
-/* --- Broadcom SiByte adapters						*/
-#define I2C_HW_SIBYTE		0x150000
-
 /* --- SGI adapters							*/
 #define I2C_HW_SGI_VINO		0x160000
 
-/* --- XSCALE on-chip adapters                          */
-#define I2C_HW_IOP3XX		0x140000
-
-/* --- Au1550 PSC adapters adapters					*/
-#define I2C_HW_AU1550_PSC	0x1b0000
-
 /* --- SMBus only adapters						*/
-#define I2C_HW_SMBUS_PIIX4	0x040000
-#define I2C_HW_SMBUS_ALI15X3	0x040001
-#define I2C_HW_SMBUS_VIA2	0x040002
-#define I2C_HW_SMBUS_I801	0x040004
-#define I2C_HW_SMBUS_AMD756	0x040005
-#define I2C_HW_SMBUS_SIS5595	0x040006
-#define I2C_HW_SMBUS_ALI1535	0x040007
-#define I2C_HW_SMBUS_SIS630	0x040008
-#define I2C_HW_SMBUS_SIS96X	0x040009
-#define I2C_HW_SMBUS_AMD8111	0x04000a
-#define I2C_HW_SMBUS_SCX200	0x04000b
-#define I2C_HW_SMBUS_NFORCE2	0x04000c
 #define I2C_HW_SMBUS_W9968CF	0x04000d
 #define I2C_HW_SMBUS_OV511	0x04000e /* OV511(+) USB 1.1 webcam ICs */
 #define I2C_HW_SMBUS_OV518	0x04000f /* OV518(+) USB 1.1 webcam ICs */
 #define I2C_HW_SMBUS_CAFE	0x040012 /* Marvell 88ALP01 "CAFE" cam  */
-#define I2C_HW_SMBUS_ALI1563	0x040013
-
-/* --- MCP107 adapter */
-#define I2C_HW_MPC107		0x0d0000
-
-/* --- Embedded adapters */
-#define I2C_HW_MV64XXX		0x190000
-#define I2C_HW_BLACKFIN		0x190001 /* ADI Blackfin I2C TWI driver */
 
 /* --- Miscellaneous adapters */
 #define I2C_HW_SAA7146		0x060000 /* SAA7146 video decoder bus */
-- 
cgit v1.2.3


From 5fb4523afbffae5a5cec4989ee4c9fbc3dbdef33 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Mon, 26 Jan 2009 21:19:57 +0100
Subject: i2c: Warn on deprecated binding model use

Let the kernel developers know that i2c_attach_client() and
i2c_detach_client() are deprecated and should no longer be used.
Drivers using these should be converted to the standard device
driver binding model (probe and remove methods.)

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Ben Dooks <ben-linux@fluff.org>
---
 include/linux/i2c.h | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 20873d402467..fcfbfea3af72 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -151,7 +151,7 @@ struct i2c_driver {
 	 * has been dynamically allocated by the driver in the function above,
 	 * it must be freed here.  (LEGACY I2C DRIVERS ONLY)
 	 */
-	int (*detach_client)(struct i2c_client *);
+	int (*detach_client)(struct i2c_client *) __deprecated;
 
 	/* Standard driver model interfaces, for "new style" i2c drivers.
 	 * With the driver model, device enumeration is NEVER done by drivers;
@@ -429,8 +429,10 @@ static inline int i2c_add_driver(struct i2c_driver *driver)
 	return i2c_register_driver(THIS_MODULE, driver);
 }
 
-extern int i2c_attach_client(struct i2c_client *);
-extern int i2c_detach_client(struct i2c_client *);
+/* These are deprecated, your driver should use the standard .probe()
+ * and .remove() methods instead. */
+extern int __deprecated i2c_attach_client(struct i2c_client *);
+extern int __deprecated i2c_detach_client(struct i2c_client *);
 
 extern struct i2c_client *i2c_use_client(struct i2c_client *client);
 extern void i2c_release_client(struct i2c_client *client);
-- 
cgit v1.2.3


From 3121a48d87a580f369eeb26aa0a075142274a353 Mon Sep 17 00:00:00 2001
From: Krzysztof Hałasa <khc@pm.waw.pl>
Date: Mon, 26 Jan 2009 12:30:12 -0800
Subject: net: Fix linux/if_frad.h's suitability for userspace.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The userspace interfaces are protected by CONFIG_* ifdefs
and that of course can't work.

Reported by Jaswinder Singh Rajput.

Signed-off-by: Krzysztof Hałasa <khc@pm.waw.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/if_frad.h | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/if_frad.h b/include/linux/if_frad.h
index 5c34240de746..60e16a551dd6 100644
--- a/include/linux/if_frad.h
+++ b/include/linux/if_frad.h
@@ -26,8 +26,6 @@
 
 #include <linux/if.h>
 
-#if defined(CONFIG_DLCI) || defined(CONFIG_DLCI_MODULE)
-
 /* Structures and constants associated with the DLCI device driver */
 
 struct dlci_add
@@ -127,6 +125,8 @@ struct frad_conf
 
 #ifdef __KERNEL__
 
+#if defined(CONFIG_DLCI) || defined(CONFIG_DLCI_MODULE)
+
 /* these are the fields of an RFC 1490 header */
 struct frhdr
 {
@@ -190,12 +190,10 @@ struct frad_local
    int               buffer;		/* current buffer for S508 firmware */
 };
 
-#endif /* __KERNEL__ */
-
 #endif /* CONFIG_DLCI || CONFIG_DLCI_MODULE */
 
-#ifdef __KERNEL__
 extern void dlci_ioctl_set(int (*hook)(unsigned int, void __user *));
-#endif
+
+#endif /* __KERNEL__ */
 
 #endif
-- 
cgit v1.2.3


From 5ee810072175042775e39bdd3eaaa68884c27805 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Mon, 26 Jan 2009 19:21:26 -0800
Subject: Fix "multiple definition of `debugfs_create_size_t'"

Introduced by 8adb711f3668b034e7b956fac951ed08b53e0d55 ("debugfs:
introduce stub for debugfs_create_size_t() when DEBUG_FS=n") and due to
a simple missing "static inline".

Reported-and-tested-by: Jeff Chua <jeff.chua.linux@gmail.com>
Acked-by: Greg KH <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/debugfs.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/debugfs.h b/include/linux/debugfs.h
index 0f5c33b0bd3e..af0e01d4c663 100644
--- a/include/linux/debugfs.h
+++ b/include/linux/debugfs.h
@@ -162,7 +162,7 @@ static inline struct dentry *debugfs_create_x32(const char *name, mode_t mode,
 	return ERR_PTR(-ENODEV);
 }
 
-struct dentry *debugfs_create_size_t(const char *name, mode_t mode,
+static inline struct dentry *debugfs_create_size_t(const char *name, mode_t mode,
 				     struct dentry *parent,
 				     size_t *value)
 {
-- 
cgit v1.2.3


From abfe2d7b915c872f3a1fd203267cedebf90daa45 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 19 Jan 2009 20:54:54 +0100
Subject: Hibernation: Introduce system_entering_hibernation

Introduce boolean function system_entering_hibernation() returning
'true' during the last phase of hibernation, in which devices are
being put into low power states and the sleep state (for example,
ACPI S4) is finally entered.

Some device drivers need such a function to check if the system is
in the final phase of hibernation.  In particular, some SATA drivers
are going to use it for blacklisting systems in which the disks
should not be spun down during the last phase of hibernation (the
BIOS will do that anyway).

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 include/linux/suspend.h |  2 ++
 kernel/power/disk.c     | 10 ++++++++++
 2 files changed, 12 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index 2b409c44db83..c7d9bb1832ba 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -237,6 +237,7 @@ extern int hibernate_nvs_alloc(void);
 extern void hibernate_nvs_free(void);
 extern void hibernate_nvs_save(void);
 extern void hibernate_nvs_restore(void);
+extern bool system_entering_hibernation(void);
 #else /* CONFIG_HIBERNATION */
 static inline int swsusp_page_is_forbidden(struct page *p) { return 0; }
 static inline void swsusp_set_page_free(struct page *p) {}
@@ -252,6 +253,7 @@ static inline int hibernate_nvs_alloc(void) { return 0; }
 static inline void hibernate_nvs_free(void) {}
 static inline void hibernate_nvs_save(void) {}
 static inline void hibernate_nvs_restore(void) {}
+static inline bool system_entering_hibernation(void) { return false; }
 #endif /* CONFIG_HIBERNATION */
 
 #ifdef CONFIG_PM_SLEEP
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index 45e8541ab7e3..432ee575c9ee 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -71,6 +71,14 @@ void hibernation_set_ops(struct platform_hibernation_ops *ops)
 	mutex_unlock(&pm_mutex);
 }
 
+static bool entering_platform_hibernation;
+
+bool system_entering_hibernation(void)
+{
+	return entering_platform_hibernation;
+}
+EXPORT_SYMBOL(system_entering_hibernation);
+
 #ifdef CONFIG_PM_DEBUG
 static void hibernation_debug_sleep(void)
 {
@@ -411,6 +419,7 @@ int hibernation_platform_enter(void)
 	if (error)
 		goto Close;
 
+	entering_platform_hibernation = true;
 	suspend_console();
 	error = device_suspend(PMSG_HIBERNATE);
 	if (error) {
@@ -445,6 +454,7 @@ int hibernation_platform_enter(void)
  Finish:
 	hibernation_ops->finish();
  Resume_devices:
+	entering_platform_hibernation = false;
 	device_resume(PMSG_RESTORE);
 	resume_console();
  Close:
-- 
cgit v1.2.3


From d7b1956fed33d30c4815e848fd7a143722916868 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 19 Jan 2009 20:55:50 +0100
Subject: DMI: Introduce dmi_first_match to make the interface more flexible

Some notebooks from HP have the problem that their BIOSes attempt to
spin down hard drives before entering ACPI system states S4 and S5.
This leads to a yo-yo effect during system power-off shutdown and the
last phase of hibernation when the disk is first spun down by the
kernel and then almost immediately turned on and off by the BIOS.
This, in turn, may result in shortening the disk's life times.

To prevent this from happening we can blacklist the affected systems
using DMI information.  However, only the on-board controlles should
be blacklisted and their PCI slot numbers can be used for this
purpose.  Unfortunately the existing interface for checking DMI
information of the system is not very convenient for this purpose,
because to use it, we would have to define special callback functions
or create a separate struct dmi_system_id table for each blacklisted
system.

To overcome this difficulty introduce a new function
dmi_first_match() returning a pointer to the first entry in an array
of struct dmi_system_id elements that matches the system DMI
information.  Then, we can use this pointer to access the entry's
.driver_data field containing the additional information, such as
the PCI slot number, allowing us to do the desired blacklisting.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/firmware/dmi_scan.c | 74 +++++++++++++++++++++++++++++++++------------
 include/linux/dmi.h         |  1 +
 2 files changed, 56 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c
index d76adfea5df7..8f0f7c449305 100644
--- a/drivers/firmware/dmi_scan.c
+++ b/drivers/firmware/dmi_scan.c
@@ -414,6 +414,29 @@ void __init dmi_scan_machine(void)
 	dmi_initialized = 1;
 }
 
+/**
+ *	dmi_matches - check if dmi_system_id structure matches system DMI data
+ *	@dmi: pointer to the dmi_system_id structure to check
+ */
+static bool dmi_matches(const struct dmi_system_id *dmi)
+{
+	int i;
+
+	WARN(!dmi_initialized, KERN_ERR "dmi check: not initialized yet.\n");
+
+	for (i = 0; i < ARRAY_SIZE(dmi->matches); i++) {
+		int s = dmi->matches[i].slot;
+		if (s == DMI_NONE)
+			continue;
+		if (dmi_ident[s]
+		    && strstr(dmi_ident[s], dmi->matches[i].substr))
+			continue;
+		/* No match */
+		return false;
+	}
+	return true;
+}
+
 /**
  *	dmi_check_system - check system DMI data
  *	@list: array of dmi_system_id structures to match against
@@ -429,31 +452,44 @@ void __init dmi_scan_machine(void)
  */
 int dmi_check_system(const struct dmi_system_id *list)
 {
-	int i, count = 0;
-	const struct dmi_system_id *d = list;
-
-	WARN(!dmi_initialized, KERN_ERR "dmi check: not initialized yet.\n");
-
-	while (d->ident) {
-		for (i = 0; i < ARRAY_SIZE(d->matches); i++) {
-			int s = d->matches[i].slot;
-			if (s == DMI_NONE)
-				continue;
-			if (dmi_ident[s] && strstr(dmi_ident[s], d->matches[i].substr))
-				continue;
-			/* No match */
-			goto fail;
+	int count = 0;
+	const struct dmi_system_id *d;
+
+	for (d = list; d->ident; d++)
+		if (dmi_matches(d)) {
+			count++;
+			if (d->callback && d->callback(d))
+				break;
 		}
-		count++;
-		if (d->callback && d->callback(d))
-			break;
-fail:		d++;
-	}
 
 	return count;
 }
 EXPORT_SYMBOL(dmi_check_system);
 
+/**
+ *	dmi_first_match - find dmi_system_id structure matching system DMI data
+ *	@list: array of dmi_system_id structures to match against
+ *		All non-null elements of the list must match
+ *		their slot's (field index's) data (i.e., each
+ *		list string must be a substring of the specified
+ *		DMI slot's string data) to be considered a
+ *		successful match.
+ *
+ *	Walk the blacklist table until the first match is found.  Return the
+ *	pointer to the matching entry or NULL if there's no match.
+ */
+const struct dmi_system_id *dmi_first_match(const struct dmi_system_id *list)
+{
+	const struct dmi_system_id *d;
+
+	for (d = list; d->ident; d++)
+		if (dmi_matches(d))
+			return d;
+
+	return NULL;
+}
+EXPORT_SYMBOL(dmi_first_match);
+
 /**
  *	dmi_get_system_info - return DMI data value
  *	@field: data index (see enum dmi_field)
diff --git a/include/linux/dmi.h b/include/linux/dmi.h
index 34161907b2f8..aea23105d3ed 100644
--- a/include/linux/dmi.h
+++ b/include/linux/dmi.h
@@ -38,6 +38,7 @@ struct dmi_device {
 #ifdef CONFIG_DMI
 
 extern int dmi_check_system(const struct dmi_system_id *list);
+const struct dmi_system_id *dmi_first_match(const struct dmi_system_id *list);
 extern const char * dmi_get_system_info(int field);
 extern const struct dmi_device * dmi_find_device(int type, const char *name,
 	const struct dmi_device *from);
-- 
cgit v1.2.3


From 2a6e58d2731dcc05dafa7f976d935e0f0627fcd7 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 19 Jan 2009 20:56:43 +0100
Subject: SATA: Blacklisting of systems that spin off disks during ACPI power
 off

Introduce new libata flags ATA_FLAG_NO_POWEROFF_SPINDOWN and
ATA_FLAG_NO_HIBERNATE_SPINDOWN that, if set, will prevent disks from
being spun off during system power off and hibernation, respectively
(to handle the hibernation case we need the new system state
SYSTEM_HIBERNATE_ENTER that can be checked against by libata, in
analogy with SYSTEM_POWER_OFF).

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-scsi.c | 20 +++++++++++++++++---
 include/linux/libata.h    |  2 ++
 2 files changed, 19 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index a1a6e6298c33..3c4c5ae277ba 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -46,6 +46,7 @@
 #include <linux/libata.h>
 #include <linux/hdreg.h>
 #include <linux/uaccess.h>
+#include <linux/suspend.h>
 
 #include "libata.h"
 
@@ -1303,6 +1304,17 @@ static unsigned int ata_scsi_start_stop_xlat(struct ata_queued_cmd *qc)
 
 		tf->command = ATA_CMD_VERIFY;	/* READ VERIFY */
 	} else {
+		/* Some odd clown BIOSen issue spindown on power off (ACPI S4
+		 * or S5) causing some drives to spin up and down again.
+		 */
+		if ((qc->ap->flags & ATA_FLAG_NO_POWEROFF_SPINDOWN) &&
+		    system_state == SYSTEM_POWER_OFF)
+			goto skip;
+
+		if ((qc->ap->flags & ATA_FLAG_NO_HIBERNATE_SPINDOWN) &&
+		     system_entering_hibernation())
+			goto skip;
+
 		/* XXX: This is for backward compatibility, will be
 		 * removed.  Read Documentation/feature-removal-schedule.txt
 		 * for more info.
@@ -1326,8 +1338,7 @@ static unsigned int ata_scsi_start_stop_xlat(struct ata_queued_cmd *qc)
 				scmd->scsi_done = qc->scsidone;
 				qc->scsidone = ata_delayed_done;
 			}
-			scmd->result = SAM_STAT_GOOD;
-			return 1;
+			goto skip;
 		}
 
 		/* Issue ATA STANDBY IMMEDIATE command */
@@ -1343,10 +1354,13 @@ static unsigned int ata_scsi_start_stop_xlat(struct ata_queued_cmd *qc)
 
 	return 0;
 
-invalid_fld:
+ invalid_fld:
 	ata_scsi_set_sense(scmd, ILLEGAL_REQUEST, 0x24, 0x0);
 	/* "Invalid field in cbd" */
 	return 1;
+ skip:
+	scmd->result = SAM_STAT_GOOD;
+	return 1;
 }
 
 
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 2c6bd66209ff..bca3ba25f52a 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -187,6 +187,8 @@ enum {
 	ATA_FLAG_PIO_POLLING	= (1 << 9), /* use polling PIO if LLD
 					     * doesn't handle PIO interrupts */
 	ATA_FLAG_NCQ		= (1 << 10), /* host supports NCQ */
+	ATA_FLAG_NO_POWEROFF_SPINDOWN = (1 << 11), /* don't spindown before poweroff */
+	ATA_FLAG_NO_HIBERNATE_SPINDOWN = (1 << 12), /* don't spindown before hibernation */
 	ATA_FLAG_DEBUGMSG	= (1 << 13),
 	ATA_FLAG_IGN_SIMPLEX	= (1 << 15), /* ignore SIMPLEX */
 	ATA_FLAG_NO_IORDY	= (1 << 16), /* controller lacks iordy */
-- 
cgit v1.2.3


From 57064d213d2e44654d4f13c66df135b5e7389a26 Mon Sep 17 00:00:00 2001
From: Seth Heasley <seth.heasley@intel.com>
Date: Fri, 23 Jan 2009 12:43:38 -0800
Subject: PCI: irq and pci_ids patch for Intel Tigerpoint DeviceIDs

This patch adds the Intel Tigerpoint LPC Controller DeviceIDs.

Signed-off-by: Seth Heasley <seth.heasley@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 arch/x86/pci/irq.c      | 1 +
 include/linux/pci_ids.h | 1 +
 2 files changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/arch/x86/pci/irq.c b/arch/x86/pci/irq.c
index 4064345cf144..fecbce6e7d7c 100644
--- a/arch/x86/pci/irq.c
+++ b/arch/x86/pci/irq.c
@@ -572,6 +572,7 @@ static __init int intel_router_probe(struct irq_router *r, struct pci_dev *route
 	case PCI_DEVICE_ID_INTEL_ICH7_1:
 	case PCI_DEVICE_ID_INTEL_ICH7_30:
 	case PCI_DEVICE_ID_INTEL_ICH7_31:
+	case PCI_DEVICE_ID_INTEL_TGP_LPC:
 	case PCI_DEVICE_ID_INTEL_ESB2_0:
 	case PCI_DEVICE_ID_INTEL_ICH8_0:
 	case PCI_DEVICE_ID_INTEL_ICH8_1:
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index d56ad9c21c09..6f260b50253b 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2421,6 +2421,7 @@
 #define PCI_DEVICE_ID_INTEL_ICH7_0	0x27b8
 #define PCI_DEVICE_ID_INTEL_ICH7_1	0x27b9
 #define PCI_DEVICE_ID_INTEL_ICH7_30	0x27b0
+#define PCI_DEVICE_ID_INTEL_TGP_LPC	0x27bc
 #define PCI_DEVICE_ID_INTEL_ICH7_31	0x27bd
 #define PCI_DEVICE_ID_INTEL_ICH7_17	0x27da
 #define PCI_DEVICE_ID_INTEL_ICH7_19	0x27dd
-- 
cgit v1.2.3


From 1cf3eb2ff6b0844c678f2f48d0053b9d12b7da67 Mon Sep 17 00:00:00 2001
From: Jeff Mahoney <jeffm@suse.com>
Date: Tue, 27 Jan 2009 23:48:59 +0200
Subject: kmalloc: return NULL instead of link failure

The SLAB kmalloc with a constant value isn't consistent with the other
implementations because it bails out with __you_cannot_kmalloc_that_much
rather than returning NULL and properly allowing the caller to fall back
to vmalloc or take other action.  This doesn't happen with a non-constant
value or with SLOB or SLUB.

Starting with 2.6.28, I've been seeing build failures on s390x.  This is
due to init_section_page_cgroup trying to allocate 2.5MB when the max size
for a kmalloc on s390x is 2MB.

It's failing because the value is constant.  The workarounds at the call
size are ugly and the caller shouldn't have to change behavior depending
on what the backend of the API is.

So, this patch eliminates the link failure and returns NULL like the other
implementations.

Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Matt Mackall <mpm@selenic.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: <stable@kernel.org>		[2.6.28.x]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
---
 include/linux/slab_def.h | 10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/slab_def.h b/include/linux/slab_def.h
index 39c3a5eb8ebe..6ca6a7b66d75 100644
--- a/include/linux/slab_def.h
+++ b/include/linux/slab_def.h
@@ -43,10 +43,7 @@ static inline void *kmalloc(size_t size, gfp_t flags)
 			i++;
 #include <linux/kmalloc_sizes.h>
 #undef CACHE
-		{
-			extern void __you_cannot_kmalloc_that_much(void);
-			__you_cannot_kmalloc_that_much();
-		}
+		return NULL;
 found:
 #ifdef CONFIG_ZONE_DMA
 		if (flags & GFP_DMA)
@@ -77,10 +74,7 @@ static inline void *kmalloc_node(size_t size, gfp_t flags, int node)
 			i++;
 #include <linux/kmalloc_sizes.h>
 #undef CACHE
-		{
-			extern void __you_cannot_kmalloc_that_much(void);
-			__you_cannot_kmalloc_that_much();
-		}
+		return NULL;
 found:
 #ifdef CONFIG_ZONE_DMA
 		if (flags & GFP_DMA)
-- 
cgit v1.2.3


From 40413dcb7b273bda681dca38e6ff0bbb3728ef11 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Thu, 22 Jan 2009 01:58:36 +0300
Subject: Fix longstanding "error: storage size of '__mod_dmi_device_table'
 isn't known"

gcc 3.4.6 doesn't like MODULE_DEVICE_TABLE(dmi, x) expansion enough to
error out.  Shut it up in a most simple way.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/mod_devicetable.h | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h
index 97b91d1abb43..fde86671f48f 100644
--- a/include/linux/mod_devicetable.h
+++ b/include/linux/mod_devicetable.h
@@ -443,6 +443,13 @@ struct dmi_system_id {
 	struct dmi_strmatch matches[4];
 	void *driver_data;
 };
+/*
+ * struct dmi_device_id appears during expansion of
+ * "MODULE_DEVICE_TABLE(dmi, x)". Compiler doesn't look inside it
+ * but this is enough for gcc 3.4.6 to error out:
+ *	error: storage size of '__mod_dmi_device_table' isn't known
+ */
+#define dmi_device_id dmi_system_id
 #endif
 
 #define DMI_MATCH(a, b)	{ a, b }
-- 
cgit v1.2.3


From 11e76ae0f3a82bbb6c06df8af2167af8b96a0584 Mon Sep 17 00:00:00 2001
From: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Date: Thu, 8 Jan 2009 12:52:19 -0800
Subject: USB: add kernel-doc for wusb_dev in struct usb_device

Reported by Randy Dunlap from a warning on the v2.6.29 merge window.

Signed-off-by: Inaky Perez-Gonzalez <inaky@linux.intel.com>
Cc: David Vrabel <david.vrabel@csr.com>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/usb.h b/include/linux/usb.h
index 85ee9be9361e..88079fd60235 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -418,6 +418,8 @@ struct usb_tt;
  * @autosuspend_disabled: autosuspend disabled by the user
  * @autoresume_disabled: autoresume disabled by the user
  * @skip_sys_resume: skip the next system resume
+ * @wusb_dev: if this is a Wireless USB device, link to the WUSB
+ *	specific data for the device.
  *
  * Notes:
  * Usbcore drivers should not set usbdev->state directly.  Instead use
-- 
cgit v1.2.3


From d8204ee2ad1c9babd7e33d4c118ec99a78a8442e Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@kernel.crashing.org>
Date: Wed, 28 Jan 2009 00:07:20 -0600
Subject: dmi: Fix build breakage

Commit d7b1956fed33d30c4815e848fd7a143722916868 ("DMI: Introduce
dmi_first_match to make the interface more flexible") introduced compile
errors like the following when !CONFIG_DMI

    drivers/ata/sata_sil.c: In function 'sil_broken_system_poweroff':
    drivers/ata/sata_sil.c:713: error: implicit declaration of function 'dmi_first_match'
    drivers/ata/sata_sil.c:713: warning: initialization makes pointer from integer without a cast

We just need a dummy version of dmi_first_match() to fix this all up.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/dmi.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/dmi.h b/include/linux/dmi.h
index aea23105d3ed..d741b9ceb0e0 100644
--- a/include/linux/dmi.h
+++ b/include/linux/dmi.h
@@ -65,6 +65,8 @@ static inline int dmi_walk(void (*decode)(const struct dmi_header *))
 	{ return -1; }
 static inline bool dmi_match(enum dmi_field f, const char *str)
 	{ return false; }
+static inline const struct dmi_system_id *
+	dmi_first_match(const struct dmi_system_id *list) { return NULL; }
 
 #endif
 
-- 
cgit v1.2.3


From dc19835df6c47ff676ad6c98722d5e529db5d74c Mon Sep 17 00:00:00 2001
From: Masami Hiramatsu <mhiramat@redhat.com>
Date: Thu, 29 Jan 2009 14:25:08 -0800
Subject: kprobes: fix module compilation error with CONFIG_KPROBES=n

Define kprobes related data structures even if CONFIG_KPROBES is not set.
This fixes compilation errors which occur if CONFIG_KPROBES is not set, in
kprobe using modules.

[akpm@linux-foundation.org: fix build for non-kprobes-supporting architectures]
Reviewed-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/kprobes.h | 47 ++++++++++++++++++++++++++---------------------
 1 file changed, 26 insertions(+), 21 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index d6ea19e314bb..32851eef48f0 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -49,6 +49,13 @@
 
 /* Attach to insert probes on any functions which should be ignored*/
 #define __kprobes	__attribute__((__section__(".kprobes.text"))) notrace
+#else /* CONFIG_KPROBES */
+typedef int kprobe_opcode_t;
+struct arch_specific_insn {
+	int dummy;
+};
+#define __kprobes	notrace
+#endif /* CONFIG_KPROBES */
 
 struct kprobe;
 struct pt_regs;
@@ -131,23 +138,6 @@ struct jprobe {
 /* For backward compatibility with old code using JPROBE_ENTRY() */
 #define JPROBE_ENTRY(handler)	(handler)
 
-DECLARE_PER_CPU(struct kprobe *, current_kprobe);
-DECLARE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
-
-#ifdef CONFIG_KRETPROBES
-extern void arch_prepare_kretprobe(struct kretprobe_instance *ri,
-				   struct pt_regs *regs);
-extern int arch_trampoline_kprobe(struct kprobe *p);
-#else /* CONFIG_KRETPROBES */
-static inline void arch_prepare_kretprobe(struct kretprobe *rp,
-					struct pt_regs *regs)
-{
-}
-static inline int arch_trampoline_kprobe(struct kprobe *p)
-{
-	return 0;
-}
-#endif /* CONFIG_KRETPROBES */
 /*
  * Function-return probe -
  * Note:
@@ -188,6 +178,25 @@ struct kprobe_blackpoint {
 	unsigned long range;
 };
 
+#ifdef CONFIG_KPROBES
+DECLARE_PER_CPU(struct kprobe *, current_kprobe);
+DECLARE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
+
+#ifdef CONFIG_KRETPROBES
+extern void arch_prepare_kretprobe(struct kretprobe_instance *ri,
+				   struct pt_regs *regs);
+extern int arch_trampoline_kprobe(struct kprobe *p);
+#else /* CONFIG_KRETPROBES */
+static inline void arch_prepare_kretprobe(struct kretprobe *rp,
+					struct pt_regs *regs)
+{
+}
+static inline int arch_trampoline_kprobe(struct kprobe *p)
+{
+	return 0;
+}
+#endif /* CONFIG_KRETPROBES */
+
 extern struct kretprobe_blackpoint kretprobe_blacklist[];
 
 static inline void kretprobe_assert(struct kretprobe_instance *ri,
@@ -264,10 +273,6 @@ void recycle_rp_inst(struct kretprobe_instance *ri, struct hlist_head *head);
 
 #else /* CONFIG_KPROBES */
 
-#define __kprobes	notrace
-struct jprobe;
-struct kretprobe;
-
 static inline struct kprobe *get_kprobe(void *addr)
 {
 	return NULL;
-- 
cgit v1.2.3


From e5d9a90c36e05dd080704ea58328c00f64facdc1 Mon Sep 17 00:00:00 2001
From: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Date: Thu, 29 Jan 2009 14:25:18 -0800
Subject: alpha: use syscall wrappers

Convert OSF syscalls and add alpha specific SYSCALL_ALIAS() macro.

Signed-off-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/alpha/Kconfig          |   1 +
 arch/alpha/kernel/entry.S   |   2 +-
 arch/alpha/kernel/osf_sys.c | 113 ++++++++++++++++++++------------------------
 arch/alpha/kernel/signal.c  |  18 +++----
 arch/alpha/kernel/systbls.S |  52 ++++++++++----------
 include/linux/syscalls.h    |   5 ++
 6 files changed, 92 insertions(+), 99 deletions(-)

(limited to 'include/linux')

diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index 6110197757a3..9fb8aae5c391 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -8,6 +8,7 @@ config ALPHA
 	select HAVE_AOUT
 	select HAVE_IDE
 	select HAVE_OPROFILE
+	select HAVE_SYSCALL_WRAPPERS
 	help
 	  The Alpha is a 64-bit general-purpose processor designed and
 	  marketed by the Digital Equipment Corporation of blessed memory,
diff --git a/arch/alpha/kernel/entry.S b/arch/alpha/kernel/entry.S
index aa2e50cf9857..e4a54b615894 100644
--- a/arch/alpha/kernel/entry.S
+++ b/arch/alpha/kernel/entry.S
@@ -933,7 +933,7 @@ sys_execve:
 osf_sigprocmask:
 	.prologue 0
 	mov	$sp, $18
-	jmp	$31, do_osf_sigprocmask
+	jmp	$31, sys_osf_sigprocmask
 .end osf_sigprocmask
 
 	.align	4
diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c
index 18a3ea1aac51..ae41f097864b 100644
--- a/arch/alpha/kernel/osf_sys.c
+++ b/arch/alpha/kernel/osf_sys.c
@@ -54,8 +54,7 @@ extern int do_pipe(int *);
  * identical to OSF as we don't return 0 on success, but doing otherwise
  * would require changes to libc.  Hopefully this is good enough.
  */
-asmlinkage unsigned long
-osf_brk(unsigned long brk)
+SYSCALL_DEFINE1(osf_brk, unsigned long, brk)
 {
 	unsigned long retval = sys_brk(brk);
 	if (brk && brk != retval)
@@ -66,9 +65,9 @@ osf_brk(unsigned long brk)
 /*
  * This is pure guess-work..
  */
-asmlinkage int
-osf_set_program_attributes(unsigned long text_start, unsigned long text_len,
-			   unsigned long bss_start, unsigned long bss_len)
+SYSCALL_DEFINE4(osf_set_program_attributes, unsigned long, text_start,
+		unsigned long, text_len, unsigned long, bss_start,
+		unsigned long, bss_len)
 {
 	struct mm_struct *mm;
 
@@ -146,9 +145,9 @@ Efault:
 	return -EFAULT;
 }
 
-asmlinkage int
-osf_getdirentries(unsigned int fd, struct osf_dirent __user *dirent,
-		  unsigned int count, long __user *basep)
+SYSCALL_DEFINE4(osf_getdirentries, unsigned int, fd,
+		struct osf_dirent __user *, dirent, unsigned int, count,
+		long __user *, basep)
 {
 	int error;
 	struct file *file;
@@ -177,9 +176,9 @@ osf_getdirentries(unsigned int fd, struct osf_dirent __user *dirent,
 
 #undef NAME_OFFSET
 
-asmlinkage unsigned long
-osf_mmap(unsigned long addr, unsigned long len, unsigned long prot,
-	 unsigned long flags, unsigned long fd, unsigned long off)
+SYSCALL_DEFINE6(osf_mmap, unsigned long, addr, unsigned long, len,
+		unsigned long, prot, unsigned long, flags, unsigned long, fd,
+		unsigned long, off)
 {
 	struct file *file = NULL;
 	unsigned long ret = -EBADF;
@@ -254,8 +253,8 @@ do_osf_statfs(struct dentry * dentry, struct osf_statfs __user *buffer,
 	return error;	
 }
 
-asmlinkage int
-osf_statfs(char __user *pathname, struct osf_statfs __user *buffer, unsigned long bufsiz)
+SYSCALL_DEFINE3(osf_statfs, char __user *, pathname,
+		struct osf_statfs __user *, buffer, unsigned long, bufsiz)
 {
 	struct path path;
 	int retval;
@@ -268,8 +267,8 @@ osf_statfs(char __user *pathname, struct osf_statfs __user *buffer, unsigned lon
 	return retval;
 }
 
-asmlinkage int
-osf_fstatfs(unsigned long fd, struct osf_statfs __user *buffer, unsigned long bufsiz)
+SYSCALL_DEFINE3(osf_fstatfs, unsigned long, fd,
+		struct osf_statfs __user *, buffer, unsigned long, bufsiz)
 {
 	struct file *file;
 	int retval;
@@ -368,8 +367,8 @@ osf_procfs_mount(char *dirname, struct procfs_args __user *args, int flags)
 	return do_mount("", dirname, "proc", flags, NULL);
 }
 
-asmlinkage int
-osf_mount(unsigned long typenr, char __user *path, int flag, void __user *data)
+SYSCALL_DEFINE4(osf_mount, unsigned long, typenr, char __user *, path,
+		int, flag, void __user *, data)
 {
 	int retval = -EINVAL;
 	char *name;
@@ -399,8 +398,7 @@ osf_mount(unsigned long typenr, char __user *path, int flag, void __user *data)
 	return retval;
 }
 
-asmlinkage int
-osf_utsname(char __user *name)
+SYSCALL_DEFINE1(osf_utsname, char __user *, name)
 {
 	int error;
 
@@ -423,14 +421,12 @@ osf_utsname(char __user *name)
 	return error;
 }
 
-asmlinkage unsigned long
-sys_getpagesize(void)
+SYSCALL_DEFINE0(getpagesize)
 {
 	return PAGE_SIZE;
 }
 
-asmlinkage unsigned long
-sys_getdtablesize(void)
+SYSCALL_DEFINE0(getdtablesize)
 {
 	return sysctl_nr_open;
 }
@@ -438,8 +434,7 @@ sys_getdtablesize(void)
 /*
  * For compatibility with OSF/1 only.  Use utsname(2) instead.
  */
-asmlinkage int
-osf_getdomainname(char __user *name, int namelen)
+SYSCALL_DEFINE2(osf_getdomainname, char __user *, name, int, namelen)
 {
 	unsigned len;
 	int i;
@@ -527,8 +522,8 @@ enum pl_code {
 	PL_DEL = 5, PL_FDEL = 6
 };
 
-asmlinkage long
-osf_proplist_syscall(enum pl_code code, union pl_args __user *args)
+SYSCALL_DEFINE2(osf_proplist_syscall, enum pl_code, code,
+		union pl_args __user *, args)
 {
 	long error;
 	int __user *min_buf_size_ptr;
@@ -567,8 +562,8 @@ osf_proplist_syscall(enum pl_code code, union pl_args __user *args)
 	return error;
 }
 
-asmlinkage int
-osf_sigstack(struct sigstack __user *uss, struct sigstack __user *uoss)
+SYSCALL_DEFINE2(osf_sigstack, struct sigstack __user *, uss,
+		struct sigstack __user *, uoss)
 {
 	unsigned long usp = rdusp();
 	unsigned long oss_sp = current->sas_ss_sp + current->sas_ss_size;
@@ -608,8 +603,7 @@ osf_sigstack(struct sigstack __user *uss, struct sigstack __user *uoss)
 	return error;
 }
 
-asmlinkage long
-osf_sysinfo(int command, char __user *buf, long count)
+SYSCALL_DEFINE3(osf_sysinfo, int, command, char __user *, buf, long, count)
 {
 	char *sysinfo_table[] = {
 		utsname()->sysname,
@@ -647,9 +641,8 @@ osf_sysinfo(int command, char __user *buf, long count)
 	return err;
 }
 
-asmlinkage unsigned long
-osf_getsysinfo(unsigned long op, void __user *buffer, unsigned long nbytes,
-	       int __user *start, void __user *arg)
+SYSCALL_DEFINE5(osf_getsysinfo, unsigned long, op, void __user *, buffer,
+		unsigned long, nbytes, int __user *, start, void __user *, arg)
 {
 	unsigned long w;
 	struct percpu_struct *cpu;
@@ -705,9 +698,8 @@ osf_getsysinfo(unsigned long op, void __user *buffer, unsigned long nbytes,
 	return -EOPNOTSUPP;
 }
 
-asmlinkage unsigned long
-osf_setsysinfo(unsigned long op, void __user *buffer, unsigned long nbytes,
-	       int __user *start, void __user *arg)
+SYSCALL_DEFINE5(osf_setsysinfo, unsigned long, op, void __user *, buffer,
+		unsigned long, nbytes, int __user *, start, void __user *, arg)
 {
 	switch (op) {
 	case SSI_IEEE_FP_CONTROL: {
@@ -880,8 +872,8 @@ jiffies_to_timeval32(unsigned long jiffies, struct timeval32 *value)
 	value->tv_sec = jiffies / HZ;
 }
 
-asmlinkage int
-osf_gettimeofday(struct timeval32 __user *tv, struct timezone __user *tz)
+SYSCALL_DEFINE2(osf_gettimeofday, struct timeval32 __user *, tv,
+		struct timezone __user *, tz)
 {
 	if (tv) {
 		struct timeval ktv;
@@ -896,8 +888,8 @@ osf_gettimeofday(struct timeval32 __user *tv, struct timezone __user *tz)
 	return 0;
 }
 
-asmlinkage int
-osf_settimeofday(struct timeval32 __user *tv, struct timezone __user *tz)
+SYSCALL_DEFINE2(osf_settimeofday, struct timeval32 __user *, tv,
+		struct timezone __user *, tz)
 {
 	struct timespec kts;
 	struct timezone ktz;
@@ -916,8 +908,7 @@ osf_settimeofday(struct timeval32 __user *tv, struct timezone __user *tz)
 	return do_sys_settimeofday(tv ? &kts : NULL, tz ? &ktz : NULL);
 }
 
-asmlinkage int
-osf_getitimer(int which, struct itimerval32 __user *it)
+SYSCALL_DEFINE2(osf_getitimer, int, which, struct itimerval32 __user *, it)
 {
 	struct itimerval kit;
 	int error;
@@ -929,8 +920,8 @@ osf_getitimer(int which, struct itimerval32 __user *it)
 	return error;
 }
 
-asmlinkage int
-osf_setitimer(int which, struct itimerval32 __user *in, struct itimerval32 __user *out)
+SYSCALL_DEFINE3(osf_setitimer, int, which, struct itimerval32 __user *, in,
+		struct itimerval32 __user *, out)
 {
 	struct itimerval kin, kout;
 	int error;
@@ -952,8 +943,8 @@ osf_setitimer(int which, struct itimerval32 __user *in, struct itimerval32 __use
 
 }
 
-asmlinkage int
-osf_utimes(char __user *filename, struct timeval32 __user *tvs)
+SYSCALL_DEFINE2(osf_utimes, char __user *, filename,
+		struct timeval32 __user *, tvs)
 {
 	struct timespec tv[2];
 
@@ -979,9 +970,8 @@ osf_utimes(char __user *filename, struct timeval32 __user *tvs)
 #define MAX_SELECT_SECONDS \
 	((unsigned long) (MAX_SCHEDULE_TIMEOUT / HZ)-1)
 
-asmlinkage int
-osf_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp,
-	   struct timeval32 __user *tvp)
+SYSCALL_DEFINE5(osf_select, int, n, fd_set __user *, inp, fd_set __user *, outp,
+		fd_set __user *, exp, struct timeval32 __user *, tvp)
 {
 	struct timespec end_time, *to = NULL;
 	if (tvp) {
@@ -1026,8 +1016,7 @@ struct rusage32 {
 	long	ru_nivcsw;		/* involuntary " */
 };
 
-asmlinkage int
-osf_getrusage(int who, struct rusage32 __user *ru)
+SYSCALL_DEFINE2(osf_getrusage, int, who, struct rusage32 __user *, ru)
 {
 	struct rusage32 r;
 
@@ -1053,9 +1042,8 @@ osf_getrusage(int who, struct rusage32 __user *ru)
 	return copy_to_user(ru, &r, sizeof(r)) ? -EFAULT : 0;
 }
 
-asmlinkage long
-osf_wait4(pid_t pid, int __user *ustatus, int options,
-	  struct rusage32 __user *ur)
+SYSCALL_DEFINE4(osf_wait4, pid_t, pid, int __user *, ustatus, int, options,
+		struct rusage32 __user *, ur)
 {
 	struct rusage r;
 	long ret, err;
@@ -1101,8 +1089,8 @@ osf_wait4(pid_t pid, int __user *ustatus, int options,
  * seems to be a timeval pointer, and I suspect the second
  * one is the time remaining.. Ho humm.. No documentation.
  */
-asmlinkage int
-osf_usleep_thread(struct timeval32 __user *sleep, struct timeval32 __user *remain)
+SYSCALL_DEFINE2(osf_usleep_thread, struct timeval32 __user *, sleep,
+		struct timeval32 __user *, remain)
 {
 	struct timeval tmp;
 	unsigned long ticks;
@@ -1155,8 +1143,7 @@ struct timex32 {
 	int  :32; int  :32; int  :32; int  :32;
 };
 
-asmlinkage int
-sys_old_adjtimex(struct timex32 __user *txc_p)
+SYSCALL_DEFINE1(old_adjtimex, struct timex32 __user *, txc_p)
 {
         struct timex txc;
 	int ret;
@@ -1267,8 +1254,8 @@ osf_fix_iov_len(const struct iovec __user *iov, unsigned long count)
 	return 0;
 }
 
-asmlinkage ssize_t
-osf_readv(unsigned long fd, const struct iovec __user * vector, unsigned long count)
+SYSCALL_DEFINE3(osf_readv, unsigned long, fd,
+		const struct iovec __user *, vector, unsigned long, count)
 {
 	if (unlikely(personality(current->personality) == PER_OSF4))
 		if (osf_fix_iov_len(vector, count))
@@ -1276,8 +1263,8 @@ osf_readv(unsigned long fd, const struct iovec __user * vector, unsigned long co
 	return sys_readv(fd, vector, count);
 }
 
-asmlinkage ssize_t
-osf_writev(unsigned long fd, const struct iovec __user * vector, unsigned long count)
+SYSCALL_DEFINE3(osf_writev, unsigned long, fd,
+		const struct iovec __user *, vector, unsigned long, count)
 {
 	if (unlikely(personality(current->personality) == PER_OSF4))
 		if (osf_fix_iov_len(vector, count))
diff --git a/arch/alpha/kernel/signal.c b/arch/alpha/kernel/signal.c
index 410af4f3140e..df65eaa84c4c 100644
--- a/arch/alpha/kernel/signal.c
+++ b/arch/alpha/kernel/signal.c
@@ -19,6 +19,7 @@
 #include <linux/tty.h>
 #include <linux/binfmts.h>
 #include <linux/bitops.h>
+#include <linux/syscalls.h>
 
 #include <asm/uaccess.h>
 #include <asm/sigcontext.h>
@@ -51,8 +52,8 @@ static void do_signal(struct pt_regs *, struct switch_stack *,
  * Note that we don't need to acquire the kernel lock for SMP
  * operation, as all of this is local to this thread.
  */
-asmlinkage unsigned long
-do_osf_sigprocmask(int how, unsigned long newmask, struct pt_regs *regs)
+SYSCALL_DEFINE3(osf_sigprocmask, int, how, unsigned long, newmask,
+		struct pt_regs *, regs)
 {
 	unsigned long oldmask = -EINVAL;
 
@@ -81,9 +82,9 @@ do_osf_sigprocmask(int how, unsigned long newmask, struct pt_regs *regs)
 	return oldmask;
 }
 
-asmlinkage int 
-osf_sigaction(int sig, const struct osf_sigaction __user *act,
-	      struct osf_sigaction __user *oact)
+SYSCALL_DEFINE3(osf_sigaction, int, sig,
+		const struct osf_sigaction __user *, act,
+		struct osf_sigaction __user *, oact)
 {
 	struct k_sigaction new_ka, old_ka;
 	int ret;
@@ -112,10 +113,9 @@ osf_sigaction(int sig, const struct osf_sigaction __user *act,
 	return ret;
 }
 
-asmlinkage long
-sys_rt_sigaction(int sig, const struct sigaction __user *act,
-		 struct sigaction __user *oact,
-		 size_t sigsetsize, void __user *restorer)
+SYSCALL_DEFINE5(rt_sigaction, int, sig, const struct sigaction __user *, act,
+		struct sigaction __user *, oact,
+		size_t, sigsetsize, void __user *, restorer)
 {
 	struct k_sigaction new_ka, old_ka;
 	int ret;
diff --git a/arch/alpha/kernel/systbls.S b/arch/alpha/kernel/systbls.S
index 9d9e3a98bb95..95c9aef1c106 100644
--- a/arch/alpha/kernel/systbls.S
+++ b/arch/alpha/kernel/systbls.S
@@ -17,7 +17,7 @@ sys_call_table:
 	.quad sys_write
 	.quad alpha_ni_syscall			/* 5 */
 	.quad sys_close
-	.quad osf_wait4
+	.quad sys_osf_wait4
 	.quad alpha_ni_syscall
 	.quad sys_link
 	.quad sys_unlink			/* 10 */
@@ -27,11 +27,11 @@ sys_call_table:
 	.quad sys_mknod
 	.quad sys_chmod				/* 15 */
 	.quad sys_chown
-	.quad osf_brk
+	.quad sys_osf_brk
 	.quad alpha_ni_syscall
 	.quad sys_lseek
 	.quad sys_getxpid			/* 20 */
-	.quad osf_mount
+	.quad sys_osf_mount
 	.quad sys_umount
 	.quad sys_setuid
 	.quad sys_getxuid
@@ -53,7 +53,7 @@ sys_call_table:
 	.quad alpha_ni_syscall			/* 40 */
 	.quad sys_dup
 	.quad sys_alpha_pipe
-	.quad osf_set_program_attributes
+	.quad sys_osf_set_program_attributes
 	.quad alpha_ni_syscall
 	.quad sys_open				/* 45 */
 	.quad alpha_ni_syscall
@@ -81,7 +81,7 @@ sys_call_table:
 	.quad sys_newlstat
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall			/* 70 */
-	.quad osf_mmap
+	.quad sys_osf_mmap
 	.quad alpha_ni_syscall
 	.quad sys_munmap
 	.quad sys_mprotect
@@ -94,17 +94,17 @@ sys_call_table:
 	.quad sys_setgroups			/* 80 */
 	.quad alpha_ni_syscall
 	.quad sys_setpgid
-	.quad osf_setitimer
+	.quad sys_osf_setitimer
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall			/* 85 */
-	.quad osf_getitimer
+	.quad sys_osf_getitimer
 	.quad sys_gethostname
 	.quad sys_sethostname
 	.quad sys_getdtablesize
 	.quad sys_dup2				/* 90 */
 	.quad sys_newfstat
 	.quad sys_fcntl
-	.quad osf_select
+	.quad sys_osf_select
 	.quad sys_poll
 	.quad sys_fsync				/* 95 */
 	.quad sys_setpriority
@@ -123,22 +123,22 @@ sys_call_table:
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall			/* 110 */
 	.quad sys_sigsuspend
-	.quad osf_sigstack
+	.quad sys_osf_sigstack
 	.quad sys_recvmsg
 	.quad sys_sendmsg
 	.quad alpha_ni_syscall			/* 115 */
-	.quad osf_gettimeofday
-	.quad osf_getrusage
+	.quad sys_osf_gettimeofday
+	.quad sys_osf_getrusage
 	.quad sys_getsockopt
 	.quad alpha_ni_syscall
 #ifdef CONFIG_OSF4_COMPAT
-	.quad osf_readv				/* 120 */
-	.quad osf_writev
+	.quad sys_osf_readv			/* 120 */
+	.quad sys_osf_writev
 #else
 	.quad sys_readv				/* 120 */
 	.quad sys_writev
 #endif
-	.quad osf_settimeofday
+	.quad sys_osf_settimeofday
 	.quad sys_fchown
 	.quad sys_fchmod
 	.quad sys_recvfrom			/* 125 */
@@ -154,7 +154,7 @@ sys_call_table:
 	.quad sys_socketpair			/* 135 */
 	.quad sys_mkdir
 	.quad sys_rmdir
-	.quad osf_utimes
+	.quad sys_osf_utimes
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall			/* 140 */
 	.quad sys_getpeername
@@ -172,16 +172,16 @@ sys_call_table:
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall			/* 155 */
-	.quad osf_sigaction
+	.quad sys_osf_sigaction
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall
-	.quad osf_getdirentries
-	.quad osf_statfs			/* 160 */
-	.quad osf_fstatfs
+	.quad sys_osf_getdirentries
+	.quad sys_osf_statfs			/* 160 */
+	.quad sys_osf_fstatfs
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall
-	.quad osf_getdomainname			/* 165 */
+	.quad sys_osf_getdomainname		/* 165 */
 	.quad sys_setdomainname
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall
@@ -224,7 +224,7 @@ sys_call_table:
 	.quad sys_semctl
 	.quad sys_semget			/* 205 */
 	.quad sys_semop
-	.quad osf_utsname
+	.quad sys_osf_utsname
 	.quad sys_lchown
 	.quad sys_shmat
 	.quad sys_shmctl			/* 210 */
@@ -258,23 +258,23 @@ sys_call_table:
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall			/* 240 */
-	.quad osf_sysinfo
+	.quad sys_osf_sysinfo
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall
-	.quad osf_proplist_syscall
+	.quad sys_osf_proplist_syscall
 	.quad alpha_ni_syscall			/* 245 */
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall			/* 250 */
-	.quad osf_usleep_thread
+	.quad sys_osf_usleep_thread
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall
 	.quad sys_sysfs
 	.quad alpha_ni_syscall			/* 255 */
-	.quad osf_getsysinfo
-	.quad osf_setsysinfo
+	.quad sys_osf_getsysinfo
+	.quad sys_osf_setsysinfo
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall
 	.quad alpha_ni_syscall			/* 260 */
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 16875f89e6a7..0eda02ff2414 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -108,9 +108,14 @@ struct old_linux_dirent;
 	asm ("\t.globl " #alias "\n\t.set " #alias ", " #name "\n"	\
 	     "\t.globl ." #alias "\n\t.set ." #alias ", ." #name)
 #else
+#ifdef CONFIG_ALPHA
+#define SYSCALL_ALIAS(alias, name)					\
+	asm ( #alias " = " #name "\n\t.globl " #alias)
+#else
 #define SYSCALL_ALIAS(alias, name)					\
 	asm ("\t.globl " #alias "\n\t.set " #alias ", " #name)
 #endif
+#endif
 
 #ifdef CONFIG_HAVE_SYSCALL_WRAPPERS
 
-- 
cgit v1.2.3


From 804b3c28a4e4fa1c224571bf76edb534b9c4b1ed Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Thu, 29 Jan 2009 14:25:21 -0800
Subject: cgroups: add cpu_relax() calls in css_tryget() and
 cgroup_clear_css_refs()

css_tryget() and cgroup_clear_css_refs() contain polling loops; these
loops should have cpu_relax calls in them to reduce cross-cache traffic.

Signed-off-by: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cgroup.h | 1 +
 kernel/cgroup.c        | 7 +++++--
 2 files changed, 6 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index e267e62827bb..e4e8e117d27d 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -99,6 +99,7 @@ static inline bool css_tryget(struct cgroup_subsys_state *css)
 	while (!atomic_inc_not_zero(&css->refcnt)) {
 		if (test_bit(CSS_REMOVED, &css->flags))
 			return false;
+		cpu_relax();
 	}
 	return true;
 }
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 0066092de19a..492215d67fa5 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -2509,7 +2509,7 @@ static int cgroup_clear_css_refs(struct cgroup *cgrp)
 	for_each_subsys(cgrp->root, ss) {
 		struct cgroup_subsys_state *css = cgrp->subsys[ss->subsys_id];
 		int refcnt;
-		do {
+		while (1) {
 			/* We can only remove a CSS with a refcnt==1 */
 			refcnt = atomic_read(&css->refcnt);
 			if (refcnt > 1) {
@@ -2523,7 +2523,10 @@ static int cgroup_clear_css_refs(struct cgroup *cgrp)
 			 * css_tryget() to spin until we set the
 			 * CSS_REMOVED bits or abort
 			 */
-		} while (atomic_cmpxchg(&css->refcnt, refcnt, 0) != refcnt);
+			if (atomic_cmpxchg(&css->refcnt, refcnt, 0) == refcnt)
+				break;
+			cpu_relax();
+		}
 	}
  done:
 	for_each_subsys(cgrp->root, ss) {
-- 
cgit v1.2.3


From 9df04e1f25effde823a600e755b51475d438f56b Mon Sep 17 00:00:00 2001
From: Davide Libenzi <davidel@xmailserver.org>
Date: Thu, 29 Jan 2009 14:25:26 -0800
Subject: epoll: drop max_user_instances and rely only on max_user_watches

Linus suggested to put limits where the money is, and max_user_watches
already does that w/out the need of max_user_instances.  That has the
advantage to mitigate the potential DoS while allowing pretty generous
default behavior.

Allowing top 4% of low memory (per user) to be allocated in epoll watches,
we have:

LOMEM    MAX_WATCHES (per user)
512MB    ~178000
1GB      ~356000
2GB      ~712000

A box with 512MB of lomem, will meet some challenge in hitting 180K
watches, socket buffers math teaches us.  No more max_user_instances
limits then.

Signed-off-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Willy Tarreau <w@1wt.eu>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: Bron Gondwana <brong@fastmail.fm>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/eventpoll.c        | 22 ++++------------------
 include/linux/sched.h |  1 -
 2 files changed, 4 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index ba2f9ec71192..011b9b8c90c6 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -234,8 +234,6 @@ struct ep_pqueue {
 /*
  * Configuration options available inside /proc/sys/fs/epoll/
  */
-/* Maximum number of epoll devices, per user */
-static int max_user_instances __read_mostly;
 /* Maximum number of epoll watched descriptors, per user */
 static int max_user_watches __read_mostly;
 
@@ -260,14 +258,6 @@ static struct kmem_cache *pwq_cache __read_mostly;
 static int zero;
 
 ctl_table epoll_table[] = {
-	{
-		.procname	= "max_user_instances",
-		.data		= &max_user_instances,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_minmax,
-		.extra1		= &zero,
-	},
 	{
 		.procname	= "max_user_watches",
 		.data		= &max_user_watches,
@@ -491,7 +481,6 @@ static void ep_free(struct eventpoll *ep)
 
 	mutex_unlock(&epmutex);
 	mutex_destroy(&ep->mtx);
-	atomic_dec(&ep->user->epoll_devs);
 	free_uid(ep->user);
 	kfree(ep);
 }
@@ -581,10 +570,6 @@ static int ep_alloc(struct eventpoll **pep)
 	struct eventpoll *ep;
 
 	user = get_current_user();
-	error = -EMFILE;
-	if (unlikely(atomic_read(&user->epoll_devs) >=
-			max_user_instances))
-		goto free_uid;
 	error = -ENOMEM;
 	ep = kzalloc(sizeof(*ep), GFP_KERNEL);
 	if (unlikely(!ep))
@@ -1141,7 +1126,6 @@ SYSCALL_DEFINE1(epoll_create1, int, flags)
 			      flags & O_CLOEXEC);
 	if (fd < 0)
 		ep_free(ep);
-	atomic_inc(&ep->user->epoll_devs);
 
 error_return:
 	DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d) = %d\n",
@@ -1366,8 +1350,10 @@ static int __init eventpoll_init(void)
 	struct sysinfo si;
 
 	si_meminfo(&si);
-	max_user_instances = 128;
-	max_user_watches = (((si.totalram - si.totalhigh) / 32) << PAGE_SHIFT) /
+	/*
+	 * Allows top 4% of lomem to be allocated for epoll watches (per user).
+	 */
+	max_user_watches = (((si.totalram - si.totalhigh) / 25) << PAGE_SHIFT) /
 		EP_ITEM_COST;
 
 	/* Initialize the structure used to perform safe poll wait head wake ups */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 02e16d207304..5a7c76388731 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -630,7 +630,6 @@ struct user_struct {
 	atomic_t inotify_devs;	/* How many inotify devs does this user have opened? */
 #endif
 #ifdef CONFIG_EPOLL
-	atomic_t epoll_devs;	/* The number of epoll descriptors currently open */
 	atomic_t epoll_watches;	/* The number of file descriptors currently watched */
 #endif
 #ifdef CONFIG_POSIX_MQUEUE
-- 
cgit v1.2.3


From 7b24fc4d7eb611da367dea3aad45473050aacd6c Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Sun, 4 Jan 2009 02:43:38 -0500
Subject: block: Don't verify integrity metadata on read error

If we get an I/O error on a read request there is no point in doing a
verify pass on the integrity buffer.  Adjust the completion path
accordingly.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 fs/bio-integrity.c  | 25 +++++++++++++++----------
 include/linux/bio.h |  1 -
 2 files changed, 15 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c
index 77ebc3c263d6..8396d741f804 100644
--- a/fs/bio-integrity.c
+++ b/fs/bio-integrity.c
@@ -465,7 +465,7 @@ static int bio_integrity_verify(struct bio *bio)
 
 		if (ret) {
 			kunmap_atomic(kaddr, KM_USER0);
-			break;
+			return ret;
 		}
 
 		sectors = bv->bv_len / bi->sector_size;
@@ -493,18 +493,13 @@ static void bio_integrity_verify_fn(struct work_struct *work)
 	struct bio_integrity_payload *bip =
 		container_of(work, struct bio_integrity_payload, bip_work);
 	struct bio *bio = bip->bip_bio;
-	int error = bip->bip_error;
+	int error;
 
-	if (bio_integrity_verify(bio)) {
-		clear_bit(BIO_UPTODATE, &bio->bi_flags);
-		error = -EIO;
-	}
+	error = bio_integrity_verify(bio);
 
 	/* Restore original bio completion handler */
 	bio->bi_end_io = bip->bip_end_io;
-
-	if (bio->bi_end_io)
-		bio->bi_end_io(bio, error);
+	bio_endio(bio, error);
 }
 
 /**
@@ -525,7 +520,17 @@ void bio_integrity_endio(struct bio *bio, int error)
 
 	BUG_ON(bip->bip_bio != bio);
 
-	bip->bip_error = error;
+	/* In case of an I/O error there is no point in verifying the
+	 * integrity metadata.  Restore original bio end_io handler
+	 * and run it.
+	 */
+	if (error) {
+		bio->bi_end_io = bip->bip_end_io;
+		bio_endio(bio, error);
+
+		return;
+	}
+
 	INIT_WORK(&bip->bip_work, bio_integrity_verify_fn);
 	queue_work(kintegrityd_wq, &bip->bip_work);
 }
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 18462c5b8fff..18fc4a281a7b 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -312,7 +312,6 @@ struct bio_integrity_payload {
 	void			*bip_buf;	/* generated integrity data */
 	bio_end_io_t		*bip_end_io;	/* saved I/O completion fn */
 
-	int			bip_error;	/* saved I/O error */
 	unsigned int		bip_size;
 
 	unsigned short		bip_pool;	/* pool the ivec came from */
-- 
cgit v1.2.3


From 16642eb68216d8e0e136a99e514e9166e7125838 Mon Sep 17 00:00:00 2001
From: Alberto Bertogli <albertito@blitiri.com.ar>
Date: Mon, 5 Jan 2009 10:18:53 +0100
Subject: Fix small typo in bio.h's documentation

Signed-off-by: Alberto Bertogli <albertito@blitiri.com.ar>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/bio.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/bio.h b/include/linux/bio.h
index 18fc4a281a7b..5175aa3103c6 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -144,7 +144,7 @@ struct bio {
  * bit 1 -- rw-ahead when set
  * bit 2 -- barrier
  *	Insert a serialization point in the IO queue, forcing previously
- *	submitted IO to be completed before this oen is issued.
+ *	submitted IO to be completed before this one is issued.
  * bit 3 -- synchronous I/O hint: the block layer will unplug immediately
  *	Note that this does NOT indicate that the IO itself is sync, just
  *	that the block layer will not postpone issue of this IO by plugging.
-- 
cgit v1.2.3


From 213d9417fec62ef4c3675621b9364a667954d4dd Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Tue, 6 Jan 2009 09:16:05 +0100
Subject: block: seperate bio/request unplug and sync bits

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/blk-core.c       |  5 ++++-
 include/linux/bio.h    | 18 +++++++++++-------
 include/linux/blkdev.h |  2 ++
 3 files changed, 17 insertions(+), 8 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-core.c b/block/blk-core.c
index a824e49c0d0a..9e2e86fb78b8 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1125,6 +1125,8 @@ void init_request_from_bio(struct request *req, struct bio *bio)
 
 	if (bio_sync(bio))
 		req->cmd_flags |= REQ_RW_SYNC;
+	if (bio_unplug(bio))
+		req->cmd_flags |= REQ_UNPLUG;
 	if (bio_rw_meta(bio))
 		req->cmd_flags |= REQ_RW_META;
 
@@ -1141,6 +1143,7 @@ static int __make_request(struct request_queue *q, struct bio *bio)
 	int el_ret, nr_sectors;
 	const unsigned short prio = bio_prio(bio);
 	const int sync = bio_sync(bio);
+	const int unplug = bio_unplug(bio);
 	int rw_flags;
 
 	nr_sectors = bio_sectors(bio);
@@ -1244,7 +1247,7 @@ get_rq:
 		blk_plug_device(q);
 	add_request(q, req);
 out:
-	if (sync || blk_queue_nonrot(q))
+	if (unplug || blk_queue_nonrot(q))
 		__generic_unplug_device(q);
 	spin_unlock_irq(q->queue_lock);
 	return 0;
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 5175aa3103c6..f53568c5852e 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -163,12 +163,15 @@ struct bio {
 #define BIO_RW		0	/* Must match RW in req flags (blkdev.h) */
 #define BIO_RW_AHEAD	1	/* Must match FAILFAST in req flags */
 #define BIO_RW_BARRIER	2
-#define BIO_RW_SYNC	3
-#define BIO_RW_META	4
-#define BIO_RW_DISCARD	5
-#define BIO_RW_FAILFAST_DEV		6
-#define BIO_RW_FAILFAST_TRANSPORT	7
-#define BIO_RW_FAILFAST_DRIVER		8
+#define BIO_RW_SYNCIO	3
+#define BIO_RW_UNPLUG	4
+#define BIO_RW_META	5
+#define BIO_RW_DISCARD	6
+#define BIO_RW_FAILFAST_DEV		7
+#define BIO_RW_FAILFAST_TRANSPORT	8
+#define BIO_RW_FAILFAST_DRIVER		9
+
+#define BIO_RW_SYNC	(BIO_RW_SYNCIO | BIO_RW_UNPLUG)
 
 /*
  * upper 16 bits of bi_rw define the io priority of this bio
@@ -194,7 +197,8 @@ struct bio {
 #define bio_segments(bio)	((bio)->bi_vcnt - (bio)->bi_idx)
 #define bio_sectors(bio)	((bio)->bi_size >> 9)
 #define bio_barrier(bio)	((bio)->bi_rw & (1 << BIO_RW_BARRIER))
-#define bio_sync(bio)		((bio)->bi_rw & (1 << BIO_RW_SYNC))
+#define bio_sync(bio)		((bio)->bi_rw & (1 << BIO_RW_SYNCIO))
+#define bio_unplug(bio)		((bio)->bi_rw & (1 << BIO_RW_UNPLUG))
 #define bio_failfast_dev(bio)	((bio)->bi_rw &	(1 << BIO_RW_FAILFAST_DEV))
 #define bio_failfast_transport(bio)	\
 	((bio)->bi_rw & (1 << BIO_RW_FAILFAST_TRANSPORT))
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 044467ef7b11..75426e4b8cdf 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -108,6 +108,7 @@ enum rq_flag_bits {
 	__REQ_RW_META,		/* metadata io request */
 	__REQ_COPY_USER,	/* contains copies of user pages */
 	__REQ_INTEGRITY,	/* integrity metadata has been remapped */
+	__REQ_UNPLUG,		/* unplug queue on submission */
 	__REQ_NR_BITS,		/* stops here */
 };
 
@@ -134,6 +135,7 @@ enum rq_flag_bits {
 #define REQ_RW_META	(1 << __REQ_RW_META)
 #define REQ_COPY_USER	(1 << __REQ_COPY_USER)
 #define REQ_INTEGRITY	(1 << __REQ_INTEGRITY)
+#define REQ_UNPLUG	(1 << __REQ_UNPLUG)
 
 #define BLK_MAX_CDB	16
 
-- 
cgit v1.2.3


From 1dfa17f4ab8543d82caf4d36636b93916a18f456 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Tue, 6 Jan 2009 09:21:49 +0100
Subject: block: add bio_rw_flagged() for testing bio->bi_rw

The existing functions for checking bio->bi_rw are badly named. So lets
mirror what we do for bio->bi_flags testing, use a properly named
function so that it's immediately obvious what is being tested.

Maintain compatability names for the old macros, eventually we'll get
rid of these.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/bio.h | 28 ++++++++++++++++++----------
 1 file changed, 18 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/bio.h b/include/linux/bio.h
index f53568c5852e..0942765cf8c0 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -173,6 +173,24 @@ struct bio {
 
 #define BIO_RW_SYNC	(BIO_RW_SYNCIO | BIO_RW_UNPLUG)
 
+#define bio_rw_flagged(bio, flag)	((bio)->bi_rw & (1 << (flag)))
+
+/*
+ * Old defines, these should eventually be replaced by direct usage of
+ * bio_rw_flagged()
+ */
+#define bio_barrier(bio)	bio_rw_flagged(bio, BIO_RW_BARRIER)
+#define bio_sync(bio)		bio_rw_flagged(bio, BIO_RW_SYNCIO)
+#define bio_unplug(bio)		bio_rw_flagged(bio, BIO_RW_UNPLUG)
+#define bio_failfast_dev(bio)	bio_rw_flagged(bio, BIO_RW_FAILFAST_DEV)
+#define bio_failfast_transport(bio)	\
+		bio_rw_flagged(bio, BIO_RW_FAILFAST_TRANSPORT)
+#define bio_failfast_driver(bio) 	\
+		bio_rw_flagged(bio, BIO_RW_FAILFAST_DRIVER)
+#define bio_rw_ahead(bio)	bio_rw_flagged(bio, BIO_RW_AHEAD)
+#define bio_rw_meta(bio)	bio_rw_flagged(bio, BIO_RW_META)
+#define bio_discard(bio)	bio_rw_flagged(bio, BIO_RW_DISCARD)
+
 /*
  * upper 16 bits of bi_rw define the io priority of this bio
  */
@@ -196,16 +214,6 @@ struct bio {
 #define bio_offset(bio)		bio_iovec((bio))->bv_offset
 #define bio_segments(bio)	((bio)->bi_vcnt - (bio)->bi_idx)
 #define bio_sectors(bio)	((bio)->bi_size >> 9)
-#define bio_barrier(bio)	((bio)->bi_rw & (1 << BIO_RW_BARRIER))
-#define bio_sync(bio)		((bio)->bi_rw & (1 << BIO_RW_SYNCIO))
-#define bio_unplug(bio)		((bio)->bi_rw & (1 << BIO_RW_UNPLUG))
-#define bio_failfast_dev(bio)	((bio)->bi_rw &	(1 << BIO_RW_FAILFAST_DEV))
-#define bio_failfast_transport(bio)	\
-	((bio)->bi_rw & (1 << BIO_RW_FAILFAST_TRANSPORT))
-#define bio_failfast_driver(bio) ((bio)->bi_rw & (1 << BIO_RW_FAILFAST_DRIVER))
-#define bio_rw_ahead(bio)	((bio)->bi_rw & (1 << BIO_RW_AHEAD))
-#define bio_rw_meta(bio)	((bio)->bi_rw & (1 << BIO_RW_META))
-#define bio_discard(bio)	((bio)->bi_rw & (1 << BIO_RW_DISCARD))
 #define bio_empty_barrier(bio)	(bio_barrier(bio) && !bio_has_data(bio) && !bio_discard(bio))
 
 static inline unsigned int bio_cur_sectors(struct bio *bio)
-- 
cgit v1.2.3


From a229fc61ef0ee3c30fd193beee0eeb87410227f1 Mon Sep 17 00:00:00 2001
From: Boaz Harrosh <bharrosh@panasas.com>
Date: Mon, 19 Jan 2009 10:37:38 +0100
Subject: include/linux: Add bsg.h to the Kernel exported headers

bsg.h in current form is perfectly suitable for user-mode
consumption. It is needed together with scsi/sg.h for applications
that want to interface with the bsg driver.

Currently the few projects that use it would copy it over into
the projects. But that is not acceptable for projects that need
to provide source and devel packages for distros.

This should also be submitted to stable 2.6.28 and 2.6.27 since bsg had
a stable API since these Kernels and distro users will need the header
for these kernels a swell

Signed-off-by: Boaz Harrosh <bharrosh@panasas.com>
Acked-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
CC: stable@kernel.org
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/Kbuild | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 12e9a2957caf..2124c063a7ef 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -41,6 +41,7 @@ header-y += baycom.h
 header-y += bfs_fs.h
 header-y += blkpg.h
 header-y += bpqether.h
+header-y += bsg.h
 header-y += can.h
 header-y += cdk.h
 header-y += chio.h
-- 
cgit v1.2.3


From bc58ba9468d94d62c56ab9b47173583ec140b165 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Fri, 23 Jan 2009 10:54:44 +0100
Subject: block: add sysfs file for controlling io stats accounting

This allows us to turn off disk stat accounting completely, for the cases
where the 0.5-1% reduction in system time is important.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/blk-core.c       | 90 ++++++++++++++++++++++++++++++--------------------
 block/blk-sysfs.c      | 28 ++++++++++++++++
 include/linux/blkdev.h |  6 ++++
 3 files changed, 88 insertions(+), 36 deletions(-)

(limited to 'include/linux')

diff --git a/block/blk-core.c b/block/blk-core.c
index ae75c047f45d..ca69f3d94100 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -64,11 +64,12 @@ static struct workqueue_struct *kblockd_workqueue;
 
 static void drive_stat_acct(struct request *rq, int new_io)
 {
+	struct gendisk *disk = rq->rq_disk;
 	struct hd_struct *part;
 	int rw = rq_data_dir(rq);
 	int cpu;
 
-	if (!blk_fs_request(rq) || !rq->rq_disk)
+	if (!blk_fs_request(rq) || !disk || !blk_queue_io_stat(disk->queue))
 		return;
 
 	cpu = part_stat_lock();
@@ -599,8 +600,7 @@ blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id)
 	q->request_fn		= rfn;
 	q->prep_rq_fn		= NULL;
 	q->unplug_fn		= generic_unplug_device;
-	q->queue_flags		= (1 << QUEUE_FLAG_CLUSTER |
-				   1 << QUEUE_FLAG_STACKABLE);
+	q->queue_flags		= QUEUE_FLAG_DEFAULT;
 	q->queue_lock		= lock;
 
 	blk_queue_segment_boundary(q, BLK_SEG_BOUNDARY_MASK);
@@ -1663,6 +1663,55 @@ void blkdev_dequeue_request(struct request *req)
 }
 EXPORT_SYMBOL(blkdev_dequeue_request);
 
+static void blk_account_io_completion(struct request *req, unsigned int bytes)
+{
+	struct gendisk *disk = req->rq_disk;
+
+	if (!disk || !blk_queue_io_stat(disk->queue))
+		return;
+
+	if (blk_fs_request(req)) {
+		const int rw = rq_data_dir(req);
+		struct hd_struct *part;
+		int cpu;
+
+		cpu = part_stat_lock();
+		part = disk_map_sector_rcu(req->rq_disk, req->sector);
+		part_stat_add(cpu, part, sectors[rw], bytes >> 9);
+		part_stat_unlock();
+	}
+}
+
+static void blk_account_io_done(struct request *req)
+{
+	struct gendisk *disk = req->rq_disk;
+
+	if (!disk || !blk_queue_io_stat(disk->queue))
+		return;
+
+	/*
+	 * Account IO completion.  bar_rq isn't accounted as a normal
+	 * IO on queueing nor completion.  Accounting the containing
+	 * request is enough.
+	 */
+	if (blk_fs_request(req) && req != &req->q->bar_rq) {
+		unsigned long duration = jiffies - req->start_time;
+		const int rw = rq_data_dir(req);
+		struct hd_struct *part;
+		int cpu;
+
+		cpu = part_stat_lock();
+		part = disk_map_sector_rcu(disk, req->sector);
+
+		part_stat_inc(cpu, part, ios[rw]);
+		part_stat_add(cpu, part, ticks[rw], duration);
+		part_round_stats(cpu, part);
+		part_dec_in_flight(part);
+
+		part_stat_unlock();
+	}
+}
+
 /**
  * __end_that_request_first - end I/O on a request
  * @req:      the request being processed
@@ -1698,16 +1747,7 @@ static int __end_that_request_first(struct request *req, int error,
 				(unsigned long long)req->sector);
 	}
 
-	if (blk_fs_request(req) && req->rq_disk) {
-		const int rw = rq_data_dir(req);
-		struct hd_struct *part;
-		int cpu;
-
-		cpu = part_stat_lock();
-		part = disk_map_sector_rcu(req->rq_disk, req->sector);
-		part_stat_add(cpu, part, sectors[rw], nr_bytes >> 9);
-		part_stat_unlock();
-	}
+	blk_account_io_completion(req, nr_bytes);
 
 	total_bytes = bio_nbytes = 0;
 	while ((bio = req->bio) != NULL) {
@@ -1787,8 +1827,6 @@ static int __end_that_request_first(struct request *req, int error,
  */
 static void end_that_request_last(struct request *req, int error)
 {
-	struct gendisk *disk = req->rq_disk;
-
 	if (blk_rq_tagged(req))
 		blk_queue_end_tag(req->q, req);
 
@@ -1800,27 +1838,7 @@ static void end_that_request_last(struct request *req, int error)
 
 	blk_delete_timer(req);
 
-	/*
-	 * Account IO completion.  bar_rq isn't accounted as a normal
-	 * IO on queueing nor completion.  Accounting the containing
-	 * request is enough.
-	 */
-	if (disk && blk_fs_request(req) && req != &req->q->bar_rq) {
-		unsigned long duration = jiffies - req->start_time;
-		const int rw = rq_data_dir(req);
-		struct hd_struct *part;
-		int cpu;
-
-		cpu = part_stat_lock();
-		part = disk_map_sector_rcu(disk, req->sector);
-
-		part_stat_inc(cpu, part, ios[rw]);
-		part_stat_add(cpu, part, ticks[rw], duration);
-		part_round_stats(cpu, part);
-		part_dec_in_flight(part);
-
-		part_stat_unlock();
-	}
+	blk_account_io_done(req);
 
 	if (req->end_io)
 		req->end_io(req, error);
diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
index b538ab8dbbd1..e29ddfc73cf4 100644
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -197,6 +197,27 @@ queue_rq_affinity_store(struct request_queue *q, const char *page, size_t count)
 	return ret;
 }
 
+static ssize_t queue_iostats_show(struct request_queue *q, char *page)
+{
+	return queue_var_show(blk_queue_io_stat(q), page);
+}
+
+static ssize_t queue_iostats_store(struct request_queue *q, const char *page,
+				   size_t count)
+{
+	unsigned long stats;
+	ssize_t ret = queue_var_store(&stats, page, count);
+
+	spin_lock_irq(q->queue_lock);
+	if (stats)
+		queue_flag_set(QUEUE_FLAG_IO_STAT, q);
+	else
+		queue_flag_clear(QUEUE_FLAG_IO_STAT, q);
+	spin_unlock_irq(q->queue_lock);
+
+	return ret;
+}
+
 static struct queue_sysfs_entry queue_requests_entry = {
 	.attr = {.name = "nr_requests", .mode = S_IRUGO | S_IWUSR },
 	.show = queue_requests_show,
@@ -249,6 +270,12 @@ static struct queue_sysfs_entry queue_rq_affinity_entry = {
 	.store = queue_rq_affinity_store,
 };
 
+static struct queue_sysfs_entry queue_iostats_entry = {
+	.attr = {.name = "iostats", .mode = S_IRUGO | S_IWUSR },
+	.show = queue_iostats_show,
+	.store = queue_iostats_store,
+};
+
 static struct attribute *default_attrs[] = {
 	&queue_requests_entry.attr,
 	&queue_ra_entry.attr,
@@ -259,6 +286,7 @@ static struct attribute *default_attrs[] = {
 	&queue_nonrot_entry.attr,
 	&queue_nomerges_entry.attr,
 	&queue_rq_affinity_entry.attr,
+	&queue_iostats_entry.attr,
 	NULL,
 };
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 75426e4b8cdf..d08c4b8219a6 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -451,6 +451,11 @@ struct request_queue
 #define QUEUE_FLAG_STACKABLE   13	/* supports request stacking */
 #define QUEUE_FLAG_NONROT      14	/* non-rotational device (SSD) */
 #define QUEUE_FLAG_VIRT        QUEUE_FLAG_NONROT /* paravirt device */
+#define QUEUE_FLAG_IO_STAT     15	/* do IO stats */
+
+#define QUEUE_FLAG_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
+				 (1 << QUEUE_FLAG_CLUSTER) |		\
+				  1 << QUEUE_FLAG_STACKABLE)
 
 static inline int queue_is_locked(struct request_queue *q)
 {
@@ -567,6 +572,7 @@ enum {
 #define blk_queue_stopped(q)	test_bit(QUEUE_FLAG_STOPPED, &(q)->queue_flags)
 #define blk_queue_nomerges(q)	test_bit(QUEUE_FLAG_NOMERGES, &(q)->queue_flags)
 #define blk_queue_nonrot(q)	test_bit(QUEUE_FLAG_NONROT, &(q)->queue_flags)
+#define blk_queue_io_stat(q)	test_bit(QUEUE_FLAG_IO_STAT, &(q)->queue_flags)
 #define blk_queue_flushing(q)	((q)->ordseq)
 #define blk_queue_stackable(q)	\
 	test_bit(QUEUE_FLAG_STACKABLE, &(q)->queue_flags)
-- 
cgit v1.2.3


From 9d6aa4c7ece26652fcbfe37bd45679eac5f69347 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 19:50:25 +0530
Subject: headers_check fix: can/bcm.h

fix the following 'make headers_check' warning:

  usr/include/linux/can/bcm.h:29: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/can/bcm.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/can/bcm.h b/include/linux/can/bcm.h
index 7f293273c444..1432b278c52d 100644
--- a/include/linux/can/bcm.h
+++ b/include/linux/can/bcm.h
@@ -14,6 +14,8 @@
 #ifndef CAN_BCM_H
 #define CAN_BCM_H
 
+#include <linux/types.h>
+
 /**
  * struct bcm_msg_head - head of messages to/from the broadcast manager
  * @opcode:    opcode, see enum below.
-- 
cgit v1.2.3


From 15cf98ad2965aaefaa2f85332535ff39e48f9f4e Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 19:53:38 +0530
Subject: headers_check fix: dvb/audio.h

fix the following 'make headers_check' warning:

  usr/include/linux/dvb/audio.h:133: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/dvb/audio.h | 5 -----
 1 file changed, 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/dvb/audio.h b/include/linux/dvb/audio.h
index 89412e18f571..bb0df2aaebfa 100644
--- a/include/linux/dvb/audio.h
+++ b/include/linux/dvb/audio.h
@@ -24,12 +24,7 @@
 #ifndef _DVBAUDIO_H_
 #define _DVBAUDIO_H_
 
-#ifdef __KERNEL__
 #include <linux/types.h>
-#else
-#include <stdint.h>
-#endif
-
 
 typedef enum {
 	AUDIO_SOURCE_DEMUX, /* Select the demux as the main source */
-- 
cgit v1.2.3


From c86629c855800071314810881d1d2fb226ca9ec9 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 19:55:32 +0530
Subject: headers_check fix: dvb/dmx.h

fix the following 'make headers_check' warnings:

  usr/include/linux/dvb/dmx.h:27: include of <linux/types.h> is preferred over <asm/types.h>
  usr/include/linux/dvb/dmx.h:90: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/dvb/dmx.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/dvb/dmx.h b/include/linux/dvb/dmx.h
index 402fb7a8d922..fef943738a24 100644
--- a/include/linux/dvb/dmx.h
+++ b/include/linux/dvb/dmx.h
@@ -24,7 +24,7 @@
 #ifndef _DVBDMX_H_
 #define _DVBDMX_H_
 
-#include <asm/types.h>
+#include <linux/types.h>
 #ifdef __KERNEL__
 #include <linux/time.h>
 #else
-- 
cgit v1.2.3


From de189f078ee4ae74944e6827dff184a3ef1fc89b Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 19:56:41 +0530
Subject: headers_check fix: dvb/frontend.h

fix the following 'make headers_check' warnings:

  usr/include/linux/dvb/frontend.h:29: include of <linux/types.h> is preferred over <asm/types.h>
  usr/include/linux/dvb/frontend.h:76: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/dvb/frontend.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/dvb/frontend.h b/include/linux/dvb/frontend.h
index 55026b1a40bd..51c8d2d49e42 100644
--- a/include/linux/dvb/frontend.h
+++ b/include/linux/dvb/frontend.h
@@ -26,8 +26,7 @@
 #ifndef _DVBFRONTEND_H_
 #define _DVBFRONTEND_H_
 
-#include <asm/types.h>
-
+#include <linux/types.h>
 
 typedef enum fe_type {
 	FE_QPSK,
-- 
cgit v1.2.3


From 8996be9de98a9362a3192b866dd8ab9930e28ad9 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 19:58:00 +0530
Subject: headers_check fix: dvb/net.h

fix the following 'make headers_check' warnings:

  usr/include/linux/dvb/net.h:27: include of <linux/types.h> is preferred over <asm/types.h>
  usr/include/linux/dvb/net.h:31: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/dvb/net.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/dvb/net.h b/include/linux/dvb/net.h
index 5be474bf0d2b..f451e7eb0b0b 100644
--- a/include/linux/dvb/net.h
+++ b/include/linux/dvb/net.h
@@ -24,8 +24,7 @@
 #ifndef _DVBNET_H_
 #define _DVBNET_H_
 
-#include <asm/types.h>
-
+#include <linux/types.h>
 
 struct dvb_net_if {
 	__u16 pid;
-- 
cgit v1.2.3


From b852d36b86902abb272b0f2dd7a56dd2d17ea88c Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 19:59:53 +0530
Subject: headers_check fix: dvb/video.h

fix the following 'make headers_check' warnings:

  usr/include/linux/dvb/video.h:29: include of <linux/types.h> is preferred over <asm/types.h>
  usr/include/linux/dvb/video.h:102: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/dvb/video.h | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/dvb/video.h b/include/linux/dvb/video.h
index 50839fe9e39e..bd49c3ebf916 100644
--- a/include/linux/dvb/video.h
+++ b/include/linux/dvb/video.h
@@ -24,17 +24,14 @@
 #ifndef _DVBVIDEO_H_
 #define _DVBVIDEO_H_
 
-#include <linux/compiler.h>
-
-#ifdef __KERNEL__
 #include <linux/types.h>
+#ifdef __KERNEL__
+#include <linux/compiler.h>
 #else
-#include <asm/types.h>
 #include <stdint.h>
 #include <time.h>
 #endif
 
-
 typedef enum {
 	VIDEO_FORMAT_4_3,     /* Select 4:3 format */
 	VIDEO_FORMAT_16_9,    /* Select 16:9 format. */
-- 
cgit v1.2.3


From 9df27bab62e60d1f786abd0599af4a5e3192a784 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:00:47 +0530
Subject: headers_check fix: netfilter/xt_conntrack.h

fix the following 'make headers_check' warning:

  usr/include/linux/netfilter/xt_conntrack.h:40: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/netfilter/xt_conntrack.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/netfilter/xt_conntrack.h b/include/linux/netfilter/xt_conntrack.h
index f3fd83e46bab..8f5345275393 100644
--- a/include/linux/netfilter/xt_conntrack.h
+++ b/include/linux/netfilter/xt_conntrack.h
@@ -5,6 +5,7 @@
 #ifndef _XT_CONNTRACK_H
 #define _XT_CONNTRACK_H
 
+#include <linux/types.h>
 #include <linux/netfilter/nf_conntrack_tuple_common.h>
 
 #define XT_CONNTRACK_STATE_BIT(ctinfo) (1 << ((ctinfo)%IP_CT_IS_REPLY+1))
-- 
cgit v1.2.3


From 3187cedf158687432cdf152eeee205f7b149f4ef Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:03:06 +0530
Subject: headers_check fix: nfsd/export.h

fix the following 'make headers_check' warning:

  usr/include/linux/nfsd/export.h:13: include of <linux/types.h> is preferred over <asm/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/nfsd/export.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/nfsd/export.h b/include/linux/nfsd/export.h
index 5431512b2757..bcd0201589f8 100644
--- a/include/linux/nfsd/export.h
+++ b/include/linux/nfsd/export.h
@@ -10,9 +10,8 @@
 #ifndef NFSD_EXPORT_H
 #define NFSD_EXPORT_H
 
-#include <asm/types.h>
-#ifdef __KERNEL__
 # include <linux/types.h>
+#ifdef __KERNEL__
 # include <linux/in.h>
 #endif
 
-- 
cgit v1.2.3


From 9e87b1e53f3c72c1196dc22cb359b5d6188a3729 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:04:03 +0530
Subject: headers_check fix: nfsd/nfsfh.h

fix the following 'make headers_check' warnings:

  usr/include/linux/nfsd/nfsfh.h:17: include of <linux/types.h> is preferred over <asm/types.h>
  usr/include/linux/nfsd/nfsfh.h:28: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/nfsd/nfsfh.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/nfsd/nfsfh.h b/include/linux/nfsd/nfsfh.h
index b2e093870bc6..fa317f6c154b 100644
--- a/include/linux/nfsd/nfsfh.h
+++ b/include/linux/nfsd/nfsfh.h
@@ -14,9 +14,8 @@
 #ifndef _LINUX_NFSD_FH_H
 #define _LINUX_NFSD_FH_H
 
-#include <asm/types.h>
-#ifdef __KERNEL__
 # include <linux/types.h>
+#ifdef __KERNEL__
 # include <linux/string.h>
 # include <linux/fs.h>
 #endif
-- 
cgit v1.2.3


From 03cf1e0c3b4ee4ef51dc7eb197a4d098ad4873af Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:05:27 +0530
Subject: headers_check fix: nfsd/syscall.h

fix the following 'make headers_check' warnings:

  usr/include/linux/nfsd/syscall.h:12: include of <linux/types.h> is preferred over <asm/types.h>
  usr/include/linux/nfsd/syscall.h:104: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/nfsd/syscall.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/nfsd/syscall.h b/include/linux/nfsd/syscall.h
index 4e439765b705..7a3b565b898f 100644
--- a/include/linux/nfsd/syscall.h
+++ b/include/linux/nfsd/syscall.h
@@ -9,9 +9,8 @@
 #ifndef NFSD_SYSCALL_H
 #define NFSD_SYSCALL_H
 
-#include <asm/types.h>
-#ifdef __KERNEL__
 # include <linux/types.h>
+#ifdef __KERNEL__
 # include <linux/in.h>
 #endif 
 #include <linux/posix_types.h>
-- 
cgit v1.2.3


From bcf74582af3feca80ec96cc21d0a26c938d1863e Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:06:44 +0530
Subject: headers_check fix: raid/md_p.h

fix the following 'make headers_check' warning:

  usr/include/linux/raid/md_p.h:85: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/raid/md_p.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/raid/md_p.h b/include/linux/raid/md_p.h
index 9491026afe66..6ba830fa8538 100644
--- a/include/linux/raid/md_p.h
+++ b/include/linux/raid/md_p.h
@@ -15,6 +15,8 @@
 #ifndef _MD_P_H
 #define _MD_P_H
 
+#include <linux/types.h>
+
 /*
  * RAID superblock.
  *
-- 
cgit v1.2.3


From 550e978aa52e2ac3c493e8a0b36b368ade6dd2b4 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:07:35 +0530
Subject: headers_check fix: spi/spidev.h

fix the following 'make headers_check' warning:

  usr/include/linux/spi/spidev.h:83: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/spi/spidev.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/spi/spidev.h b/include/linux/spi/spidev.h
index c93ef9d42a01..95251ccd5a07 100644
--- a/include/linux/spi/spidev.h
+++ b/include/linux/spi/spidev.h
@@ -22,6 +22,7 @@
 #ifndef SPIDEV_H
 #define SPIDEV_H
 
+#include <linux/types.h>
 
 /* User space versions of kernel symbols for SPI clocking modes,
  * matching <linux/spi/spi.h>
-- 
cgit v1.2.3


From 2d594c0c8aa46beb21be1c5c2b7141f89d206313 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:10:05 +0530
Subject: headers_check fix: tc_act/tc_gact.h

fix the following 'make headers_check' warning:

  usr/include/linux/tc_act/tc_gact.h:19: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/tc_act/tc_gact.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/tc_act/tc_gact.h b/include/linux/tc_act/tc_gact.h
index 23a03eb630db..e895c0a39629 100644
--- a/include/linux/tc_act/tc_gact.h
+++ b/include/linux/tc_act/tc_gact.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_TC_GACT_H
 #define __LINUX_TC_GACT_H
 
+#include <linux/types.h>
 #include <linux/pkt_cls.h>
 
 #define TCA_ACT_GACT 5
-- 
cgit v1.2.3


From 9c536d275823b8a6281894f4f8c2687f60578253 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:10:44 +0530
Subject: headers_check fix: tc_act/tc_mirred.h

fix the following 'make headers_check' warning:

  usr/include/linux/tc_act/tc_mirred.h:16: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/tc_act/tc_mirred.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/tc_act/tc_mirred.h b/include/linux/tc_act/tc_mirred.h
index 71d63409d568..0a99ab60d610 100644
--- a/include/linux/tc_act/tc_mirred.h
+++ b/include/linux/tc_act/tc_mirred.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_TC_MIR_H
 #define __LINUX_TC_MIR_H
 
+#include <linux/types.h>
 #include <linux/pkt_cls.h>
 
 #define TCA_ACT_MIRRED 8
-- 
cgit v1.2.3


From 5dbbf3bcae2f6b5dee1c33b3eeced00bcb6c4f71 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:11:22 +0530
Subject: headers_check fix: tc_act/tc_pedit.h

fix the following 'make headers_check' warning:

  usr/include/linux/tc_act/tc_pedit.h:19: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/tc_act/tc_pedit.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/tc_act/tc_pedit.h b/include/linux/tc_act/tc_pedit.h
index 83e56e32e8e0..54ce9064115a 100644
--- a/include/linux/tc_act/tc_pedit.h
+++ b/include/linux/tc_act/tc_pedit.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_TC_PED_H
 #define __LINUX_TC_PED_H
 
+#include <linux/types.h>
 #include <linux/pkt_cls.h>
 
 #define TCA_ACT_PEDIT 7
-- 
cgit v1.2.3


From ba3a51e3b899c1bd34c18f84a1c6f7e5f99be850 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:12:01 +0530
Subject: headers_check fix: tc_ematch/tc_em_cmp.h

fix the following 'make headers_check' warning:

  usr/include/linux/tc_ematch/tc_em_cmp.h:8: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/tc_ematch/tc_em_cmp.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/tc_ematch/tc_em_cmp.h b/include/linux/tc_ematch/tc_em_cmp.h
index c7f4d43618fd..38e7f7b25ec2 100644
--- a/include/linux/tc_ematch/tc_em_cmp.h
+++ b/include/linux/tc_ematch/tc_em_cmp.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_TC_EM_CMP_H
 #define __LINUX_TC_EM_CMP_H
 
+#include <linux/types.h>
 #include <linux/pkt_cls.h>
 
 struct tcf_em_cmp
-- 
cgit v1.2.3


From 9976007a13e67b973f94c8d472ed615ac4cf8078 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:12:59 +0530
Subject: headers_check fix: tc_ematch/tc_em_meta.h

fix the following 'make headers_check' warning:

  usr/include/linux/tc_ematch/tc_em_meta.h:18: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/tc_ematch/tc_em_meta.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/tc_ematch/tc_em_meta.h b/include/linux/tc_ematch/tc_em_meta.h
index c50d2ba5caf0..dcfb733fa1f6 100644
--- a/include/linux/tc_ematch/tc_em_meta.h
+++ b/include/linux/tc_ematch/tc_em_meta.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_TC_EM_META_H
 #define __LINUX_TC_EM_META_H
 
+#include <linux/types.h>
 #include <linux/pkt_cls.h>
 
 enum
-- 
cgit v1.2.3


From ac836c6f1b17f674e35a7e2784541bb8ab0bce38 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:13:41 +0530
Subject: headers_check fix: tc_ematch/tc_em_nbyte.h

fix the following 'make headers_check' warning:

  usr/include/linux/tc_ematch/tc_em_nbyte.h:8: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/tc_ematch/tc_em_nbyte.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/tc_ematch/tc_em_nbyte.h b/include/linux/tc_ematch/tc_em_nbyte.h
index f19d1f58ec9d..9ed8c2e58488 100644
--- a/include/linux/tc_ematch/tc_em_nbyte.h
+++ b/include/linux/tc_ematch/tc_em_nbyte.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_TC_EM_NBYTE_H
 #define __LINUX_TC_EM_NBYTE_H
 
+#include <linux/types.h>
 #include <linux/pkt_cls.h>
 
 struct tcf_em_nbyte
-- 
cgit v1.2.3


From 30f410a6d372f067df3d02e3db328720bf421c81 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:14:36 +0530
Subject: headers_check fix: tc_ematch/tc_em_text.h

fix the following 'make headers_check' warning:

  usr/include/linux/tc_ematch/tc_em_text.h:11: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/tc_ematch/tc_em_text.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/tc_ematch/tc_em_text.h b/include/linux/tc_ematch/tc_em_text.h
index 7cd43e99c7f5..d12a73a225fc 100644
--- a/include/linux/tc_ematch/tc_em_text.h
+++ b/include/linux/tc_ematch/tc_em_text.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_TC_EM_TEXT_H
 #define __LINUX_TC_EM_TEXT_H
 
+#include <linux/types.h>
 #include <linux/pkt_cls.h>
 
 #define TC_EM_TEXT_ALGOSIZ	16
-- 
cgit v1.2.3


From d8151585690d824ac5b60a94ef86f8bfd61478fa Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:15:47 +0530
Subject: headers_check fix: usb/cdc.h

fix the following 'make headers_check' warning:

  usr/include/linux/usb/cdc.h:50: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/usb/cdc.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/usb/cdc.h b/include/linux/usb/cdc.h
index 18a729343ffa..3c86ed25a04c 100644
--- a/include/linux/usb/cdc.h
+++ b/include/linux/usb/cdc.h
@@ -9,6 +9,8 @@
 #ifndef __LINUX_USB_CDC_H
 #define __LINUX_USB_CDC_H
 
+#include <linux/types.h>
+
 #define USB_CDC_SUBCLASS_ACM			0x02
 #define USB_CDC_SUBCLASS_ETHERNET		0x06
 #define USB_CDC_SUBCLASS_WHCM			0x08
-- 
cgit v1.2.3


From 4c866d444078d931579c50c9ce3133709390287b Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:16:33 +0530
Subject: headers_check fix: usb/gadgetfs.h

fix the following 'make headers_check' warning:

  usr/include/linux/usb/gadgetfs.h:21: include of <linux/types.h> is preferred over <asm/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/usb/gadgetfs.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/usb/gadgetfs.h b/include/linux/usb/gadgetfs.h
index ea45f265ec05..612102e4d75e 100644
--- a/include/linux/usb/gadgetfs.h
+++ b/include/linux/usb/gadgetfs.h
@@ -18,7 +18,7 @@
 #ifndef __LINUX_USB_GADGETFS_H
 #define __LINUX_USB_GADGETFS_H
 
-#include <asm/types.h>
+#include <linux/types.h>
 #include <asm/ioctl.h>
 
 #include <linux/usb/ch9.h>
-- 
cgit v1.2.3


From bd247b348aaa9f28a53a64df06c69d6f40ff2280 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:20:10 +0530
Subject: headers_check fix: linux/aio_abi.h

fix the following 'make headers_check' warning:

  usr/include/linux/aio_abi.h:58: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/aio_abi.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/aio_abi.h b/include/linux/aio_abi.h
index 9e0172931315..2c8731664180 100644
--- a/include/linux/aio_abi.h
+++ b/include/linux/aio_abi.h
@@ -27,6 +27,7 @@
 #ifndef __LINUX__AIO_ABI_H
 #define __LINUX__AIO_ABI_H
 
+#include <linux/types.h>
 #include <asm/byteorder.h>
 
 typedef unsigned long	aio_context_t;
-- 
cgit v1.2.3


From 85c09569e563cbb9376f10da20ada42107dfef98 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:26:25 +0530
Subject: headers_check fix: linux/atalk.h

fix the following 'make headers_check' warning:

  usr/include/linux/atalk.h:15: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/atalk.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/atalk.h b/include/linux/atalk.h
index e9ebac2e2ecc..d34c187432ed 100644
--- a/include/linux/atalk.h
+++ b/include/linux/atalk.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_ATALK_H__
 #define __LINUX_ATALK_H__
 
+#include <linux/types.h>
 #include <asm/byteorder.h>
 
 /*
-- 
cgit v1.2.3


From f757f603f7d52254120cbfcd967f67f663264c64 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:29:11 +0530
Subject: headers_check fix: linux/atmbr2684.h

fix the following 'make headers_check' warning:

  usr/include/linux/atmbr2684.h:88: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/atmbr2684.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/atmbr2684.h b/include/linux/atmbr2684.h
index 52bf72affbba..fdb2629b6189 100644
--- a/include/linux/atmbr2684.h
+++ b/include/linux/atmbr2684.h
@@ -1,6 +1,7 @@
 #ifndef _LINUX_ATMBR2684_H
 #define _LINUX_ATMBR2684_H
 
+#include <linux/types.h>
 #include <linux/atm.h>
 #include <linux/if.h>		/* For IFNAMSIZ */
 
-- 
cgit v1.2.3


From 5d461bfebe4be9ae8d25d4570d4eaa415ca76f0f Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:31:19 +0530
Subject: headers_check fix: linux/auto_fs4.h

fix the following 'make headers_check' warning:

  usr/include/linux/auto_fs4.h:132: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/auto_fs4.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/auto_fs4.h b/include/linux/auto_fs4.h
index 55fa478bd639..8b49ac48a5b7 100644
--- a/include/linux/auto_fs4.h
+++ b/include/linux/auto_fs4.h
@@ -12,6 +12,7 @@
 #define _LINUX_AUTO_FS4_H
 
 /* Include common v3 definitions */
+#include <linux/types.h>
 #include <linux/auto_fs.h>
 
 /* autofs v4 definitions */
-- 
cgit v1.2.3


From 1da9ebd5abb2e960c4ca4d49f7587e6c76b16ac0 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:34:15 +0530
Subject: headers_check fix: linux/bfs_fs.h

fix the following 'make headers_check' warning:

  usr/include/linux/bfs_fs.h:24: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/bfs_fs.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/bfs_fs.h b/include/linux/bfs_fs.h
index 8ed6dfdcd783..1c0b355aa515 100644
--- a/include/linux/bfs_fs.h
+++ b/include/linux/bfs_fs.h
@@ -6,6 +6,8 @@
 #ifndef _LINUX_BFS_FS_H
 #define _LINUX_BFS_FS_H
 
+#include <linux/types.h>
+
 #define BFS_BSIZE_BITS		9
 #define BFS_BSIZE		(1<<BFS_BSIZE_BITS)
 
@@ -17,7 +19,6 @@
 #define BFS_VDIR 2L
 #define BFS_VREG 1L
 
-
 /* BFS inode layout on disk */
 struct bfs_inode {
 	__le16 i_ino;
-- 
cgit v1.2.3


From 4502b80e44f1fc9af33f66053c6c99ae9dba32a6 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:36:52 +0530
Subject: headers_check fix: linux/blktrace_api.h

fix the following 'make headers_check' warning:

  usr/include/linux/blktrace_api.h:96: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/blktrace_api.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h
index 1dba3493d520..25379cba2370 100644
--- a/include/linux/blktrace_api.h
+++ b/include/linux/blktrace_api.h
@@ -1,6 +1,7 @@
 #ifndef BLKTRACE_H
 #define BLKTRACE_H
 
+#include <linux/types.h>
 #ifdef __KERNEL__
 #include <linux/blkdev.h>
 #include <linux/relay.h>
-- 
cgit v1.2.3


From 9fa91d99bfdd9582e43b6b9ab97678c51373c4ae Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:39:30 +0530
Subject: headers_check fix: linux/capability.h

fix the following 'make headers_check' warning:

  usr/include/linux/capability.h:73: extern's make no sense in userspace

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/capability.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/capability.h b/include/linux/capability.h
index 02bdb768d43b..1b9872556131 100644
--- a/include/linux/capability.h
+++ b/include/linux/capability.h
@@ -69,10 +69,6 @@ typedef struct __user_cap_data_struct {
 #define VFS_CAP_U32             VFS_CAP_U32_2
 #define VFS_CAP_REVISION	VFS_CAP_REVISION_2
 
-#ifdef CONFIG_SECURITY_FILE_CAPABILITIES
-extern int file_caps_enabled;
-#endif
-
 struct vfs_cap_data {
 	__le32 magic_etc;            /* Little endian */
 	struct {
@@ -96,6 +92,10 @@ struct vfs_cap_data {
 #define _KERNEL_CAPABILITY_VERSION _LINUX_CAPABILITY_VERSION_3
 #define _KERNEL_CAPABILITY_U32S    _LINUX_CAPABILITY_U32S_3
 
+#ifdef CONFIG_SECURITY_FILE_CAPABILITIES
+extern int file_caps_enabled;
+#endif
+
 typedef struct kernel_cap_struct {
 	__u32 cap[_KERNEL_CAPABILITY_U32S];
 } kernel_cap_t;
-- 
cgit v1.2.3


From 960066a919f1db57817df6d02e72b01542f1deed Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:41:51 +0530
Subject: headers_check fix: linux/cdrom.h

fix the following 'make headers_check' warning:

  usr/include/linux/cdrom.h:155: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/cdrom.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/cdrom.h b/include/linux/cdrom.h
index 0b49e08d3cb0..78e904796622 100644
--- a/include/linux/cdrom.h
+++ b/include/linux/cdrom.h
@@ -11,6 +11,7 @@
 #ifndef	_LINUX_CDROM_H
 #define	_LINUX_CDROM_H
 
+#include <linux/types.h>
 #include <asm/byteorder.h>
 
 /*******************************************************
-- 
cgit v1.2.3


From 59e4cf19ede2d2725c1b336707c1077afdd3cf85 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:43:31 +0530
Subject: headers_check fix: linux/cgroupstats.h

fix the following 'make headers_check' warning:

  usr/include/linux/cgroupstats.h:31: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/cgroupstats.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/cgroupstats.h b/include/linux/cgroupstats.h
index 4f53abf6855d..3753c33160d1 100644
--- a/include/linux/cgroupstats.h
+++ b/include/linux/cgroupstats.h
@@ -15,6 +15,7 @@
 #ifndef _LINUX_CGROUPSTATS_H
 #define _LINUX_CGROUPSTATS_H
 
+#include <linux/types.h>
 #include <linux/taskstats.h>
 
 /*
-- 
cgit v1.2.3


From 37eb1f4c3320ed505fbe59a916635b2342c740e4 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:51:30 +0530
Subject: headers_check fix: linux/dlm_plock.h

fix the following 'make headers_check' warning:

  usr/include/linux/dlm_plock.h:25: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/dlm_plock.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/dlm_plock.h b/include/linux/dlm_plock.h
index 18d5fdbceb74..2dd21243104f 100644
--- a/include/linux/dlm_plock.h
+++ b/include/linux/dlm_plock.h
@@ -9,6 +9,8 @@
 #ifndef __DLM_PLOCK_DOT_H__
 #define __DLM_PLOCK_DOT_H__
 
+#include <linux/types.h>
+
 #define DLM_PLOCK_MISC_NAME		"dlm_plock"
 
 #define DLM_PLOCK_VERSION_MAJOR	1
-- 
cgit v1.2.3


From 57d1780fab89d3736de0d24189129c17178448f0 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:52:54 +0530
Subject: headers_check fix: linux/dn.h

fix the following 'make headers_check' warning:

  usr/include/linux/dn.h:75: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/dn.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/dn.h b/include/linux/dn.h
index 02bba040fcfb..fe9990823193 100644
--- a/include/linux/dn.h
+++ b/include/linux/dn.h
@@ -1,6 +1,8 @@
 #ifndef _LINUX_DN_H
 #define _LINUX_DN_H
 
+#include <linux/types.h>
+
 /*
 
 	DECnet Data Structures and Constants
-- 
cgit v1.2.3


From 4144147081b9d08e69055a780888fcbb7cfcbb8e Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 20:54:27 +0530
Subject: headers_check fix: linux/edd.h

fix the following 'make headers_check' warning:

  usr/include/linux/edd.h:70: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/edd.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/edd.h b/include/linux/edd.h
index 5d747c5cd0fe..4cbd0fe9df08 100644
--- a/include/linux/edd.h
+++ b/include/linux/edd.h
@@ -30,6 +30,8 @@
 #ifndef _LINUX_EDD_H
 #define _LINUX_EDD_H
 
+#include <linux/types.h>
+
 #define EDDNR 0x1e9		/* addr of number of edd_info structs at EDDBUF
 				   in boot_params - treat this as 1 byte  */
 #define EDDBUF	0xd00		/* addr of edd_info structs in boot_params */
-- 
cgit v1.2.3


From bd71b5f734c66ad0134e308036b13d122907b8c6 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:01:11 +0530
Subject: headers_check fix: linux/efs_fs_sb.h

fix the following 'make headers_check' warning:

  usr/include/linux/efs_fs_sb.h:49: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/efs_fs_sb.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/efs_fs_sb.h b/include/linux/efs_fs_sb.h
index ff1945e37790..a01be90c58cc 100644
--- a/include/linux/efs_fs_sb.h
+++ b/include/linux/efs_fs_sb.h
@@ -9,6 +9,7 @@
 #ifndef __EFS_FS_SB_H__
 #define __EFS_FS_SB_H__
 
+#include <linux/types.h>
 #include <linux/magic.h>
 
 /* EFS superblock magic numbers */
-- 
cgit v1.2.3


From 177a858ff8d71a8e7f8b0ef53ff49441e29c8fb1 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:04:44 +0530
Subject: headers_check fix: linux/elf-fdpic.h

fix the following 'make headers_check' warning:

  usr/include/linux/elf-fdpic.h:62: extern's make no sense in userspace

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/elf-fdpic.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/elf-fdpic.h b/include/linux/elf-fdpic.h
index 9f5b7456bff3..7cd2e80cebc8 100644
--- a/include/linux/elf-fdpic.h
+++ b/include/linux/elf-fdpic.h
@@ -58,11 +58,13 @@ struct elf_fdpic_params {
 #define ELF_FDPIC_FLAG_PRESENT		0x80000000	/* T if this object is present */
 };
 
+#ifdef __KERNEL__
 #ifdef CONFIG_MMU
 extern void elf_fdpic_arch_lay_out_mm(struct elf_fdpic_params *exec_params,
 				      struct elf_fdpic_params *interp_params,
 				      unsigned long *start_stack,
 				      unsigned long *start_brk);
 #endif
+#endif /* __KERNEL__ */
 
 #endif /* _LINUX_ELF_FDPIC_H */
-- 
cgit v1.2.3


From f4aa1c30255278b7b50a1cd273c7b4a46f099a90 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:05:50 +0530
Subject: headers_check fix: linux/elf.h

fix the following 'make headers_check' warnings:

  usr/include/linux/elf.h:379: extern's make no sense in userspace
  usr/include/linux/elf.h:387: extern's make no sense in userspace
  usr/include/linux/elf.h:401: extern's make no sense in userspace
  usr/include/linux/elf.h:402: extern's make no sense in userspace

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/elf.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/elf.h b/include/linux/elf.h
index 0b61ca41a044..45a937be6d38 100644
--- a/include/linux/elf.h
+++ b/include/linux/elf.h
@@ -377,6 +377,7 @@ typedef struct elf64_note {
   Elf64_Word n_type;	/* Content type */
 } Elf64_Nhdr;
 
+#ifdef __KERNEL__
 #if ELF_CLASS == ELFCLASS32
 
 extern Elf32_Dyn _DYNAMIC [];
@@ -404,5 +405,5 @@ static inline int elf_coredump_extra_notes_write(struct file *file,
 extern int elf_coredump_extra_notes_size(void);
 extern int elf_coredump_extra_notes_write(struct file *file, loff_t *foffset);
 #endif
-
+#endif /* __KERNEL__ */
 #endif /* _LINUX_ELF_H */
-- 
cgit v1.2.3


From 93c1c0e310b56acbd366a43b15260a1775481f24 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:09:18 +0530
Subject: headers_check fix: linux/errqueue.h

fix the following 'make headers_check' warning:

  usr/include/linux/errqueue.h:6: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/errqueue.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/errqueue.h b/include/linux/errqueue.h
index 92f8d4fab32b..ceb1454b6977 100644
--- a/include/linux/errqueue.h
+++ b/include/linux/errqueue.h
@@ -1,6 +1,8 @@
 #ifndef _LINUX_ERRQUEUE_H
 #define _LINUX_ERRQUEUE_H 1
 
+#include <linux/types.h>
+
 struct sock_extended_err
 {
 	__u32	ee_errno;	
-- 
cgit v1.2.3


From 985f302cb42e18912c88a3d2f9d9008844045ee3 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:10:52 +0530
Subject: headers_check fix: linux/genetlink.h

fix the following 'make headers_check' warning:

  usr/include/linux/genetlink.h:12: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/genetlink.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/genetlink.h b/include/linux/genetlink.h
index 7da02c93002b..b834ef6d59fa 100644
--- a/include/linux/genetlink.h
+++ b/include/linux/genetlink.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_GENERIC_NETLINK_H
 #define __LINUX_GENERIC_NETLINK_H
 
+#include <linux/types.h>
 #include <linux/netlink.h>
 
 #define GENL_NAMSIZ	16	/* length of family name */
-- 
cgit v1.2.3


From 237416fe05067237f0bcc6370d84c09b52fb776a Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:12:38 +0530
Subject: headers_check fix: linux/gfs2_ondisk.h

fix the following 'make headers_check' warning:

  usr/include/linux/gfs2_ondisk.h:109: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/gfs2_ondisk.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/gfs2_ondisk.h b/include/linux/gfs2_ondisk.h
index 14d0df0b5749..c56b4bce56d0 100644
--- a/include/linux/gfs2_ondisk.h
+++ b/include/linux/gfs2_ondisk.h
@@ -10,6 +10,8 @@
 #ifndef __GFS2_ONDISK_DOT_H__
 #define __GFS2_ONDISK_DOT_H__
 
+#include <linux/types.h>
+
 #define GFS2_MAGIC		0x01161970
 #define GFS2_BASIC_BLOCK	512
 #define GFS2_BASIC_BLOCK_SHIFT	9
-- 
cgit v1.2.3


From b08ead0527bcfdcab39a347b531701289485b484 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:15:34 +0530
Subject: headers_check fix: linux/hid.h

fix the following 'make headers_check' warnings:

  usr/include/linux/hid.h:69: extern's make no sense in userspace
  usr/include/linux/hid.h:76: extern's make no sense in userspace

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/hid.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/hid.h b/include/linux/hid.h
index 81aa84d60c6b..fa8ee9cef7be 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -791,6 +791,7 @@ dbg_hid(const char *fmt, ...)
 		__FILE__ , ## arg)
 #endif /* HID_FF */
 
+#ifdef __KERNEL__
 #ifdef CONFIG_HID_COMPAT
 #define HID_COMPAT_LOAD_DRIVER(name)	\
 /* prototype to avoid sparse warning */	\
@@ -804,6 +805,7 @@ EXPORT_SYMBOL(hid_compat_##name)
 	extern void hid_compat_##name(void);	\
 	hid_compat_##name();			\
 } while (0)
+#endif /* __KERNEL__ */
 
 #endif
 
-- 
cgit v1.2.3


From c244ae5b16dc31b5bea67e6d6e9d6ff654aee781 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:17:06 +0530
Subject: headers_check fix: linux/hiddev.h

fix the following 'make headers_check' warning:

  usr/include/linux/hiddev.h:40: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/hiddev.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/hiddev.h b/include/linux/hiddev.h
index c760ae0eb6a1..bb6f58baf319 100644
--- a/include/linux/hiddev.h
+++ b/include/linux/hiddev.h
@@ -27,6 +27,8 @@
  * Vojtech Pavlik, Ucitelska 1576, Prague 8, 182 00 Czech Republic
  */
 
+#include <linux/types.h>
+
 /*
  * The event structure itself
  */
-- 
cgit v1.2.3


From 1cc49ae2e6d241e5cfc2c52e3329f5ef8dd42f18 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:18:37 +0530
Subject: headers_check fix: linux/icmpv6.h

fix the following 'make headers_check' warning:

  usr/include/linux/icmpv6.h:8: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/icmpv6.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h
index a93a8dd33118..10d701eec484 100644
--- a/include/linux/icmpv6.h
+++ b/include/linux/icmpv6.h
@@ -1,6 +1,7 @@
 #ifndef _LINUX_ICMPV6_H
 #define _LINUX_ICMPV6_H
 
+#include <linux/types.h>
 #include <asm/byteorder.h>
 
 struct icmp6hdr {
-- 
cgit v1.2.3


From 680ee0bd2a9625965812c1209476168fd0704a00 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:21:01 +0530
Subject: headers_check fix: linux/if_addr.h

fix the following 'make headers_check' warning:

  usr/include/linux/if_addr.h:8: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/if_addr.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/if_addr.h b/include/linux/if_addr.h
index 43f3bedaafd3..a60c821be44c 100644
--- a/include/linux/if_addr.h
+++ b/include/linux/if_addr.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_IF_ADDR_H
 #define __LINUX_IF_ADDR_H
 
+#include <linux/types.h>
 #include <linux/netlink.h>
 
 struct ifaddrmsg
-- 
cgit v1.2.3


From 1759cb994c3ff51e69268379da1cdd96048a8268 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:25:00 +0530
Subject: headers_check fix: linux/if_addrlabel.h

fix the following 'make headers_check' warning:

  usr/include/linux/if_addrlabel.h:15: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/if_addrlabel.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/if_addrlabel.h b/include/linux/if_addrlabel.h
index 9fe79c95dd28..89571f65d6de 100644
--- a/include/linux/if_addrlabel.h
+++ b/include/linux/if_addrlabel.h
@@ -10,6 +10,8 @@
 #ifndef __LINUX_IF_ADDRLABEL_H
 #define __LINUX_IF_ADDRLABEL_H
 
+#include <linux/types.h>
+
 struct ifaddrlblmsg
 {
 	__u8		ifal_family;		/* Address family */
-- 
cgit v1.2.3


From ba7161387e82fbbdc4b49533aa1345bb7befda13 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:26:35 +0530
Subject: headers_check fix: linux/if_fc.h

fix the following 'make headers_check' warning:

  usr/include/linux/if_fc.h:37: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/if_fc.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/if_fc.h b/include/linux/if_fc.h
index 376a34ea4723..6ed7f1bf35c8 100644
--- a/include/linux/if_fc.h
+++ b/include/linux/if_fc.h
@@ -20,6 +20,7 @@
 #ifndef _LINUX_IF_FC_H
 #define _LINUX_IF_FC_H
 
+#include <linux/types.h>
 
 #define FC_ALEN	6		/* Octets in one ethernet addr	 */
 #define FC_HLEN   (sizeof(struct fch_hdr)+sizeof(struct fcllc))
-- 
cgit v1.2.3


From b06e936939931c5acb1ca5dfe1d02b4d2f7cb11f Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:27:59 +0530
Subject: headers_check fix: linux/if_hippi.h

fix the following 'make headers_check' warning:

  usr/include/linux/if_hippi.h:82: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/if_hippi.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/if_hippi.h b/include/linux/if_hippi.h
index f0f23516bb59..4a7c9940b080 100644
--- a/include/linux/if_hippi.h
+++ b/include/linux/if_hippi.h
@@ -22,6 +22,7 @@
 #ifndef _LINUX_IF_HIPPI_H
 #define _LINUX_IF_HIPPI_H
 
+#include <linux/types.h>
 #include <asm/byteorder.h>
 
 /*
-- 
cgit v1.2.3


From 85db53102dbf0816e9c5426c9322a64759e7166b Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:29:20 +0530
Subject: headers_check fix: linux/if_link.h

fix the following 'make headers_check' warning:

  usr/include/linux/if_link.h:9: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/if_link.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/if_link.h b/include/linux/if_link.h
index f9032c88716a..176c5182c515 100644
--- a/include/linux/if_link.h
+++ b/include/linux/if_link.h
@@ -1,6 +1,7 @@
 #ifndef _LINUX_IF_LINK_H
 #define _LINUX_IF_LINK_H
 
+#include <linux/types.h>
 #include <linux/netlink.h>
 
 /* The struct should be in sync with struct net_device_stats */
-- 
cgit v1.2.3


From 0fe5a8fe0c145a6ff8f3daacd32f1824d0c021a9 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:31:29 +0530
Subject: headers_check fix: linux/if_ppp.h

fix the following 'make headers_check' warning:

  usr/include/linux/if_ppp.h:96: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/if_ppp.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/if_ppp.h b/include/linux/if_ppp.h
index c3b1f8562709..fcef103aa3f6 100644
--- a/include/linux/if_ppp.h
+++ b/include/linux/if_ppp.h
@@ -33,6 +33,7 @@
 #ifndef _IF_PPP_H_
 #define _IF_PPP_H_
 
+#include <linux/types.h>
 #include <linux/compiler.h>
 
 /*
-- 
cgit v1.2.3


From 84ad40ebbaeb22fc665b1f307d32128c46e8d42d Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:32:52 +0530
Subject: headers_check fix: linux/if_strip.h

fix the following 'make headers_check' warning:

  usr/include/linux/if_strip.h:22: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/if_strip.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/if_strip.h b/include/linux/if_strip.h
index fb5c5c98442f..6526a6235832 100644
--- a/include/linux/if_strip.h
+++ b/include/linux/if_strip.h
@@ -18,6 +18,8 @@
 #ifndef __LINUX_STRIP_H
 #define __LINUX_STRIP_H
 
+#include <linux/types.h>
+
 typedef struct {
     __u8 c[6];
 } MetricomAddress;
-- 
cgit v1.2.3


From 65863dbc0833e06b905679f61450f05a68bae4c2 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:34:36 +0530
Subject: headers_check fix: linux/if_tr.h

fix the following 'make headers_check' warning:

  usr/include/linux/if_tr.h:37: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/if_tr.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/if_tr.h b/include/linux/if_tr.h
index 5bcec8b2c5e2..fc23aeb0f201 100644
--- a/include/linux/if_tr.h
+++ b/include/linux/if_tr.h
@@ -19,6 +19,7 @@
 #ifndef _LINUX_IF_TR_H
 #define _LINUX_IF_TR_H
 
+#include <linux/types.h>
 #include <asm/byteorder.h>	/* For __be16 */
 
 /* IEEE 802.5 Token-Ring magic constants.  The frame sizes omit the preamble
-- 
cgit v1.2.3


From de8b0bcafabfb4400aa028282293ce7d52307433 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:36:04 +0530
Subject: headers_check fix: linux/igmp.h

fix the following 'make headers_check' warning:

  usr/include/linux/igmp.h:31: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/igmp.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/igmp.h b/include/linux/igmp.h
index f734a0ba0698..92fbd8cbd68f 100644
--- a/include/linux/igmp.h
+++ b/include/linux/igmp.h
@@ -16,6 +16,7 @@
 #ifndef _LINUX_IGMP_H
 #define _LINUX_IGMP_H
 
+#include <linux/types.h>
 #include <asm/byteorder.h>
 
 /*
-- 
cgit v1.2.3


From 2df005b75ab910f789f099f81bb70b3aa37203a7 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:42:47 +0530
Subject: headers_check fix: linux/inet_diag.h

fix the following 'make headers_check' warning:

  usr/include/linux/inet_diag.h:16: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/inet_diag.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/inet_diag.h b/include/linux/inet_diag.h
index 6e8bc548635a..bc8c49022084 100644
--- a/include/linux/inet_diag.h
+++ b/include/linux/inet_diag.h
@@ -1,6 +1,8 @@
 #ifndef _INET_DIAG_H_
 #define _INET_DIAG_H_ 1
 
+#include <linux/types.h>
+
 /* Just some random number */
 #define TCPDIAG_GETSOCK 18
 #define DCCPDIAG_GETSOCK 19
-- 
cgit v1.2.3


From 217a2291570b1e4c28cb6e4cd099707e456a09b8 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:44:19 +0530
Subject: headers_check fix: linux/ip6_tunnel.h

fix the following 'make headers_check' warning:

  include/linux/ip6_tunnel.h:21: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/ip6_tunnel.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/ip6_tunnel.h b/include/linux/ip6_tunnel.h
index 1e7cc4af40de..acb9ad684d63 100644
--- a/include/linux/ip6_tunnel.h
+++ b/include/linux/ip6_tunnel.h
@@ -1,6 +1,8 @@
 #ifndef _IP6_TUNNEL_H
 #define _IP6_TUNNEL_H
 
+#include <linux/types.h>
+
 #define IPV6_TLV_TNL_ENCAP_LIMIT 4
 #define IPV6_DEFAULT_TNL_ENCAP_LIMIT 4
 
-- 
cgit v1.2.3


From 5c6aa2badf1b97ead5ffec8094f0c6236e0c07c5 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:46:43 +0530
Subject: headers_check fix: linux/ipv6.h

fix the following 'make headers_check' warning:

  usr/include/linux/ipv6.h:26: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/ipv6.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 0b816cae533e..476d9464ac82 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -1,6 +1,7 @@
 #ifndef _IPV6_H
 #define _IPV6_H
 
+#include <linux/types.h>
 #include <linux/in6.h>
 #include <asm/byteorder.h>
 
-- 
cgit v1.2.3


From e5144de521417b0f0eea74ece89acd437ecd32c9 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:47:52 +0530
Subject: headers_check fix: linux/ipv6_route.h

fix the following 'make headers_check' warning:

  usr/include/linux/ipv6_route.h:42: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/ipv6_route.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/ipv6_route.h b/include/linux/ipv6_route.h
index b323ff577967..1e7d8af2defe 100644
--- a/include/linux/ipv6_route.h
+++ b/include/linux/ipv6_route.h
@@ -13,6 +13,8 @@
 #ifndef _LINUX_IPV6_ROUTE_H
 #define _LINUX_IPV6_ROUTE_H
 
+#include <linux/types.h>
+
 #define RTF_DEFAULT	0x00010000	/* default - learned via ND	*/
 #define RTF_ALLONLINK	0x00020000	/* (deprecated and will be removed)
 					   fallback, no routers on link */
-- 
cgit v1.2.3


From d6d20f54847e27ed886e8285c208368ef3d42abb Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:49:07 +0530
Subject: headers_check fix: linux/ipx.h

fix the following 'make headers_check' warning:

  usr/include/linux/ipx.h:13: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/ipx.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/ipx.h b/include/linux/ipx.h
index eb19b4ea84f4..aabb1d294025 100644
--- a/include/linux/ipx.h
+++ b/include/linux/ipx.h
@@ -1,5 +1,6 @@
 #ifndef _IPX_H_
 #define _IPX_H_
+#include <linux/types.h>
 #include <linux/sockios.h>
 #include <linux/socket.h>
 #define IPX_NODE_LEN	6
-- 
cgit v1.2.3


From df9c04ed3ff455aa5cb7c4bcddf4544fe54cfa33 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:50:39 +0530
Subject: headers_check fix: linux/irda.h

fix the following 'make headers_check' warning:

  usr/include/linux/irda.h:127: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/irda.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/irda.h b/include/linux/irda.h
index 28f88ecba344..00bdad0e8515 100644
--- a/include/linux/irda.h
+++ b/include/linux/irda.h
@@ -25,6 +25,8 @@
 #ifndef KERNEL_IRDA_H
 #define KERNEL_IRDA_H
 
+#include <linux/types.h>
+
 /* Please do *not* add any #include in this file, this file is
  * included as-is in user space.
  * Please fix the calling file to properly included needed files before
-- 
cgit v1.2.3


From 4b7ae34277608a30346d076beb44cbc466aa73e5 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:54:23 +0530
Subject: headers_check fix: linux/minix_fs.h

fix the following 'make headers_check' warning:

  usr/include/linux/minix_fs.h:34: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/minix_fs.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/minix_fs.h b/include/linux/minix_fs.h
index 0e39745f5111..13fe09e0576a 100644
--- a/include/linux/minix_fs.h
+++ b/include/linux/minix_fs.h
@@ -1,6 +1,7 @@
 #ifndef _LINUX_MINIX_FS_H
 #define _LINUX_MINIX_FS_H
 
+#include <linux/types.h>
 #include <linux/magic.h>
 
 /*
-- 
cgit v1.2.3


From 8ef342021a55e4237e593c7f6304d0caa7bf1232 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:56:32 +0530
Subject: headers_check fix: linux/msdos_fs.h

fix the following 'make headers_check' warning:

  usr/include/linux/msdos_fs.h:100: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/msdos_fs.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/msdos_fs.h b/include/linux/msdos_fs.h
index e0a9b207920d..ce38f1caa5e1 100644
--- a/include/linux/msdos_fs.h
+++ b/include/linux/msdos_fs.h
@@ -1,6 +1,7 @@
 #ifndef _LINUX_MSDOS_FS_H
 #define _LINUX_MSDOS_FS_H
 
+#include <linux/types.h>
 #include <linux/magic.h>
 #include <asm/byteorder.h>
 
-- 
cgit v1.2.3


From ee79a6415f911801eb7804704ac130088281b2d8 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:58:19 +0530
Subject: headers_check fix: linux/neighbour.h

fix the following 'make headers_check' warning:

  usr/include/linux/neighbour.h:8: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/neighbour.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/neighbour.h b/include/linux/neighbour.h
index bd3bbf668cdb..8730d5dae1bc 100644
--- a/include/linux/neighbour.h
+++ b/include/linux/neighbour.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_NEIGHBOUR_H
 #define __LINUX_NEIGHBOUR_H
 
+#include <linux/types.h>
 #include <linux/netlink.h>
 
 struct ndmsg
-- 
cgit v1.2.3


From a81184c1f8cf8589a00894c20422982defc3f056 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 21:59:48 +0530
Subject: headers_check fix: linux/nfs_idmap.h

fix the following 'make headers_check' warning:

  usr/include/linux/nfs_idmap.h:55: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/nfs_idmap.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/nfs_idmap.h b/include/linux/nfs_idmap.h
index 15a9f3b7289a..91a1c24e0cbf 100644
--- a/include/linux/nfs_idmap.h
+++ b/include/linux/nfs_idmap.h
@@ -37,6 +37,8 @@
 #ifndef NFS_IDMAP_H
 #define NFS_IDMAP_H
 
+#include <linux/types.h>
+
 /* XXX from bits/utmp.h  */
 #define IDMAP_NAMESZ  128
 
-- 
cgit v1.2.3


From 06f43adba62f99de101616ffc5d0564aab237111 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:03:25 +0530
Subject: headers_check fix: linux/phonet.h

fix the following 'make headers_check' warning:

  usr/include/linux/phonet.h:50: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/phonet.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/phonet.h b/include/linux/phonet.h
index 4157faa857b6..ee5e3c9e2bca 100644
--- a/include/linux/phonet.h
+++ b/include/linux/phonet.h
@@ -23,6 +23,8 @@
 #ifndef LINUX_PHONET_H
 #define LINUX_PHONET_H
 
+#include <linux/types.h>
+
 /* Automatic protocol selection */
 #define PN_PROTO_TRANSPORT	0
 /* Phonet datagram socket */
-- 
cgit v1.2.3


From ed307444d8f276d7052400c47d9f4c5393997c42 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:05:32 +0530
Subject: headers_check fix: linux/pkt_cls.h

fix the following 'make headers_check' warning:

  linux/pkt_cls.h:122: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/pkt_cls.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/pkt_cls.h b/include/linux/pkt_cls.h
index e6aa8482ad7a..3c842edff388 100644
--- a/include/linux/pkt_cls.h
+++ b/include/linux/pkt_cls.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_PKT_CLS_H
 #define __LINUX_PKT_CLS_H
 
+#include <linux/types.h>
 #include <linux/pkt_sched.h>
 
 /* I think i could have done better macros ; for now this is stolen from
-- 
cgit v1.2.3


From b8adfd3c753b47c47f626e032da7999530c316f0 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:07:05 +0530
Subject: headers_check fix: linux/pkt_sched.h

fix the following 'make headers_check' warning:

  usr/include/linux/pkt_sched.h:32: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/pkt_sched.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h
index e3f133adba78..b2648e8e4987 100644
--- a/include/linux/pkt_sched.h
+++ b/include/linux/pkt_sched.h
@@ -1,6 +1,8 @@
 #ifndef __LINUX_PKT_SCHED_H
 #define __LINUX_PKT_SCHED_H
 
+#include <linux/types.h>
+
 /* Logical priority bands not depending on specific packet scheduler.
    Every scheduler will map them to real traffic classes, if it has
    no more precise mechanism to classify packets.
-- 
cgit v1.2.3


From 7260a91666a3149181e7b78bbf73beebbb04f8fa Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:09:17 +0530
Subject: headers_check fix: linux/ppp_defs.h

fix the following 'make headers_check' warning:

  usr/include/linux/ppp_defs.h:50: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/ppp_defs.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/ppp_defs.h b/include/linux/ppp_defs.h
index 6e8adc77522c..1c866bda2018 100644
--- a/include/linux/ppp_defs.h
+++ b/include/linux/ppp_defs.h
@@ -25,6 +25,8 @@
  * OR MODIFICATIONS.
  */
 
+#include <linux/types.h>
+
 /*
  *  ==FILEVERSION 20000114==
  *
-- 
cgit v1.2.3


From 68622c61dc7971382f5d69cd5d881e618ea30414 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:11:32 +0530
Subject: headers_check fix: linux/random.h

fix the following 'make headers_check' warning:

  usr/include/linux/random.h:39: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/random.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/random.h b/include/linux/random.h
index 407ea3646f8f..25d02fe5c9b5 100644
--- a/include/linux/random.h
+++ b/include/linux/random.h
@@ -7,6 +7,7 @@
 #ifndef _LINUX_RANDOM_H
 #define _LINUX_RANDOM_H
 
+#include <linux/types.h>
 #include <linux/ioctl.h>
 #include <linux/irqnr.h>
 
-- 
cgit v1.2.3


From a788fd53aec9a439f6b8bf57888c30aea1176e1b Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:14:02 +0530
Subject: headers_check fix: linux/signalfd.h

fix the following 'make headers_check' warning:

  usr/include/linux/signalfd.h:19: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/signalfd.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/signalfd.h b/include/linux/signalfd.h
index bef0c46d4713..b363b916c909 100644
--- a/include/linux/signalfd.h
+++ b/include/linux/signalfd.h
@@ -8,6 +8,7 @@
 #ifndef _LINUX_SIGNALFD_H
 #define _LINUX_SIGNALFD_H
 
+#include <linux/types.h>
 /* For O_CLOEXEC and O_NONBLOCK */
 #include <linux/fcntl.h>
 
-- 
cgit v1.2.3


From e6faa002be269233bf1e8961e7e0a79ca3f2db8b Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:17:53 +0530
Subject: headers_check fix: linux/sound.h

fix the following 'make headers_check' warnings:

  usr/include/linux/sound.h:33: extern's make no sense in userspace
  usr/include/linux/sound.h:34: extern's make no sense in userspace
  usr/include/linux/sound.h:35: extern's make no sense in userspace
  usr/include/linux/sound.h:36: extern's make no sense in userspace
  usr/include/linux/sound.h:37: extern's make no sense in userspace
  usr/include/linux/sound.h:39: extern's make no sense in userspace
  usr/include/linux/sound.h:40: extern's make no sense in userspace
  usr/include/linux/sound.h:41: extern's make no sense in userspace
  usr/include/linux/sound.h:42: extern's make no sense in userspace

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/sound.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/sound.h b/include/linux/sound.h
index 9e2a94feed6b..44dcf0570432 100644
--- a/include/linux/sound.h
+++ b/include/linux/sound.h
@@ -25,6 +25,7 @@
 #define SND_DEV_AMIDI		13	/* Like /dev/midi (obsolete) */
 #define SND_DEV_ADMMIDI		14	/* Like /dev/dmmidi (onsolete) */
 
+#ifdef __KERNEL__
 /*
  *	Sound core interface functions
  */
@@ -40,3 +41,4 @@ extern void unregister_sound_special(int unit);
 extern void unregister_sound_mixer(int unit);
 extern void unregister_sound_midi(int unit);
 extern void unregister_sound_dsp(int unit);
+#endif /* __KERNEL__ */
-- 
cgit v1.2.3


From 6b6bcd0ed953ae0ed73af4759788fb8384bbaeed Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:20:04 +0530
Subject: headers_check fix: linux/synclink.h

fix the following 'make headers_check' warning:

  usr/include/linux/synclink.h:209: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/synclink.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/synclink.h b/include/linux/synclink.h
index c844a229acc9..99b8bdb17b2b 100644
--- a/include/linux/synclink.h
+++ b/include/linux/synclink.h
@@ -13,6 +13,8 @@
 #define _SYNCLINK_H_
 #define SYNCLINK_H_VERSION 3.6
 
+#include <linux/types.h>
+
 #define BIT0	0x0001
 #define BIT1	0x0002
 #define BIT2	0x0004
-- 
cgit v1.2.3


From 448314fc968252b0b95f74bbe63fdcaf41e6413d Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:21:19 +0530
Subject: headers_check fix: linux/taskstats.h

fix the following 'make headers_check' warning:

  usr/include/linux/taskstats.h:44: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/taskstats.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/taskstats.h b/include/linux/taskstats.h
index 18269e956a71..341dddb55090 100644
--- a/include/linux/taskstats.h
+++ b/include/linux/taskstats.h
@@ -16,6 +16,8 @@
 #ifndef _LINUX_TASKSTATS_H
 #define _LINUX_TASKSTATS_H
 
+#include <linux/types.h>
+
 /* Format for per-task data returned to userland when
  *	- a task exits
  *	- listener requests stats for a task
-- 
cgit v1.2.3


From 8b1e3a2f7f84484a8c208671adac39eb148c7d61 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:22:51 +0530
Subject: headers_check fix: linux/video_decoder.h

fix the following 'make headers_check' warning:

  usr/include/linux/video_decoder.h:7: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/video_decoder.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/video_decoder.h b/include/linux/video_decoder.h
index 121e26da2c18..e26c0c86a6ea 100644
--- a/include/linux/video_decoder.h
+++ b/include/linux/video_decoder.h
@@ -1,6 +1,8 @@
 #ifndef _LINUX_VIDEO_DECODER_H
 #define _LINUX_VIDEO_DECODER_H
 
+#include <linux/types.h>
+
 #define HAVE_VIDEO_DECODER 1
 
 struct video_decoder_capability { /* this name is too long */
-- 
cgit v1.2.3


From a4c1d7c8c61969667a853d08b039507669463807 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:24:09 +0530
Subject: headers_check fix: linux/video_encoder.h

fix the following 'make headers_check' warning:

  usr/include/linux/video_encoder.h:5: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/video_encoder.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/video_encoder.h b/include/linux/video_encoder.h
index 4b0e6907a7b4..b7b6423bbb8a 100644
--- a/include/linux/video_encoder.h
+++ b/include/linux/video_encoder.h
@@ -1,6 +1,8 @@
 #ifndef _LINUX_VIDEO_ENCODER_H
 #define _LINUX_VIDEO_ENCODER_H
 
+#include <linux/types.h>
+
 struct video_encoder_capability { /* this name is too long */
 	__u32	flags;
 #define	VIDEO_ENCODER_PAL	1	/* can encode PAL signal */
-- 
cgit v1.2.3


From 98be96b85398499212bc77ae3076a69e20368428 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:26:01 +0530
Subject: headers_check fix: linux/videodev.h

fix the following 'make headers_check' warning:

  usr/include/linux/videodev.h:53: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/videodev.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/videodev.h b/include/linux/videodev.h
index 15a653d41132..837f392fbe97 100644
--- a/include/linux/videodev.h
+++ b/include/linux/videodev.h
@@ -12,6 +12,7 @@
 #ifndef __LINUX_VIDEODEV_H
 #define __LINUX_VIDEODEV_H
 
+#include <linux/types.h>
 #include <linux/ioctl.h>
 #include <linux/videodev2.h>
 
-- 
cgit v1.2.3


From 982f8184f9a9251ba4e5c6d79ec32d25c0ad3cc8 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:27:58 +0530
Subject: headers_check fix: linux/virtio_blk.h

fix the following 'make headers_check' warning:

  usr/include/linux/virtio_blk.h:21: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/virtio_blk.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/virtio_blk.h b/include/linux/virtio_blk.h
index c1aef85243bf..94c56d29869d 100644
--- a/include/linux/virtio_blk.h
+++ b/include/linux/virtio_blk.h
@@ -2,6 +2,7 @@
 #define _LINUX_VIRTIO_BLK_H
 /* This header is BSD licensed so anyone can use the definitions to implement
  * compatible drivers/servers. */
+#include <linux/types.h>
 #include <linux/virtio_config.h>
 
 /* The ID for virtio_block */
-- 
cgit v1.2.3


From 8697325408d9be18fa24346c346b23fa56c3b190 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:29:33 +0530
Subject: headers_check fix: linux/virtio_console.h

fix the following 'make headers_check' warning:

  usr/include/linux/virtio_console.h:15: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/virtio_console.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/virtio_console.h b/include/linux/virtio_console.h
index 7615ffcdd555..dc161115ae35 100644
--- a/include/linux/virtio_console.h
+++ b/include/linux/virtio_console.h
@@ -1,5 +1,6 @@
 #ifndef _LINUX_VIRTIO_CONSOLE_H
 #define _LINUX_VIRTIO_CONSOLE_H
+#include <linux/types.h>
 #include <linux/virtio_config.h>
 /* This header, excluding the #ifdef __KERNEL__ part, is BSD licensed so
  * anyone can use the definitions to implement compatible drivers/servers. */
-- 
cgit v1.2.3


From 9a0e0ac21ca2af4715808b97bd600f0aecd87240 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:31:05 +0530
Subject: headers_check fix: linux/virtio_net.h

fix the following 'make headers_check' warning:

  usr/include/linux/virtio_net.h:28: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/virtio_net.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/virtio_net.h b/include/linux/virtio_net.h
index 5cdd0aa8bde9..3efa86c3ecb3 100644
--- a/include/linux/virtio_net.h
+++ b/include/linux/virtio_net.h
@@ -2,6 +2,7 @@
 #define _LINUX_VIRTIO_NET_H
 /* This header is BSD licensed so anyone can use the definitions to implement
  * compatible drivers/servers. */
+#include <linux/types.h>
 #include <linux/virtio_config.h>
 
 /* The ID for virtio_net */
-- 
cgit v1.2.3


From d5c72d7842c71403bc3d57ca05a8a1f96d81e262 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:39:17 +0530
Subject: headers_check fix: linux/nubus.h

fix the following 'make headers_check' warning:

  usr/include/linux/nubus.h:232: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/nubus.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/nubus.h b/include/linux/nubus.h
index c4355076d1a5..7382af374731 100644
--- a/include/linux/nubus.h
+++ b/include/linux/nubus.h
@@ -12,6 +12,7 @@
 #ifndef LINUX_NUBUS_H
 #define LINUX_NUBUS_H
 
+#include <linux/types.h>
 #ifdef __KERNEL__
 #include <asm/nubus.h>
 #endif
-- 
cgit v1.2.3


From 541c94f1d5ac2665fd15f1b827416f8c0b2f55cb Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Fri, 30 Jan 2009 22:40:32 +0530
Subject: headers_check fix: linux/rtnetlink.h

fix the following 'make headers_check' warning:

  usr/include/linux/rtnetlink.h:328: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 include/linux/rtnetlink.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index e88f7058b3a1..1e5f6730ff31 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -1,6 +1,7 @@
 #ifndef __LINUX_RTNETLINK_H
 #define __LINUX_RTNETLINK_H
 
+#include <linux/types.h>
 #include <linux/netlink.h>
 #include <linux/if_link.h>
 #include <linux/if_addr.h>
-- 
cgit v1.2.3


From 94df7de0289bc2df3d6e85cd2ece52bf42682f45 Mon Sep 17 00:00:00 2001
From: Sebastien Dugue <sebastien.dugue@bull.net>
Date: Mon, 1 Dec 2008 14:09:07 +0100
Subject: hrtimers: allow the hot-unplugging of all cpus

Impact: fix CPU hotplug hang on Power6 testbox

On architectures that support offlining all cpus (at least powerpc/pseries),
hot-unpluging the tick_do_timer_cpu can result in a system hang.

This comes from the fact that if the cpu going down happens to be the
cpu doing the tick, then as the tick_do_timer_cpu handover happens after the
cpu is dead (via the CPU_DEAD notification), we're left without ticks,
jiffies are frozen and any task relying on timers (msleep, ...) is stuck.
That's particularly the case for the cpu looping in __cpu_die() waiting
for the dying cpu to be dead.

This patch addresses this by having the tick_do_timer_cpu handover happen
earlier during the CPU_DYING notification. For this, a new clockevent
notification type is introduced (CLOCK_EVT_NOTIFY_CPU_DYING) which is triggered
in hrtimer_cpu_notify().

Signed-off-by: Sebastien Dugue <sebastien.dugue@bull.net>
Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/clockchips.h |  1 +
 kernel/hrtimer.c           |  4 ++++
 kernel/time/tick-common.c  | 26 +++++++++++++++++++-------
 3 files changed, 24 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h
index cea153697ec7..3a1dbba4d3ae 100644
--- a/include/linux/clockchips.h
+++ b/include/linux/clockchips.h
@@ -36,6 +36,7 @@ enum clock_event_nofitiers {
 	CLOCK_EVT_NOTIFY_BROADCAST_EXIT,
 	CLOCK_EVT_NOTIFY_SUSPEND,
 	CLOCK_EVT_NOTIFY_RESUME,
+	CLOCK_EVT_NOTIFY_CPU_DYING,
 	CLOCK_EVT_NOTIFY_CPU_DEAD,
 };
 
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 8fea312ca36c..647a40e2fea1 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -1608,6 +1608,10 @@ static int __cpuinit hrtimer_cpu_notify(struct notifier_block *self,
 		break;
 
 #ifdef CONFIG_HOTPLUG_CPU
+	case CPU_DYING:
+	case CPU_DYING_FROZEN:
+		clockevents_notify(CLOCK_EVT_NOTIFY_CPU_DYING, &scpu);
+		break;
 	case CPU_DEAD:
 	case CPU_DEAD_FROZEN:
 	{
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index 63e05d423a09..21a5ca849514 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -273,6 +273,21 @@ out_bc:
 	return ret;
 }
 
+/*
+ * Transfer the do_timer job away from a dying cpu.
+ *
+ * Called with interrupts disabled.
+ */
+static void tick_handover_do_timer(int *cpup)
+{
+	if (*cpup == tick_do_timer_cpu) {
+		int cpu = cpumask_first(cpu_online_mask);
+
+		tick_do_timer_cpu = (cpu < nr_cpu_ids) ? cpu :
+			TICK_DO_TIMER_NONE;
+	}
+}
+
 /*
  * Shutdown an event device on a given cpu:
  *
@@ -297,13 +312,6 @@ static void tick_shutdown(unsigned int *cpup)
 		clockevents_exchange_device(dev, NULL);
 		td->evtdev = NULL;
 	}
-	/* Transfer the do_timer job away from this cpu */
-	if (*cpup == tick_do_timer_cpu) {
-		int cpu = cpumask_first(cpu_online_mask);
-
-		tick_do_timer_cpu = (cpu < nr_cpu_ids) ? cpu :
-			TICK_DO_TIMER_NONE;
-	}
 	spin_unlock_irqrestore(&tick_device_lock, flags);
 }
 
@@ -357,6 +365,10 @@ static int tick_notify(struct notifier_block *nb, unsigned long reason,
 		tick_broadcast_oneshot_control(reason);
 		break;
 
+	case CLOCK_EVT_NOTIFY_CPU_DYING:
+		tick_handover_do_timer(dev);
+		break;
+
 	case CLOCK_EVT_NOTIFY_CPU_DEAD:
 		tick_shutdown_broadcast_oneshot(dev);
 		tick_shutdown_broadcast(dev);
-- 
cgit v1.2.3


From 0648e10d71c8e510d80772c4cb4220c97e9c34c7 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Mon, 2 Feb 2009 08:43:48 +0100
Subject: block: fix inconsistent parenthesisation of QUEUE_FLAG_DEFAULT

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/blkdev.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d08c4b8219a6..dcaa0fd84b02 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -455,7 +455,7 @@ struct request_queue
 
 #define QUEUE_FLAG_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
 				 (1 << QUEUE_FLAG_CLUSTER) |		\
-				  1 << QUEUE_FLAG_STACKABLE)
+				 (1 << QUEUE_FLAG_STACKABLE))
 
 static inline int queue_is_locked(struct request_queue *q)
 {
-- 
cgit v1.2.3


From 20b636bf7c946da260391cd4570b16506f140a2c Mon Sep 17 00:00:00 2001
From: Alberto Bertogli <albertito@blitiri.com.ar>
Date: Mon, 2 Feb 2009 12:41:07 +0100
Subject: Fix misleading comment in bio.h

The comment says "remember to add offset!", but the function already adds
it.

Signed-off-by: Alberto Bertogli <albertito@blitiri.com.ar>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/bio.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/bio.h b/include/linux/bio.h
index 0942765cf8c0..99728320cc05 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -451,8 +451,8 @@ extern struct biovec_slab bvec_slabs[BIOVEC_NR_POOLS] __read_mostly;
 
 #ifdef CONFIG_HIGHMEM
 /*
- * remember to add offset! and never ever reenable interrupts between a
- * bvec_kmap_irq and bvec_kunmap_irq!!
+ * remember never ever reenable interrupts between a bvec_kmap_irq and
+ * bvec_kunmap_irq!
  *
  * This function MUST be inlined - it plays with the CPU interrupt flags.
  */
-- 
cgit v1.2.3


From c52440a69df22dca69794936a91e2fb529a707fb Mon Sep 17 00:00:00 2001
From: Alberto Bertogli <albertito@blitiri.com.ar>
Date: Mon, 2 Feb 2009 12:41:07 +0100
Subject: bio.h: If they MUST be inlined, then use __always_inline

bvec_kmap_irq() and bvec_kunmap_irq() comments say they MUST be inlined,
so mark them as __always_inline.

Signed-off-by: Alberto Bertogli <albertito@blitiri.com.ar>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/bio.h | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/bio.h b/include/linux/bio.h
index 99728320cc05..2aa283ab062b 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -456,7 +456,8 @@ extern struct biovec_slab bvec_slabs[BIOVEC_NR_POOLS] __read_mostly;
  *
  * This function MUST be inlined - it plays with the CPU interrupt flags.
  */
-static inline char *bvec_kmap_irq(struct bio_vec *bvec, unsigned long *flags)
+static __always_inline char *bvec_kmap_irq(struct bio_vec *bvec,
+		unsigned long *flags)
 {
 	unsigned long addr;
 
@@ -472,7 +473,8 @@ static inline char *bvec_kmap_irq(struct bio_vec *bvec, unsigned long *flags)
 	return (char *) addr + bvec->bv_offset;
 }
 
-static inline void bvec_kunmap_irq(char *buffer, unsigned long *flags)
+static __always_inline void bvec_kunmap_irq(char *buffer,
+		unsigned long *flags)
 {
 	unsigned long ptr = (unsigned long) buffer & PAGE_MASK;
 
-- 
cgit v1.2.3


From 0afd4a21ba7d75e93fa79cf05d7a21774e149c0f Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Mon, 2 Feb 2009 13:27:44 -0800
Subject: net: Fix userland breakage wrt. linux/if_tunnel.h

Reported by Andrew Walrond <andrew@walrond.org>

Changeset c19e654ddbe3831252f61e76a74d661e1a755530
("gre: Add netlink interface") added an include
of linux/ip.h to linux/if_tunnel.h

We can't really let that get exposed to userspace
because this conflicts with types defined in netinet/ip.h
which userland is almost certainly going to have included
either explicitly or implicitly.

So guard this include with a __KERNEL__ ifdef.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/Kbuild      | 2 +-
 include/linux/if_tunnel.h | 3 +++
 2 files changed, 4 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 12e9a2957caf..6a9bb9753235 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -89,7 +89,6 @@ header-y += if_ppp.h
 header-y += if_slip.h
 header-y += if_strip.h
 header-y += if_tun.h
-header-y += if_tunnel.h
 header-y += in_route.h
 header-y += ioctl.h
 header-y += ip6_tunnel.h
@@ -235,6 +234,7 @@ unifdef-y += if_phonet.h
 unifdef-y += if_pppol2tp.h
 unifdef-y += if_pppox.h
 unifdef-y += if_tr.h
+unifdef-y += if_tunnel.h
 unifdef-y += if_vlan.h
 unifdef-y += igmp.h
 unifdef-y += inet_diag.h
diff --git a/include/linux/if_tunnel.h b/include/linux/if_tunnel.h
index aeab2cb32a9c..82c43624c067 100644
--- a/include/linux/if_tunnel.h
+++ b/include/linux/if_tunnel.h
@@ -2,7 +2,10 @@
 #define _IF_TUNNEL_H_
 
 #include <linux/types.h>
+
+#ifdef __KERNEL__
 #include <linux/ip.h>
+#endif
 
 #define SIOCGETTUNNEL   (SIOCDEVPRIVATE + 0)
 #define SIOCADDTUNNEL   (SIOCDEVPRIVATE + 1)
-- 
cgit v1.2.3


From 720eba31f47aeade8ec130ca7f4353223c49170f Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Tue, 3 Feb 2009 13:31:36 +1030
Subject: modules: Use a better scheme for refcounting

Current refcounting for modules (done if CONFIG_MODULE_UNLOAD=y) is
using a lot of memory.

Each 'struct module' contains an [NR_CPUS] array of full cache lines.

This patch uses existing infrastructure (percpu_modalloc() &
percpu_modfree()) to allocate percpu space for the refcount storage.

Instead of wasting NR_CPUS*128 bytes (on i386), we now use
nr_cpu_ids*sizeof(local_t) bytes.

On a typical distro, where NR_CPUS=8, shiping 2000 modules, we reduce
size of module files by about 2 Mbytes. (1Kb per module)

Instead of having all refcounters in the same memory node - with TLB misses
because of vmalloc() - this new implementation permits to have better
NUMA properties, since each  CPU will use storage on its preferred node,
thanks to percpu storage.

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/module.h | 25 ++++++++++++++++---------
 kernel/module.c        | 35 +++++++++++++++++++++++++----------
 2 files changed, 41 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/module.h b/include/linux/module.h
index 4f7ea12463d3..f3b8329eb5b8 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -219,11 +219,6 @@ void *__symbol_get_gpl(const char *symbol);
 
 #endif
 
-struct module_ref
-{
-	local_t count;
-} ____cacheline_aligned;
-
 enum module_state
 {
 	MODULE_STATE_LIVE,
@@ -344,8 +339,11 @@ struct module
 	/* Destruction function. */
 	void (*exit)(void);
 
-	/* Reference counts */
-	struct module_ref ref[NR_CPUS];
+#ifdef CONFIG_SMP
+	char *refptr;
+#else
+	local_t ref;
+#endif
 #endif
 };
 #ifndef MODULE_ARCH_INIT
@@ -395,13 +393,22 @@ void __symbol_put(const char *symbol);
 #define symbol_put(x) __symbol_put(MODULE_SYMBOL_PREFIX #x)
 void symbol_put_addr(void *addr);
 
+static inline local_t *__module_ref_addr(struct module *mod, int cpu)
+{
+#ifdef CONFIG_SMP
+	return (local_t *) (mod->refptr + per_cpu_offset(cpu));
+#else
+	return &mod->ref;
+#endif
+}
+
 /* Sometimes we know we already have a refcount, and it's easier not
    to handle the error case (which only happens with rmmod --wait). */
 static inline void __module_get(struct module *module)
 {
 	if (module) {
 		BUG_ON(module_refcount(module) == 0);
-		local_inc(&module->ref[get_cpu()].count);
+		local_inc(__module_ref_addr(module, get_cpu()));
 		put_cpu();
 	}
 }
@@ -413,7 +420,7 @@ static inline int try_module_get(struct module *module)
 	if (module) {
 		unsigned int cpu = get_cpu();
 		if (likely(module_is_live(module)))
-			local_inc(&module->ref[cpu].count);
+			local_inc(__module_ref_addr(module, cpu));
 		else
 			ret = 0;
 		put_cpu();
diff --git a/kernel/module.c b/kernel/module.c
index e8b51d41dd72..ba22484a987e 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -573,13 +573,13 @@ static char last_unloaded_module[MODULE_NAME_LEN+1];
 /* Init the unload section of the module. */
 static void module_unload_init(struct module *mod)
 {
-	unsigned int i;
+	int cpu;
 
 	INIT_LIST_HEAD(&mod->modules_which_use_me);
-	for (i = 0; i < NR_CPUS; i++)
-		local_set(&mod->ref[i].count, 0);
+	for_each_possible_cpu(cpu)
+		local_set(__module_ref_addr(mod, cpu), 0);
 	/* Hold reference count during initialization. */
-	local_set(&mod->ref[raw_smp_processor_id()].count, 1);
+	local_set(__module_ref_addr(mod, raw_smp_processor_id()), 1);
 	/* Backwards compatibility macros put refcount during init. */
 	mod->waiter = current;
 }
@@ -717,10 +717,11 @@ static int try_stop_module(struct module *mod, int flags, int *forced)
 
 unsigned int module_refcount(struct module *mod)
 {
-	unsigned int i, total = 0;
+	unsigned int total = 0;
+	int cpu;
 
-	for (i = 0; i < NR_CPUS; i++)
-		total += local_read(&mod->ref[i].count);
+	for_each_possible_cpu(cpu)
+		total += local_read(__module_ref_addr(mod, cpu));
 	return total;
 }
 EXPORT_SYMBOL(module_refcount);
@@ -894,7 +895,7 @@ void module_put(struct module *module)
 {
 	if (module) {
 		unsigned int cpu = get_cpu();
-		local_dec(&module->ref[cpu].count);
+		local_dec(__module_ref_addr(module, cpu));
 		/* Maybe they're waiting for us to drop reference? */
 		if (unlikely(!module_is_live(module)))
 			wake_up_process(module->waiter);
@@ -1464,7 +1465,10 @@ static void free_module(struct module *mod)
 	kfree(mod->args);
 	if (mod->percpu)
 		percpu_modfree(mod->percpu);
-
+#if defined(CONFIG_MODULE_UNLOAD) && defined(CONFIG_SMP)
+	if (mod->refptr)
+		percpu_modfree(mod->refptr);
+#endif
 	/* Free lock-classes: */
 	lockdep_free_key_range(mod->module_core, mod->core_size);
 
@@ -2011,6 +2015,14 @@ static noinline struct module *load_module(void __user *umod,
 	if (err < 0)
 		goto free_mod;
 
+#if defined(CONFIG_MODULE_UNLOAD) && defined(CONFIG_SMP)
+	mod->refptr = percpu_modalloc(sizeof(local_t), __alignof__(local_t),
+				      mod->name);
+	if (!mod->refptr) {
+		err = -ENOMEM;
+		goto free_mod;
+	}
+#endif
 	if (pcpuindex) {
 		/* We have a special allocation for this section. */
 		percpu = percpu_modalloc(sechdrs[pcpuindex].sh_size,
@@ -2018,7 +2030,7 @@ static noinline struct module *load_module(void __user *umod,
 					 mod->name);
 		if (!percpu) {
 			err = -ENOMEM;
-			goto free_mod;
+			goto free_percpu;
 		}
 		sechdrs[pcpuindex].sh_flags &= ~(unsigned long)SHF_ALLOC;
 		mod->percpu = percpu;
@@ -2282,6 +2294,9 @@ static noinline struct module *load_module(void __user *umod,
  free_percpu:
 	if (percpu)
 		percpu_modfree(percpu);
+#if defined(CONFIG_MODULE_UNLOAD) && defined(CONFIG_SMP)
+	percpu_modfree(mod->refptr);
+#endif
  free_mod:
 	kfree(args);
  free_hdr:
-- 
cgit v1.2.3


From 2999b58b795ad81f10e34bdbbfd2742172f247e4 Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Date: Sun, 1 Feb 2009 20:46:39 +0400
Subject: ide/libata: fix ata_id_is_cfa() (take 4)

When checking for the CFA feature set support, ata_id_is_cfa() tests bit 2 in
word 82 of the identify data instead the word 83;  it also checks the ATA/PI
version support in the word 80 (which the CompactFlash specifications have as
reserved), this having no slightest chance to work on the modern CF cards that
don't have 0x848A in the word 0...

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 include/linux/ata.h | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ata.h b/include/linux/ata.h
index a53318b8cbd0..08a86d5cdf1b 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -731,12 +731,17 @@ static inline int ata_id_current_chs_valid(const u16 *id)
 
 static inline int ata_id_is_cfa(const u16 *id)
 {
-	if (id[ATA_ID_CONFIG] == 0x848A)	/* Standard CF */
+	if (id[ATA_ID_CONFIG] == 0x848A)	/* Traditional CF */
 		return 1;
-	/* Could be CF hiding as standard ATA */
-	if (ata_id_major_version(id) >= 3 &&
-	    id[ATA_ID_COMMAND_SET_1] != 0xFFFF &&
-	   (id[ATA_ID_COMMAND_SET_1] & (1 << 2)))
+	/*
+	 * CF specs don't require specific value in the word 0 anymore and yet
+	 * they forbid to report the ATA version in the word 80 and require the
+	 * CFA feature set support to be indicated in the word 83 in this case.
+	 * Unfortunately, some cards only follow either of this requirements,
+	 * and while those that don't indicate CFA feature support need some
+	 * sort of quirk list, it seems impractical for the ones that do...
+	 */
+	if ((id[ATA_ID_COMMAND_SET_2] & 0xC004) == 0x4004)
 		return 1;
 	return 0;
 }
-- 
cgit v1.2.3


From 99cf610aa4840d822cdc67d194b23b55010ca9bd Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Thu, 29 Jan 2009 20:31:32 +0900
Subject: libata: clear dev->ering in smarter way

dev->ering used to be cleared together with the rest of ata_device in
ata_dev_init() which is called whenever a probing event occurs.
dev->ering is about to be used to track probing failures so it needs
to remain persistent over multiple porbing events.  This patch
achieves this by doing the following.

* Instead of CLEAR_OFFSET, define CLEAR_BEGIN and CLEAR_END and only
  clear between BEGIN and END.  ering is moved after END.  The split
  of persistent area is to allow hotter items remain at the head.

* ering is explicitly cleared on ata_dev_disable() and when device
  attach succeeds.  So, ering is persistent throug a device's life
  time (unless explicitly cleared of course) and also through periods
  inbetween disablement of an attached device and successful detection
  of the next one.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c |  4 ++--
 drivers/ata/libata-eh.c   |  7 +++++++
 include/linux/libata.h    | 18 ++++++++++--------
 3 files changed, 19 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index d006e5c4768c..564c03c4ebb3 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -5404,8 +5404,8 @@ void ata_dev_init(struct ata_device *dev)
 	dev->horkage = 0;
 	spin_unlock_irqrestore(ap->lock, flags);
 
-	memset((void *)dev + ATA_DEVICE_CLEAR_OFFSET, 0,
-	       sizeof(*dev) - ATA_DEVICE_CLEAR_OFFSET);
+	memset((void *)dev + ATA_DEVICE_CLEAR_BEGIN, 0,
+	       ATA_DEVICE_CLEAR_END - ATA_DEVICE_CLEAR_BEGIN);
 	dev->pio_mask = UINT_MAX;
 	dev->mwdma_mask = UINT_MAX;
 	dev->udma_mask = UINT_MAX;
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index aafe82bf5e2e..90092c1aae53 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -1194,6 +1194,11 @@ void ata_dev_disable(struct ata_device *dev)
 	ata_acpi_on_disable(dev);
 	ata_down_xfermask_limit(dev, ATA_DNXFER_FORCE_PIO0 | ATA_DNXFER_QUIET);
 	dev->class++;
+
+	/* From now till the next successful probe, ering is used to
+	 * track probe failures.  Clear accumulated device error info.
+	 */
+	ata_ering_clear(&dev->ering);
 }
 
 /**
@@ -2765,6 +2770,8 @@ static int ata_eh_revalidate_and_attach(struct ata_link *link,
 						     readid_flags, dev->id);
 			switch (rc) {
 			case 0:
+				/* clear error info accumulated during probe */
+				ata_ering_clear(&dev->ering);
 				new_mask |= 1 << dev->devno;
 				break;
 			case -ENOENT:
diff --git a/include/linux/libata.h b/include/linux/libata.h
index bca3ba25f52a..9dcefee5843c 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -580,7 +580,7 @@ struct ata_device {
 	acpi_handle		acpi_handle;
 	union acpi_object	*gtf_cache;
 #endif
-	/* n_sector is used as CLEAR_OFFSET, read comment above CLEAR_OFFSET */
+	/* n_sector is CLEAR_BEGIN, read comment above CLEAR_BEGIN */
 	u64			n_sectors;	/* size of device, if ATA */
 	unsigned int		class;		/* ATA_DEV_xxx */
 	unsigned long		unpark_deadline;
@@ -605,20 +605,22 @@ struct ata_device {
 	u16			heads;		/* Number of heads */
 	u16			sectors;	/* Number of sectors per track */
 
-	/* error history */
-	int			spdn_cnt;
-	struct ata_ering	ering;
-
 	union {
 		u16		id[ATA_ID_WORDS]; /* IDENTIFY xxx DEVICE data */
 		u32		gscr[SATA_PMP_GSCR_DWORDS]; /* PMP GSCR block */
 	};
+
+	/* error history */
+	int			spdn_cnt;
+	/* ering is CLEAR_END, read comment above CLEAR_END */
+	struct ata_ering	ering;
 };
 
-/* Offset into struct ata_device.  Fields above it are maintained
- * acress device init.  Fields below are zeroed.
+/* Fields between ATA_DEVICE_CLEAR_BEGIN and ATA_DEVICE_CLEAR_END are
+ * cleared to zero on ata_dev_init().
  */
-#define ATA_DEVICE_CLEAR_OFFSET		offsetof(struct ata_device, n_sectors)
+#define ATA_DEVICE_CLEAR_BEGIN		offsetof(struct ata_device, n_sectors)
+#define ATA_DEVICE_CLEAR_END		offsetof(struct ata_device, ering)
 
 struct ata_eh_info {
 	struct ata_device	*dev;		/* offending device */
-- 
cgit v1.2.3


From 9062712fa9ed13b531dfc2228086650b8bd6a255 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Thu, 29 Jan 2009 20:31:36 +0900
Subject: libata: implement HORKAGE_1_5_GBPS and apply it to WD My Book

3Gbps is often much more prone to transmission failures.  It's usually
okay to let EH handle speed down after transmission failures but some
WD My Book drives completely shutdown after certain transmission
failures and after it only power cycling can revive them.  Combined
with the fact that external drives often end up with cable assembly
which is longer than usual and more likely to have intervening gender,
this makes these drives very likely to shutdown under certain
configurations virtually rendering them unusable.

This patch implements HOARKGE_1_5_GBPS and applies it to WD My Book
such that 1.5Gbps is forced once the device is identified.

Please take a look at the following bz for related reports.

  http://bugzilla.kernel.org/show_bug.cgi?id=9913

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c | 41 +++++++++++++++++++++++++++++++++++++++++
 include/linux/libata.h    |  1 +
 2 files changed, 42 insertions(+)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 3c5965d56c47..9fbf0595f3d4 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -2232,6 +2232,40 @@ retry:
 	return rc;
 }
 
+static int ata_do_link_spd_horkage(struct ata_device *dev)
+{
+	struct ata_link *plink = ata_dev_phys_link(dev);
+	u32 target, target_limit;
+
+	if (!sata_scr_valid(plink))
+		return 0;
+
+	if (dev->horkage & ATA_HORKAGE_1_5_GBPS)
+		target = 1;
+	else
+		return 0;
+
+	target_limit = (1 << target) - 1;
+
+	/* if already on stricter limit, no need to push further */
+	if (plink->sata_spd_limit <= target_limit)
+		return 0;
+
+	plink->sata_spd_limit = target_limit;
+
+	/* Request another EH round by returning -EAGAIN if link is
+	 * going faster than the target speed.  Forward progress is
+	 * guaranteed by setting sata_spd_limit to target_limit above.
+	 */
+	if (plink->sata_spd > target) {
+		ata_dev_printk(dev, KERN_INFO,
+			       "applying link speed limit horkage to %s\n",
+			       sata_spd_string(target));
+		return -EAGAIN;
+	}
+	return 0;
+}
+
 static inline u8 ata_dev_knobble(struct ata_device *dev)
 {
 	struct ata_port *ap = dev->link->ap;
@@ -2322,6 +2356,10 @@ int ata_dev_configure(struct ata_device *dev)
 		return 0;
 	}
 
+	rc = ata_do_link_spd_horkage(dev);
+	if (rc)
+		return rc;
+
 	/* let ACPI work its magic */
 	rc = ata_acpi_on_devcfg(dev);
 	if (rc)
@@ -4223,6 +4261,9 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = {
 	/* Devices that do not need bridging limits applied */
 	{ "MTRON MSP-SATA*",		NULL,	ATA_HORKAGE_BRIDGE_OK, },
 
+	/* Devices which aren't very happy with higher link speeds */
+	{ "WD My Book",			NULL,	ATA_HORKAGE_1_5_GBPS, },
+
 	/* End Marker */
 	{ }
 };
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 9dcefee5843c..5d87bc09a1f5 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -380,6 +380,7 @@ enum {
 	ATA_HORKAGE_ATAPI_MOD16_DMA = (1 << 11), /* use ATAPI DMA for commands
 						    not multiple of 16 bytes */
 	ATA_HORKAGE_FIRMWARE_WARN = (1 << 12),	/* firwmare update warning */
+	ATA_HORKAGE_1_5_GBPS	= (1 << 13),	/* force 1.5 Gbps */
 
 	 /* DMA mask for user DMA control: User visible values; DO NOT
 	    renumber */
-- 
cgit v1.2.3


From 35626129abcd6a7547e84c817ef5b6eff7a8758b Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Mon, 2 Feb 2009 16:00:29 -0800
Subject: sched: add missing kernel-doc in sched.h

Add kernel-doc notation for @lock:

include/linux/sched.h:457: No description found for parameter 'lock'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h | 1 +
 1 file changed, 1 insertion(+)

(limited to 'include/linux')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 5a7c76388731..2127e959e0f4 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -443,6 +443,7 @@ struct pacct_struct {
  * @utime:		time spent in user mode, in &cputime_t units
  * @stime:		time spent in kernel mode, in &cputime_t units
  * @sum_exec_runtime:	total time spent on the CPU, in nanoseconds
+ * @lock:		lock for fields in this struct
  *
  * This structure groups together three kinds of CPU time that are
  * tracked for threads and thread groups.  Most things considering
-- 
cgit v1.2.3


From ac7b9004909d03d67016368093e81d37cae72895 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 4 Feb 2009 15:11:59 -0800
Subject: generic swap(): don't return a value from swap()

The swap() macro is accidentally retuning the value of its first argument.
Change it into a doesn't-return-anything macro before someone goes and
relies upon this behaviour.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wu Fengguang <wfg@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/kernel.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 343df9ef2412..7fa371898e3e 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -480,7 +480,8 @@ static inline char *pack_hex_byte(char *buf, u8 byte)
 /*
  * swap - swap value of @a and @b
  */
-#define swap(a, b) ({ typeof(a) __tmp = (a); (a) = (b); (b) = __tmp; })
+#define swap(a, b) \
+	do { typeof(a) __tmp = (a); (a) = (b); (b) = __tmp; } while (0)
 
 /**
  * container_of - cast a member of a structure out to the containing structure
-- 
cgit v1.2.3


From 1f5e31d7e55ac7fbd4ec5e5b20c8868b0e4564c9 Mon Sep 17 00:00:00 2001
From: Andrea Righi <righi.andrea@gmail.com>
Date: Wed, 4 Feb 2009 15:12:03 -0800
Subject: fbmem: don't call copy_from/to_user() with mutex held

Avoid calling copy_from/to_user() with fb_info->lock mutex held in fbmem
ioctl().

fb_mmap() is called under mm->mmap_sem (A) held, that also acquires
fb_info->lock (B); fb_ioctl() takes fb_info->lock (B) and does
copy_from/to_user() that might acquire mm->mmap_sem (A), causing a
deadlock.

NOTE: it doesn't push down the fb_info->lock in each own driver's
fb_ioctl(), so there are still potential deadlocks elsewhere.

Signed-off-by: Andrea Righi <righi.andrea@gmail.com>
Cc: Dave Jones <davej@redhat.com>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Johannes Weiner <hannes@saeurebad.de>
Cc: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/fbcmap.c |  20 +++++---
 drivers/video/fbmem.c  | 135 +++++++++++++++++++++++++------------------------
 include/linux/fb.h     |  15 ++++++
 3 files changed, 98 insertions(+), 72 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/video/fbcmap.c b/drivers/video/fbcmap.c
index 91b78e691505..f53b9f1d6aba 100644
--- a/drivers/video/fbcmap.c
+++ b/drivers/video/fbcmap.c
@@ -250,10 +250,6 @@ int fb_set_user_cmap(struct fb_cmap_user *cmap, struct fb_info *info)
 	int rc, size = cmap->len * sizeof(u16);
 	struct fb_cmap umap;
 
-	if (cmap->start < 0 || (!info->fbops->fb_setcolreg &&
-			        !info->fbops->fb_setcmap))
-		return -EINVAL;
-
 	memset(&umap, 0, sizeof(struct fb_cmap));
 	rc = fb_alloc_cmap(&umap, cmap->len, cmap->transp != NULL);
 	if (rc)
@@ -262,11 +258,23 @@ int fb_set_user_cmap(struct fb_cmap_user *cmap, struct fb_info *info)
 	    copy_from_user(umap.green, cmap->green, size) ||
 	    copy_from_user(umap.blue, cmap->blue, size) ||
 	    (cmap->transp && copy_from_user(umap.transp, cmap->transp, size))) {
-		fb_dealloc_cmap(&umap);
-		return -EFAULT;
+		rc = -EFAULT;
+		goto out;
 	}
 	umap.start = cmap->start;
+	if (!lock_fb_info(info)) {
+		rc = -ENODEV;
+		goto out;
+	}
+	if (cmap->start < 0 || (!info->fbops->fb_setcolreg &&
+				!info->fbops->fb_setcmap)) {
+		rc = -EINVAL;
+		goto out1;
+	}
 	rc = fb_set_cmap(&umap, info);
+out1:
+	unlock_fb_info(info);
+out:
 	fb_dealloc_cmap(&umap);
 	return rc;
 }
diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c
index 756efeb91abc..cfd9dce1ce0b 100644
--- a/drivers/video/fbmem.c
+++ b/drivers/video/fbmem.c
@@ -1013,132 +1013,139 @@ static long do_fb_ioctl(struct fb_info *info, unsigned int cmd,
 	struct fb_var_screeninfo var;
 	struct fb_fix_screeninfo fix;
 	struct fb_con2fbmap con2fb;
+	struct fb_cmap cmap_from;
 	struct fb_cmap_user cmap;
 	struct fb_event event;
 	void __user *argp = (void __user *)arg;
 	long ret = 0;
 
-	fb = info->fbops;
-	if (!fb)
-		return -ENODEV;
-
 	switch (cmd) {
 	case FBIOGET_VSCREENINFO:
-		ret = copy_to_user(argp, &info->var,
-				    sizeof(var)) ? -EFAULT : 0;
+		if (!lock_fb_info(info))
+			return -ENODEV;
+		var = info->var;
+		unlock_fb_info(info);
+
+		ret = copy_to_user(argp, &var, sizeof(var)) ? -EFAULT : 0;
 		break;
 	case FBIOPUT_VSCREENINFO:
-		if (copy_from_user(&var, argp, sizeof(var))) {
-			ret =  -EFAULT;
-			break;
-		}
+		if (copy_from_user(&var, argp, sizeof(var)))
+			return -EFAULT;
+		if (!lock_fb_info(info))
+			return -ENODEV;
 		acquire_console_sem();
 		info->flags |= FBINFO_MISC_USEREVENT;
 		ret = fb_set_var(info, &var);
 		info->flags &= ~FBINFO_MISC_USEREVENT;
 		release_console_sem();
-		if (ret == 0 && copy_to_user(argp, &var, sizeof(var)))
+		unlock_fb_info(info);
+		if (!ret && copy_to_user(argp, &var, sizeof(var)))
 			ret = -EFAULT;
 		break;
 	case FBIOGET_FSCREENINFO:
-		ret = copy_to_user(argp, &info->fix,
-				    sizeof(fix)) ? -EFAULT : 0;
+		if (!lock_fb_info(info))
+			return -ENODEV;
+		fix = info->fix;
+		unlock_fb_info(info);
+
+		ret = copy_to_user(argp, &fix, sizeof(fix)) ? -EFAULT : 0;
 		break;
 	case FBIOPUTCMAP:
 		if (copy_from_user(&cmap, argp, sizeof(cmap)))
-			ret = -EFAULT;
-		else
-			ret = fb_set_user_cmap(&cmap, info);
+			return -EFAULT;
+		ret = fb_set_user_cmap(&cmap, info);
 		break;
 	case FBIOGETCMAP:
 		if (copy_from_user(&cmap, argp, sizeof(cmap)))
-			ret = -EFAULT;
-		else
-			ret = fb_cmap_to_user(&info->cmap, &cmap);
+			return -EFAULT;
+		if (!lock_fb_info(info))
+			return -ENODEV;
+		cmap_from = info->cmap;
+		unlock_fb_info(info);
+		ret = fb_cmap_to_user(&cmap_from, &cmap);
 		break;
 	case FBIOPAN_DISPLAY:
-		if (copy_from_user(&var, argp, sizeof(var))) {
-			ret = -EFAULT;
-			break;
-		}
+		if (copy_from_user(&var, argp, sizeof(var)))
+			return -EFAULT;
+		if (!lock_fb_info(info))
+			return -ENODEV;
 		acquire_console_sem();
 		ret = fb_pan_display(info, &var);
 		release_console_sem();
+		unlock_fb_info(info);
 		if (ret == 0 && copy_to_user(argp, &var, sizeof(var)))
-			ret = -EFAULT;
+			return -EFAULT;
 		break;
 	case FBIO_CURSOR:
 		ret = -EINVAL;
 		break;
 	case FBIOGET_CON2FBMAP:
 		if (copy_from_user(&con2fb, argp, sizeof(con2fb)))
-			ret = -EFAULT;
-		else if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES)
-			ret = -EINVAL;
-		else {
-			con2fb.framebuffer = -1;
-			event.info = info;
-			event.data = &con2fb;
-			fb_notifier_call_chain(FB_EVENT_GET_CONSOLE_MAP,
-								&event);
-			ret = copy_to_user(argp, &con2fb,
-				    sizeof(con2fb)) ? -EFAULT : 0;
-		}
+			return -EFAULT;
+		if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES)
+			return -EINVAL;
+		con2fb.framebuffer = -1;
+		event.data = &con2fb;
+
+		if (!lock_fb_info(info))
+			return -ENODEV;
+		event.info = info;
+		fb_notifier_call_chain(FB_EVENT_GET_CONSOLE_MAP, &event);
+		unlock_fb_info(info);
+
+		ret = copy_to_user(argp, &con2fb, sizeof(con2fb)) ? -EFAULT : 0;
 		break;
 	case FBIOPUT_CON2FBMAP:
-		if (copy_from_user(&con2fb, argp, sizeof(con2fb))) {
-			ret = -EFAULT;
-			break;
-		}
-		if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES) {
-			ret = -EINVAL;
-			break;
-		}
-		if (con2fb.framebuffer < 0 || con2fb.framebuffer >= FB_MAX) {
-			ret = -EINVAL;
-			break;
-		}
+		if (copy_from_user(&con2fb, argp, sizeof(con2fb)))
+			return -EFAULT;
+		if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES)
+			return -EINVAL;
+		if (con2fb.framebuffer < 0 || con2fb.framebuffer >= FB_MAX)
+			return -EINVAL;
 		if (!registered_fb[con2fb.framebuffer])
 			request_module("fb%d", con2fb.framebuffer);
 		if (!registered_fb[con2fb.framebuffer]) {
 			ret = -EINVAL;
 			break;
 		}
-		event.info = info;
 		event.data = &con2fb;
+		if (!lock_fb_info(info))
+			return -ENODEV;
+		event.info = info;
 		ret = fb_notifier_call_chain(FB_EVENT_SET_CONSOLE_MAP,
 					      &event);
+		unlock_fb_info(info);
 		break;
 	case FBIOBLANK:
+		if (!lock_fb_info(info))
+			return -ENODEV;
 		acquire_console_sem();
 		info->flags |= FBINFO_MISC_USEREVENT;
 		ret = fb_blank(info, arg);
 		info->flags &= ~FBINFO_MISC_USEREVENT;
 		release_console_sem();
-		break;;
+		unlock_fb_info(info);
+		break;
 	default:
-		if (fb->fb_ioctl == NULL)
-			ret = -ENOTTY;
-		else
+		if (!lock_fb_info(info))
+			return -ENODEV;
+		fb = info->fbops;
+		if (fb->fb_ioctl)
 			ret = fb->fb_ioctl(info, cmd, arg);
+		else
+			ret = -ENOTTY;
+		unlock_fb_info(info);
 	}
 	return ret;
 }
 
 static long fb_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
-__acquires(&info->lock)
-__releases(&info->lock)
 {
 	struct inode *inode = file->f_path.dentry->d_inode;
 	int fbidx = iminor(inode);
-	struct fb_info *info;
-	long ret;
+	struct fb_info *info = registered_fb[fbidx];
 
-	info = registered_fb[fbidx];
-	mutex_lock(&info->lock);
-	ret = do_fb_ioctl(info, cmd, arg);
-	mutex_unlock(&info->lock);
-	return ret;
+	return do_fb_ioctl(info, cmd, arg);
 }
 
 #ifdef CONFIG_COMPAT
@@ -1257,8 +1264,6 @@ static int fb_get_fscreeninfo(struct fb_info *info, unsigned int cmd,
 
 static long fb_compat_ioctl(struct file *file, unsigned int cmd,
 			    unsigned long arg)
-__acquires(&info->lock)
-__releases(&info->lock)
 {
 	struct inode *inode = file->f_path.dentry->d_inode;
 	int fbidx = iminor(inode);
@@ -1266,7 +1271,6 @@ __releases(&info->lock)
 	struct fb_ops *fb = info->fbops;
 	long ret = -ENOIOCTLCMD;
 
-	mutex_lock(&info->lock);
 	switch(cmd) {
 	case FBIOGET_VSCREENINFO:
 	case FBIOPUT_VSCREENINFO:
@@ -1292,7 +1296,6 @@ __releases(&info->lock)
 			ret = fb->fb_compat_ioctl(info, cmd, arg);
 		break;
 	}
-	mutex_unlock(&info->lock);
 	return ret;
 }
 #endif
diff --git a/include/linux/fb.h b/include/linux/fb.h
index 818fe21257e8..31527e17076b 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -960,6 +960,21 @@ extern struct fb_info *registered_fb[FB_MAX];
 extern int num_registered_fb;
 extern struct class *fb_class;
 
+static inline int lock_fb_info(struct fb_info *info)
+{
+	mutex_lock(&info->lock);
+	if (!info->fbops) {
+		mutex_unlock(&info->lock);
+		return 0;
+	}
+	return 1;
+}
+
+static inline void unlock_fb_info(struct fb_info *info)
+{
+	mutex_unlock(&info->lock);
+}
+
 static inline void __fb_pad_aligned_buffer(u8 *dst, u32 d_pitch,
 					   u8 *src, u32 s_pitch, u32 height)
 {
-- 
cgit v1.2.3


From 777c6c5f1f6e757ae49ecca2ed72d6b1f523c007 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@cmpxchg.org>
Date: Wed, 4 Feb 2009 15:12:14 -0800
Subject: wait: prevent exclusive waiter starvation

With exclusive waiters, every process woken up through the wait queue must
ensure that the next waiter down the line is woken when it has finished.

Interruptible waiters don't do that when aborting due to a signal.  And if
an aborting waiter is concurrently woken up through the waitqueue, noone
will ever wake up the next waiter.

This has been observed with __wait_on_bit_lock() used by
lock_page_killable(): the first contender on the queue was aborting when
the actual lock holder woke it up concurrently.  The aborted contender
didn't acquire the lock and therefor never did an unlock followed by
waking up the next waiter.

Add abort_exclusive_wait() which removes the process' wait descriptor from
the waitqueue, iff still queued, or wakes up the next waiter otherwise.
It does so under the waitqueue lock.  Racing with a wake up means the
aborting process is either already woken (removed from the queue) and will
wake up the next waiter, or it will remove itself from the queue and the
concurrent wake up will apply to the next waiter after it.

Use abort_exclusive_wait() in __wait_event_interruptible_exclusive() and
__wait_on_bit_lock() when they were interrupted by other means than a wake
up through the queue.

[akpm@linux-foundation.org: coding-style fixes]
Reported-by: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Mentored-by: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Chuck Lever <cel@citi.umich.edu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: <stable@kernel.org>		["after some testing"]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/wait.h | 11 ++++++++--
 kernel/sched.c       |  4 ++--
 kernel/wait.c        | 59 +++++++++++++++++++++++++++++++++++++++++++++-------
 3 files changed, 63 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/wait.h b/include/linux/wait.h
index ef609f842fac..a210ede73b56 100644
--- a/include/linux/wait.h
+++ b/include/linux/wait.h
@@ -132,6 +132,8 @@ static inline void __remove_wait_queue(wait_queue_head_t *head,
 	list_del(&old->task_list);
 }
 
+void __wake_up_common(wait_queue_head_t *q, unsigned int mode,
+			int nr_exclusive, int sync, void *key);
 void __wake_up(wait_queue_head_t *q, unsigned int mode, int nr, void *key);
 extern void __wake_up_locked(wait_queue_head_t *q, unsigned int mode);
 extern void __wake_up_sync(wait_queue_head_t *q, unsigned int mode, int nr);
@@ -333,16 +335,19 @@ do {									\
 	for (;;) {							\
 		prepare_to_wait_exclusive(&wq, &__wait,			\
 					TASK_INTERRUPTIBLE);		\
-		if (condition)						\
+		if (condition) {					\
+			finish_wait(&wq, &__wait);			\
 			break;						\
+		}							\
 		if (!signal_pending(current)) {				\
 			schedule();					\
 			continue;					\
 		}							\
 		ret = -ERESTARTSYS;					\
+		abort_exclusive_wait(&wq, &__wait, 			\
+				TASK_INTERRUPTIBLE, NULL);		\
 		break;							\
 	}								\
-	finish_wait(&wq, &__wait);					\
 } while (0)
 
 #define wait_event_interruptible_exclusive(wq, condition)		\
@@ -431,6 +436,8 @@ extern long interruptible_sleep_on_timeout(wait_queue_head_t *q,
 void prepare_to_wait(wait_queue_head_t *q, wait_queue_t *wait, int state);
 void prepare_to_wait_exclusive(wait_queue_head_t *q, wait_queue_t *wait, int state);
 void finish_wait(wait_queue_head_t *q, wait_queue_t *wait);
+void abort_exclusive_wait(wait_queue_head_t *q, wait_queue_t *wait,
+			unsigned int mode, void *key);
 int autoremove_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key);
 int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key);
 
diff --git a/kernel/sched.c b/kernel/sched.c
index 242d0d47a70d..8ee437a5ec1d 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4697,8 +4697,8 @@ EXPORT_SYMBOL(default_wake_function);
  * started to run but is not in state TASK_RUNNING. try_to_wake_up() returns
  * zero in this (rare) case, and we handle it by continuing to scan the queue.
  */
-static void __wake_up_common(wait_queue_head_t *q, unsigned int mode,
-			     int nr_exclusive, int sync, void *key)
+void __wake_up_common(wait_queue_head_t *q, unsigned int mode,
+			int nr_exclusive, int sync, void *key)
 {
 	wait_queue_t *curr, *next;
 
diff --git a/kernel/wait.c b/kernel/wait.c
index cd87131f2fc2..42a2dbc181c8 100644
--- a/kernel/wait.c
+++ b/kernel/wait.c
@@ -91,6 +91,15 @@ prepare_to_wait_exclusive(wait_queue_head_t *q, wait_queue_t *wait, int state)
 }
 EXPORT_SYMBOL(prepare_to_wait_exclusive);
 
+/*
+ * finish_wait - clean up after waiting in a queue
+ * @q: waitqueue waited on
+ * @wait: wait descriptor
+ *
+ * Sets current thread back to running state and removes
+ * the wait descriptor from the given waitqueue if still
+ * queued.
+ */
 void finish_wait(wait_queue_head_t *q, wait_queue_t *wait)
 {
 	unsigned long flags;
@@ -117,6 +126,39 @@ void finish_wait(wait_queue_head_t *q, wait_queue_t *wait)
 }
 EXPORT_SYMBOL(finish_wait);
 
+/*
+ * abort_exclusive_wait - abort exclusive waiting in a queue
+ * @q: waitqueue waited on
+ * @wait: wait descriptor
+ * @state: runstate of the waiter to be woken
+ * @key: key to identify a wait bit queue or %NULL
+ *
+ * Sets current thread back to running state and removes
+ * the wait descriptor from the given waitqueue if still
+ * queued.
+ *
+ * Wakes up the next waiter if the caller is concurrently
+ * woken up through the queue.
+ *
+ * This prevents waiter starvation where an exclusive waiter
+ * aborts and is woken up concurrently and noone wakes up
+ * the next waiter.
+ */
+void abort_exclusive_wait(wait_queue_head_t *q, wait_queue_t *wait,
+			unsigned int mode, void *key)
+{
+	unsigned long flags;
+
+	__set_current_state(TASK_RUNNING);
+	spin_lock_irqsave(&q->lock, flags);
+	if (!list_empty(&wait->task_list))
+		list_del_init(&wait->task_list);
+	else if (waitqueue_active(q))
+		__wake_up_common(q, mode, 1, 0, key);
+	spin_unlock_irqrestore(&q->lock, flags);
+}
+EXPORT_SYMBOL(abort_exclusive_wait);
+
 int autoremove_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key)
 {
 	int ret = default_wake_function(wait, mode, sync, key);
@@ -177,17 +219,20 @@ int __sched
 __wait_on_bit_lock(wait_queue_head_t *wq, struct wait_bit_queue *q,
 			int (*action)(void *), unsigned mode)
 {
-	int ret = 0;
-
 	do {
+		int ret;
+
 		prepare_to_wait_exclusive(wq, &q->wait, mode);
-		if (test_bit(q->key.bit_nr, q->key.flags)) {
-			if ((ret = (*action)(q->key.flags)))
-				break;
-		}
+		if (!test_bit(q->key.bit_nr, q->key.flags))
+			continue;
+		ret = action(q->key.flags);
+		if (!ret)
+			continue;
+		abort_exclusive_wait(wq, &q->wait, mode, &q->key);
+		return ret;
 	} while (test_and_set_bit(q->key.bit_nr, q->key.flags));
 	finish_wait(wq, &q->wait);
-	return ret;
+	return 0;
 }
 EXPORT_SYMBOL(__wait_on_bit_lock);
 
-- 
cgit v1.2.3